Science.gov

Sample records for artificial neural nets

  1. Squeezing the turnip with artificial neural nets.

    PubMed

    Francl, Leonard J

    2004-09-01

    ABSTRACT Modeling in epidemiology has followed many different strategies and philosophies. Artificial neural networks (ANNs) comprise a family of highly flexible and adaptive models that have shown promise for application to modeling disease phenomena in general and plant disease forecasting in particular. ANN modeling requires the availability of representative, robust input data and exhaustive testing of model aptness and optimization; meanwhile, ANNs sacrifice much of the biological insight often derived through other model forms. On the other hand, ANNs may extract previously undetected and possibly complex relationships, which can increase prediction accuracy over mainstream statistical methods, usually in an incremental manner.

  2. Hadamard design and artificial neural nets

    SciTech Connect

    Kuerten, K.E. Universitaet Wien ); Klingen, N. )

    1993-04-01

    Hadamard theory is shown to play an important role in the generation of Boolean decision functions, a fundamental tool in the field of artificial neural network design. Based on a group-theoretic introduction of a complete set of Hadamard vectors, whose matrices are of the order of a power of two, the authors classify subsets according to the degree of their linear dependence. They show in the thermodynamic limit that essentially the whole Hadamard space is occupied by representatives with defect not exceeding two or three. 15 refs., 1 fig.

  3. Using artificial neural nets to predict building energy parameters

    SciTech Connect

    Stevenson, W.J.

    1994-12-31

    Artificial neural nets were used as nonlinear function approximators on two data sets of building energy parameters and solar radiation data. During the modeling (training) phase, the data to be predicted were unavailable, providing a ``blind`` test of the technique. The first time series consisted of building energy ``inputs`` (such as solar radiation and temperature) for September--December 1989 and required the prediction of energy use for January--February 1990. The extrapolation was performed with only the data immediately at hand. Although results for chilled and hot-water use were acceptable, the prediction of electricity use would have benefited markedly from easily available additional information, such as working and nonworking days. The second time series required the prediction of beam solar insolation from four global directional measurements. This was an interpolation problem, and good predictions were achieved for this data set. Conjugate gradient and cascade correlation neural net programs were used.

  4. Artificial neural nets for K-complex detection.

    PubMed

    Jansen, B H

    1990-01-01

    An explorative study was initiated to determine whether artificial neural nets (ANNs) can be used to detect K-complexes in EEGs (electroencephalograms). K-complexes are relatively large waves with a duration of between 500 and 1500 ms often seen during sleep stage 2. Sleep spindles (bursts of rhythmic activity with a frequency of 12 to 16 Hz) are almost always observed in the neighborhood of K-complexes. The data and methods used to analyze K-complex are described. In all cases, a multilayer backpropagation ANN was used. The number of input nodes and hidden layers varied. Two different strategies were used to prepare the input to the ANN, and results for both are presented. The results indicate that the neural net approaches used are not adequate for the detection of K-complexes.

  5. Applications of artificial neural nets in structural mechanics

    NASA Technical Reports Server (NTRS)

    Berke, Laszlo; Hajela, Prabhat

    1990-01-01

    A brief introduction to the fundamental of Neural Nets is given, followed by two applications in structural optimization. In the first case, the feasibility of simulating with neural nets the many structural analyses performed during optimization iterations was studied. In the second case, the concept of using neural nets to capture design expertise was studied.

  6. Applications of artificial neural nets in structural mechanics

    NASA Technical Reports Server (NTRS)

    Berke, L.; Hajela, P.

    1992-01-01

    A brief introduction to the fundamental of Neural Nets is given, followed by two applications in structural optimization. In the first case, the feasibility of simulating with neural nets the many structural analyses performed during optimization iterations was studied. In the second case, the concept of using neural nets to capture design expertise was studied.

  7. Applications of artificial neural nets in clinical biomechanics.

    PubMed

    Schöllhorn, W I

    2004-11-01

    The purpose of this article is to provide an overview of current applications of artificial neural networks in the area of clinical biomechanics. The body of literature on artificial neural networks grew intractably vast during the last 15 years. Conventional statistical models may present certain limitations that can be overcome by neural networks. Artificial neural networks in general are introduced, some limitations, and some proven benefits are discussed.

  8. Surface daytime net radiation estimation using artificial neural networks

    DOE PAGESBeta

    Jiang, Bo; Zhang, Yi; Liang, Shunlin; Zhang, Xiaotong; Xiao, Zhiqiang

    2014-11-11

    Net all-wave surface radiation (Rn) is one of the most important fundamental parameters in various applications. However, conventional Rn measurements are difficult to collect because of the high cost and ongoing maintenance of recording instruments. Therefore, various empirical Rn estimation models have been developed. This study presents the results of two artificial neural network (ANN) models (general regression neural networks (GRNN) and Neuroet) to estimate Rn globally from multi-source data, including remotely sensed products, surface measurements, and meteorological reanalysis products. Rn estimates provided by the two ANNs were tested against in-situ radiation measurements obtained from 251 global sites between 1991–2010more » both in global mode (all data were used to fit the models) and in conditional mode (the data were divided into four subsets and the models were fitted separately). Based on the results obtained from extensive experiments, it has been proved that the two ANNs were superior to linear-based empirical models in both global and conditional modes and that the GRNN performed better and was more stable than Neuroet. The GRNN estimates had a determination coefficient (R2) of 0.92, a root mean square error (RMSE) of 34.27 W·m–2 , and a bias of –0.61 W·m–2 in global mode based on the validation dataset. In conclusion, ANN methods are a potentially powerful tool for global Rn estimation.« less

  9. Surface daytime net radiation estimation using artificial neural networks

    SciTech Connect

    Jiang, Bo; Zhang, Yi; Liang, Shunlin; Zhang, Xiaotong; Xiao, Zhiqiang

    2014-11-11

    Net all-wave surface radiation (Rn) is one of the most important fundamental parameters in various applications. However, conventional Rn measurements are difficult to collect because of the high cost and ongoing maintenance of recording instruments. Therefore, various empirical Rn estimation models have been developed. This study presents the results of two artificial neural network (ANN) models (general regression neural networks (GRNN) and Neuroet) to estimate Rn globally from multi-source data, including remotely sensed products, surface measurements, and meteorological reanalysis products. Rn estimates provided by the two ANNs were tested against in-situ radiation measurements obtained from 251 global sites between 1991–2010 both in global mode (all data were used to fit the models) and in conditional mode (the data were divided into four subsets and the models were fitted separately). Based on the results obtained from extensive experiments, it has been proved that the two ANNs were superior to linear-based empirical models in both global and conditional modes and that the GRNN performed better and was more stable than Neuroet. The GRNN estimates had a determination coefficient (R2) of 0.92, a root mean square error (RMSE) of 34.27 W·m–2 , and a bias of –0.61 W·m–2 in global mode based on the validation dataset. In conclusion, ANN methods are a potentially powerful tool for global Rn estimation.

  10. A comparison of polynomial approximations and artificial neural nets as response surfaces

    NASA Technical Reports Server (NTRS)

    Carpenter, William C.; Barthelemy, Jean-Francois M.

    1992-01-01

    Artificial neural nets and polynomial approximations were used to develop response surfaces for several test problems. Based on the number of functional evaluations required to build the approximations and the number of undetermined parameters associated with the approximations, the performance of the two types of approximations was found to be comparable. A rule of thumb is developed for determining the number of nodes to be used on a hidden layer of an artificial neural net, and the number of designs needed to train an approximation is discussed.

  11. [A method of recognizing biology surface spectrum using cascade-connection artificial neural nets].

    PubMed

    Shi, Wei-Jie; Yao, Yong; Zhang, Tie-Qiang; Meng, Xian-Jiang

    2008-05-01

    A method of recognizing the visible spectrum of micro-areas on the biological surface with cascade-connection artificial neural nets is presented in the present paper. The visible spectra of spots on apples' pericarp, ranging from 500 to 730 nm, were obtained with a fiber-probe spectrometer, and a new spectrum recognition system consisting of three-level cascade-connection neural nets was set up. The experiments show that the spectra of rotten, scar and bumped spot on an apple's pericarp can be recognized by the spectrum recognition system, and the recognition accuracy is higher than 85% even when noise level is 15%. The new recognition system overcomes the disadvantages of poor accuracy and poor anti-noise with the traditional system based on single cascade neural nets. Finally, a new method of expression of recognition results was proved. The method is based on the conception of degree of membership in fuzzing mathematics, and through it the recognition results can be expressed exactly and objectively.

  12. Estimation of Surface Net Radiation from Operational Meteorological Measurements Using Artificial Neural Networks

    NASA Astrophysics Data System (ADS)

    Geraldo Ferreira, A.; Lopez-Baeza, Ernesto; Soria-Olivas, Emilio; Serrano Lopez, Antonio J.

    2012-07-01

    The study of net radiation at the surface (Rn) is of fundamental importance because this parameter defines the total amount of energy that is available for the physical and biological processes such as air and soil warming and evapotranspiration, the latter being used to optimize the quality and yield of crops, water resources planning, weather forecasting, etc. Despite its importance, the net radiation is measured only in a very few number of standard weather stations. This work presents a methodology based on artificial neural networks (ANN), by modeling the relationships between net radiation and common meteorological variables measured at meteorological stations. The meteorological parameters used as input to the ANN models were: wind velocity and direction, surface and air temperature, relative humidity, and soil moisture and temperature. The output parameter was Rn. A comparison has been made between Rn estimates provided by the neural models and in situ measured Rn values. The statistical results given in terms of low root mean square error and mean absolute error show that the proposed methodology is suitable to estimate net radiation at surface from common meteorological variables.

  13. Artificial awareness for robots using artificial neural nets to monitor robotic workcells

    SciTech Connect

    Tucker, S.D.; Ray, L.P.

    1997-04-01

    Current robotic systems are unable to recognize most unexpected changes in the work environment, such as tool breakage, workpiece motion, or sensor failure. Unless halted by a human operator, they are likely to continue actions that are at best inappropriate, and at worst may cause damage to the workpiece or robot. This project investigated use of Artificial Neural Networks (ANNs) to learn the expected characteristics of sensor data during normal operations, recognize when data no longer is consistent with normal operation, suspend operations and alert a human operator. Data on force and torque applied at the robot tool tip were collected from two workcells: a robotic deburring system and a robot material-handling system. Data were collected for normal operations and for operations in which a fault condition was introduced. Data simulating sensor failure and excessive sensor noise were generated. Artificial Neural Networks (ANN) were trained to classify operating conditions; several ANN architectures were tested. The selected ANNs were able to correctly classify all valid operating conditions and the majority of fault conditions over the entire range of operating conditions, having {open_quotes}learned{close_quotes} the expected force/torque data. Most faults introduced appreciable error in the data; these were correctly classified. However, a small minority of faults did not give rise to a detectable difference in force and torque data. It is believed that these faults could be detected using other sensors. The computational workload varies with the implementation, but is moderate: up to 2.3 megaflops. This makes implementation of a real-time workcell monitor feasible.

  14. Accelerometer signal-based human activity recognition using augmented autoregressive model coefficients and artificial neural nets.

    PubMed

    Khan, A M; Lee, Y K; Kim, T S

    2008-01-01

    Automatic recognition of human activities is one of the important and challenging research areas in proactive and ubiquitous computing. In this work, we present some preliminary results of recognizing human activities using augmented features extracted from the activity signals measured using a single triaxial accelerometer sensor and artificial neural nets. The features include autoregressive (AR) modeling coefficients of activity signals, signal magnitude areas (SMA), and title angles (TA). We have recognized four human activities using AR coefficients (ARC) only, ARC with SMA, and ARC with SMA and TA. With the last augmented features, we have achieved the recognition rate above 99% for all four activities including lying, standing, walking, and running. With our proposed technique, real time recognition of some human activities is possible.

  15. Discussion of using artificial neural nets to identify the well-test interpretation model

    SciTech Connect

    Yeung, K. ); Chakrabarty, C. ); Wu, S. )

    1994-09-01

    Use of artificial neural nets (ANN's) to identify noisy and apparently unrecognizable patterns is common for many real-world problems, ranging from applications such as speech recognition to stock market prediction. ANN approaches are often good candidates for recognizing patterns when rigid mathematical models do not exist or are insufficient to meet a full-scale identification requirement. Al-Kaabi and Lee's proposal of using ANN's to identify the well-test interpretation model is appropriate because well-test data is often highly nonlinear and noisy. The purpose of this discussion is to present some of the authors results in a similar study and to suggest a simple technique that would enhance the use of ANN's in Al-Kaabi and Lee's approach.

  16. Neural timing nets.

    PubMed

    Cariani, P A

    2001-01-01

    Formulations of artificial neural networks are directly related to assumptions about neural coding in the brain. Traditional connectionist networks assume channel-based rate coding, while time-delay networks convert temporally-coded inputs into rate-coded outputs. Neural timing nets that operate on time structured input spike trains to produce meaningful time-structured outputs are proposed. Basic computational properties of simple feedforward and recurrent timing nets are outlined and applied to auditory computations. Feed-forward timing nets consist of arrays of coincidence detectors connected via tapped delay lines. These temporal sieves extract common spike patterns in their inputs that can subserve extraction of common fundamental frequencies (periodicity pitch) and common spectrum (timbre). Feedforward timing nets can also be used to separate time-shifted patterns, fusing patterns with similar internal temporal structure and spatially segregating different ones. Simple recurrent timing nets consisting of arrays of delay loops amplify and separate recurring time patterns. Single- and multichannel recurrent timing nets are presented that demonstrate the separation of concurrent, double vowels. Timing nets constitute a new and general neural network strategy for performing temporal computations on neural spike trains: extraction of common periodicities, detection of recurring temporal patterns, and formation and separation of invariant spike patterns that subserve auditory objects.

  17. Application of artificial aging techniques to samples of rum and comparison with traditionally aged rums by analysis with artificial neural nets.

    PubMed

    Quesada Granados, J; Merelo Guervós, J J; Oliveras López, M J; González Peñalver, J; Olalla Herrera, M; Blanca Herrera, R; López Martinez, M C

    2002-03-13

    Artificial aging techniques were applied to samples of rum. These samples were then compared, by artificial neural nets, with traditionally aged rums. Analysis was based on the phenolic and furanic composition of each sample. There were found to be few statistical differences between samples, thus confirming the possibility of applying artificial aging techniques to obtain rum with phenolic and furanic characteristics that are similar to those of rum obtained by traditional methods.

  18. Comparison of polynomial approximations and artificial neural nets for response surfaces in engineering optimization

    NASA Technical Reports Server (NTRS)

    Carpenter, William C.

    1991-01-01

    Engineering optimization problems involve minimizing some function subject to constraints. In areas such as aircraft optimization, the constraint equations may be from numerous disciplines such as transfer of information between these disciplines and the optimization algorithm. They are also suited to problems which may require numerous re-optimizations such as in multi-objective function optimization or to problems where the design space contains numerous local minima, thus requiring repeated optimizations from different initial designs. Their use has been limited, however, by the fact that development of response surfaces randomly selected or preselected points in the design space. Thus, they have been thought to be inefficient compared to algorithms to the optimum solution. A development has taken place in the last several years which may effect the desirability of using response surfaces. It may be possible that artificial neural nets are more efficient in developing response surfaces than polynomial approximations which have been used in the past. This development is the concern of the work.

  19. Smallest artificial molecular neural-net for collective and emergent information processing

    NASA Astrophysics Data System (ADS)

    Bandyopadhyay, Anirban; Sahu, Satyajit; Fujita, Daisuke

    2009-09-01

    While exploring the random diffusion of 2 bit molecular switches (we define as molecular neuron) on an atomic flat Au (111) substrate, we have found that at least four molecules are required to construct a functional neural net. Surface electron density wave enables communication of one to many molecules at a time—a prerequisite for the parallel processing. Here we have shown that in a neural net of several molecules, some of them could dynamically store information as memory and consistently replicate the fundamental relationship that is found only in a collective and emergent computing system like our brain.

  20. Neural nets.

    PubMed

    Hejnol, Andreas; Rentzsch, Fabian

    2015-09-21

    Although modern evolutionary biology has abandoned the use of 'lower' or 'higher' for animals, the quote of G.H. Parker captures quite well the current understanding of the nerve net as the evolutionarily oldest organization of the nervous system, the major organ system responsible for processing information and coordinating animal behaviour. The degree of complexity of a nervous system - in particular its organization into substructures such as brains and nerve cords - shows fascinating variations between animals. Even within an individual, the nervous system can show parallel existing types of organizations that are only partially connected, illustrated by the well-known central and peripheral nervous system. In general, the architecture of the nervous system is adapted to the specific needs and lifestyle of the individual species. How these diverse and complex nervous systems evolved is an ongoing debate among zoologists and evolutionary biologists.

  1. Artificial neural nets: dependence of the EEG amplitude's probability distribution on statistical parameters.

    PubMed

    Anninos, P; Zenone, S; Elul, R

    1983-08-01

    The statistical laws governing the output of a population of unitary generators are not explicit with regard to the effect of population size and properties of the individual generators on the summed activity. Experimental work was therefore undertaken with artificial nerve nets, the activity of which simulates with a high degree of realism individual nerve cells and the electroencephalogram. It was found that the summed activity is not affected by the statistical properties of single generators even in nets of only 200-1000 elements. On the other hand, the output of the net is highly sensitive to the level of connectivity between individual generators. When connectivity is low, the summed output is distributed in normal (Gaussian) fashion. The output of the net becomes less and less Gaussian with increase in coupling between the generators.

  2. Using artificial neural nets to identify the well-test interpretation model

    SciTech Connect

    Alkaabi, A.U.; Lee, W.J. )

    1993-09-01

    In a pressure-transient test, a signal of pressure vs. time is recorded. When this signal is plotted with specialized plotting functions, diagnostic plots, such as derivative or Horner plots, are produced that often are used in the interpretation process. The signal on these plots is deformed and shaped by underlying mechanisms in the formation and wellbore. These mechanisms are the well-test interpretation model. The objective of this work is to identify these mechanisms from the signatures on the derivative plot. Identifying the well-test interpretation model is described in the literature as an inverse problem. The traditional way of solving an inverse problem is with inverse theory techniques (e.g., regression analysis). A serious disadvantage of such techniques is that one has to assume an interpretation model. The inverse theory provides estimates of the model parameters but not of the model itself. Because more than one interpretation model can produce the same signal, this approach can lead to misleading results. The authors seek the model itself instead of its parameters in this study. In this study, they trained a neural net simulator to identify the well-test interpretation model from the derivative plot. The neural net simulator can be part of a well-test expert system or a computer-enhanced well-test interpretation.

  3. Neural Network Development Tool (NETS)

    NASA Technical Reports Server (NTRS)

    Baffes, Paul T.

    1990-01-01

    Artificial neural networks formed from hundreds or thousands of simulated neurons, connected in manner similar to that in human brain. Such network models learning behavior. Using NETS involves translating problem to be solved into input/output pairs, designing network configuration, and training network. Written in C.

  4. Artificial neural nets in computer-aided macro motor unit potential classification.

    PubMed

    Schizas, C N; Pattichis, C S; Schofield, I S; Fawcett, P R; Middleton, L T

    1990-01-01

    The use of macro electromyography to obtain a macro motor unit potential (MMUP) is described. At least 20 potentials are measured from a single muscle to obtain a reasonable estimate of the parameters of an average motor unit potential. The MMUP data are analyzed by means of the peak-to-peak amplitude and the integral of the central 50 ms of the signal. The possibility of using artificial neural networks (ANNs) to analyze the macro data in a way that makes no assumptions about the relationships between the parameters and without recourse to conventional modeling methods is discussed. The results of an analysis carried out on 820 MMUPs recorded from 41 subjects who were classified on the basis of a clinical opinion and the appearance of a muscle biopsy are presented and discussed.

  5. A modular artificial neural net for controlling a six-legged walking system.

    PubMed

    Cruse, H; Bartling, C; Cymbalyuk, G; Dean, J; Dreifert, M

    1995-01-01

    A system that controls the leg movement of an animal or a robot walking over irregular ground has to ensure stable support for the body and at the same time propel it forward. To do so, it has to react adaptively to unpredictable features of the environment. As part of our study of the underlying mechanisms, we present here a model for the control of the leg movement of a 6-legged walking system. The model is based on biological data obtained from the stick insect. It represents a combined treatment of realistic kinematics and biologically motivated, adaptive gait generation. The model extends a previous algorithmic model by substituting simple networks of artificial neurons for the algorithms previously used to control leg state and interleg coordination. Each system controlling an individual leg consists of three subnets. A hierarchically superior net contains two sensory and two 'premotor' units; it rhythmically suppresses the output of one or the other of the two subordinate nets. These are continuously active. They might be called the 'swing module' and the 'stance module' because they are responsible for controlling the swing (return stroke) and the stance (power stroke) movements, respectively. The swing module consists of three motor units and seven sensory units. It can produce appropriate return stroke movements for a broad range of initial and final positions, can cope with mechanical disturbances of the leg movement, and is able to react to an obstacle which hinders the normal performance of the swing movement. The complete model is able to walk at different speeds over irregular surfaces. The control system rapidly reestablishes a stable gait when the movement of the legs is disturbed.

  6. A comparison of artificial neural net and inductive decision tree learning applied to the diagnosis of coronary artery disease

    SciTech Connect

    Silver, D.L.; Hurwitz, G.A.; Cradduck, T.D.

    1994-05-01

    A variety of artificial intelligence systems are available for applications within nuclear medicine. It is important to understand the strengths and weaknesses of these systems and the class of problems for which each is best. Two supervised machine learning systems, a back propagation neural network and an inductive decision tree, were applied to the classification of coronary artery disease given a set of diagnostic input parameters. A comparison indicates that both paradigms perform well depending upon the requirements of the user. We examined the setup complexity, learning and classification speed, training accuracy, ability to generalize to previously unseen cases, and the explanatory power of the internal representations generated by the learning systems. A database of 503 patient records composed of ten parameters was used for the analysis. The target response was a binary value of disease or no disease. The results indicate that the inductive decision tree learning system is the better choice for this class of problem. It is easier to setup and training takes less time. It has good explanatory power since it produces a printed decision tree of the internal representation of acquired knowledge. On the other hand, the artificial neural net provides better generalization for new test cases, and has greater classification accuracy.

  7. Use of genetic programming, logistic regression, and artificial neural nets to predict readmission after coronary artery bypass surgery.

    PubMed

    Engoren, Milo; Habib, Robert H; Dooner, John J; Schwann, Thomas A

    2013-08-01

    As many as 14 % of patients undergoing coronary artery bypass surgery are readmitted within 30 days. Readmission is usually the result of morbidity and may lead to death. The purpose of this study is to develop and compare statistical and genetic programming models to predict readmission. Patients were divided into separate Construction and Validation populations. Using 88 variables, logistic regression, genetic programs, and artificial neural nets were used to develop predictive models. Models were first constructed and tested on the Construction populations, then validated on the Validation population. Areas under the receiver operator characteristic curves (AU ROC) were used to compare the models. Two hundred and two patients (7.6 %) in the 2,644 patient Construction group and 216 (8.0 %) of the 2,711 patient Validation group were re-admitted within 30 days of CABG surgery. Logistic regression predicted readmission with AU ROC = .675 ± .021 in the Construction group. Genetic programs significantly improved the accuracy, AU ROC = .767 ± .001, p < .001). Artificial neural nets were less accurate with AU ROC = 0.597 ± .001 in the Construction group. Predictive accuracy of all three techniques fell in the Validation group. However, the accuracy of genetic programming (AU ROC = .654 ± .001) was still trivially but statistically non-significantly better than that of the logistic regression (AU ROC = .644 ± .020, p = .61). Genetic programming and logistic regression provide alternative methods to predict readmission that are similarly accurate.

  8. Invariance and neural nets.

    PubMed

    Barnard, E; Casasent, D

    1991-01-01

    Application of neural nets to invariant pattern recognition is considered. The authors study various techniques for obtaining this invariance with neural net classifiers and identify the invariant-feature technique as the most suitable for current neural classifiers. A novel formulation of invariance in terms of constraints on the feature values leads to a general method for transforming any given feature space so that it becomes invariant to specified transformations. A case study using range imagery is used to exemplify these ideas, and good performance is obtained.

  9. A multi-views multi-learners approach towards dysarthric speech recognition using multi-nets artificial neural networks.

    PubMed

    Shahamiri, Seyed Reza; Salim, Siti Salwah Binti

    2014-09-01

    Automatic speech recognition (ASR) can be very helpful for speakers who suffer from dysarthria, a neurological disability that damages the control of motor speech articulators. Although a few attempts have been made to apply ASR technologies to sufferers of dysarthria, previous studies show that such ASR systems have not attained an adequate level of performance. In this study, a dysarthric multi-networks speech recognizer (DM-NSR) model is provided using a realization of multi-views multi-learners approach called multi-nets artificial neural networks, which tolerates variability of dysarthric speech. In particular, the DM-NSR model employs several ANNs (as learners) to approximate the likelihood of ASR vocabulary words and to deal with the complexity of dysarthric speech. The proposed DM-NSR approach was presented as both speaker-dependent and speaker-independent paradigms. In order to highlight the performance of the proposed model over legacy models, multi-views single-learner models of the DM-NSRs were also provided and their efficiencies were compared in detail. Moreover, a comparison among the prominent dysarthric ASR methods and the proposed one is provided. The results show that the DM-NSR recorded improved recognition rate by up to 24.67% and the error rate was reduced by up to 8.63% over the reference model.

  10. Application of neural nets in structural optimization

    NASA Technical Reports Server (NTRS)

    Berke, Laszlo; Hajela, Prabhat

    1993-01-01

    The biological motivation for Artificial Neural Net developments is briefly discussed, and the most popular paradigm, the feedforward supervised learning net with error back propagation training algorithm, is introduced. Possible approaches for utilization in structural optimization is illustrated through simple examples. Other currently ongoing developments for application in structural mechanics are also mentioned.

  11. Quantum Neural Nets

    NASA Technical Reports Server (NTRS)

    Zak, Michail; Williams, Colin P.

    1997-01-01

    The capacity of classical neurocomputers is limited by the number of classical degrees of freedom which is roughly proportional to the size of the computer. By Contrast, a Hypothetical quantum neurocomputer can implement an exponentially large number of the degrees of freedom within the same size. In this paper an attempt is made to reconcile linear reversible structure of quantum evolution with nonlinear irreversible dynamics for neural nets.

  12. Neural nets on the MPP

    NASA Technical Reports Server (NTRS)

    Hastings, Harold M.; Waner, Stefan

    1987-01-01

    The Massively Parallel Processor (MPP) is an ideal machine for computer experiments with simulated neural nets as well as more general cellular automata. Experiments using the MPP with a formal model neural network are described. The results on problem mapping and computational efficiency apply equally well to the neural nets of Hopfield, Hinton et al., and Geman and Geman.

  13. Author's reply to discussion of using artificial neural nets to identify the well-test interpretation model

    SciTech Connect

    Al-Kaabi, A.U.; Lee, W.J. )

    1994-09-01

    The authors thank Yeung et al. for their discussion about their original paper. They agree with Yeung et al. that their proposed scaling method, when applied to patterns with distinct subparts such as the one shown, represents an improvement on the method they proposed. This is particularly true because Yeung et al.'s method eliminates the need to train the artificial neural networks (ANN's) on different sizes (scales) of the same pattern of a specific interpretation model. This paper presents the following comments for discussion and suggestions for further improvement.

  14. Synchronization in neural nets

    NASA Technical Reports Server (NTRS)

    Vidal, Jacques J.; Haggerty, John

    1988-01-01

    The paper presents an artificial neural network concept (the Synchronizable Oscillator Networks) where the instants of individual firings in the form of point processes constitute the only form of information transmitted between joining neurons. In the model, neurons fire spontaneously and regularly in the absence of perturbation. When interaction is present, the scheduled firings are advanced or delayed by the firing of neighboring neurons. Networks of such neurons become global oscillators which exhibit multiple synchronizing attractors. From arbitrary initial states, energy minimization learning procedures can make the network converge to oscillatory modes that satisfy multi-dimensional constraints. Such networks can directly represent routing and scheduling problems that consist of ordering sequences of events.

  15. Generalized Adaptive Artificial Neural Networks

    NASA Technical Reports Server (NTRS)

    Tawel, Raoul

    1993-01-01

    Mathematical model of supervised learning by artificial neural network provides for simultaneous adjustments of both temperatures of neurons and synaptic weights, and includes feedback as well as feedforward synaptic connections. Extension of mathematical model described in "Adaptive Neurons For Artificial Neural Networks" (NPO-17803). Dynamics of neural network represented in new model by less-restrictive continuous formalism.

  16. Binary classification by stochastic neural nets.

    PubMed

    Nadas, A

    1995-01-01

    We classify points in R(d) (feature vector space) by functions related to feedforward artificial neural networks. These functions, dubbed "stochastic neural nets", arise in a natural way from probabilistic as well as from statistical considerations. The probabilistic idea is to define a classifying bit locally by using the sign of a hidden state-dependent noisy linear function of the feature vector as a new (d+1)th coordinate of the vector. This (d+1)-dimensional distribution is approximated by a mixture distribution. The statistical idea is that the approximating mixtures, and hence the a posteriori class probability functions (stochastic neural nets) defined by them, can be conveniently trained either by maximum likelihood or by a Bayes criterion through the use of an appropriate expectation-maximization algorithm.

  17. [Artificial neural networks in Neurosciences].

    PubMed

    Porras Chavarino, Carmen; Salinas Martínez de Lecea, José María

    2011-11-01

    This article shows that artificial neural networks are used for confirming the relationships between physiological and cognitive changes. Specifically, we explore the influence of a decrease of neurotransmitters on the behaviour of old people in recognition tasks. This artificial neural network recognizes learned patterns. When we change the threshold of activation in some units, the artificial neural network simulates the experimental results of old people in recognition tasks. However, the main contributions of this paper are the design of an artificial neural network and its operation inspired by the nervous system and the way the inputs are coded and the process of orthogonalization of patterns.

  18. Weakly connected neural nets

    NASA Technical Reports Server (NTRS)

    Zak, Michail

    1990-01-01

    A new neural network architecture is proposed based upon effects of non-Lipschitzian dynamics. The network is fully connected, but these connections are active only during vanishingly short time periods. The advantages of this architecture are discussed.

  19. Optimization for training neural nets.

    PubMed

    Barnard, E

    1992-01-01

    Various techniques of optimizing criterion functions to train neural-net classifiers are investigated. These techniques include three standard deterministic techniques (variable metric, conjugate gradient, and steepest descent), and a new stochastic technique. It is found that the stochastic technique is preferable on problems with large training sets and that the convergence rates of the variable metric and conjugate gradient techniques are similar.

  20. Implementing neural nets with programmable logic

    NASA Technical Reports Server (NTRS)

    Vidal, Jacques J.

    1988-01-01

    Networks of Boolean programmable logic modules are presented as one purely digital class of artificial neural nets. The approach contrasts with the continuous analog framework usually suggested. Programmable logic networks are capable of handling many neural-net applications. They avoid some of the limitations of threshold logic networks and present distinct opportunities. The network nodes are called dynamically programmable logic modules. They can be implemented with digitally controlled demultiplexers. Each node performs a Boolean function of its inputs which can be dynamically assigned. The overall network is therefore a combinational circuit and its outputs are Boolean global functions of the network's input variables. The approach offers definite advantages for VLSI implementation, namely, a regular architecture with limited connectivity, simplicity of the control machinery, natural modularity, and the support of a mature technology.

  1. Adaptive-clustering optical neural net.

    PubMed

    Casasent, D P; Barnard, E

    1990-06-10

    Pattern recognition techniques (for clustering and linear discriminant function selection) are combined with neural net methods (that provide an automated method to combine linear discriminant functions into piecewise linear discriminant surfaces). The resulting adaptive-clustering neural net is suitable for optical implementation and has certain desirable properties in comparison with other neural nets. Simulation results are provided.

  2. Survival analysis and neural nets.

    PubMed

    Liestøl, K; Andersen, P K; Andersen, U

    1994-06-30

    We consider feed-forward neural nets and their relation to regression models for survival data. We show how the back-propagation algorithm may be used to obtain maximum likelihood estimates in certain standard regression models for survival data, as well as in various generalizations of these. Examples concerning malignant melanoma and post-partum amenorrhoea during lactation are used as illustration. We conclude that although problems with the substantial number of parameters and their interpretation remain, the feed-forward neural network models are flexible extensions to the standard regression models and thereby candidates for use in prediction and exploratory analyses in larger data sets.

  3. Symmetry breaking in neural nets.

    PubMed

    Pessa, E

    1988-01-01

    In this paper two well-known homogeneous models of neural nets undergoing symmetry-breaking transitions are studied in order to see if, after the transition, there is the appearance of Goldstone modes. These have been found only in an approximate way; there are indications, however, that they can play a prominent role when the tissue is subjected to external inputs, constraining it to be slaved to the characteristics of those. This circumstance should be essential in explaining how a structured net can store complex inputs and give subsequently ordered outputs.

  4. Multifunctional hybrid optical/digital neural net

    NASA Astrophysics Data System (ADS)

    Casasent, David P.

    1990-08-01

    A multi-functional hybrid neural net is described. It is hybrid since it uses a digital hardware Hecht-Nielsen Corporation (HNC) neural net for adaptive learning and an optical neural net for on-line processing/classification. It is also hybrid in its combination of pattern recognition and neural net techniques. The system is multi-functional. It can function as an optimization and adaptive pattern recognition neural net as well as an auto and heteroassociative processor. I . W. JTRODUCTION Neural nets (NNs) have recently received enormous attention [1 -2] with increasing attention to the use of optical processors and a variety of new learning algorithms. Section 2 describes our hybrid NN with attention to Its fabrication and the role for optical and digital processors. Section 3 details Its use as an associative processor. Section 4 highlights is use in 3 optimization NN problems (a mixture NN a multitarget tracker (MTT) NN and a matrix inversion NN). Section 5 briefly notes it use as a production NN system and symbolic NN. Section 6 describes its use as an adaptive pattern recognition (PR) NN (that marries PR and NN techniques). 2. HYBRID ARCHITECTURE Figure 1 shows our basic hybrid NN [3]. The optical portion of the system is a matrix-vector (M-V) processor whose vector output P3 is the product of the vector at P1 and the matrix at P2. An HNC digital hardware NN is used during learning determine the interconnection weights forP2. If P2 is a spatial light modulator (SLM) its contents can be updated (using gated learning) from thedigital NN. The operations in most adaptive PR NN learning algorithms are sufficiently complex thatthey are best implemented digitally. In addition the learning operations required are often not well suited for optical realization for optimization NNs the weights are fixed and in adaptive learning learning is off-line and once completed the weights can often be fixed. Four gates are shown that determine the final output or the new P1

  5. From Brains to Neural Nets to Brains.

    PubMed

    Harth, Erich

    1997-10-01

    The paper traces theoretical work concerning the understanding and simulation of brain functions from early studies of artificial neural nets to present considerations of human consciousness. The emphasis is on work carried out since about 1963 at my laboratory in collaboration with my students. The discussion centers on sensory, especially visual, information processing, some of the cerebral mechanisms involved, and current approaches to an understanding of conscious perception. The sketchpad model, in which the ubiquitous feedback pathways in the brain play a dominant role, is described, together with a discussion of the meaning and applicability of scientific reductionism to the problem of consciousness.

  6. Multiscale optimization in neural nets.

    PubMed

    Mjolsness, E; Garrett, C D; Miranker, W L

    1991-01-01

    One way to speed up convergence in a large optimization problem is to introduce a smaller, approximate version of the problem at a coarser scale and to alternate between relaxation steps for the fine-scale and coarse-scale problems. Such an optimization method for neural networks governed by quite general objective functions is presented. At the coarse scale, there is a smaller approximating neural net which, like the original net, is nonlinear and has a nonquadratic objective function. The transitions and information flow from fine to coarse scale and back do not disrupt the optimization, and the user need only specify a partition of the original fine-scale variables. Thus, the method can be applied easily to many problems and networks. There is generally about a fivefold improvement in estimated cost under the multiscale method. In the networks to which it was applied, a nontrivial speedup by a constant factor of between two and five was observed, independent of problem size. Further improvements in computational cost are very likely to be available, especially for problem-specific multiscale neural net methods.

  7. Are artificial neural networks black boxes?

    PubMed

    Benitez, J M; Castro, J L; Requena, I

    1997-01-01

    Artificial neural networks are efficient computing models which have shown their strengths in solving hard problems in artificial intelligence. They have also been shown to be universal approximators. Notwithstanding, one of the major criticisms is their being black boxes, since no satisfactory explanation of their behavior has been offered. In this paper, we provide such an interpretation of neural networks so that they will no longer be seen as black boxes. This is stated after establishing the equality between a certain class of neural nets and fuzzy rule-based systems. This interpretation is built with fuzzy rules using a new fuzzy logic operator which is defined after introducing the concept of f-duality. In addition, this interpretation offers an automated knowledge acquisition procedure.

  8. A practical guide to neural nets

    SciTech Connect

    Nelson, M.M.; Illingworth, W.T.

    1991-01-01

    The concept of neural networks, their operation, and applications are reviewed. Topics discussed include definitions, terminology, and concepts of neural networks, the principal issues and problems addressed by neural network technology, recent developments in the field of artificial intelligence, characteristics and limitations of neural networks, and various neural network architectures. Other topics covered include the basic learning mechanisms of neural networks, examples of neural network applications, implementations of neural networks, some current problems in neural network research, and suggestions for future research. 126 refs.

  9. Noisy neural nets exhibiting epileptic features.

    PubMed

    Kokkinidis, M; Anninos, P

    1985-04-01

    On the basis of our previous studies of noisy neural nets we propose a model for the explanation of epileptic phenomena. Our neural net model is capable of exhibiting epileptic features if the number of spontaneously firing neurons is periodically increased beyond a certain threshold. Some alternative epileptogenic mechanisms are also discussed. The epileptic behavior of the neural net is determined by a combination of certain parameters of its phase diagram. The general features of the model are consistent with several experimental observations and explain some poorly understood clinical phenomena. The differences between normal and epileptic neural nets are explained in terms of the structural properties of the model.

  10. Artificial neural networks in medicine

    SciTech Connect

    Keller, P.E.

    1994-07-01

    This Technology Brief provides an overview of artificial neural networks (ANN). A definition and explanation of an ANN is given and situations in which an ANN is used are described. ANN applications to medicine specifically are then explored and the areas in which it is currently being used are discussed. Included are medical diagnostic aides, biochemical analysis, medical image analysis and drug development.

  11. Artificial nets from superconducting nanogranules

    SciTech Connect

    Ovchinnikov, Yu. N.; Kresin, V. Z.

    2012-06-15

    We show that a large transport current can flow through superconducting nets composed of nano-clusters. Although thermal and quantum fluctuations lead to a finite value of dissipation, this value can be very small in one- and two-dimensional systems for realistic parameters of the nanoclusters and distances between them. The value of the action for vortex tunneling at zero temperature can be made sufficiently large to make the dissipation negligibly small. We estimate the temperature T{sub 0} of the transition from the thermal activation to quantum tunneling.

  12. Introduction to Concepts in Artificial Neural Networks

    NASA Technical Reports Server (NTRS)

    Niebur, Dagmar

    1995-01-01

    This introduction to artificial neural networks summarizes some basic concepts of computational neuroscience and the resulting models of artificial neurons. The terminology of biological and artificial neurons, biological and machine learning and neural processing is introduced. The concepts of supervised and unsupervised learning are explained with examples from the power system area. Finally, a taxonomy of different types of neurons and different classes of artificial neural networks is presented.

  13. Intensity encoding in unsupervised neural nets.

    PubMed

    Parkinson, Alan M.; Parpia, Dawood Y.

    1998-06-01

    The requirement of input vector normalisation in unsupervised neural nets results in a loss of information about the intensity of the signal contained in the input datastream. We show through a simple algebraic analysis that the introduction of an additional input channel encoding the root-mean-square intensity in the signals cannot restore this information if the input vectors have to be, nevertheless, all of the same length. We suggest an alternative method of encoding the input vectors where each of the input channels is split into two components in such a way that the resultant input vector is then of fixed length and retains information of the intensity in the signals. We further demonstrate, by using synthetic data, that a Kohonen Net is capable of forming topological maps of signals of different intensity, where an adjacency relationship is maintained both among the signals of the same frequency composition at different intensities and between signals of different frequency compositions at the same intensity. A second experiment reported here shows the same behaviour for less artificial inputs (based on a cochlear model) and additionally demonstrates that the trained network can respond appropriately to signals not previously encountered.

  14. Neural Net Safety Monitor Design

    NASA Technical Reports Server (NTRS)

    Larson, Richard R.

    2007-01-01

    The National Aeronautics and Space Administration (NASA) at the Dryden Flight Research Center (DFRC) has been conducting flight-test research using an F-15 aircraft (figure 1). This aircraft has been specially modified to interface a neural net (NN) controller as part of a single-string Airborne Research Test System (ARTS) computer with the existing quad-redundant flight control system (FCC) shown in figure 2. The NN commands are passed to FCC channels 2 and 4 and are cross channel data linked (CCDL) to the other computers as shown. Numerous types of fault-detection monitors exist in the FCC when the NN mode is engaged; these monitors would cause an automatic disengagement of the NN in the event of a triggering fault. Unfortunately, these monitors still may not prevent a possible NN hard-over command from coming through to the control laws. Therefore, an additional and unique safety monitor was designed for a single-string source that allows authority at maximum actuator rates but protects the pilot and structural loads against excessive g-limits in the case of a NN hard-over command input. This additional monitor resides in the FCCs and is executed before the control laws are computed. This presentation describes a floating limiter (FL) concept1 that was developed and successfully test-flown for this program (figure 3). The FL computes the rate of change of the NN commands that are input to the FCC from the ARTS. A window is created with upper and lower boundaries, which is constantly floating and trying to stay centered as the NN command rates are changing. The limiter works by only allowing the window to move at a much slower rate than those of the NN commands. Anywhere within the window, however, full rates are allowed. If a rate persists in one direction, it will eventually hit the boundary and be rate-limited to the floating limiter rate. When this happens, a persistent counter begins and after a limit is reached, a NN disengage command is generated. The

  15. Noisy neural nets exhibiting memory domains.

    PubMed

    Anninos, P; Kokkinidis, M; Skouras, A

    1984-08-21

    Previous studies with probabilistic neural nets in which the neural connections are set up by means of chemical markers, revealed the existence of multiple memory domains. We generalized these studies by considering the intrinsic noise of the systems, caused by the spontaneous release of synaptic transmitter substance. A simple mathematical model is developed, which yields characteristics of multiple memory domains analogous to those occurring in noiseless nets.

  16. CDMA and TDMA based neural nets.

    PubMed

    Herrero, J C

    2001-06-01

    CDMA and TDMA telecommunication techniques were established long time ago, but they have acquired a renewed presence due to the rapidly increasing mobile phones demand. In this paper, we are going to see they are suitable for neural nets, if we leave the concept "connection" between processing units and we adopt the concept "messages" exchanged between them. This may open the door to neural nets with a higher number of processing units and flexible configuration.

  17. Correcting wave predictions with artificial neural networks

    NASA Astrophysics Data System (ADS)

    Makarynskyy, O.; Makarynska, D.

    2003-04-01

    The predictions of wind waves with different lead times are necessary in a large scope of coastal and open ocean activities. Numerical wave models, which usually provide this information, are based on deterministic equations that do not entirely account for the complexity and uncertainty of the wave generation and dissipation processes. An attempt to improve wave parameters short-term forecasts based on artificial neural networks is reported. In recent years, artificial neural networks have been used in a number of coastal engineering applications due to their ability to approximate the nonlinear mathematical behavior without a priori knowledge of interrelations among the elements within a system. The common multilayer feed-forward networks, with a nonlinear transfer functions in the hidden layers, were developed and employed to forecast the wave characteristics over one hour intervals starting from one up to 24 hours, and to correct these predictions. Three non-overlapping data sets of wave characteristics, both from a buoy, moored roughly 60 miles west of the Aran Islands, west coast of Ireland, were used to train and validate the neural nets involved. The networks were trained with error back propagation algorithm. Time series plots and scatterplots of the wave characteristics as well as tables with statistics show an improvement of the results achieved due to the correction procedure employed.

  18. Devices and circuits for nanoelectronic implementation of artificial neural networks

    NASA Astrophysics Data System (ADS)

    Turel, Ozgur

    Biological neural networks perform complicated information processing tasks at speeds better than conventional computers based on conventional algorithms. This has inspired researchers to look into the way these networks function, and propose artificial networks that mimic their behavior. Unfortunately, most artificial neural networks, either software or hardware, do not provide either the speed or the complexity of a human brain. Nanoelectronics, with high density and low power dissipation that it provides, may be used in developing more efficient artificial neural networks. This work consists of two major contributions in this direction. First is the proposal of the CMOL concept, hybrid CMOS-molecular hardware [1-8]. CMOL may circumvent most of the problems in posed by molecular devices, such as low yield, vet provide high active device density, ˜1012/cm 2. The second contribution is CrossNets, artificial neural networks that are based on CMOL. We showed that CrossNets, with their fault tolerance, exceptional speed (˜ 4 to 6 orders of magnitude faster than biological neural networks) can perform any task any artificial neural network can perform. Moreover, there is a hope that if their integration scale is increased to that of human cerebral cortex (˜ 1010 neurons and ˜ 1014 synapses), they may be capable of performing more advanced tasks.

  19. Plant Growth Models Using Artificial Neural Networks

    NASA Technical Reports Server (NTRS)

    Bubenheim, David

    1997-01-01

    In this paper, we descrive our motivation and approach to devloping models and the neural network architecture. Initial use of the artificial neural network for modeling the single plant process of transpiration is presented.

  20. Quantum dissipation and neural net dynamics.

    PubMed

    Pessa, E; Vitiello, G

    1999-05-01

    Inspired by the dissipative quantum model of brain, we model the states of neural nets in terms of collective modes by the help of the formalism of Quantum Field Theory. We exhibit an explicit neural net model which allows to memorize a sequence of several informations without reciprocal destructive interference, namely we solve the overprinting problem in such a way last registered information does not destroy the ones previously registered. Moreover, the net is able to recall not only the last registered information in the sequence, but also anyone of those previously registered.

  1. Artificial neural networks in neurosurgery.

    PubMed

    Azimi, Parisa; Mohammadi, Hasan Reza; Benzel, Edward C; Shahzadi, Sohrab; Azhari, Shirzad; Montazeri, Ali

    2015-03-01

    Artificial neural networks (ANNs) effectively analyze non-linear data sets. The aimed was A review of the relevant published articles that focused on the application of ANNs as a tool for assisting clinical decision-making in neurosurgery. A literature review of all full publications in English biomedical journals (1993-2013) was undertaken. The strategy included a combination of key words 'artificial neural networks', 'prognostic', 'brain', 'tumor tracking', 'head', 'tumor', 'spine', 'classification' and 'back pain' in the title and abstract of the manuscripts using the PubMed search engine. The major findings are summarized, with a focus on the application of ANNs for diagnostic and prognostic purposes. Finally, the future of ANNs in neurosurgery is explored. A total of 1093 citations were identified and screened. In all, 57 citations were found to be relevant. Of these, 50 articles were eligible for inclusion in this review. The synthesis of the data showed several applications of ANN in neurosurgery, including: (1) diagnosis and assessment of disease progression in low back pain, brain tumours and primary epilepsy; (2) enhancing clinically relevant information extraction from radiographic images, intracranial pressure processing, low back pain and real-time tumour tracking; (3) outcome prediction in epilepsy, brain metastases, lumbar spinal stenosis, lumbar disc herniation, childhood hydrocephalus, trauma mortality, and the occurrence of symptomatic cerebral vasospasm in patients with aneurysmal subarachnoid haemorrhage; (4) the use in the biomechanical assessments of spinal disease. ANNs can be effectively employed for diagnosis, prognosis and outcome prediction in neurosurgery.

  2. Genetic algorithms for genetic neural nets. Research report

    SciTech Connect

    Sharp, D.H.; Reinitz, J.; Mjolsness, E.

    1991-01-01

    In contrast to most synthetic neural nets, biological neural networks have a strong component of genetic determination which acts before and during experiential learning. Three broad levels of phenomena are present: long-term evolution, involving crossover as well as point mutation; a developmental process mapping genetic information to a set of cells and their internal states of gene expression (genotype to phenotype); and the subsequent synaptogenesis. We describe a very simple mathematical idealization of these three levels which combines the crossover search method of genetic algorithms with the developmental models used in our previous work on 'genetic' or 'recursively generated' artificial neural nets and elaborated into a connectionist model of biological development. Despite incorporating all three levels (evolution on genes; development of cells; synapse formation) the model may actually be far cheaper to compute with than a comparable search directly in synaptic weight space.

  3. Accelerator diagnosis and control by Neural Nets

    SciTech Connect

    Spencer, J.E.

    1989-01-01

    Neural Nets (NN) have been described as a solution looking for a problem. In the last conference, Artificial Intelligence (AI) was considered in the accelerator context. While good for local surveillance and control, its use for large complex systems (LCS) was much more restricted. By contrast, NN provide a good metaphor for LCS. It can be argued that they are logically equivalent to multi-loop feedback/forward control of faulty systems, and therefore provide an ideal adaptive control system. Thus, where AI may be good for maintaining a 'golden orbit,' NN should be good for obtaining it via a quantitative approach to 'look and adjust' methods like operator tweaking which use pattern recognition to deal with hardware and software limitations, inaccuracies or errors as well as imprecise knowledge or understanding of effects like annealing and hysteresis. Further, insights from NN allow one to define feasibility conditions for LCS in terms of design constraints and tolerances. Hardware and software implications are discussed and several LCS of current interest are compared and contrasted. 15 refs., 5 figs.

  4. Neural net based MRAC for a class of nonlinear plants.

    PubMed

    Ahmed, M S

    2000-01-01

    A neural net based state feedback model reference adaptive control scheme is presented for a class of nonlinear plants. The proposed scheme employs a time varying pseudo-linear state feedback control, where the state feedback gain being generated from the output of artificial neural networks. The plant need not be in a feedback linearizable form. Both regulation and tracking control have been considered. Global stability of the scheme has been proved through Lyapunov theory. The extension of the scheme to MIMO plants is also included. Simulation studies have been conducted on an industrial robot to validate and illustrate the proposed method.

  5. Creativity in design and artificial neural networks

    SciTech Connect

    Neocleous, C.C.; Esat, I.I.; Schizas, C.N.

    1996-12-31

    The creativity phase is identified as an integral part of the design phase. The characteristics of creative persons which are relevant to designing artificial neural networks manifesting aspects of creativity, are identified. Based on these identifications, a general framework of artificial neural network characteristics to implement such a goal are proposed.

  6. Modular, Hierarchical Learning By Artificial Neural Networks

    NASA Technical Reports Server (NTRS)

    Baldi, Pierre F.; Toomarian, Nikzad

    1996-01-01

    Modular and hierarchical approach to supervised learning by artificial neural networks leads to neural networks more structured than neural networks in which all neurons fully interconnected. These networks utilize general feedforward flow of information and sparse recurrent connections to achieve dynamical effects. The modular organization, sparsity of modular units and connections, and fact that learning is much more circumscribed are all attractive features for designing neural-network hardware. Learning streamlined by imitating some aspects of biological neural networks.

  7. Analysis of clinical data using neural nets.

    PubMed

    Minor, J M; Namini, H

    1996-03-01

    Clinical studies investigate the interdependence of dosing regimen, efficacy, and side effects. These relationships often involve complex dynamical functions. The study of phenomena intermediate between dosing and efficacy, e.g., pharmacokinetics (PK), helps one identify and understand these interdependences. However, efficacy still tends to be a complicated function of PK parameters, and indeed these parameters are becoming more complex as a function of dosing regimen, viz., studies involving immunosuppressants, biotechnology drugs, sophisticated delivery systems, and dosing strategies. Stationary and time-dependent neural nets can help one identify and model such unknown complex dynamical functions with few assumptions and limited data (1-7). Neural nets can relate dosing directly to efficacy, dosing to PK, PK to efficacy, or any component in the complex associations among treatments, pharmacodynamics, efficacy, and side effects. Neural nets can also assist one in the design of clinical trials involving complex and sophisticated procedures, e.g., randomized controlled clinical trials.

  8. Neural nets and eddy-current testing

    SciTech Connect

    Allen, J.D. Jr.; Dodd, C.V.; Pate, J.R.; Schell, F.M.

    1990-01-01

    Artificial neural networks of a novel type have been trained and tested on a variety of eddy-current flaw signals commonly occurring in nuclear reactor steam generators with the ultimate goal of emulating, at least crudely, the vision and reasoning capabilities of the human analyst. The network methodology itself was that of Allen and Schell developed originally for studies of such biologically relevant neural properties as cognitive complementarity and concept formation. Because they are so important to the results obtained, we discuss the general characteristics of the approach in the Preamble. In Section I we describe the relevant aspects of the neural network configuration presently in use and in Section II the method by which the artificial neural systems have been trained. Finally, we discuss results of the training process for systems explored.

  9. The power of neural nets

    NASA Technical Reports Server (NTRS)

    Ryan, J. P.; Shah, B. H.

    1987-01-01

    Implementation of the Hopfield net which is used in the image processing type of applications where only partial information about the image may be available is discussed. The image classification type of algorithm of Hopfield and other learning algorithms, such as the Boltzmann machine and the back-propagation training algorithm, have many vital applications in space.

  10. Introduction to artificial neural networks.

    PubMed

    Grossi, Enzo; Buscema, Massimo

    2007-12-01

    The coupling of computer science and theoretical bases such as nonlinear dynamics and chaos theory allows the creation of 'intelligent' agents, such as artificial neural networks (ANNs), able to adapt themselves dynamically to problems of high complexity. ANNs are able to reproduce the dynamic interaction of multiple factors simultaneously, allowing the study of complexity; they can also draw conclusions on individual basis and not as average trends. These tools can offer specific advantages with respect to classical statistical techniques. This article is designed to acquaint gastroenterologists with concepts and paradigms related to ANNs. The family of ANNs, when appropriately selected and used, permits the maximization of what can be derived from available data and from complex, dynamic, and multidimensional phenomena, which are often poorly predictable in the traditional 'cause and effect' philosophy. PMID:17998827

  11. Making real neural nets: design criteria.

    PubMed

    Curtis, A S; Breckenridge, L; Connolly, P; Dow, J A; Wildinson, C D; Wilson, R

    1992-07-01

    Neural nets may be assembled with living nerve cells in vitro to test theories about neural processing and the ways in which patterns develop in the nervous system, and to test ideas about plasticity and learning in processing systems. This may benefit the design of computer systems and prosthetic devices. Extracting information from such nets can be achieved by means of intracellular and extracellular electrodes and fluorescent dyes. Patterning of cells may be achieved using microfabrication techniques, and extracellular electrodes can be combined within the patterned substrate.

  12. Artificial neural networks and their use in quantitative pathology.

    PubMed

    Dytch, H E; Wied, G L

    1990-12-01

    A brief general introduction to artificial neural networks is presented, examining in detail the structure and operation of a prototype net developed for the solution of a simple pattern recognition problem in quantitative pathology. The process by which a neural network learns through example and gradually embodies its knowledge as a distributed representation is discussed, using this example. The application of neurocomputer technology to problems in quantitative pathology is explored, using real-world and illustrative examples. Included are examples of the use of artificial neural networks for pattern recognition, database analysis and machine vision. In the context of these examples, characteristics of neural nets, such as their ability to tolerate ambiguous, noisy and spurious data and spontaneously generalize from known examples to handle unfamiliar cases, are examined. Finally, the strengths and deficiencies of a connectionist approach are compared to those of traditional symbolic expert system methodology. It is concluded that artificial neural networks, used in conjunction with other nonalgorithmic artificial intelligence techniques and traditional algorithmic processing, may provide useful software engineering tools for the development of systems in quantitative pathology.

  13. Analysis of torsional oscillations using an artificial neural network

    SciTech Connect

    Hsu, Y.Y.; Jeng, L,H. )

    1992-12-01

    In this paper, a novel approach using an artificial neural network (ANN) is proposed for the analysis of torsional oscillations in a power system. In the developed artificial neural network, those system variables such as generator loadings and capacitor compensation ratio which have major impacts on the damping characteristics of torsional oscillatio modes are employed as the inputs. The outputs of the neural net provide the desired eigenvalues for torsional modes. Once the connection weights of the neural network have been learned using a set of training data derived off-line, the neural network can be applied to torsional analysis in real-time situations. To demonstrate the effectiveness of the proposed neural net, torsional analysis is performed on the IEEE First Benchmark Model. It is concluded from the test results that accurate assessment of the torsional mode eigenvalues can be achieved by the neural network in a very efficient manner. Thereofore, the proposed neural network approach can serve as a valuable tool to system operators in conducting SSR analysis in operational planning.

  14. Learning in Artificial Neural Systems

    NASA Technical Reports Server (NTRS)

    Matheus, Christopher J.; Hohensee, William E.

    1987-01-01

    This paper presents an overview and analysis of learning in Artificial Neural Systems (ANS's). It begins with a general introduction to neural networks and connectionist approaches to information processing. The basis for learning in ANS's is then described, and compared with classical Machine learning. While similar in some ways, ANS learning deviates from tradition in its dependence on the modification of individual weights to bring about changes in a knowledge representation distributed across connections in a network. This unique form of learning is analyzed from two aspects: the selection of an appropriate network architecture for representing the problem, and the choice of a suitable learning rule capable of reproducing the desired function within the given network. The various network architectures are classified, and then identified with explicit restrictions on the types of functions they are capable of representing. The learning rules, i.e., algorithms that specify how the network weights are modified, are similarly taxonomized, and where possible, the limitations inherent to specific classes of rules are outlined.

  15. Detection of Structural Abnormalities Using Neural Nets

    NASA Technical Reports Server (NTRS)

    Zak, M.; Maccalla, A.; Daggumati, V.; Gulati, S.; Toomarian, N.

    1996-01-01

    This paper describes a feed-forward neural net approach for detection of abnormal system behavior based upon sensor data analyses. A new dynamical invariant representing structural parameters of the system is introduced in such a way that any structural abnormalities in the system behavior are detected from the corresponding changes to the invariant.

  16. Neural net simulation of the corpus callosum.

    PubMed

    Anninos, P A; Cook, N D

    1988-02-01

    The effects of simulated anatomical and physiological parameters were investigated in a "neural net" model, where two neural nets corresponding to two small patches of cerebral cortex were connected by a simulated "corpus callosum." The isolated neural nets have previously been shown to exhibit oscillatory activity similar to the raw EEG. By connecting the nets with fibers which have a specified percentage of inhibition and a specified percentage of homotopicity, the effects of such parameters on the cyclic activity of the nets were studied. It was found that, regardless of the inhibitory-excitatory nature of the simulated corpus callosum, the cyclic activity established in one hemisphere is more readily transferred to the contralateral hemisphere, the greater the percentage of homotopic callosal fibers. Learning was more rapid when the effect of the corpus callosum was primarily excitatory, but learning also took place over inhibitory or mixed callosal tracts. The simulation does not therefore resolve the issue of the predominant physiological effect of the corpus callosum, but does indicate that, given the assumptions of the simulation, "learning" can occur regardless of the percentage of excitatory or inhibitory fibers. It is noteworthy that homotopicity was more important for learning across an inhibitory tract than across an excitatory tract.

  17. Neural-net Processed Electronic Holography for Rotating Machines

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J.

    2003-01-01

    This report presents the results of an R&D effort to apply neural-net processed electronic holography to NDE of rotors. Electronic holography was used to generate characteristic patterns or mode shapes of vibrating rotors and rotor components. Artificial neural networks were trained to identify damage-induced changes in the characteristic patterns. The development and optimization of a neural-net training method were the most significant contributions of this work, and the training method and its optimization are discussed in detail. A second positive result was the assembly and testing of a fiber-optic holocamera. A major disappointment was the inadequacy of the high-speed-holography hardware selected for this effort, but the use of scaled holograms to match the low effective resolution of an image intensifier was one interesting attempt to compensate. This report also discusses in some detail the physics and environmental requirements for rotor electronic holography. The major conclusions were that neural-net and electronic-holography inspections of stationary components in the laboratory and the field are quite practical and worthy of continuing development, but that electronic holography of moving rotors is still an expensive high-risk endeavor.

  18. The use of artificial neural networks in experimental data acquisition and aerodynamic design

    NASA Technical Reports Server (NTRS)

    Meade, Andrew J., Jr.

    1991-01-01

    It is proposed that an artificial neural network be used to construct an intelligent data acquisition system. The artificial neural networks (ANN) model has a potential for replacing traditional procedures as well as for use in computational fluid dynamics validation. Potential advantages of the ANN model are listed. As a proof of concept, the author modeled a NACA 0012 airfoil at specific conditions, using the neural network simulator NETS, developed by James Baffes of the NASA Johnson Space Center. The neural network predictions were compared to the actual data. It is concluded that artificial neural networks can provide an elegant and valuable class of mathematical tools for data analysis.

  19. A mixture neural net for multispectral imaging spectrometer processing

    NASA Technical Reports Server (NTRS)

    Casasent, David; Slagle, Timothy

    1990-01-01

    Each spatial region viewed by an imaging spectrometer contains various elements in a mixture. The elements present and the amount of each are to be determined. A neural net solution is considered. Initial optical neural net hardware is described. The first simulations on the component requirements of a neural net are considered. The pseudoinverse solution is shown to not suffice, i.e. a neural net solution is required.

  20. Artificial Neural Networks and Instructional Technology.

    ERIC Educational Resources Information Center

    Carlson, Patricia A.

    1991-01-01

    Artificial neural networks (ANN), part of artificial intelligence, are discussed. Such networks are fed sample cases (training sets), learn how to recognize patterns in the sample data, and use this experience in handling new cases. Two cognitive roles for ANNs (intelligent filters and spreading, associative memories) are examined. Prototypes…

  1. Artificial astrocytes improve neural network performance.

    PubMed

    Porto-Pazos, Ana B; Veiguela, Noha; Mesejo, Pablo; Navarrete, Marta; Alvarellos, Alberto; Ibáñez, Oscar; Pazos, Alejandro; Araque, Alfonso

    2011-01-01

    Compelling evidence indicates the existence of bidirectional communication between astrocytes and neurons. Astrocytes, a type of glial cells classically considered to be passive supportive cells, have been recently demonstrated to be actively involved in the processing and regulation of synaptic information, suggesting that brain function arises from the activity of neuron-glia networks. However, the actual impact of astrocytes in neural network function is largely unknown and its application in artificial intelligence remains untested. We have investigated the consequences of including artificial astrocytes, which present the biologically defined properties involved in astrocyte-neuron communication, on artificial neural network performance. Using connectionist systems and evolutionary algorithms, we have compared the performance of artificial neural networks (NN) and artificial neuron-glia networks (NGN) to solve classification problems. We show that the degree of success of NGN is superior to NN. Analysis of performances of NN with different number of neurons or different architectures indicate that the effects of NGN cannot be accounted for an increased number of network elements, but rather they are specifically due to astrocytes. Furthermore, the relative efficacy of NGN vs. NN increases as the complexity of the network increases. These results indicate that artificial astrocytes improve neural network performance, and established the concept of Artificial Neuron-Glia Networks, which represents a novel concept in Artificial Intelligence with implications in computational science as well as in the understanding of brain function. PMID:21526157

  2. Artificial Astrocytes Improve Neural Network Performance

    PubMed Central

    Porto-Pazos, Ana B.; Veiguela, Noha; Mesejo, Pablo; Navarrete, Marta; Alvarellos, Alberto; Ibáñez, Oscar; Pazos, Alejandro; Araque, Alfonso

    2011-01-01

    Compelling evidence indicates the existence of bidirectional communication between astrocytes and neurons. Astrocytes, a type of glial cells classically considered to be passive supportive cells, have been recently demonstrated to be actively involved in the processing and regulation of synaptic information, suggesting that brain function arises from the activity of neuron-glia networks. However, the actual impact of astrocytes in neural network function is largely unknown and its application in artificial intelligence remains untested. We have investigated the consequences of including artificial astrocytes, which present the biologically defined properties involved in astrocyte-neuron communication, on artificial neural network performance. Using connectionist systems and evolutionary algorithms, we have compared the performance of artificial neural networks (NN) and artificial neuron-glia networks (NGN) to solve classification problems. We show that the degree of success of NGN is superior to NN. Analysis of performances of NN with different number of neurons or different architectures indicate that the effects of NGN cannot be accounted for an increased number of network elements, but rather they are specifically due to astrocytes. Furthermore, the relative efficacy of NGN vs. NN increases as the complexity of the network increases. These results indicate that artificial astrocytes improve neural network performance, and established the concept of Artificial Neuron-Glia Networks, which represents a novel concept in Artificial Intelligence with implications in computational science as well as in the understanding of brain function. PMID:21526157

  3. SEU fault tolerance in artificial neural networks

    SciTech Connect

    Velazco, R.; Assoum, A.; Radi, N.E.; Ecoffet, R.; Botey, X.

    1995-12-01

    In this paper the authors investigate the robustness of Artificial Neural Networks when encountering transient modification of information bits related to the network operation. These kinds of faults are likely to occur as a consequence of interaction with radiation. Results of tests performed to evaluate the fault tolerance properties of two different digital neural circuits are presented.

  4. Cognitive reasoning using fuzzy neural nets.

    PubMed

    Pal, S; Konar, A

    1996-01-01

    The paper presents a new model for cognitive reasoning using fuzzy neural nets. The analysis of the proposed model yields guaranteed stability of the temporal fuzzy inferences, derived from the network and conditional stability of the structure of the cognitive map, framed by the arcs of the network. The results arrived at in the paper have been illustrated with reference to a typical weather forecast system.

  5. Neural net forecasting for geomagnetic activity

    NASA Technical Reports Server (NTRS)

    Hernandez, J. V.; Tajima, T.; Horton, W.

    1993-01-01

    We use neural nets to construct nonlinear models to forecast the AL index given solar wind and interplanetary magnetic field (IMF) data. We follow two approaches: (1) the state space reconstruction approach, which is a nonlinear generalization of autoregressive-moving average models (ARMA) and (2) the nonlinear filter approach, which reduces to a moving average model (MA) in the linear limit. The database used here is that of Bargatze et al. (1985).

  6. Neural Net Gains Estimation Based on an Equivalent Model.

    PubMed

    Aguilar Cruz, Karen Alicia; Medel Juárez, José de Jesús; Fernández Muñoz, José Luis; Esmeralda Vigueras Velázquez, Midory

    2016-01-01

    A model of an Equivalent Artificial Neural Net (EANN) describes the gains set, viewed as parameters in a layer, and this consideration is a reproducible process, applicable to a neuron in a neural net (NN). The EANN helps to estimate the NN gains or parameters, so we propose two methods to determine them. The first considers a fuzzy inference combined with the traditional Kalman filter, obtaining the equivalent model and estimating in a fuzzy sense the gains matrix A and the proper gain K into the traditional filter identification. The second develops a direct estimation in state space, describing an EANN using the expected value and the recursive description of the gains estimation. Finally, a comparison of both descriptions is performed; highlighting the analytical method describes the neural net coefficients in a direct form, whereas the other technique requires selecting into the Knowledge Base (KB) the factors based on the functional error and the reference signal built with the past information of the system.

  7. Neural Net Gains Estimation Based on an Equivalent Model.

    PubMed

    Aguilar Cruz, Karen Alicia; Medel Juárez, José de Jesús; Fernández Muñoz, José Luis; Esmeralda Vigueras Velázquez, Midory

    2016-01-01

    A model of an Equivalent Artificial Neural Net (EANN) describes the gains set, viewed as parameters in a layer, and this consideration is a reproducible process, applicable to a neuron in a neural net (NN). The EANN helps to estimate the NN gains or parameters, so we propose two methods to determine them. The first considers a fuzzy inference combined with the traditional Kalman filter, obtaining the equivalent model and estimating in a fuzzy sense the gains matrix A and the proper gain K into the traditional filter identification. The second develops a direct estimation in state space, describing an EANN using the expected value and the recursive description of the gains estimation. Finally, a comparison of both descriptions is performed; highlighting the analytical method describes the neural net coefficients in a direct form, whereas the other technique requires selecting into the Knowledge Base (KB) the factors based on the functional error and the reference signal built with the past information of the system. PMID:27366146

  8. Neural Net Gains Estimation Based on an Equivalent Model

    PubMed Central

    Aguilar Cruz, Karen Alicia; Medel Juárez, José de Jesús; Fernández Muñoz, José Luis; Esmeralda Vigueras Velázquez, Midory

    2016-01-01

    A model of an Equivalent Artificial Neural Net (EANN) describes the gains set, viewed as parameters in a layer, and this consideration is a reproducible process, applicable to a neuron in a neural net (NN). The EANN helps to estimate the NN gains or parameters, so we propose two methods to determine them. The first considers a fuzzy inference combined with the traditional Kalman filter, obtaining the equivalent model and estimating in a fuzzy sense the gains matrix A and the proper gain K into the traditional filter identification. The second develops a direct estimation in state space, describing an EANN using the expected value and the recursive description of the gains estimation. Finally, a comparison of both descriptions is performed; highlighting the analytical method describes the neural net coefficients in a direct form, whereas the other technique requires selecting into the Knowledge Base (KB) the factors based on the functional error and the reference signal built with the past information of the system. PMID:27366146

  9. The cerebral hemispheres and bilateral neural nets.

    PubMed

    Cook, N D; Beech, A R

    1990-06-01

    A high-level cognitive dichotomy ("language and context") is reviewed in relation to empirical findings concerning the functions of the human cerebral hemispheres. We argue that the right hemisphere's involvement in the generation of connotative and contextual information in parallel with the denotative and literal language functions of the left hemisphere provides an important insight into the organization of viable cognitive systems. The role of the corpus callosum in producing the dichotomy is discussed. Finally, the generation of asymmetrical activity in structurally symmetrical, bilateral neural nets is described. The model demonstrates how complementary memory states can be generated in bilateral nets without assuming different modes of information processing, provided that the nets have inhibitory, homotopic connections. Unlike excitatory connections, inhibitory connections are sufficient to generate asymmetric hemispheric activity without postulating intrinsic differences between the cerebral hemispheres.

  10. Document analysis with neural net circuits

    NASA Technical Reports Server (NTRS)

    Graf, Hans Peter

    1994-01-01

    Document analysis is one of the main applications of machine vision today and offers great opportunities for neural net circuits. Despite more and more data processing with computers, the number of paper documents is still increasing rapidly. A fast translation of data from paper into electronic format is needed almost everywhere, and when done manually, this is a time consuming process. Markets range from small scanners for personal use to high-volume document analysis systems, such as address readers for the postal service or check processing systems for banks. A major concern with present systems is the accuracy of the automatic interpretation. Today's algorithms fail miserably when noise is present, when print quality is poor, or when the layout is complex. A common approach to circumvent these problems is to restrict the variations of the documents handled by a system. In our laboratory, we had the best luck with circuits implementing basic functions, such as convolutions, that can be used in many different algorithms. To illustrate the flexibility of this approach, three applications of the NET32K circuit are described in this short viewgraph presentation: locating address blocks, cleaning document images by removing noise, and locating areas of interest in personal checks to improve image compression. Several of the ideas realized in this circuit that were inspired by neural nets, such as analog computation with a low resolution, resulted in a chip that is well suited for real-world document analysis applications and that compares favorably with alternative, 'conventional' circuits.

  11. A neural net model for multiple memory domains.

    PubMed

    Anninos, P; Kokkinidis, M

    1984-07-01

    Previous studies with neural nets constructed of discrete populations of formal neurons have assumed that all neurons have the same probability of connection with any other neuron in the net. However, in this new study we incorporate the behavior of the neural systems in which the neural connections can be set up by means of chemical markers carried by the individual cells. With this new approach we studied the dynamics of isolated neural nets again as well as the dynamics of neural nets with sustained inputs. Results obtained with this approach show simple and multiple hysteresis phenomena. Such hysteresis loops may be considered to represent the basis for short-term memory.

  12. Wavelets, self-organizing maps and artificial neural nets for predicting energy use and estimating uncertainties in energy savings in commercial buildings

    NASA Astrophysics Data System (ADS)

    Lei, Yafeng

    This dissertation develops a "neighborhood" based neural network model utilizing wavelet analysis and Self-organizing Map (SOM) to predict building baseline energy use. Wavelet analysis was used for feature extraction of the daily weather profiles. The resulting few significant wavelet coefficients represent not only average but also variation of the weather components. A SOM is used for clustering and projecting high-dimensional data into usually a one or two dimensional map to reveal the data structure which is not clear by visual inspection. In this study, neighborhoods that contain days with similar meteorological conditions are classified by a SOM using significant wavelet coefficients; a baseline model is then developed for each neighborhood. In each neighborhood, modeling is more robust without unnecessary compromises that occur in global predictor regression models. This method was applied to the Energy Predictor Shootout II dataset and compared with the winning entries for hourly energy use predictions. A comparison between the "neighborhood" based linear regression model and the change-point model for daily energy use prediction was also performed. We also studied the application of the non-parametric nearest neighborhood points approach in determining the uncertainty of energy use prediction. The uncertainty from "local" system behavior rather than from global statistical indices such as root mean square error and other measures is shown to be more realistic and credible than the statistical approaches currently used. In general, a baseline model developed by local system behavior is more reliable than a global baseline model. The "neighborhood" based neural network model was found to predict building baseline energy use more accurately and achieve more reliable estimation of energy savings as well as the associated uncertainties in energy savings from building retrofits.

  13. Alpha spectral analysis via artificial neural networks

    SciTech Connect

    Kangas, L.J.; Hashem, S.; Keller, P.E.; Kouzes, R.T.; Troyer, G.L.

    1994-10-01

    An artificial neural network system that assigns quality factors to alpha particle energy spectra is discussed. The alpha energy spectra are used to detect plutonium contamination in the work environment. The quality factors represent the levels of spectral degradation caused by miscalibration and foreign matter affecting the instruments. A set of spectra was labeled with a quality factor by an expert and used in training the artificial neural network expert system. The investigation shows that the expert knowledge of alpha spectra quality factors can be transferred to an ANN system.

  14. Optimization of Training Sets for Neural-Net Processing of Characteristic Patterns from Vibrating Solids

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J.

    2001-01-01

    Artificial neural networks have been used for a number of years to process holography-generated characteristic patterns of vibrating structures. This technology depends critically on the selection and the conditioning of the training sets. A scaling operation called folding is discussed for conditioning training sets optimally for training feed-forward neural networks to process characteristic fringe patterns. Folding allows feed-forward nets to be trained easily to detect damage-induced vibration-displacement-distribution changes as small as 10 nm. A specific application to aerospace of neural-net processing of characteristic patterns is presented to motivate the conditioning and optimization effort.

  15. Multitarget tracking with cubic energy optical neural nets.

    PubMed

    Barnard, E; Casasent, D P

    1989-02-15

    A neural net processor and its optical realization are described for a multitarget tracking application. A cubic energy function results and a new optical neural processor is required. Initial simulation data are presented.

  16. Psychometric Measurement Models and Artificial Neural Networks

    ERIC Educational Resources Information Center

    Sese, Albert; Palmer, Alfonso L.; Montano, Juan J.

    2004-01-01

    The study of measurement models in psychometrics by means of dimensionality reduction techniques such as Principal Components Analysis (PCA) is a very common practice. In recent times, an upsurge of interest in the study of artificial neural networks apt to computing a principal component extraction has been observed. Despite this interest, the…

  17. Artificial neural interfaces for bionic cardiovascular treatments.

    PubMed

    Kawada, Toru; Sugimachi, Masaru

    2009-01-01

    An artificial nerve, in the broad sense, may be conceptualized as a physical and logical interface system that reestablishes the information traffic between the central nervous system and peripheral organs. Studies on artificial nerves targeting the autonomic nervous system are in progress to explore new treatment strategies for several cardiovascular diseases. In this article, we will review our research targeting the autonomic nervous system to treat cardiovascular diseases. First, we identified the rule for decoding native sympathetic nerve activity into a heart rate using transfer function analysis, and established a framework for a neurally regulated cardiac pacemaker. Second, we designed a bionic baroreflex system to restore the baroreflex buffering function using electrical stimulation of the celiac ganglion in a rat model of orthostatic hypotension. Third, based on the hypothesis that autonomic imbalance aggravates chronic heart failure, we implanted a neural interface into the right vagal nerve and demonstrated that intermittent vagal stimulation significantly improved the survival rate in rats with chronic heart failure following myocardial infarction. Although several practical problems need to be resolved, such as those relating to the development of electrodes feasible for long-term nerve activity recording, studies of artificial neural interfaces with the autonomic nervous system have great possibilities in the field of cardiovascular treatment. We expect further development of artificial neural interfaces as novel strategies to cope with cardiovascular diseases resistant to conventional therapeutics.

  18. Artificial neural network simulation of battery performance

    SciTech Connect

    O`Gorman, C.C.; Ingersoll, D.; Jungst, R.G.; Paez, T.L.

    1998-12-31

    Although they appear deceptively simple, batteries embody a complex set of interacting physical and chemical processes. While the discrete engineering characteristics of a battery such as the physical dimensions of the individual components, are relatively straightforward to define explicitly, their myriad chemical and physical processes, including interactions, are much more difficult to accurately represent. Within this category are the diffusive and solubility characteristics of individual species, reaction kinetics and mechanisms of primary chemical species as well as intermediates, and growth and morphology characteristics of reaction products as influenced by environmental and operational use profiles. For this reason, development of analytical models that can consistently predict the performance of a battery has only been partially successful, even though significant resources have been applied to this problem. As an alternative approach, the authors have begun development of a non-phenomenological model for battery systems based on artificial neural networks. Both recurrent and non-recurrent forms of these networks have been successfully used to develop accurate representations of battery behavior. The connectionist normalized linear spline (CMLS) network has been implemented with a self-organizing layer to model a battery system with the generalized radial basis function net. Concurrently, efforts are under way to use the feedforward back propagation network to map the {open_quotes}state{close_quotes} of a battery system. Because of the complexity of battery systems, accurate representation of the input and output parameters has proven to be very important. This paper describes these initial feasibility studies as well as the current models and makes comparisons between predicted and actual performance.

  19. Neural-net based real-time economic dispatch for thermal power plants

    SciTech Connect

    Djukanovic, M.; Milosevic, B.; Calovic, M.; Sobajic, D.J.

    1996-12-01

    This paper proposes the application of artificial neural networks to real-time optimal generation dispatch of thermal units. The approach can take into account the operational requirements and network losses. The proposed economic dispatch uses an artificial neural network (ANN) for generation of penalty factors, depending on the input generator powers and identified system load change. Then, a few additional iterations are performed within an iterative computation procedure for the solution of coordination equations, by using reference-bus penalty-factors derived from the Newton-Raphson load flow. A coordination technique for environmental and economic dispatch of pure thermal systems, based on the neural-net theory for simplified solution algorithms and improved man-machine interface is introduced. Numerical results on two test examples show that the proposed algorithm can efficiently and accurately develop optimal and feasible generator output trajectories, by applying neural-net forecasts of system load patterns.

  20. Fast training algorithms for multilayer neural nets.

    PubMed

    Brent, R P

    1991-01-01

    An algorithm that is faster than back-propagation and for which it is not necessary to specify the number of hidden units in advance is described. The relationship with other fast pattern-recognition algorithms, such as algorithms based on k-d trees, is discussed. The algorithm has been implemented and tested on artificial problems, such as the parity problem, and on real problems arising in speech recognition. Experimental results, including training times and recognition accuracy, are given. Generally, the algorithm achieves accuracy as good as or better than nets trained using back-propagation. Accuracy is comparable to that for the nearest-neighbor algorithm, which is slower and requires more storage space.

  1. [Medical use of artificial neural networks].

    PubMed

    Molnár, B; Papik, K; Schaefer, R; Dombóvári, Z; Fehér, J; Tulassay, Z

    1998-01-01

    The main aim of the research in medical diagnostics is to develop more exact, cost-effective and handsome systems, procedures and methods for supporting the clinicians. In their paper the authors introduce a new method that recently came into the focus referred to as artificial neural networks. Based on the literature of the past 5-6 years they give a brief review--highlighting the most important ones--showing the idea behind neural networks, what they are used for in the medical field. The definition, structure and operation of neural networks are discussed. In the application part they collect examples in order to give an insight in the neural network application research. It is emphasised that in the near future basically new diagnostic equipment can be developed based on this new technology in the field of ECG, EEG and macroscopic and microscopic image analysis systems.

  2. Porosity Log Prediction Using Artificial Neural Network

    NASA Astrophysics Data System (ADS)

    Dwi Saputro, Oki; Lazuardi Maulana, Zulfikar; Dzar Eljabbar Latief, Fourier

    2016-08-01

    Well logging is important in oil and gas exploration. Many physical parameters of reservoir is derived from well logging measurement. Geophysicists often use well logging to obtain reservoir properties such as porosity, water saturation and permeability. Most of the time, the measurement of the reservoir properties are considered expensive. One of method to substitute the measurement is by conducting a prediction using artificial neural network. In this paper, artificial neural network is performed to predict porosity log data from other log data. Three well from ‘yy’ field are used to conduct the prediction experiment. The log data are sonic, gamma ray, and porosity log. One of three well is used as training data for the artificial neural network which employ the Levenberg-Marquardt Backpropagation algorithm. Through several trials, we devise that the most optimal input training is sonic log data and gamma ray log data with 10 hidden layer. The prediction result in well 1 has correlation of 0.92 and mean squared error of 5.67 x10-4. Trained network apply to other well data. The result show that correlation in well 2 and well 3 is 0.872 and 0.9077 respectively. Mean squared error in well 2 and well 3 is 11 x 10-4 and 9.539 x 10-4. From the result we can conclude that sonic log and gamma ray log could be good combination for predicting porosity with neural network.

  3. Second-order neural nets for constrained optimization.

    PubMed

    Zhang, S; Zhu, X; Zou, L H

    1992-01-01

    Analog neural nets for constrained optimization are proposed as an analogue of Newton's algorithm in numerical analysis. The neural model is globally stable and can converge to the constrained stationary points. Nonlinear neurons are introduced into the net, making it possible to solve optimization problems where the variables take discrete values, i.e., combinatorial optimization.

  4. Extraction of shoreline features by neural nets and image processing

    SciTech Connect

    Ryan, T.W.; Sementilli, P.J.; Yuen, P.; Hunt, B.R. )

    1991-07-01

    This paper demonstrates the capability of using neural networks as a tool for delineation of shorelines. The neural nets used are multilayer perceptrons, i.e., feed-forward nets with one or more layers of nodes between the input and output nodes. The back-propagation learning algorithm is used as the adaptation rule. 24 refs.

  5. The ANNIGMA-wrapper approach to fast feature selection for neural nets.

    PubMed

    Hsu, Chun-Nan; Huang, Hung-Ju; Dietrich, S

    2002-01-01

    This paper presents a novel feature selection approach for backpropagation neural networks (NNs). Previously, a feature selection technique known as the wrapper model was shown effective for decision trees induction. However, it is prohibitively expensive when applied to real-world neural net training characterized by large volumes of data and many feature choices. Our approach incorporates a weight analysis-based heuristic called artificial neural net input gain measurement approximation (ANNIGMA) to direct the search in the wrapper model and allows effective feature selection feasible for neural net applications. Experimental results on standard datasets show that this approach can efficiently reduce the number of features while maintaining or even improving the accuracy. We also report two successful applications of our approach in the helicopter maintenance applications.

  6. Catheter-manometer system damped blood pressures detected by neural nets.

    PubMed

    Prentza, A; Wesseling, K H

    1995-07-01

    Degraded catheter-manometer systems cause distortion of blood pressure waveforms, often leading to erroneously resonant or damped waveforms, requiring waveforms quality control. We have tried multilayer perceptron back-propagation trained neural nets of varying architecture to detect damping on sets of normal and artificially damped brachial arterial pressure waves. A second-order digital simulation of a catheter-manometer system is used to cause waveform distortion. Each beat in the waveforms is represented by an 11 parameter input vector. From a group of normotensive or (borderline) hypertensive subjects, pressure waves are used to statistically test and train the neural nets. For each patient and category 5-10 waves are available. The best neural nets correctly classify about 75-85% of the individual beats as either adequate or damped. Using a single majority vote classification per subject per damped or adequate situation, the best neural nets correctly classify at least 16 of the 18 situations in nine test subjects (binomial P = 0.001). More importantly, these neural nets can always detect damping before clinically relevant parameters such as systolic pressure and computed stroke volume are reduced by more than 2%. Neural nets seem remarkably well adapted to solving such subtle problems as detecting a slight damping of arterial pressure waves before it affects waveforms to a clinically relevant degree.

  7. Class of continuous level associative memory neural nets.

    PubMed

    Marks Ii, R J

    1987-05-15

    A neural net capable of restoring continuous level library vectors from memory is considered. As with Hopfield's neural net content addressable memory, the vectors in the memory library are used to program the neural interconnects. Given a portion of one of the library vectors, the net extrapolates the remainder. Sufficient conditions for convergence are stated. Effects of processor inexactitude and net faults are discussed. A more efficient computational technique for performing the memory extrapolation (at the cost of fault tolerance) is derived. The special case of table lookup memories is addressed specifically.

  8. Complete and partial fault tolerance of feedforward neural nets.

    PubMed

    Phatak, D S; Koren, I

    1995-01-01

    A method is proposed to estimate the fault tolerance (FT) of feedforward artificial neural nets (ANNs) and synthesize robust nets. The fault model abstracts a variety of failure modes for permanent stuck-at type faults. A procedure is developed to build FT ANNs by replicating the hidden units. It exploits the intrinsic weighted summation operation performed by the processing units to overcome faults. Metrics are devised to quantify the FT as a function of redundancy. A lower bound on the redundancy required to tolerate all possible single faults is analytically derived. Less than triple modular redundancy (TMR) cannot provide complete FT for all possible single faults. The actual redundancy needed to synthesize a completely FT net is specific to the problem at hand and is usually much higher than that dictated by the general lower bound. The conventional TMR scheme of triplication and majority voting is the best way to achieve complete FT in most ANNs. Although the redundancy needed for complete FT is substantial, the ANNs exhibit good partial FT to begin with and degrade gracefully. The first replication yields maximum enhancement in partial FT compared with later successive replications. For large nets, exhaustive testing of all possible single faults is prohibitive, so the strategy of randomly testing a small fraction of the total number of links is adopted. It yields partial FT estimates that are very close to those obtained by exhaustive testing. When the fraction of links tested is held fixed, the accuracy of the estimate generated by random testing is seen to improve as the net size grows.

  9. Development of programmable artificial neural networks

    NASA Technical Reports Server (NTRS)

    Meade, Andrew J.

    1993-01-01

    Conventionally programmed digital computers can process numbers with great speed and precision, but do not easily recognize patterns or imprecise or contradictory data. Instead of being programmed in the conventional sense, artificial neural networks are capable of self-learning through exposure to repeated examples. However, the training of an ANN can be a time consuming and unpredictable process. A general method is being developed to mate the adaptability of the ANN with the speed and precision of the digital computer. This method was successful in building feedforward networks that can approximate functions and their partial derivatives from examples in a single iteration. The general method also allows the formation of feedforward networks that can approximate the solution to nonlinear ordinary and partial differential equations to desired accuracy without the need of examples. It is believed that continued research will produce artificial neural networks that can be used with confidence in practical scientific computing and engineering applications.

  10. Artificial neural network cardiopulmonary modeling and diagnosis

    DOEpatents

    Kangas, L.J.; Keller, P.E.

    1997-10-28

    The present invention is a method of diagnosing a cardiopulmonary condition in an individual by comparing data from a progressive multi-stage test for the individual to a non-linear multi-variate model, preferably a recurrent artificial neural network having sensor fusion. The present invention relies on a cardiovascular model developed from physiological measurements of an individual. Any differences between the modeled parameters and the parameters of an individual at a given time are used for diagnosis. 12 figs.

  11. Artificial neural network cardiopulmonary modeling and diagnosis

    DOEpatents

    Kangas, Lars J.; Keller, Paul E.

    1997-01-01

    The present invention is a method of diagnosing a cardiopulmonary condition in an individual by comparing data from a progressive multi-stage test for the individual to a non-linear multi-variate model, preferably a recurrent artificial neural network having sensor fusion. The present invention relies on a cardiovascular model developed from physiological measurements of an individual. Any differences between the modeled parameters and the parameters of an individual at a given time are used for diagnosis.

  12. Evaluating atmospheric CO2 effects on gross primary productivity and net ecosystem exchanges of terrestrial ecosystems in the conterminous United States using the AmeriFlux data and an artificial neural network approach

    DOE PAGESBeta

    Liu, Shaoqing; Zhuang, Qianlai; He, Yujie; Noormets, Asko; Chen, Jiquan; Gu, Lianhong

    2016-01-21

    Quantitative understanding of regional gross primary productivity (GPP) and net ecosystem exchanges (NEE) and their responses to environmental changes are critical to quantifying the feedbacks of ecosystems to the global climate system. Numerous studies have used the eddy flux data to upscale the eddy covariance derived carbon fluxes from stand scales to regional and global scales. However, few studies incorporated atmospheric carbon dioxide (CO2) concentrations into those extrapolations. In this study, we consider the effect of atmospheric CO2 using an artificial neural network (ANN) approach to upscale the AmeriFlux tower of NEE and the derived GPP to the conterminous Unitedmore » States. Two ANN models incorporating remote sensing variables at an 8-day time step were developed. One included CO2 as an explanatory variable and the other did not. The models were first trained, validated using eddy flux data, and then extrapolated to the region at a 0.05° × 0.05° (latitude × longitude) resolution from 2001 to 2006. We found that both models performed well in simulating site-level carbon fluxes. The spatially-averaged annual GPP with and without considering the atmospheric CO2 were 789 and 788 g C m-2 yr-1, respectively (for NEE, the values were -112 and -109 g C m-2 yr-1, respectively). Model predictions were comparable with previous published results and MODIS GPP products. However, the difference in GPP between the two models exhibited a great spatial and seasonal variability, with an annual difference of 200 g C m-2 yr-1. Further analysis suggested that air temperature played an important role in determining the atmospheric CO2 effects on carbon fluxes. In addition, the simulation that did not consider atmospheric CO2 failed to detect ecosystem responses to droughts in part of the US in 2006. In conclusion, we suggest that the spatially and temporally varied atmospheric CO2 concentrations should be factored into carbon quantification when scaling eddy flux

  13. Evaluating atmospheric CO2 effects on gross primary productivity and net ecosystem exchanges of terrestrial ecosystems in the conterminous United States using the AmeriFlux data and an artificial neural network approach

    SciTech Connect

    Liu, Shaoqing; Zhuang, Qianlai; He, Yujie; Noormets, Asko; Chen, Jiquan; Gu, Lianhong

    2016-01-01

    Quantitative understanding of regional gross primary productivity (GPP) and net ecosystem exchanges (NEE) and their responses to environmental changes are critical to quantifying the feedbacks of ecosystems to the global climate system. Numerous studies have used the eddy flux data to upscale the eddy covariance derived carbon fluxes from stand scales to regional and global scales. However, few studies incorporated atmospheric carbon dioxide (CO2) concentrations into those extrapolations. Here, we consider the effect of atmospheric CO2 using an artificial neural network (ANN) approach to upscale the AmeriFlux tower of NEE and the derived GPP to the conterminous United States. Two ANN models incorporating remote sensing variables at an 8-day time step were developed. One included CO2 as an explanatory variable and the other did not. The models were first trained, validated using eddy flux data, and then extrapolated to the region at a 0.05 degrees x 0.05 degrees (latitude x longitude) resolution from 2001 to 2006. We found that both models performed well in simulating site-level carbon fluxes. The spatially averaged annual GPP with and without considering the atmospheric CO2 were 789 and 788 g Cm-2 yr(-1), respectively (for NEE, the values were 112 and 109 g Cm-2 yr(-1), respectively). Model predictions were comparable with previous published results and MODIS GPP products. However, the difference in GPP between the two models exhibited a great spatial and seasonal variability, with an annual difference of 200 g Cm-2 yr(-1). Further analysis suggested that air temperature played an important role in determining the atmospheric CO2 effects on carbon fluxes. In addition, the simulation that did not consider atmospheric CO2 failed to detect ecosystem responses to droughts in part of the US in 2006. The study suggests that the spatially and temporally varied atmospheric CO2 concentrations should be factored into carbon quantification when scaling eddy flux data to a

  14. Predicting lithologic parameters using artificial neural networks

    SciTech Connect

    Link, C.A.; Wideman, C.J.; Hanneman, D.L.

    1995-06-01

    Artificial neural networks (ANNs) are becoming increasingly popular as a method for parameter classification and as a tool for recognizing complex relationships in a variety of data types. The power of ANNs lies in their ability to {open_quotes}learn{close_quotes} from a set of training data and then being able to {open_quotes}generalize{close_quotes} to new data sets. In addition, ANNs are able to incorporate data over a large range of scales and are robust in the presence of noise. A back propagation artificial neural network has proved to be a useful tool for predicting sequence boundaries from well logs in a Cenozoic basin. The network was trained using the following log set: neutron porosity, bulk density, pef, and interpreted paleosol horizons from a well in the Deer Lodge Valley, southwestern Montana. After successful training, this network was applied to the same set of well logs from a nearby well minus the interpreted paleosol horizons. The trained neural network was able to produce reasonable predictions for paleosol sequence boundaries in the test well based on the previous training. In an ongoing oil reservoir characterization project, a back propagation neural network is being used to produce estimates of porosity and permeability for subsequent input into a reservoir simulator. A combination of core, well log, geological, and 3-D seismic data serves as input to a back propagation network which outputs estimates of the spatial distribution of porosity and permeability away from the well.

  15. Scaling Properties of Topological Neural Nets

    NASA Astrophysics Data System (ADS)

    Hubler, Alfred; Jun, Joseph

    2006-03-01

    We study the agglomeration of metallic particles in an electric field. Earlier it has been shown that this system is a hardware implementation of a neural net [1]. In this paper we study the growth and topological properties of the emerging networks. In contrast to other networks the conductivity of the connections has a fixed value, but the completeness and number of connections depends on the training patterns. We find that the patterns grow in three stages: growth of shooters, ramification, and expansion [2]. The emerging patterns are hierarchical. For the limiting patterns certain properties are highly reproducible, such as the number of end points and the number of branching points, while other properties are not well reproducible, such as the number of tree structures. Further there are power law relations between the mass and the number of branching points and the number of end points. [1] M. Sperl, A. Chang, N. Weber, and A. Hubler, Hebbian Learning in the Agglomeration of Conducting Particles, Phys.Rev.E. 59, 3165-3168 (1999). [2] J. K. Jun and A. Hubler, Formation and structure of ramified charge transportation networks in an electromechanical system, PNAS 102, 536--540 (2005).

  16. Real-time applications of neural nets

    SciTech Connect

    Spencer, J.E.

    1989-05-01

    Producing, accelerating and colliding very high power, low emittance beams for long periods is a formidable problem in real-time control. As energy has grown exponentially in time so has the complexity of the machines and their control systems. Similar growth rates have occurred in many areas, e.g., improved integrated circuits have been paid for with comparable increases in complexity. However, in this case, reliability, capability and cost have improved due to reduced size, high production and increased integration which allow various kinds of feedback. In contrast, most large complex systems (LCS) are perceived to lack such possibilities because only one copy is made. Neural nets, as a metaphor for LCS, suggest ways to circumvent such limitations. It is argued that they are logically equivalent to multi-loop feedback/forward control of faulty systems. While complimentary to AI, they mesh nicely with characteristics desired for real-time systems. Such issues are considered, examples given and possibilities discussed. 21 refs., 6 figs.

  17. Informational properties of neural nets performing algorithmic and logical tasks.

    PubMed

    Ritz, B M; Hofacker, G L

    1996-06-01

    It is argued that the genetic information necessary to encode an algorithmic neural processor tutoring an otherwise randomly connected biological neural net is represented by the entropy of the analogous minimal Turing machine. Such a near-minimal machine is constructed performing the whole range of bivalent propositional logic in n variables. Neural nets computing the same task are presented; their informational entropy can be gauged with reference to the analogous Turing machine. It is also shown that nets with one hidden layer can be trained to perform algorithms solving propositional logic by error back-propagation. PMID:8672562

  18. Informational properties of neural nets performing algorithmic and logical tasks.

    PubMed

    Ritz, B M; Hofacker, G L

    1996-06-01

    It is argued that the genetic information necessary to encode an algorithmic neural processor tutoring an otherwise randomly connected biological neural net is represented by the entropy of the analogous minimal Turing machine. Such a near-minimal machine is constructed performing the whole range of bivalent propositional logic in n variables. Neural nets computing the same task are presented; their informational entropy can be gauged with reference to the analogous Turing machine. It is also shown that nets with one hidden layer can be trained to perform algorithms solving propositional logic by error back-propagation.

  19. Artificial neural networks for classifying olfactory signals.

    PubMed

    Linder, R; Pöppl, S J

    2000-01-01

    For practical applications, artificial neural networks have to meet several requirements: Mainly they should learn quick, classify accurate and behave robust. Programs should be user-friendly and should not need the presence of an expert for fine tuning diverse learning parameters. The present paper demonstrates an approach using an oversized network topology, adaptive propagation (APROP), a modified error function, and averaging outputs of four networks described for the first time. As an example, signals from different semiconductor gas sensors of an electronic nose were classified. The electronic nose smelt different types of edible oil with extremely different a-priori-probabilities. The fully-specified neural network classifier fulfilled the above mentioned demands. The new approach will be helpful not only for classifying olfactory signals automatically but also in many other fields in medicine, e.g. in data mining from medical databases.

  20. Synchronous machine steady-state stability analysis using an artificial neural network

    SciTech Connect

    Chen, C.R.; Hsu, Y.Y. . Dept. of Electrical Engineering)

    1991-03-01

    A new type of artificial neural network is proposed for the steady-state stability analysis of a synchronous generator. In the developed artificial neutral network, those system variables which play an important role in steady-state stability such as generator outputs and power system stabilizer parameters are employed as the inputs. The output of the neural net provides the information on steady-state stability. Once the connection weights of the neural network have been learned using a set of training data derived off-line, the neural net can be applied to analyze the steady-state stability of the system time. To demonstrate the effectiveness of the proposed neural net, steady-state stability analysis is performed on a synchronous generator connected to a large power system. It is found that the proposed neural net requires much less training time than the multilayer feedforward network with backpropagation-momentum learning algorithm. It is also concluded from the test results that correct stability assessment can be achieved by the neural network.

  1. Digital Image Compression Using Artificial Neural Networks

    NASA Technical Reports Server (NTRS)

    Serra-Ricart, M.; Garrido, L.; Gaitan, V.; Aloy, A.

    1993-01-01

    The problem of storing, transmitting, and manipulating digital images is considered. Because of the file sizes involved, large amounts of digitized image information are becoming common in modern projects. Our goal is to described an image compression transform coder based on artificial neural networks techniques (NNCTC). A comparison of the compression results obtained from digital astronomical images by the NNCTC and the method used in the compression of the digitized sky survey from the Space Telescope Science Institute based on the H-transform is performed in order to assess the reliability of the NNCTC.

  2. Higher-order artificial neural networks

    SciTech Connect

    Bengtsson, M.

    1990-12-01

    The report investigates the storage capacity of an artificial neural network where the state of each neuron depends on quadratic correlations of all other neurons, i.e. a third order network. This is in contrast to a standard Hopfield network where the state of each single neuron depends on the state on every other neuron, without any correlations. The storage capacity of a third order network is larger than that for standard Hopfield by one order of N. However, the number of connections is also larger by an order of N. It is shown that the storage capacity per connection is identical for standard Hopfield and for this third order network.

  3. Examples of Current and Future Uses of Neural-Net Image Processing for Aerospace Applications

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J.

    2004-01-01

    Feed forward artificial neural networks are very convenient for performing correlated interpolation of pairs of complex noisy data sets as well as detecting small changes in image data. Image-to-image, image-to-variable and image-to-index applications have been tested at Glenn. Early demonstration applications are summarized including image-directed alignment of optics, tomography, flow-visualization control of wind-tunnel operations and structural-model-trained neural networks. A practical application is reviewed that employs neural-net detection of structural damage from interference fringe patterns. Both sensor-based and optics-only calibration procedures are available for this technique. These accomplishments have generated the knowledge necessary to suggest some other applications for NASA and Government programs. A tomography application is discussed to support Glenn's Icing Research tomography effort. The self-regularizing capability of a neural net is shown to predict the expected performance of the tomography geometry and to augment fast data processing. Other potential applications involve the quantum technologies. It may be possible to use a neural net as an image-to-image controller of an optical tweezers being used for diagnostics of isolated nano structures. The image-to-image transformation properties also offer the potential for simulating quantum computing. Computer resources are detailed for implementing the black box calibration features of the neural nets.

  4. Molnets: An Artificial Chemistry Based on Neural Networks

    NASA Technical Reports Server (NTRS)

    Colombano, Silvano; Luk, Johnny; Segovia-Juarez, Jose L.; Lohn, Jason; Clancy, Daniel (Technical Monitor)

    2002-01-01

    The fundamental problem in the evolution of matter is to understand how structure-function relationships are formed and increase in complexity from the molecular level all the way to a genetic system. We have created a system where structure-function relationships arise naturally and without the need of ad hoc function assignments to given structures. The idea was inspired by neural networks, where the structure of the net embodies specific computational properties. In this system networks interact with other networks to create connections between the inputs of one net and the outputs of another. The newly created net then recomputes its own synaptic weights, based on anti-hebbian rules. As a result some connections may be cut, and multiple nets can emerge as products of a 'reaction'. The idea is to study emergent reaction behaviors, based on simple rules that constitute a pseudophysics of the system. These simple rules are parameterized to produce behaviors that emulate chemical reactions. We find that these simple rules show a gradual increase in the size and complexity of molecules. We have been building a virtual artificial chemistry laboratory for discovering interesting reactions and for testing further ideas on the evolution of primitive molecules. Some of these ideas include the potential effect of membranes and selective diffusion according to molecular size.

  5. Identification and interpretation of patterns in rocket engine data: Artificial intelligence and neural network approaches

    NASA Technical Reports Server (NTRS)

    Ali, Moonis; Whitehead, Bruce; Gupta, Uday K.; Ferber, Harry

    1995-01-01

    This paper describes an expert system which is designed to perform automatic data analysis, identify anomalous events and determine the characteristic features of these events. We have employed both artificial intelligence and neural net approaches in the design of this expert system.

  6. Use of artifical neural nets to predict permeability in Hugoton Field

    SciTech Connect

    Thompson, K.A.; Franklin, M.H.; Olson, T.M.

    1996-12-31

    One of the most difficult tasks in petrophysics is establishing a quantitative relationship between core permeability and wireline logs. This is a tough problem in Hugoton Field, where a complicated mix of carbonates and clastics further obscure the correlation. One can successfully model complex relationships such as permeability-to-logs using artificial neural networks. Mind and Vision, Inc.`s neural net software was used because of its orientation toward depth-related data (such as logs) and its ability to run on a variety of log analysis platforms. This type of neural net program allows the expert geologist to select a few (10-100) points of control to train the {open_quotes}brainstate{close_quotes} using logs as predicters and core permeability as {open_quotes}truth{close_quotes}. In Hugoton Field, the brainstate provides an estimate of permeability at each depth in 474 logged wells. These neural net-derived permeabilities are being used in reservoir characterization models for fluid saturations. Other applications of this artificial neural network technique include deterministic relationships of logs to: core lithology, core porosity, pore type, and other wireline logs (e.g., predicting a sonic log from a density log).

  7. Use of artifical neural nets to predict permeability in Hugoton Field

    SciTech Connect

    Thompson, K.A.; Franklin, M.H.; Olson, T.M. )

    1996-01-01

    One of the most difficult tasks in petrophysics is establishing a quantitative relationship between core permeability and wireline logs. This is a tough problem in Hugoton Field, where a complicated mix of carbonates and clastics further obscure the correlation. One can successfully model complex relationships such as permeability-to-logs using artificial neural networks. Mind and Vision, Inc.'s neural net software was used because of its orientation toward depth-related data (such as logs) and its ability to run on a variety of log analysis platforms. This type of neural net program allows the expert geologist to select a few (10-100) points of control to train the [open quotes]brainstate[close quotes] using logs as predicters and core permeability as [open quotes]truth[close quotes]. In Hugoton Field, the brainstate provides an estimate of permeability at each depth in 474 logged wells. These neural net-derived permeabilities are being used in reservoir characterization models for fluid saturations. Other applications of this artificial neural network technique include deterministic relationships of logs to: core lithology, core porosity, pore type, and other wireline logs (e.g., predicting a sonic log from a density log).

  8. Context-dependent neural nets--structures and learning.

    PubMed

    Ciskowski, Piotr; Rafajłowicz, Ewaryst

    2004-11-01

    A novel approach toward neural networks modeling is presented in the paper. It is unique in the fact that allows nets' weights to change according to changes of some environmental factors even after completing the learning process. The models of context-dependent (cd) neuron, one- and multilayer feedforward net are presented, with basic learning algorithms and examples of functioning. The Vapnik-Chervonenkis (VC) dimension of a cd neuron is derived, as well as VC dimension of multilayer feedforward nets. Cd nets' properties are discussed and compared with the properties of traditional nets. Possibilities of applications to classification and control problems are also outlined and an example presented.

  9. Automation of Some Operations of a Wind Tunnel Using Artificial Neural Networks

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J.; Buggele, Alvin E.

    1996-01-01

    Artificial neural networks were used successfully to sequence operations in a small, recently modernized, supersonic wind tunnel at NASA-Lewis Research Center. The neural nets generated correct estimates of shadowgraph patterns, pressure sensor readings and mach numbers for conditions occurring shortly after startup and extending to fully developed flow. Artificial neural networks were trained and tested for estimating: sensor readings from shadowgraph patterns, shadowgraph patterns from shadowgraph patterns and sensor readings from sensor readings. The 3.81 by 10 in. (0.0968 by 0.254 m) tunnel was operated with its mach 2.0 nozzle, and shadowgraph was recorded near the nozzle exit. These results support the thesis that artificial neural networks can be combined with current workstation technology to automate wind tunnel operations.

  10. Functional expansion representations of artificial neural networks

    NASA Technical Reports Server (NTRS)

    Gray, W. Steven

    1992-01-01

    In the past few years, significant interest has developed in using artificial neural networks to model and control nonlinear dynamical systems. While there exists many proposed schemes for accomplishing this and a wealth of supporting empirical results, most approaches to date tend to be ad hoc in nature and rely mainly on heuristic justifications. The purpose of this project was to further develop some analytical tools for representing nonlinear discrete-time input-output systems, which when applied to neural networks would give insight on architecture selection, pruning strategies, and learning algorithms. A long term goal is to determine in what sense, if any, a neural network can be used as a universal approximator for nonliner input-output maps with memory (i.e., realized by a dynamical system). This property is well known for the case of static or memoryless input-output maps. The general architecture under consideration in this project was a single-input, single-output recurrent feedforward network.

  11. Artificial neural networks in predicting current in electric arc furnaces

    NASA Astrophysics Data System (ADS)

    Panoiu, M.; Panoiu, C.; Iordan, A.; Ghiormez, L.

    2014-03-01

    The paper presents a study of the possibility of using artificial neural networks for the prediction of the current and the voltage of Electric Arc Furnaces. Multi-layer perceptron and radial based functions Artificial Neural Networks implemented in Matlab were used. The study is based on measured data items from an Electric Arc Furnace in an industrial plant in Romania.

  12. NETS

    NASA Technical Reports Server (NTRS)

    Baffes, Paul T.

    1993-01-01

    NETS development tool provides environment for simulation and development of neural networks - computer programs that "learn" from experience. Written in ANSI standard C, program allows user to generate C code for implementation of neural network.

  13. A neural net approach to space vehicle guidance

    NASA Technical Reports Server (NTRS)

    Caglayan, Alper K.; Allen, Scott M.

    1990-01-01

    The space vehicle guidance problem is formulated using a neural network approach, and the appropriate neural net architecture for modeling optimum guidance trajectories is investigated. In particular, an investigation is made of the incorporation of prior knowledge about the characteristics of the optimal guidance solution into the neural network architecture. The online classification performance of the developed network is demonstrated using a synthesized network trained with a database of optimum guidance trajectories. Such a neural-network-based guidance approach can readily adapt to environment uncertainties such as those encountered by an AOTV during atmospheric maneuvers.

  14. Artificial neural network for multifunctional areas.

    PubMed

    Riccioli, Francesco; El Asmar, Toufic; El Asmar, Jean-Pierre; Fagarazzi, Claudio; Casini, Leonardo

    2016-01-01

    The issues related to the appropriate planning of the territory are particularly pronounced in highly inhabited areas (urban areas), where in addition to protecting the environment, it is important to consider an anthropogenic (urban) development placed in the context of sustainable growth. This work aims at mathematically simulating the changes in the land use, by implementing an artificial neural network (ANN) model. More specifically, it will analyze how the increase of urban areas will develop and whether this development would impact on areas with particular socioeconomic and environmental value, defined as multifunctional areas. The simulation is applied to the Chianti Area, located in the province of Florence, in Italy. Chianti is an area with a unique landscape, and its territorial planning requires a careful examination of the territory in which it is inserted. PMID:26718948

  15. A new approach to artificial neural networks.

    PubMed

    Baptista Filho, B D; Cabral, E L; Soares, A J

    1998-01-01

    A novel approach to artificial neural networks is presented. The philosophy of this approach is based on two aspects: the design of task-specific networks, and a new neuron model with multiple synapses. The synapses' connective strengths are modified through selective and cumulative processes conducted by axo-axonic connections from a feedforward circuit. This new concept was applied to the position control of a planar two-link manipulator exhibiting excellent results on learning capability and generalization when compared with a conventional feedforward network. In the present paper, the example shows only a network developed from a neuronal reflexive circuit with some useful artifices, nevertheless without the intention of covering all possibilities devised.

  16. Dynamic artificial neural networks with affective systems.

    PubMed

    Schuman, Catherine D; Birdwell, J Douglas

    2013-01-01

    Artificial neural networks (ANNs) are processors that are trained to perform particular tasks. We couple a computational ANN with a simulated affective system in order to explore the interaction between the two. In particular, we design a simple affective system that adjusts the threshold values in the neurons of our ANN. The aim of this paper is to demonstrate that this simple affective system can control the firing rate of the ensemble of neurons in the ANN, as well as to explore the coupling between the affective system and the processes of long term potentiation (LTP) and long term depression (LTD), and the effect of the parameters of the affective system on its performance. We apply our networks with affective systems to a simple pole balancing example and briefly discuss the effect of affective systems on network performance.

  17. Galaxies, human eyes, and artificial neural networks.

    PubMed

    Lahav, O; Naim, A; Buta, R J; Corwin, H G; de Vaucouleurs, G; Dressler, A; Huchra, J P; van den Bergh, S; Raychaudhury, S; Sodré, L; Storrie-Lombardi, M C

    1995-02-10

    The quantitative morphological classification of galaxies is important for understanding the origin of type frequency and correlations with environment. However, galaxy morphological classification is still mainly done visually by dedicated individuals, in the spirit of Hubble's original scheme and its modifications. The rapid increase in data on galaxy images at low and high redshift calls for a re-examination of the classification schemes and for automatic methods. Here are shown results from a systematic comparison of the dispersion among human experts classifying a uniformly selected sample of more than 800 digitized galaxy images. These galaxy images were then classified by six of the authors independently. The human classifications are compared with each other and with an automatic classification by an artificial neural network, which replicates the classification by a human expert to the same degree of agreement as that between two human experts. PMID:17813914

  18. Artificial neural network circuits with Josephson devices

    SciTech Connect

    Harada, Y.; Goto, E. )

    1991-03-01

    This article describes a new approach of Josephson devices for computer applications. With an artificial neural network scheme Josephson devices is expected to develop a new paradigm for future computer systems. Here the authors discuss circuit configuration for a neuron with Josephson devices. The authors proposed a combination of a variable bias source and Josephson devices for a synapse circuit. The bias source signal is steered by the Josephson device input signal and becomes the synapse output signal. These output signals are summed up at the specific resistor or inductor to produce the weighted sum of Josephson devices input signals. According to the error signal, the bias source value is corrected. This corresponds to the learning procedure.

  19. Artificial neural network for multifunctional areas.

    PubMed

    Riccioli, Francesco; El Asmar, Toufic; El Asmar, Jean-Pierre; Fagarazzi, Claudio; Casini, Leonardo

    2016-01-01

    The issues related to the appropriate planning of the territory are particularly pronounced in highly inhabited areas (urban areas), where in addition to protecting the environment, it is important to consider an anthropogenic (urban) development placed in the context of sustainable growth. This work aims at mathematically simulating the changes in the land use, by implementing an artificial neural network (ANN) model. More specifically, it will analyze how the increase of urban areas will develop and whether this development would impact on areas with particular socioeconomic and environmental value, defined as multifunctional areas. The simulation is applied to the Chianti Area, located in the province of Florence, in Italy. Chianti is an area with a unique landscape, and its territorial planning requires a careful examination of the territory in which it is inserted.

  20. Use of multilayer feedforward neural nets as a display method for multidimensional distributions.

    PubMed

    Garrido, L; Gaitan, V; Serra-Ricart, M; Calbet, X

    1995-09-01

    We present a new method based on multilayer feedforward neural nets for displaying an n-dimensional distribution in a projected space of 1, 2 or 3 dimensions. A fully nonlinear net with several hidden layers is used. Efficient learning is achieved using multi-seed backpropagation. As a principal component analysis (PCA), the proposed method is useful for extracting information on the structure of the data set, but unlike the PCA, the transformation between the original distribution and the projected one is not restricted to be linear. Artificial examples and a real application are presented in order to show the reliability and potential of the method.

  1. Application of artificial neural networks to the design optimization of aerospace structural components

    NASA Technical Reports Server (NTRS)

    Berke, Laszlo; Patnaik, Surya N.; Murthy, Pappu L. N.

    1993-01-01

    The application of artificial neural networks to capture structural design expertise is demonstrated. The principal advantage of a trained neural network is that it requires trivial computational effort to produce an acceptable new design. For the class of problems addressed, the development of a conventional expert system would be extremely difficult. In the present effort, a structural optimization code with multiple nonlinear programming algorithms and an artificial neural network code NETS were used. A set of optimum designs for a ring and two aircraft wings for static and dynamic constraints were generated by using the optimization codes. The optimum design data were processed to obtain input and output pairs, which were used to develop a trained artificial neural network with the code NETS. Optimum designs for new design conditions were predicted by using the trained network. Neural net prediction of optimum designs was found to be satisfactory for most of the output design parameters. However, results from the present study indicate that caution must be exercised to ensure that all design variables are within selected error bounds.

  2. Spatial predictive mapping using artificial neural networks

    NASA Astrophysics Data System (ADS)

    Noack, S.; Knobloch, A.; Etzold, S. H.; Barth, A.; Kallmeier, E.

    2014-11-01

    The modelling or prediction of complex geospatial phenomena (like formation of geo-hazards) is one of the most important tasks for geoscientists. But in practice it faces various difficulties, caused mainly by the complexity of relationships between the phenomena itself and the controlling parameters, as well by limitations of our knowledge about the nature of physical/ mathematical relationships and by restrictions regarding accuracy and availability of data. In this situation methods of artificial intelligence, like artificial neural networks (ANN) offer a meaningful alternative modelling approach compared to the exact mathematical modelling. In the past, the application of ANN technologies in geosciences was primarily limited due to difficulties to integrate it into geo-data processing algorithms. In consideration of this background, the software advangeo® was developed to provide a normal GIS user with a powerful tool to use ANNs for prediction mapping and data preparation within his standard ESRI ArcGIS environment. In many case studies, such as land use planning, geo-hazards analysis and prevention, mineral potential mapping, agriculture & forestry advangeo® has shown its capabilities and strengths. The approach is able to add considerable value to existing data.

  3. Vibrational Analysis of Engine Components Using Neural-Net Processing and Electronic Holography

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J.; Fite, E. Brian; Mehmed, Oral; Thorp, Scott A.

    1998-01-01

    The use of computational-model trained artificial neural networks to acquire damage specific information from electronic holograms is discussed. A neural network is trained to transform two time-average holograms into a pattern related to the bending-induced-strain distribution of the vibrating component. The bending distribution is very sensitive to component damage unlike the characteristic fringe pattern or the displacement amplitude distribution. The neural network processor is fast for real-time visualization of damage. The two-hologram limit makes the processor more robust to speckle pattern decorrelation. Undamaged and cracked cantilever plates serve as effective objects for testing the combination of electronic holography and neural-net processing. The requirements are discussed for using finite-element-model trained neural networks for field inspections of engine components. The paper specifically discusses neural-network fringe pattern analysis in the presence of the laser speckle effect and the performances of two limiting cases of the neural-net architecture.

  4. Vibrational Analysis of Engine Components Using Neural-Net Processing and Electronic Holography

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J.; Fite, E. Brian; Mehmed, Oral; Thorp, Scott A.

    1997-01-01

    The use of computational-model trained artificial neural networks to acquire damage specific information from electronic holograms is discussed. A neural network is trained to transform two time-average holograms into a pattern related to the bending-induced-strain distribution of the vibrating component. The bending distribution is very sensitive to component damage unlike the characteristic fringe pattern or the displacement amplitude distribution. The neural network processor is fast for real-time visualization of damage. The two-hologram limit makes the processor more robust to speckle pattern decorrelation. Undamaged and cracked cantilever plates serve as effective objects for testing the combination of electronic holography and neural-net processing. The requirements are discussed for using finite-element-model trained neural networks for field inspections of engine components. The paper specifically discusses neural-network fringe pattern analysis in the presence of the laser speckle effect and the performances of two limiting cases of the neural-net architecture.

  5. Stochastic architecture for Hopfield neural nets

    NASA Technical Reports Server (NTRS)

    Pavel, Sandy

    1992-01-01

    An expandable stochastic digital architecture for recurrent (Hopfield like) neural networks is proposed. The main features and basic principles of stochastic processing are presented. The stochastic digital architecture is based on a chip with n full interconnected neurons with a pipeline, bit processing structure. For large applications, a flexible way to interconnect many such chips is provided.

  6. [How can an otolaryngologist benefit from artificial neural networks?].

    PubMed

    Szaleniec, Joanna; Składzień, Jacek; Tadeusiewicz, Ryszard; Oleś, Krzysztof; Konior, Marcin; Przeklasa, Robert

    2012-01-01

    Artificial neural networks are informatic systems that have unique computational capabilities. The principle of their functioning is based on the rules of data processing in the brain. This article discusses the most important features of the artificial neural networks with reference to their applications in otolaryngology. The cited studies concern the fields of rhinology, audiology, phoniatrics, vestibulology, oncology, sleep apnea and salivary gland diseases. The authors also refer to their own experience with predictive neural models designed in the Department of Otolaryngology of the Jagiellonian University Medical College in Krakow. The applications of artificial neural networks in clinical diagnosis, automated signal interpretation and outcome prediction are presented. Moreover, the article explains how the artificial neural networks work and how the otolaryngologists can use them in their clinical practice and research.

  7. Modelling blood-brain barrier partitioning using Bayesian neural nets.

    PubMed

    Winkler, David A; Burden, Frank R

    2004-07-01

    We have employed three families of molecular molecular descriptors, together with Bayesian regularized neural nets, to model the partitioning of a diverse range of drugs and other small molecules across the blood-brain barrier (BBB). The relative efficacy of each descriptors class is compared, and the advantages of flexible, parsimonious, model free mapping methods, like Bayesian neural nets, illustrated. The relative importance of the molecular descriptors for the most predictive BBB model were determined by use of automatic relevance determination (ARD), and compared with the important descriptors from other literature models of BBB partitioning.

  8. A real time neural net estimator of fatigue life

    NASA Technical Reports Server (NTRS)

    Troudet, T.; Merrill, W.

    1990-01-01

    A neural net architecture is proposed to estimate, in real-time, the fatigue life of mechanical components, as part of the Intelligent Control System for Reusable Rocket Engines. Arbitrary component loading values were used as input to train a two hidden-layer feedforward neural net to estimate component fatigue damage. The ability of the net to learn, based on a local strain approach, the mapping between load sequence and fatigue damage has been demonstrated for a uniaxial specimen. Because of its demonstrated performance, the neural computation may be extended to complex cases where the loads are biaxial or triaxial, and the geometry of the component is complex (e.g., turbopump blades). The generality of the approach is such that load/damage mappings can be directly extracted from experimental data without requiring any knowledge of the stress/strain profile of the component. In addition, the parallel network architecture allows real-time life calculations even for high frequency vibrations. Owing to its distributed nature, the neural implementation will be robust and reliable, enabling its use in hostile environments such as rocket engines. This neural net estimator of fatigue life is seen as the enabling technology to achieve component life prognosis, and therefore would be an important part of life extending control for reusable rocket engines.

  9. Optical neural net for classifying imaging spectrometer data

    NASA Technical Reports Server (NTRS)

    Barnard, Etienne; Casasent, David P.

    1989-01-01

    The problem of determining the composition of an unknown input mixture from its measured spectrum, given the spectra of a number of elements, is studied. The Hopfield minimization procedure was used to express the determination of the compositions as a problem suitable for solution by neural nets. A mathematical description of the problem was developed and used as a basis for a neural network solution and an optical implementation.

  10. Artificial Neural Network applied to lightning flashes

    NASA Astrophysics Data System (ADS)

    Gin, R. B.; Guedes, D.; Bianchi, R.

    2013-05-01

    The development of video cameras enabled cientists to study lightning discharges comportment with more precision. The main goal of this project is to create a system able to detect images of lightning discharges stored in videos and classify them using an Artificial Neural Network (ANN)using C Language and OpenCV libraries. The developed system, can be split in two different modules: detection module and classification module. The detection module uses OpenCV`s computer vision libraries and image processing techniques to detect if there are significant differences between frames in a sequence, indicating that something, still not classified, occurred. Whenever there is a significant difference between two consecutive frames, two main algorithms are used to analyze the frame image: brightness and shape algorithms. These algorithms detect both shape and brightness of the event, removing irrelevant events like birds, as well as detecting the relevant events exact position, allowing the system to track it over time. The classification module uses a neural network to classify the relevant events as horizontal or vertical lightning, save the event`s images and calculates his number of discharges. The Neural Network was implemented using the backpropagation algorithm, and was trained with 42 training images , containing 57 lightning events (one image can have more than one lightning). TheANN was tested with one to five hidden layers, with up to 50 neurons each. The best configuration achieved a success rate of 95%, with one layer containing 20 neurons (33 test images with 42 events were used in this phase). This configuration was implemented in the developed system to analyze 20 video files, containing 63 lightning discharges previously manually detected. Results showed that all the lightning discharges were detected, many irrelevant events were unconsidered, and the event's number of discharges was correctly computed. The neural network used in this project achieved a

  11. A Design of Neural-Net Based Decouplers

    NASA Astrophysics Data System (ADS)

    Tokuda, Makoto; Yamamoto, Toru; Monden, Yoshimi

    In process industries such as the chemical plants, a good control performance cannot be obtained by simply using the linear controllers, since most processes are nonlinear multivariable systems with mutual interactions. And now, in various fields, the neural networks are well known as the representative schemes to describe the nonlinear elements included in the systems. Also, many types of neural-net based control systems have been proposed, since they have the ability of function approximation, the training ability and versatility. However, the neural networks tend to require great deal of training iteration or careful adjustments of user-specified parameters. In this paper, a design method of neural-net based decouplers is proposed for nonlinear multivariable systems. Here, the decoupler is generated by the sum of a static decoupler and a neural-net based decoupler. The former is used so that the influence of mutual interactions is roughly removed, and the latter plays a role of compensating the nonlinearities and decoupling the remaining mutual interactions. Thus, by designing the control system as the hybrid system, the burden in training the neural networks can be considerably reduced. Finally, the effectiveness of the proposed control scheme is evaluated on a simulation example.

  12. Programmable synaptic devices for electronic neural nets

    NASA Technical Reports Server (NTRS)

    Moopenn, A.; Thakoor, A. P.

    1990-01-01

    The architecture, design, and operational characteristics of custom VLSI and thin film synaptic devices are described. The devices include CMOS-based synaptic chips containing 1024 reprogrammable synapses with a 6-bit dynamic range, and nonvolatile, write-once, binary synaptic arrays based on memory switching in hydrogenated amorphous silicon films. Their suitability for embodiment of fully parallel and analog neural hardware is discussed. Specifically, a neural network solution to an assignment problem of combinatorial global optimization, implemented in fully parallel hardware using the synaptic chips, is described. The network's ability to provide optimal and near optimal solutions over a time scale of few neuron time constants has been demonstrated and suggests a speedup improvement of several orders of magnitude over conventional search methods.

  13. Toward Real Time Neural Net Flight Controllers

    NASA Technical Reports Server (NTRS)

    Jorgensen, C. C.; Mah, R. W.; Ross, J.; Lu, Henry, Jr. (Technical Monitor)

    1994-01-01

    NASA Ames Research Center has an ongoing program in neural network control technology targeted toward real time flight demonstrations using a modified F-15 which permits direct inner loop control of actuators, rapid switching between alternative control designs, and substitutable processors. An important part of this program is the ACTIVE flight project which is examining the feasibility of using neural networks in the design, control, and system identification of new aircraft prototypes. This paper discusses two research applications initiated with this objective in mind: utilization of neural networks for wind tunnel aircraft model identification and rapid learning algorithms for on line reconfiguration and control. The first application involves the identification of aerodynamic flight characteristics from analysis of wind tunnel test data. This identification is important in the early stages of aircraft design because complete specification of control architecture's may not be possible even though concept models at varying scales are available for aerodynamic wind tunnel testing. Testing of this type is often a long and expensive process involving measurement of aircraft lift, drag, and moment of inertia at varying angles of attack and control surface configurations. This information in turn can be used in the design of the flight control systems by applying the derived lookup tables to generate piece wise linearized controllers. Thus, reduced costs in tunnel test times and the rapid transfer of wind tunnel insights into prototype controllers becomes an important factor in more efficient generation and testing of new flight systems. NASA Ames Research Center is successfully applying modular neural networks as one way of anticipating small scale aircraft model performances prior to testing, thus reducing the number of in tunnel test hours and potentially, the number of intermediate scaled models required for estimation of surface flow effects.

  14. Load distribution of articular cartilage from MR-images by neural nets.

    PubMed

    Seidel, Peter; Hanke, Göran; Gründer, Wilfried

    2005-01-01

    Artificial neural nets were used to determine the Young's modulus and spatial load distribution in articular cartilage by means of T2-weighted MR imaging. MR images were obtained in vitro (ex vivo?) from the joints of sheep of different ages (3 months, 9 months, 15 months, 1.5 years, 5 years, 5.5 years) and pigs (4 and 6 months) with a Bruker AMX 300 (7 T) spectrometer equipped with a micro-imaging unit. The knee of a 29-year-old male volunteer was studied in vivo under mechanical load using a clinical Siemens Vision MRT (1.5 T). The load of the cartilage is understood as a non-linear image transformation of loaded versus unloaded images. The artificial neural net was used to recognize given reference pixels of the unloaded cartilage within the image of the loaded cartilage. The Young's modulus was calculated from the local strain and the external pressure using the Hooke's law. With this method, the average Young's modulus was obtained in relationship to the biological age of the cartilage. The investigated age interval showed a progressive increase of 0.5 +/- 0.3 MPa per year. These results are consistent with published results. As shown in this pilot study, the method of neural nets allows the visualization of the spatial load distribution within the articular cartilage.

  15. Forecasting Zakat collection using artificial neural network

    NASA Astrophysics Data System (ADS)

    Sy Ahmad Ubaidillah, Sh. Hafizah; Sallehuddin, Roselina

    2013-04-01

    'Zakat', "that which purifies" or "alms", is the giving of a fixed portion of one's wealth to charity, generally to the poor and needy. It is one of the five pillars of Islam, and must be paid by all practicing Muslims who have the financial means (nisab). 'Nisab' is the minimum level to determine whether there is a 'zakat' to be paid on the assets. Today, in most Muslim countries, 'zakat' is collected through a decentralized and voluntary system. Under this voluntary system, 'zakat' committees are established, which are tasked with the collection and distribution of 'zakat' funds. 'Zakat' promotes a more equitable redistribution of wealth, and fosters a sense of solidarity amongst members of the 'Ummah'. The Malaysian government has established a 'zakat' center at every state to facilitate the management of 'zakat'. The center has to have a good 'zakat' management system to effectively execute its functions especially in the collection and distribution of 'zakat'. Therefore, a good forecasting model is needed. The purpose of this study is to develop a forecasting model for Pusat Zakat Pahang (PZP) to predict the total amount of collection from 'zakat' of assets more precisely. In this study, two different Artificial Neural Network (ANN) models using two different learning algorithms are developed; Back Propagation (BP) and Levenberg-Marquardt (LM). Both models are developed and compared in terms of their accuracy performance. The best model is determined based on the lowest mean square error and the highest correlations values. Based on the results obtained from the study, BP neural network is recommended as the forecasting model to forecast the collection from 'zakat' of assets for PZP.

  16. Application of artificial neural networks to composite ply micromechanics

    NASA Technical Reports Server (NTRS)

    Brown, D. A.; Murthy, P. L. N.; Berke, L.

    1991-01-01

    Artificial neural networks can provide improved computational efficiency relative to existing methods when an algorithmic description of functional relationships is either totally unavailable or is complex in nature. For complex calculations, significant reductions in elapsed computation time are possible. The primary goal is to demonstrate the applicability of artificial neural networks to composite material characterization. As a test case, a neural network was trained to accurately predict composite hygral, thermal, and mechanical properties when provided with basic information concerning the environment, constituent materials, and component ratios used in the creation of the composite. A brief introduction on neural networks is provided along with a description of the project itself.

  17. Neural nets for adaptive filtering and adaptive pattern recognition

    SciTech Connect

    Widrow, B.; Winter, R.

    1988-03-01

    The fields of adaptive signal processing and adaptive neural networks have been developing independently but have that adaptive linear combiner (ALC) in common. With its inputs connected to a tapped delay line, the ALC becomes a key component of an adaptive filter. With its output connected to a quantizer, the ALC becomes an adaptive threshold element of adaptive neuron. Adaptive threshold elements, on the other hand, are the building blocks of neural networks. Today neural nets are the focus of widespread research interest. Areas of investigation include pattern recognition and trainable logic. Neural network systems have not yet had the commercial impact of adaptive filtering. The commonality of the ALC to adaptive signal processing and adaptive neural networks suggests the two fields have much to share with each other. This article describes practical applications of the ALC in signal processing and pattern recognition.

  18. Detection of Wildfires with Artificial Neural Networks

    NASA Astrophysics Data System (ADS)

    Umphlett, B.; Leeman, J.; Morrissey, M. L.

    2011-12-01

    Currently fire detection for the National Oceanic and Atmospheric Administration (NOAA) using satellite data is accomplished with algorithms and error checking human analysts. Artificial neural networks (ANNs) have been shown to be more accurate than algorithms or statistical methods for applications dealing with multiple datasets of complex observed data in the natural sciences. ANNs also deal well with multiple data sources that are not all equally reliable or equally informative to the problem. An ANN was tested to evaluate its accuracy in detecting wildfires utilizing polar orbiter numerical data from the Advanced Very High Resolution Radiometer (AVHRR). Datasets containing locations of known fires were gathered from the NOAA's polar orbiting satellites via the Comprehensive Large Array-data Stewardship System (CLASS). The data was then calibrated and navigation corrected using the Environment for Visualizing Images (ENVI). Fires were located with the aid of shapefiles generated via ArcGIS. Afterwards, several smaller ten pixel by ten pixel datasets were created for each fire (using the ENVI corrected data). Several datasets were created for each fire in order to vary fire position and avoid training the ANN to look only at fires in the center of an image. Datasets containing no fires were also created. A basic pattern recognition neural network was established with the MATLAB neural network toolbox. The datasets were then randomly separated into categories used to train, validate, and test the ANN. To prevent over fitting of the data, the mean squared error (MSE) of the network was monitored and training was stopped when the MSE began to rise. Networks were tested using each channel of the AVHRR data independently, channels 3a and 3b combined, and all six channels. The number of hidden neurons for each input set was also varied between 5-350 in steps of 5 neurons. Each configuration was run 10 times, totaling about 4,200 individual network evaluations. Thirty

  19. Comparing and Contrasting Neural Net Solutions to Classical Statistical Solutions.

    ERIC Educational Resources Information Center

    Van Nelson, C.; Neff, Kathryn J.

    Data from two studies in which subjects were classified as successful or unsuccessful were analyzed using neural net technology after being analyzed with a linear regression function. Data were obtained from admission records of 201 students admitted to undergraduate and 285 students admitted to graduate programs. Data included grade point…

  20. Neural nets identify sensory receptors from somal spikes.

    PubMed

    Rose, R D; Karnavas, W J

    1993-12-10

    Retrospective analysis of somal electrophysiology from intracellularly recorded, physiologically identified afferents demonstrates that neural nets can be readily trained to identify the type of peripheral receptor supplied. Specifically, cat spinal ganglion somata could be identified as innervating muscle spindles, hairs or high-threshold mechanoreceptors. Further, both hair afferents and high-threshold mechanoreceptors could be separated into three distinct subclasses. The neural net sorting reported here utilizes only the electrophysiological properties of the somata plus conduction velocity and can with this information alone predict the functional properties of the sensory endings. Interestingly, neural net sorting could also distinguish between different types of hair afferents (or nociceptors), even when conduction velocity information was ignored. It is suggested that neural nets, in combination with computer-controlled data-acquisition systems, could greatly increase investigator efficiency and decrease the number of animals needed to demonstrate specific phenomena, such as drug effects on particular cell types. A double-edged sword of increased investigator efficiency and decreased animal usage may be of particular usefulness in the present socio-political research arena.

  1. Neural nets identify sensory receptors from somal spikes.

    PubMed

    Rose, R D; Karnavas, W J

    1993-12-10

    Retrospective analysis of somal electrophysiology from intracellularly recorded, physiologically identified afferents demonstrates that neural nets can be readily trained to identify the type of peripheral receptor supplied. Specifically, cat spinal ganglion somata could be identified as innervating muscle spindles, hairs or high-threshold mechanoreceptors. Further, both hair afferents and high-threshold mechanoreceptors could be separated into three distinct subclasses. The neural net sorting reported here utilizes only the electrophysiological properties of the somata plus conduction velocity and can with this information alone predict the functional properties of the sensory endings. Interestingly, neural net sorting could also distinguish between different types of hair afferents (or nociceptors), even when conduction velocity information was ignored. It is suggested that neural nets, in combination with computer-controlled data-acquisition systems, could greatly increase investigator efficiency and decrease the number of animals needed to demonstrate specific phenomena, such as drug effects on particular cell types. A double-edged sword of increased investigator efficiency and decreased animal usage may be of particular usefulness in the present socio-political research arena. PMID:8118704

  2. Neural net diagnostics for VLSI test

    NASA Technical Reports Server (NTRS)

    Lin, T.; Tseng, H.; Wu, A.; Dogan, N.; Meador, J.

    1990-01-01

    This paper discusses the application of neural network pattern analysis algorithms to the IC fault diagnosis problem. A fault diagnostic is a decision rule combining what is known about an ideal circuit test response with information about how it is distorted by fabrication variations and measurement noise. The rule is used to detect fault existence in fabricated circuits using real test equipment. Traditional statistical techniques may be used to achieve this goal, but they can employ unrealistic a priori assumptions about measurement data. Our approach to this problem employs an adaptive pattern analysis technique based on feedforward neural networks. During training, a feedforward network automatically captures unknown sample distributions. This is important because distributions arising from the nonlinear effects of process variation can be more complex than is typically assumed. A feedforward network is also able to extract measurement features which contribute significantly to making a correct decision. Traditional feature extraction techniques employ matrix manipulations which can be particularly costly for large measurement vectors. In this paper we discuss a software system which we are developing that uses this approach. We also provide a simple example illustrating the use of the technique for fault detection in an operational amplifier.

  3. Groundwater remediation optimization using artificial neural networks

    SciTech Connect

    Rogers, L. L., LLNL

    1998-05-01

    One continuing point of research in optimizing groundwater quality management is reduction of computational burden which is particularly limiting in field-scale applications. Often evaluation of a single pumping strategy, i.e. one call to the groundwater flow and transport model (GFTM) may take several hours on a reasonably fast workstation. For computational flexibility and efficiency, optimal groundwater remediation design at Lawrence Livermore National Laboratory (LLNL) has relied on artificial neural networks (ANNS) trained to approximate the outcome of 2-D field-scale, finite difference/finite element GFTMs. The search itself has been directed primarily by the genetic algorithm (GA) or the simulated annealing (SA) algorithm. This approach has advantages of (1) up to a million fold increase in speed of remediation pattern assessment during the searches and sensitivity analyses for the 2-D LLNL work, (2) freedom from sequential runs of the GFTM (enables workstation farming), and (3) recycling of the knowledge base (i.e. runs of the GFTM necessary to train the ANNS). Reviewed here are the background and motivation for such work, recent applications, and continuing issues of research.

  4. Geophysical phenomena classification by artificial neural networks

    NASA Technical Reports Server (NTRS)

    Gough, M. P.; Bruckner, J. R.

    1995-01-01

    Space science information systems involve accessing vast data bases. There is a need for an automatic process by which properties of the whole data set can be assimilated and presented to the user. Where data are in the form of spectrograms, phenomena can be detected by pattern recognition techniques. Presented are the first results obtained by applying unsupervised Artificial Neural Networks (ANN's) to the classification of magnetospheric wave spectra. The networks used here were a simple unsupervised Hamming network run on a PC and a more sophisticated CALM network run on a Sparc workstation. The ANN's were compared in their geophysical data recognition performance. CALM networks offer such qualities as fast learning, superiority in generalizing, the ability to continuously adapt to changes in the pattern set, and the possibility to modularize the network to allow the inter-relation between phenomena and data sets. This work is the first step toward an information system interface being developed at Sussex, the Whole Information System Expert (WISE). Phenomena in the data are automatically identified and provided to the user in the form of a data occurrence morphology, the Whole Information System Data Occurrence Morphology (WISDOM), along with relationships to other parameters and phenomena.

  5. Automated Wildfire Detection Through Artificial Neural Networks

    NASA Technical Reports Server (NTRS)

    Miller, Jerry; Borne, Kirk; Thomas, Brian; Huang, Zhenping; Chi, Yuechen

    2005-01-01

    We have tested and deployed Artificial Neural Network (ANN) data mining techniques to analyze remotely sensed multi-channel imaging data from MODIS, GOES, and AVHRR. The goal is to train the ANN to learn the signatures of wildfires in remotely sensed data in order to automate the detection process. We train the ANN using the set of human-detected wildfires in the U.S., which are provided by the Hazard Mapping System (HMS) wildfire detection group at NOAA/NESDIS. The ANN is trained to mimic the behavior of fire detection algorithms and the subjective decision- making by N O M HMS Fire Analysts. We use a local extremum search in order to isolate fire pixels, and then we extract a 7x7 pixel array around that location in 3 spectral channels. The corresponding 147 pixel values are used to populate a 147-dimensional input vector that is fed into the ANN. The ANN accuracy is tested and overfitting is avoided by using a subset of the training data that is set aside as a test data set. We have achieved an automated fire detection accuracy of 80-92%, depending on a variety of ANN parameters and for different instrument channels among the 3 satellites. We believe that this system can be deployed worldwide or for any region to detect wildfires automatically in satellite imagery of those regions. These detections can ultimately be used to provide thermal inputs to climate models.

  6. Geophysical phenomena classification by artificial neural networks

    SciTech Connect

    Gough, M.P.; Bruckner, J.R.

    1995-01-01

    Space science information systems involve accessing vast data bases. There is a need for an automatic process by which properties of the whole data set can be assimilated and presented to the user. Where data are in the form of spectrograms, phenomena can be detected by pattern recognition techniques. Presented are the first results obtained by applying unsupervised Artificial Neural Networks (ANN`s) to the classification of magnetospheric wave spectra. The networks used here were a simple unsupervised Hamming network run on a PC and a more sophisticated CALM network run on a Sparc workstation. The ANN`s were compared in their geophysical data recognition performance. CALM networks offer such qualities as fast learning, superiority in generalizing, the ability to continuously adapt to changes in the pattern set, and the possibility to modularize the network to allow the inter-relation between phenomena and data sets. This work is the first step toward an information system interface being developed at Sussex, the Whole Information System Expert (WISE). Phenomena in the data are automatically identified and provided to the user in the form of a data occurrence morphology, the Whole Information System Data Occurrence Morphology (WISDOM), along with relationships to other parameters and phenomena.

  7. DEM interpolation based on artificial neural networks

    NASA Astrophysics Data System (ADS)

    Jiao, Limin; Liu, Yaolin

    2005-10-01

    This paper proposed a systemic resolution scheme of Digital Elevation model (DEM) interpolation based on Artificial Neural Networks (ANNs). In this paper, we employ BP network to fit terrain surface, and then detect and eliminate the samples with gross errors. This paper uses Self-organizing Feature Map (SOFM) to cluster elevation samples. The study area is divided into many more homogenous tiles after clustering. BP model is employed to interpolate DEM in each cluster. Because error samples are eliminated and clusters are built, interpolation result is better. The case study indicates that ANN interpolation scheme is feasible. It also shows that ANN can get a more accurate result by comparing ANN with polynomial and spline interpolation. ANN interpolation doesn't need to determine the interpolation function beforehand, so manmade influence is lessened. The ANN interpolation is more automatic and intelligent. At the end of the paper, we propose the idea of constructing ANN surface model. This model can be used in multi-scale DEM visualization, and DEM generalization, etc.

  8. Doubly stochastic Poisson processes in artificial neural learning.

    PubMed

    Card, H C

    1998-01-01

    This paper investigates neuron activation statistics in artificial neural networks employing stochastic arithmetic. It is shown that a doubly stochastic Poisson process is an appropriate model for the signals in these circuits.

  9. Decision net, directed graph, and neural net processing of imaging spectrometer data

    NASA Technical Reports Server (NTRS)

    Casasent, David; Liu, Shiaw-Dong; Yoneyama, Hideyuki; Barnard, Etienne

    1989-01-01

    A decision-net solution involving a novel hierarchical classifier and a set of multiple directed graphs, as well as a neural-net solution, are respectively presented for large-class problem and mixture problem treatments of imaging spectrometer data. The clustering method for hierarchical classifier design, when used with multiple directed graphs, yields an efficient decision net. New directed-graph rules for reducing local maxima as well as the number of perturbations required, and the new starting-node rules for extending the reachability and reducing the search time of the graphs, are noted to yield superior results, as indicated by an illustrative 500-class imaging spectrometer problem.

  10. Artificial Neural Networks: A New Approach to Predicting Application Behavior.

    ERIC Educational Resources Information Center

    Gonzalez, Julie M. Byers; DesJardins, Stephen L.

    2002-01-01

    Applied the technique of artificial neural networks to predict which students were likely to apply to one research university. Compared the results to the traditional analysis tool, logistic regression modeling. Found that the addition of artificial intelligence models was a useful new tool for predicting student application behavior. (EV)

  11. Study on the identifying of meat's visible spectrum based on BP artificial neural network

    NASA Astrophysics Data System (ADS)

    Li, Xiaotian; Zhang, Tieqiang; Li, Bo; Jiang, Yongheng; Liu, Binghui; Li, Zhaokai

    2006-01-01

    A method to identify different meat by the visible and reflected spectra of meat with BP artificial neural net (BP-ANN) was introduced in this paper. The visible and reflected spectra (from 420 to 535nm) of different meat (beef and pork) were measured with fiber sensor spectrometer. A kind of ANN with a double-hidden layer was created to identify the different meat automatically. Its right ratio reaches 92.71%.

  12. Advances in Artificial Neural Networks - Methodological Development and Application

    Technology Transfer Automated Retrieval System (TEKTRAN)

    Artificial neural networks as a major soft-computing technology have been extensively studied and applied during the last three decades. Research on backpropagation training algorithms for multilayer perceptron networks has spurred development of other neural network training algorithms for other ne...

  13. Multiple image sensor data fusion through artificial neural networks

    Technology Transfer Automated Retrieval System (TEKTRAN)

    With multisensor data fusion technology, the data from multiple sensors are fused in order to make a more accurate estimation of the environment through measurement, processing and analysis. Artificial neural networks are the computational models that mimic biological neural networks. With high per...

  14. Computation and control with neural nets

    SciTech Connect

    Corneliusen, A.; Terdal, P.; Knight, T.; Spencer, J.

    1989-10-04

    As energies have increased exponentially with time so have the size and complexity of accelerators and control systems. NN may offer the kinds of improvements in computation and control that are needed to maintain acceptable functionality. For control their associative characteristics could provide signal conversion or data translation. Because they can do any computation such as least squares, they can close feedback loops autonomously to provide intelligent control at the point of action rather than at a central location that requires transfers, conversions, hand-shaking and other costly repetitions like input protection. Both computation and control can be integrated on a single chip, printed circuit or an optical equivalent that is also inherently faster through full parallel operation. For such reasons one expects lower costs and better results. Such systems could be optimized by integrating sensor and signal processing functions. Distributed nets of such hardware could communicate and provide global monitoring and multiprocessing in various ways e.g. via token, slotted or parallel rings (or Steiner trees) for compatibility with existing systems. Problems and advantages of this approach such as an optimal, real-time Turing machine are discussed. Simple examples are simulated and hardware implemented using discrete elements that demonstrate some basic characteristics of learning and parallelism. Future microprocessors' are predicted and requested on this basis. 19 refs., 18 figs.

  15. A neural net representation of experienced and nonexperienced users during manual wheelchair propulsion.

    PubMed

    Patterson, P; Draper, S

    1998-01-01

    A neural net approach was used to classify and analyze combinations of the physiological and kinematic responses (the factor patterns) of experienced and novice individuals during wheelchair propulsion, and to determine the key characteristics (individual factors) used in making this determination. A sequence of artificial neural networks (ANN) was developed and used to classify differences between eight nonimpaired controls and seven individuals using wheelchairs, who ranged in age from 24 to 36 years. The subjects propelled a wheelchair on a specially constructed dynamometer at three different velocity levels during which stroke pattern, force, energy, and efficiency data were collected. The data from 10 subjects (5 from each group) were used to train a net, with the data from the remaining 5 subjects used to test the resulting net. The nets correctly classified the training subjects in all 10 cases and correctly classified all 5 test subjects, indicating that the developed networks were able to generalize to new data sets. It was concluded that a minimal net consisting of only three variables, peak VO2 at the high velocity, hand force on the rim at the low velocity, and push angle at the high velocity, could accurately represent the differences between these groups.

  16. Neural net learning issues in classification of free text documents

    SciTech Connect

    Dasigi, V.R.; Mann, R.C.

    1996-03-01

    In intelligent analysis of large amounts of text, not any single clue indicates reliably that a pattern of interest has been found. When using multiple clues, it is not known how these should be integrated into a decision. In the context of this investigation, we have been using neural nets as parameterized mappings that allow for fusion of higher level clues extracted from free text. By using higher level clues and features, we avoid very large networks. By using the dominant singular values computed by Latent Semantic Indexing (LSI) and applying neural network algorithms for integrating these values and the outputs from other ``sensors,`` we have obtained preliminary encouraging results with text classification.

  17. Automatic voice recognition using traditional and artificial neural network approaches

    NASA Technical Reports Server (NTRS)

    Botros, Nazeih M.

    1989-01-01

    The main objective of this research is to develop an algorithm for isolated-word recognition. This research is focused on digital signal analysis rather than linguistic analysis of speech. Features extraction is carried out by applying a Linear Predictive Coding (LPC) algorithm with order of 10. Continuous-word and speaker independent recognition will be considered in future study after accomplishing this isolated word research. To examine the similarity between the reference and the training sets, two approaches are explored. The first is implementing traditional pattern recognition techniques where a dynamic time warping algorithm is applied to align the two sets and calculate the probability of matching by measuring the Euclidean distance between the two sets. The second is implementing a backpropagation artificial neural net model with three layers as the pattern classifier. The adaptation rule implemented in this network is the generalized least mean square (LMS) rule. The first approach has been accomplished. A vocabulary of 50 words was selected and tested. The accuracy of the algorithm was found to be around 85 percent. The second approach is in progress at the present time.

  18. Automatic labeling and characterization of objects using artificial neural networks

    NASA Technical Reports Server (NTRS)

    Campbell, William J.; Hill, Scott E.; Cromp, Robert F.

    1989-01-01

    Existing NASA supported scientific data bases are usually developed, managed and populated in a tedious, error prone and self-limiting way in terms of what can be described in a relational Data Base Management System (DBMS). The next generation Earth remote sensing platforms, i.e., Earth Observation System, (EOS), will be capable of generating data at a rate of over 300 Mbs per second from a suite of instruments designed for different applications. What is needed is an innovative approach that creates object-oriented databases that segment, characterize, catalog and are manageable in a domain-specific context and whose contents are available interactively and in near-real-time to the user community. Described here is work in progress that utilizes an artificial neural net approach to characterize satellite imagery of undefined objects into high-level data objects. The characterized data is then dynamically allocated to an object-oriented data base where it can be reviewed and assessed by a user. The definition, development, and evolution of the overall data system model are steps in the creation of an application-driven knowledge-based scientific information system.

  19. Stabilization of Hebbian neural nets by inhibitory learning.

    PubMed

    Easton, P; Gordon, P E

    1984-01-01

    In Hebbian neural models synaptic reinforcement occurs when the pre- and post-synaptic neurons are simultaneously active. This causes an instability toward unlimited growth of excitatory synapses. The system can be stabilized by recurrent inhibition via modifiable inhibitory synapses. When this process is included, it is possible to dispense with the non-linear normalization or cut-off conditions which were necessary for stability in previous models. The present formulation is response-linear if synaptic changes are slow. It is self-consistent because the stabilizing effects will tend to keep most neural activity in the middle range, where neural response is approximately linear. The linearized equations are tensor invariant under a class of rotations of the state space. Using this, the response to stimulation may be derived as a set of independent modes of activity distributed over the net, which may be identified with cell assemblies. A continuously infinite set of equivalent solutions exists.

  20. A real time neural net estimator of fatigue life

    NASA Technical Reports Server (NTRS)

    Troudet, T.; Merrill, W.

    1990-01-01

    A neural network architecture is proposed to estimate, in real-time, the fatigue life of mechanical components, as part of the intelligent Control System for Reusable Rocket Engines. Arbitrary component loading values were used as input to train a two hidden-layer feedforward neural net to estimate component fatigue damage. The ability of the net to learn, based on a local strain approach, the mapping between load sequence and fatigue damage has been demonstrated for a uniaxial specimen. Because of its demonstrated performance, the neural computation may be extended to complex cases where the loads are biaxial or triaxial, and the geometry of the component is complex (e.g., turbopumps blades). The generality of the approach is such that load/damage mappings can be directly extracted from experimental data without requiring any knowledge of the stress/strain profile of the component. In addition, the parallel network architecture allows real-time life calculations even for high-frequency vibrations. Owing to its distributed nature, the neural implementation will be robust and reliable, enabling its use in hostile environments such as rocket engines.

  1. Impact of descriptor vector scaling on the classification of drugs and nondrugs with artificial neural networks.

    PubMed

    Givehchi, Alireza; Schneider, Gisbert

    2004-06-01

    The influence of preprocessing of molecular descriptor vectors for solving classification tasks was analyzed for drug/nondrug classification by artificial neural networks. Molecular properties were used to form descriptor vectors. Two types of neural networks were used, supervised multilayer neural nets trained with the back-propagation algorithm, and unsupervised self-organizing maps (Kohonen maps). Data were preprocessed by logistic scaling and histogram equalization. For both types of neural networks, the preprocessing step significantly improved classification compared to nonstandardized data. Classification accuracy was measured as prediction mean square error and Matthews correlation coefficient in the case of supervised learning, and quantization error in the case of unsupervised learning. The results demonstrate that appropriate data preprocessing is an essential step in solving classification tasks.

  2. Medical image analysis with artificial neural networks.

    PubMed

    Jiang, J; Trundle, P; Ren, J

    2010-12-01

    Given that neural networks have been widely reported in the research community of medical imaging, we provide a focused literature survey on recent neural network developments in computer-aided diagnosis, medical image segmentation and edge detection towards visual content analysis, and medical image registration for its pre-processing and post-processing, with the aims of increasing awareness of how neural networks can be applied to these areas and to provide a foundation for further research and practical development. Representative techniques and algorithms are explained in detail to provide inspiring examples illustrating: (i) how a known neural network with fixed structure and training procedure could be applied to resolve a medical imaging problem; (ii) how medical images could be analysed, processed, and characterised by neural networks; and (iii) how neural networks could be expanded further to resolve problems relevant to medical imaging. In the concluding section, a highlight of comparisons among many neural network applications is included to provide a global view on computational intelligence with neural networks in medical imaging.

  3. Adaptive Neurons For Artificial Neural Networks

    NASA Technical Reports Server (NTRS)

    Tawel, Raoul

    1990-01-01

    Training time decreases dramatically. In improved mathematical model of neural-network processor, temperature of neurons (in addition to connection strengths, also called weights, of synapses) varied during supervised-learning phase of operation according to mathematical formalism and not heuristic rule. Evidence that biological neural networks also process information at neuronal level.

  4. Optoelectronic analogs of self-programming neural nets - Architecture and methodologies for implementing fast stochastic learning by simulated annealing

    NASA Technical Reports Server (NTRS)

    Farhat, Nabil H.

    1987-01-01

    Self-organization and learning is a distinctive feature of neural nets and processors that sets them apart from conventional approaches to signal processing. It leads to self-programmability which alleviates the problem of programming complexity in artificial neural nets. In this paper architectures for partitioning an optoelectronic analog of a neural net into distinct layers with prescribed interconnectivity pattern to enable stochastic learning by simulated annealing in the context of a Boltzmann machine are presented. Stochastic learning is of interest because of its relevance to the role of noise in biological neural nets. Practical considerations and methodologies for appreciably accelerating stochastic learning in such a multilayered net are described. These include the use of parallel optical computing of the global energy of the net, the use of fast nonvolatile programmable spatial light modulators to realize fast plasticity, optical generation of random number arrays, and an adaptive noisy thresholding scheme that also makes stochastic learning more biologically plausible. The findings reported predict optoelectronic chips that can be used in the realization of optical learning machines.

  5. Optoelectronic analogs of self-programming neural nets: architecture and methodologies for implementing fast stochastic learning by simulated annealing.

    PubMed

    Farhat, N H

    1987-12-01

    Self-organization and learning is a distinctive feature of neural nets and processors that sets them apart from conventional approaches to signal processing. It leads to self-programmability which alleviates the problem of programming complexity in artificial neural nets. In this paper architectures for partitioning an optoelectronic analog of a neural net into distinct layers with prescribed interconnectivity pattern to enable stochastic learning by simulated annealing in the context of a Boltzmann machine are presented. Stochastic learning is of interest because of its relevance to the role of noise in biological neural nets. Practical considerations and methodologies for appreciably accelerating stochastic learning in such a multilayered net are described. These include the use of parallel optical computing of the global energy of the net, the use of fast nonvolatile programmable spatial light modulators to realize fast plasticity, optical generation of random number arrays, and an adaptive noisy thresholding scheme that also makes stochastic learning more biologically plausible. The findings reported predict optoelectronic chips that can be used in the realization of optical learning machines.

  6. Classifying epilepsy diseases using artificial neural networks and genetic algorithm.

    PubMed

    Koçer, Sabri; Canal, M Rahmi

    2011-08-01

    In this study, FFT analysis is applied to the EEG signals of the normal and patient subjects and the obtained FFT coefficients are used as inputs in Artificial Neural Network (ANN). The differences shown by the non-stationary random signals such as EEG signals in cases of health and sickness (epilepsy) were evaluated and tried to be analyzed under computer-supported conditions by using artificial neural networks. Multi-Layer Perceptron (MLP) architecture is used Levenberg-Marquardt (LM), Quickprop (QP), Delta-bar delta (DBD), Momentum and Conjugate gradient (CG) learning algorithms, and the best performance was tried to be attained by ensuring the optimization with the use of genetic algorithms of the weights, learning rates, neuron numbers of hidden layer in the training process. This study shows that the artificial neural network increases the classification performance using genetic algorithm.

  7. Artificial neural networks for short term electrical load forecasting

    SciTech Connect

    Reinschmidt, K.F.

    1995-10-01

    The accurate prediction of hourly electrical demand one or more days ahead is of great economic importance to electric utilities for generation unit dispatch and unit commitment. Artificial neural networks for pattern recognition are developed to identify days in the historical record that are most similar to the days being forecasted, to use for load prediction. Artificial neural networks are also used to generate linear and nonlinear multivariate time series models, to project demands forward in time. The genetic algorithm is used to select the optimal set of independent variables for forecasting. Techniques are developed to combine forecasts derived from independent methods, to achieve better accuracy than any single forecast. In this way, artificial neural networks can be used to generate practical, accurate short-term electrical load forecasts.

  8. Non-US artificial neural network research. FASAC special study

    SciTech Connect

    Davidson, R. B.

    1991-10-01

    This assessment was undertaken to examine for US Government research and development sponsors (and for Government policy-makers and analysts who must be aware of foreign scientific and technological capabilities) the recent range, quality, and accomplishments of non-US artificial neural network research and development activities. It records the project's initial assessments of major artificial neural network research and development activities in Western Europe, Japan, and the Soviet Union, where the largest, most organized efforts are proceeding or where potential military applications are of interest to US policy-makers. We plan to issue updated versions of this report periodically, as more research and development activities (in more places) are examined and as the efforts assessed in this report succeed or fail. This report focuses on artificial neural networks as an information processing technology, the goal of which is design and production of powerful computers for appropriate applications.

  9. Perineuronal net, CSPG receptor and their regulation of neural plasticity.

    PubMed

    Miao, Qing-Long; Ye, Qian; Zhang, Xiao-Hui

    2014-08-25

    Perineuronal nets (PNNs) are reticular structures resulting from the aggregation of extracellular matrix (ECM) molecules around the cell body and proximal neurite of specific population of neurons in the central nervous system (CNS). Since the first description of PNNs by Camillo Golgi in 1883, the molecular composition, developmental formation and potential functions of these specialized extracellular matrix structures have only been intensively studied over the last few decades. The main components of PNNs are hyaluronan (HA), chondroitin sulfate proteoglycans (CSPGs) of the lectican family, link proteins and tenascin-R. PNNs appear late in neural development, inversely correlating with the level of neural plasticity. PNNs have long been hypothesized to play a role in stabilizing the extracellular milieu, which secures the characteristic features of enveloped neurons and protects them from the influence of malicious agents. Aberrant PNN signaling can lead to CNS dysfunctions like epilepsy, stroke and Alzheimer's disease. On the other hand, PNNs create a barrier which constrains the neural plasticity and counteracts the regeneration after nerve injury. Digestion of PNNs with chondroitinase ABC accelerates functional recovery from the spinal cord injury and restores activity-dependent mechanisms for modifying neuronal connections in the adult animals, indicating that PNN is an important regulator of neural plasticity. Here, we review recent progress in the studies on the formation of PNNs during early development and the identification of CSPG receptor - an essential molecular component of PNN signaling, along with a discussion on their unique regulatory roles in neural plasticity.

  10. Wood Defect Identification Based on Artificial Neural Network

    NASA Astrophysics Data System (ADS)

    Zhu, Xiao-Dong; Cao, Jun; Wang, Feng-Hu; Sun, Jian-Ping; Liu, Yu

    Defects in wooden material reduce the value of timber. In order to save and improve the utilization of the timber, many studies are carried out on the ways to detect defects in wood. The recent development of computer technology, data processing technology and signal processing technology provides researchers with more damage identification problem solution ideas and methods. This article studies the vibration characteristics of wood. With an exploration of the wavelet analysis and artificial neural network for the wood composite material defects based on non-destructive testing, an artificial neural network model is established for wood-based composite materials non-destructive testing technology.

  11. Functional approximation using artificial neural networks in structural mechanics

    NASA Technical Reports Server (NTRS)

    Alam, Javed; Berke, Laszlo

    1993-01-01

    The artificial neural networks (ANN) methodology is an outgrowth of research in artificial intelligence. In this study, the feed-forward network model that was proposed by Rumelhart, Hinton, and Williams was applied to the mapping of functions that are encountered in structural mechanics problems. Several different network configurations were chosen to train the available data for problems in materials characterization and structural analysis of plates and shells. By using the recall process, the accuracy of these trained networks was assessed.

  12. AUTOMATED DEFECT CLASSIFICATION USING AN ARTIFICIAL NEURAL NETWORK

    SciTech Connect

    Chady, T.; Caryk, M.; Piekarczyk, B.

    2009-03-03

    The automated defect classification algorithm based on artificial neural network with multilayer backpropagation structure was utilized. The selected features of flaws were used as input data. In order to train the neural network it is necessary to prepare learning data which is representative database of defects. Database preparation requires the following steps: image acquisition and pre-processing, image enhancement, defect detection and feature extraction. The real digital radiographs of welded parts of a ship were used for this purpose.

  13. Artificial neural networks for decision support in clinical medicine.

    PubMed

    Forsström, J J; Dalton, K J

    1995-10-01

    Connectionist models such as neural networks are alternatives to linear, parametric statistical methods. Neural networks are computer-based pattern recognition methods with loose similarities with the nervous system. Individual variables of the network, usually called 'neurones', can receive inhibitory and excitatory inputs from other neurones. The networks can define relationships among input data that are not apparent when using other approaches, and they can use these relationships to improve accuracy. Thus, neural nets have substantial power to recognize patterns even in complex datasets. Neural network methodology has outperformed classical statistical methods in cases where the input variables are interrelated. Because clinical measurements usually derive from multiple interrelated systems it is evident that neural networks might be more accurate than classical methods in multivariate analysis of clinical data. This paper reviews the use of neural networks in medical decision support. A short introduction to the basics of neural networks is given, and some practical issues in applying the networks are highlighted. The current use of neural networks in image analysis, signal processing and laboratory medicine is reviewed. It is concluded that neural networks have an important role in image analysis and in signal processing. However, further studies are needed to determine the value of neural networks in the analysis of laboratory data.

  14. Using neural nets to measure ocular refractive errors: a proposal

    NASA Astrophysics Data System (ADS)

    Netto, Antonio V.; Ferreira de Oliveira, Maria C.

    2002-12-01

    We propose the development of a functional system for diagnosing and measuring ocular refractive errors in the human eye (astigmatism, hypermetropia and myopia) by automatically analyzing images of the human ocular globe acquired with the Hartmann-Schack (HS) technique. HS images are to be input into a system capable of recognizing the presence of a refractive error and outputting a measure of such an error. The system should pre-process and image supplied by the acquisition technique and then use artificial neural networks combined with fuzzy logic to extract the necessary information and output an automated diagnosis of the refractive errors that may be present in the ocular globe under exam.

  15. Application of artificial neural networks (ANNs) in wine technology.

    PubMed

    Baykal, Halil; Yildirim, Hatice Kalkan

    2013-01-01

    In recent years, neural networks have turned out as a powerful method for numerous practical applications in a wide variety of disciplines. In more practical terms neural networks are one of nonlinear statistical data modeling tools. They can be used to model complex relationships between inputs and outputs or to find patterns in data. In food technology artificial neural networks (ANNs) are useful for food safety and quality analyses, predicting chemical, functional and sensory properties of various food products during processing and distribution. In wine technology, ANNs have been used for classification and for predicting wine process conditions. This review discusses the basic ANNs technology and its possible applications in wine technology.

  16. Face recognition: Eigenface, elastic matching, and neural nets

    SciTech Connect

    Zhang, J.; Yan, Y.; Lades, M.

    1997-09-01

    This paper is a comparative study of three recently proposed algorithms for face recognition: eigenface, autoassociation and classification neural nets, and elastic matching. After these algorithms were analyzed under a common statistical decision framework, they were evaluated experimentally on four individual data bases, each with a moderate subject size, and a combined data base with more than a hundred different subjects. Analysis and experimental results indicate that the eigenface algorithm, which is essentially a minimum distance classifier, works well when lighting variation is small. Its performance deteriorates significantly as lighting variation increases. The elastic matching algorithm, on the other hand, is insensitive to lighting, face position, and expression variations and therefore is more versatile. The performance of the autoassociation and classification nets is upper bounded by that of the eigenface but is more difficult to implement in practice.

  17. Contact prediction using mutual information and neural nets.

    PubMed

    Shackelford, George; Karplus, Kevin

    2007-01-01

    Prediction of protein structures continues to be a difficult problem, particularly when there are no solved structures for homologous proteins to use as templates. Local structure prediction (secondary structure and burial) is fairly reliable, but does not provide enough information to produce complete three-dimensional structures. Residue-residue contact prediction, though still not highly reliable, may provide a useful guide for assembling local structure prediction into full tertiary prediction. We develop a neural network which is applied to pairs of residue positions and outputs a probability of contact between the positions. One of the neural net inputs is a novel statistic for detecting correlated mutations: the statistical significance of the mutual information between the corresponding columns of a multiple sequence alignment. This statistic, combined with a second statistic based on the propensity of two amino acid types being in contact, results in a simple neural network that is a good predictor of contacts. Adding more features from amino-acid distributions and local structure predictions, the final neural network predicts contacts better than other submitted contact predictions at CASP7, including contact predictions derived from fragment-based tertiary models on free-modeling domains. It is still not known if contact predictions can improve tertiary models on free-modeling domains. Available at http://www.soe.ucsc.edu/research/compbio/SAM_T06/T06-query.html.

  18. Backpropagation neural nets with one and two hidden layers.

    PubMed

    de Villiers, J; Barnard, E

    1993-01-01

    The differences in classification and training performance of three- and four-layer (one- and two-hidden-layer) fully interconnected feedforward neural nets are investigated. To obtain results which do not merely reflect performance on a particular data set, the networks are trained on various distributions, which are themselves drawn from a distribution of distributions. Experimental results indicate that four-layered networks are more prone to fall into bad local minima, but that three- and four-layered networks perform similarly in all other respects.

  19. Compensation type algorithms for neural nets: stability and convergence.

    PubMed

    Cromme, L J; Dammasch, I E

    1989-01-01

    Plasticity of synaptic connections plays an important role in the temporal development of neural networks which are the basis of memory and behavior. The conditions for successful functional performance of these nerve nets have to be either guaranteed genetically or developed during ontogenesis. In the latter case, a general law of this development may be the successive compensation of disturbances. A compensation type algorithm is analyzed here that changes the connectivity of a given network such that deviations from each neuron's equilibrium state are reduced. The existence of compensated networks is proven, the convergence and stability of simulations are investigated, and implications for cognitive systems are discussed.

  20. Webs, cell assemblies, and chunking in neural nets: introduction.

    PubMed

    Wickelgren, W A

    1999-03-01

    This introduction to Wickelgren (1992), describes a theory of idea representation and learning in the cerebral cortex and seven properties of Hebb's (1949) formulation of cell assemblies that have played a major role in all such neural net models. Ideas are represented in the cerebral cortex by webs (innate cell assemblies), using sparse coding with sparse, all-or-none, innate linking. Recruiting a web to represent a new idea is called chunking. The innate links that bind the neurons of a web are basal dendritic synapses. Learning modifies the apical dendritic synapses that associate neurons in one web to neurons in another web.

  1. Artificial neural networks: theoretical background and pharmaceutical applications: a review.

    PubMed

    Wesolowski, Marek; Suchacz, Bogdan

    2012-01-01

    In recent times, there has been a growing interest in artificial neural networks, which are a rough simulation of the information processing ability of the human brain, as modern and vastly sophisticated computational techniques. This interest has also been reflected in the pharmaceutical sciences. This paper presents a review of articles on the subject of the application of neural networks as effective tools assisting the solution of various problems in science and the pharmaceutical industry, especially those characterized by multivariate and nonlinear dependencies. After a short description of theoretical background and practical basics concerning the computations performed by means of neural networks, the most important pharmaceutical applications of neural networks, with suitable references, are demonstrated. The huge role played by neural networks in pharmaceutical analysis, pharmaceutical technology, and searching for the relationships between the chemical structure and the properties of newly synthesized compounds as candidates for drugs is discussed.

  2. Introducing Artificial Neural Networks through a Spreadsheet Model

    ERIC Educational Resources Information Center

    Rienzo, Thomas F.; Athappilly, Kuriakose K.

    2012-01-01

    Business students taking data mining classes are often introduced to artificial neural networks (ANN) through point and click navigation exercises in application software. Even if correct outcomes are obtained, students frequently do not obtain a thorough understanding of ANN processes. This spreadsheet model was created to illuminate the roles of…

  3. Artificial Neural Networks in Policy Research: A Current Assessment.

    ERIC Educational Resources Information Center

    Woelfel, Joseph

    1993-01-01

    Suggests that artificial neural networks (ANNs) exhibit properties that promise usefulness for policy researchers. Notes that ANNs have found extensive use in areas once reserved for multivariate statistical programs such as regression and multiple classification analysis and are developing an extensive community of advocates for processing text…

  4. Artificial Neural Networks for Modeling Knowing and Learning in Science.

    ERIC Educational Resources Information Center

    Roth, Wolff-Michael

    2000-01-01

    Advocates artificial neural networks as models for cognition and development. Provides an example of how such models work in the context of a well-known Piagetian developmental task and school science activity: balance beam problems. (Contains 59 references.) (Author/WRM)

  5. Recurrent Artificial Neural Networks and Finite State Natural Language Processing.

    ERIC Educational Resources Information Center

    Moisl, Hermann

    It is argued that pessimistic assessments of the adequacy of artificial neural networks (ANNs) for natural language processing (NLP) on the grounds that they have a finite state architecture are unjustified, and that their adequacy in this regard is an empirical issue. First, arguments that counter standard objections to finite state NLP on the…

  6. Distribution feeder loss computation by artificial neural network

    SciTech Connect

    Kau, S.W.; Cho, M.Y.

    1995-12-31

    This paper proposes an artificial neural network (ANN) based feeder loss calculation model for distribution system analysis. In this paper, the functional-link network model is examined to form the artificial neural network architecture to derive the various loss calculation models for feeders with different configuration. Such artificial neural network is a feedforward network that uses standard back-propagation algorithm to adjust weights on the connection path between any two processing elements (PEs). Feeder daily load curve on each season are derived by field test data. Three-phase load flow program is executed to create the training sets with exact loss calculation results. A sensitivity analysis is executed to determine the key factors included power factor, feeder loading, primary conductors, secondary conductors, and transformer capacity as the variables for components located at input layer. By artificial neural network with the pattern recognition ability, this study has developed seasonal and yearly loss calculation models for overhead and underground feeder configuration. Two practical feeders with both overhead and underground configuration in Taiwan Power Company (TPC or Taipower) distribution system are selected for computer simulation to demonstrate the effectiveness and accuracy of the proposed models. As comparing with models derived by the conventional regression technique, results indicate that the proposed models provide more efficient tool to District engineer for feeder loss calculation.

  7. An artificial neural network approach to transformer fault diagnosis

    SciTech Connect

    Zhang, Y.; Ding, X.; Liu, Y.; Griffin, P.J.

    1996-10-01

    This paper presents an artificial neural network (ANN) approach to diagnose and detect faults in oil-filled power transformers based on dissolved gas-in-oil analysis. A two-step ANN method is used to detect faults with or without cellulose involved. Good diagnosis accuracy is obtained with the proposed approach.

  8. Artificial-neural-network-based failure detection and isolation

    NASA Astrophysics Data System (ADS)

    Sadok, Mokhtar; Gharsalli, Imed; Alouani, Ali T.

    1998-03-01

    This paper presents the design of a systematic failure detection and isolation system that uses the concept of failure sensitive variables (FSV) and artificial neural networks (ANN). The proposed approach was applied to tube leak detection in a utility boiler system. Results of the experimental testing are presented in the paper.

  9. An artificial neural network controller for intelligent transportation systems applications

    SciTech Connect

    Vitela, J.E.; Hanebutte, U.R.; Reifman, J.

    1996-04-01

    An Autonomous Intelligent Cruise Control (AICC) has been designed using a feedforward artificial neural network, as an example for utilizing artificial neural networks for nonlinear control problems arising in intelligent transportation systems applications. The AICC is based on a simple nonlinear model of the vehicle dynamics. A Neural Network Controller (NNC) code developed at Argonne National Laboratory to control discrete dynamical systems was used for this purpose. In order to test the NNC, an AICC-simulator containing graphical displays was developed for a system of two vehicles driving in a single lane. Two simulation cases are shown, one involving a lead vehicle with constant velocity and the other a lead vehicle with varying acceleration. More realistic vehicle dynamic models will be considered in future work.

  10. Use of artificial neural networks as estimators and controllers

    NASA Astrophysics Data System (ADS)

    Concilio, Antonio; Sorrentino, A.

    1996-04-01

    Active noise control is one among the most promising applications of the so-called Smart Structures, because it ensures, or promises, lower weight, lower cost, more effectiveness and all what is desirable in a vehicle design process, with respect to the current solutions. More and more attention in the research world has been devoting to this argument, pushed by both political, economical and environmental reasons, the one connected to the others. Piezoceramic actuators, integrated into the structure, seem to offer the most fashionable and practical solutions among all the proposed architectures, [1-2]. As sensors, microphones demonstrated to be the most performing, above all because they give the most suitable representation of the field that has to be cancelled, [3-4]. This approach is known as Acousto-Structural Active Control, ASAC, [5]. However, according to Fuller's definition, [6] , an intelligent controller is needed to ensure the development of an "Intelligent Structure" . Its main characteristic should be represented by the capability of learning by examples, of following the structure during its evolution, of being the system "brain" . This peculiarity may be offered by Artificial Neural Networks (ANN's), [7-8]. They present other important features, like the capability, in principle, of treating non-linear as well as linear problems, [9], of identifying dynamic systems, [10], of properly acting as a controller. Then, such a net could integrate in itself the function of "system estimator" or "observer" ,and of interpolator - extrapolator and controller, contemporarily. The authors have been working on such subjects for a long time, proposing for instance ANN's as time-domain structural parameters estimators on a simple 2D element ( a framed plate), [11], as noise and vibration controllers in a FF system, [12-13], as materials damping parameters extractors from experimental data, [14]. All these applications were aimed at noise reduction problems. The

  11. Artificial neural networks for 3-D motion analysis-Part II: Nonrigid motion.

    PubMed

    Chen, T; Lin, W C; Chen, C T

    1995-01-01

    For pt. I see ibid., p. 1386-93 (1995). An approach applying artificial neural net techniques to 3D nonrigid motion analysis is proposed. The 3D nonrigid motion of the left ventricle of a human heart is examined using biplanar cineangiography data, consisting of 3D coordinates of 30 coronary artery bifurcation points of the left ventricle and the correspondences of these points taken over 10 time instants during the heart cardiac cycle. The motion is decomposed into global rigid motion and a set of local nonrigid deformations which are coupled with the global motion. The global rigid motion can be estimated precisely as a translation vecto and a rotation matrix. Local nonrigid deformation estimation is discussed. A set of neural nets similar in structure and dynamics but different in physical size is proposed to tackle the problem of nonrigidity. These neural networks are interconnected through feedbacks. The activation function of the output layer is selected so that a feedback is involved in the output updating. The constraints are specified to ensure stable and globally consistent estimation. The objective is to find the optimal deformation matrices that satisfy the constraints for all coronary artery bifurcation points of the left ventricle. The proposed neural networks differ from other existing neural network models in their unique structure and dynamics.

  12. Convex quadratic optimization on artificial neural networks

    SciTech Connect

    Adler, I.; Verma, S.

    1994-12-31

    We present continuous-valued Hopfield recurrent neural networks on which we map convex quadratic optimization problems. We consider two different convex quadratic programs, each of which is mapped to a different neural network. Activation functions are shown to play a key role in the mapping under each model. The class of activation functions which can be used in this mapping is characterized in terms of the properties needed. It is shown that the first derivatives of penalty as well as barrier functions belong to this class. The trajectories of dynamics under the first model are shown to be closely related to affine-scaling trajectories of interior-point methods. On the other hand, the trajectories of dynamics under the second model correspond to projected steepest descent pathways.

  13. Evaluating neural networks and artificial intelligence systems

    NASA Astrophysics Data System (ADS)

    Alberts, David S.

    1994-02-01

    Systems have no intrinsic value in and of themselves, but rather derive value from the contributions they make to the missions, decisions, and tasks they are intended to support. The estimation of the cost-effectiveness of systems is a prerequisite for rational planning, budgeting, and investment documents. Neural network and expert system applications, although similar in their incorporation of a significant amount of decision-making capability, differ from each other in ways that affect the manner in which they can be evaluated. Both these types of systems are, by definition, evolutionary systems, which also impacts their evaluation. This paper discusses key aspects of neural network and expert system applications and their impact on the evaluation process. A practical approach or methodology for evaluating a certain class of expert systems that are particularly difficult to measure using traditional evaluation approaches is presented.

  14. A neutron spectrum unfolding computer code based on artificial neural networks

    NASA Astrophysics Data System (ADS)

    Ortiz-Rodríguez, J. M.; Reyes Alfaro, A.; Reyes Haro, A.; Cervantes Viramontes, J. M.; Vega-Carrillo, H. R.

    2014-02-01

    The Bonner Spheres Spectrometer consists of a thermal neutron sensor placed at the center of a number of moderating polyethylene spheres of different diameters. From the measured readings, information can be derived about the spectrum of the neutron field where measurements were made. Disadvantages of the Bonner system are the weight associated with each sphere and the need to sequentially irradiate the spheres, requiring long exposure periods. Provided a well-established response matrix and adequate irradiation conditions, the most delicate part of neutron spectrometry, is the unfolding process. The derivation of the spectral information is not simple because the unknown is not given directly as a result of the measurements. The drawbacks associated with traditional unfolding procedures have motivated the need of complementary approaches. Novel methods based on Artificial Intelligence, mainly Artificial Neural Networks, have been widely investigated. In this work, a neutron spectrum unfolding code based on neural nets technology is presented. This code is called Neutron Spectrometry and Dosimetry with Artificial Neural networks unfolding code that was designed in a graphical interface. The core of the code is an embedded neural network architecture previously optimized using the robust design of artificial neural networks methodology. The main features of the code are: easy to use, friendly and intuitive to the user. This code was designed for a Bonner Sphere System based on a 6LiI(Eu) neutron detector and a response matrix expressed in 60 energy bins taken from an International Atomic Energy Agency compilation. The main feature of the code is that as entrance data, for unfolding the neutron spectrum, only seven rate counts measured with seven Bonner spheres are required; simultaneously the code calculates 15 dosimetric quantities as well as the total flux for radiation protection purposes. This code generates a full report with all information of the unfolding in

  15. Design and development of an artificial neural network for estimation of formation permeability

    SciTech Connect

    Mohaghegh, S.; Arefi, R.; Ameri, S.; Rose, D.

    1994-12-31

    Permeability is one of the most important characteristics of hydrocarbon bearing formations. An accurate knowledge of permeability gives petroleum engineers a tool for efficiently managing the production processes of a field. It is also one of the most important pieces of information in the design and management of enhanced recovery operations. Formation permeability is often measured in the laboratory from cores or evaluated from well test data. Core analysis and well test data, however, are only available from a few wells in a field. On the other hand, almost all wells are logged. In this study an artificial neural network has been designed that is able to predict the permeability of the formations using the data provided by geophysical well logs with good accuracy. Artificial neural network, a biologically inspired computing method, with its ability to learn, self-adjust, and be trained provide a powerful tool to solve problems that involve pattern recognition. Using well logs to predict permeability has been attempted in the past. The problems with previous approaches were mainly two-fold, namely, the number of variables used (only one variable-porosity), and using regression analysis as the main tool for correlations. The approach introduced in this paper is an attempt to overcome these short comings. This is done, first, by using many variables from well logs that may provide information about the permeability. Second, by recognizing the existence of possible patterns between these variables and formation permeability using artificial neural networks. Neural nets are analog, inherently parallel and distributive systems. These characteristics, which will be discussed in the paper, are the main characteristics that enable artificial neural networks to be successful in predicting the permeability in rocks using well log information.

  16. Data fusion with artificial neural networks (ANN) for classification of earth surface from microwave satellite measurements

    NASA Technical Reports Server (NTRS)

    Lure, Y. M. Fleming; Grody, Norman C.; Chiou, Y. S. Peter; Yeh, H. Y. Michael

    1993-01-01

    A data fusion system with artificial neural networks (ANN) is used for fast and accurate classification of five earth surface conditions and surface changes, based on seven SSMI multichannel microwave satellite measurements. The measurements include brightness temperatures at 19, 22, 37, and 85 GHz at both H and V polarizations (only V at 22 GHz). The seven channel measurements are processed through a convolution computation such that all measurements are located at same grid. Five surface classes including non-scattering surface, precipitation over land, over ocean, snow, and desert are identified from ground-truth observations. The system processes sensory data in three consecutive phases: (1) pre-processing to extract feature vectors and enhance separability among detected classes; (2) preliminary classification of Earth surface patterns using two separate and parallely acting classifiers: back-propagation neural network and binary decision tree classifiers; and (3) data fusion of results from preliminary classifiers to obtain the optimal performance in overall classification. Both the binary decision tree classifier and the fusion processing centers are implemented by neural network architectures. The fusion system configuration is a hierarchical neural network architecture, in which each functional neural net will handle different processing phases in a pipelined fashion. There is a total of around 13,500 samples for this analysis, of which 4 percent are used as the training set and 96 percent as the testing set. After training, this classification system is able to bring up the detection accuracy to 94 percent compared with 88 percent for back-propagation artificial neural networks and 80 percent for binary decision tree classifiers. The neural network data fusion classification is currently under progress to be integrated in an image processing system at NOAA and to be implemented in a prototype of a massively parallel and dynamically reconfigurable Modular

  17. Identification and interpretation of patterns in rocket engine data: Artificial intelligence and neural network approaches

    NASA Technical Reports Server (NTRS)

    Ali, Moonis; Whitehead, Bruce; Gupta, Uday K.; Ferber, Harry

    1989-01-01

    This paper describes an expert system which is designed to perform automatic data analysis, identify anomalous events, and determine the characteristic features of these events. We have employed both artificial intelligence and neural net approaches in the design of this expert system. The artificial intelligence approach is useful because it provides (1) the use of human experts' knowledge of sensor behavior and faulty engine conditions in interpreting data; (2) the use of engine design knowledge and physical sensor locations in establishing relationships among the events of multiple sensors; (3) the use of stored analysis of past data of faulty engine conditions; and (4) the use of knowledge-based reasoning in distinguishing sensor failure from actual faults. The neural network approach appears promising because neural nets (1) can be trained on extremely noisy data and produce classifications which are more robust under noisy conditions than other classification techniques; (2) avoid the necessity of noise removal by digital filtering and therefore avoid the need to make assumptions about frequency bands or other signal characteristics of anomalous behavior; (3) can, in effect, generate their own feature detectors based on the characteristics of the sensor data used in training; and (4) are inherently parallel and therefore are potentially implementable in special-purpose parallel hardware.

  18. A developmental model for the evolution of artificial neural networks.

    PubMed

    Astor, J C; Adami, C

    2000-01-01

    We present a model of decentralized growth and development for artificial neural networks (ANNs), inspired by developmental biology and the physiology of nervous systems. In this model, each individual artificial neuron is an autonomous unit whose behavior is determined only by the genetic information it harbors and local concentrations of substrates. The chemicals and substrates, in turn, are modeled by a simple artificial chemistry. While the system is designed to allow for the evolution of complex networks, we demonstrate the power of the artificial chemistry by analyzing engineered (handwritten) genomes that lead to the growth of simple networks with behaviors known from physiology. To evolve more complex structures, a Java-based, platform-independent, asynchronous, distributed genetic algorithm (GA) has been implemented that allows users to participate in evolutionary experiments via the World Wide Web.

  19. Design of speaker recognition system based on artificial neural network

    NASA Astrophysics Data System (ADS)

    Chen, Yanhong; Wang, Li; Lin, Han; Li, Jinlong

    2012-10-01

    Speaker recognition is to recognize speaker's identity from its voice which contains physiological and behavioral characteristics unique to each individual. In this paper, the artificial neural network model, which has very good capacity of non-linear division in characteristic space, is used for pattern matching. The speaker's sample characteristic domain is built for his mixed voice characteristic signals based on Kmeanlbg algorithm. Then the dimension of the inputting eigenvector is reduced, and the redundant information is got rid of. On this basis, BP neural network is used to divide capacity area for characteristic space nonlinearly, and the BP neural network acts as a classifier for the speaker. Finally, a speaker recognition system based on the neural network is realized and the experiment results validate the recognition performance and robustness of the system.

  20. Artificial neural network for risk assessment in preterm neonates

    PubMed Central

    Zernikow, B; Holtmannspoetter, K; Michel, E; Pielemeier, W; Hornschuh, F; Westermann, A; Hennecke, K

    1998-01-01

    AIM—To predict the individual neonatal mortality risk of preterm infants using an artificial neural network "trained" on admission data.
METHODS—A total of 890 preterm neonates (<32 weeks gestational age and/or <1500 g birthweight) were enrolled in our retrospective study. The neural network trained on infants born between 1990and 1993. The predictive value was tested on infants born in the successive three years.
RESULTS—The artificial neural network performed significantly better than a logistic regression model (area under the receiver operator curve 0.95 vs 0.92). Survival was associated with high morbidity if the predicted mortality risk was greater than 0.50. There were no preterm infants with a predicted mortality risk of greater than 0.80. The mortality risks of two non-survivors with birthweights >2000 g and severe congenital disease had largely been underestimated.
CONCLUSION—An artificial neural network trained on admission data can accurately predict the mortality risk for most preterm infants. However, the significant number of prediction failures renders it unsuitable for individual treatment decisions.

 PMID:9828740

  1. SU-F-BRD-11: Prediction of Dosimetric Endpoints From Patient Geometry Using Neural Nets

    SciTech Connect

    O'Connell, D; Chow, P; Agazaryan, N; Jani, S; Low, D; Lamb, J

    2014-06-15

    Purpose: The previously-published overlap volume histogram (OVH) technique lends itself naturally to prediction of the dose received by a given volume of tissue (e.g. D90) in intensity-modulated radiotherapy (IMRT) treatment plans. Here we extend the OVH technique using artificial neural networks in order to predict the volume of tissue receiving a given dose (e.g. V90) in both prostate IMRT and conventional breast radiotherapy. Methods: Twenty-nine prostate treatment plans and forty-three breast treatment plans were analyzed. The spatial relationships between the prostate and rectum and between the breast and ipsilateral lung were characterized using OVHs. The OVH is a cumulative histogram representing the fractional volume of the risk organ overlapped by a series of isotropic expansions of the planning target volume (PTV). Seven cases were identified as outliers and replanned. OVH points were used as inputs to a one hidden layer feed forward artificial neural network with quality parameters of the corresponding plan, such as the rectum V50, as targets. A 3-fold cross-validation was used to estimate the prediction error. Results: The root mean square (RMS) error between the predicted rectum V50s and the planned values was 2.3, which was 35% of the standard deviation of V50 for the twenty-nine plans. The RMS error of prediction of V20 of the ipsilateral lung in breast cases was 3.9, which was 90% of the standard deviation of the V20 values in the breast plan database. Conclusion: This study demonstrates that artificial neural nets can be used to extend the OVH technique to predict dosimetric endpoints taking the form of a volume receiving a given dose, rather than the minimum dose received by a given volume. Prediction of ipsilateral lung dose in breast radiotherapy using the OVH technique remains a work in progress.

  2. Neural networks: A versatile tool from artificial intelligence

    SciTech Connect

    Yama, B.R.; Lineberry, G.T.

    1996-12-31

    Artificial Intelligence research has produced several tools for commercial application in recent years. Artificial Neural Networks (ANNs), Fuzzy Logic, and Expert Systems are some of the techniques that are widely used today in various fields of engineering and business. Among these techniques, ANNs are gaining popularity due to their learning and other brain-like capabilities. Within the mining industry, ANN technology is being utilized with large payoffs for real-time process control applications. In this paper, a brief introduction to ANNs and the associated terminology is given. The neural network development process is outlined, followed by the back-propagation learning algorithm. Next, the development of two multi-layer, feed-forward neural networks is described and the results axe presented. One network is developed for prediction of strength of intact rock specimens, and another network is developed for prediction of mineral concentrations. Preliminary results indicate a predictive error less than 10% using cross-validation on a limited data set. The performance of the neural network for prediction of mineral concentrations was compared with kriging. It was found that the neural network performed not only satisfactorily, but in some cases performed better than, the kriging model.

  3. Connectionist and neural net implementations of a robotic grasp generator

    SciTech Connect

    Stansfield, S.A.

    1992-01-06

    This paper presents two parallel implementations of a knowledge-based robotic grasp generator. The grasp generator, originally developed as a rule-based system, embodies a knowledge of the association between the features of an object and the set of valid hand shapes/arm configurations which may be used to grasp it. Objects are assumed to be unknown, with no a priori models available. The first part of this paper presents a parallelization'' of this rule base using the connectionist paradigm. Rules are mapped into a set of nodes and connections which represent knowledge about object features, grasps, and the required conditions for a given grasp to be valid for a given set of features. Having shown that the object and knowledge representations lend themselves to this parallel recasting, the second part of the paper presents a back propagation neural net implementation of the system that allows the robot to learn the association between object features and appropriate grasps. 12 refs.

  4. Connectionist and neural net implementations of a robotic grasp generator

    SciTech Connect

    Stansfield, S.A.

    1992-01-06

    This paper presents two parallel implementations of a knowledge-based robotic grasp generator. The grasp generator, originally developed as a rule-based system, embodies a knowledge of the association between the features of an object and the set of valid hand shapes/arm configurations which may be used to grasp it. Objects are assumed to be unknown, with no a priori models available. The first part of this paper presents a ``parallelization`` of this rule base using the connectionist paradigm. Rules are mapped into a set of nodes and connections which represent knowledge about object features, grasps, and the required conditions for a given grasp to be valid for a given set of features. Having shown that the object and knowledge representations lend themselves to this parallel recasting, the second part of the paper presents a back propagation neural net implementation of the system that allows the robot to learn the association between object features and appropriate grasps. 12 refs.

  5. Numerical solution of differential equations by artificial neural networks

    NASA Technical Reports Server (NTRS)

    Meade, Andrew J., Jr.

    1995-01-01

    Conventionally programmed digital computers can process numbers with great speed and precision, but do not easily recognize patterns or imprecise or contradictory data. Instead of being programmed in the conventional sense, artificial neural networks (ANN's) are capable of self-learning through exposure to repeated examples. However, the training of an ANN can be a time consuming and unpredictable process. A general method is being developed by the author to mate the adaptability of the ANN with the speed and precision of the digital computer. This method has been successful in building feedforward networks that can approximate functions and their partial derivatives from examples in a single iteration. The general method also allows the formation of feedforward networks that can approximate the solution to nonlinear ordinary and partial differential equations to desired accuracy without the need of examples. It is believed that continued research will produce artificial neural networks that can be used with confidence in practical scientific computing and engineering applications.

  6. Artificial neural network based permanent magnet DC motor drives

    SciTech Connect

    Hoque, M.A. Zaman, M.R.; Rahman, M.A.

    1995-12-31

    A novel scheme for the speed control of a permanent magnet (PM) dc motor drive incorporating artificial neural network (ANN) is proposed. The drive system includes an ANN speed controller, micro-processor based dc-dc converter and a laboratory PM dc motor. A multi-layer artificial neural network structure with a feedback loop is designed in order to precisely operate the control circuit for the dc-dc converter. The complete drive system is simulated and implemented in real time. Both the simulation and experimental results prove the inherent capability of the ANN which makes it possible to maintain desired speed control in the presence of parameter variations and load disturbances. The performances of the ANN based PM dc drive system are compared with the simulated results of the conventionally controlled drive system. This clearly indicates the better performance of the ANN based PM dc motor drive system, particularly in case of parameter and load variations.

  7. Artificial neural networks technology for neutron spectrometry and dosimetry.

    PubMed

    Vega-Carrillo, H R; Hernández-Dávila, V M; Manzanares-Acuña, E; Gallego, E; Lorente, A; Iñiguez, M P

    2007-01-01

    Artificial Neural Network Technology has been applied to unfold neutron spectra and to calculate 13 dosimetric quantities using seven count rates from a Bonner Sphere Spectrometer with a (6)LiI(Eu). Two different networks, one for spectrometry and another for dosimetry, were designed. To train and test both networks, 177 neutron spectra from the IAEA compilation were utilised. Spectra were re-binned into 31 energy groups, and the dosimetric quantities were calculated using the MCNP code and the fluence-to-dose conversion coefficients from ICRP 74. Neutron spectra and UTA4 response matrix were used to calculate the expected count rates in the Bonner spectrometer. Spectra and H(10) of (239)PuBe and (241)AmBe were experimentally obtained and compared with those determined with the artificial neural networks. PMID:17522034

  8. Automated Wildfire Detection Through Artificial Neural Networks

    NASA Technical Reports Server (NTRS)

    Miller, Jerry; Borne, Kirk; Thomas, Brian; Huang, Zhenping; Chi, Yuechen

    2005-01-01

    Wildfires have a profound impact upon the biosphere and our society in general. They cause loss of life, destruction of personal property and natural resources and alter the chemistry of the atmosphere. In response to the concern over the consequences of wildland fire and to support the fire management community, the National Oceanic and Atmospheric Administration (NOAA), National Environmental Satellite, Data and Information Service (NESDIS) located in Camp Springs, Maryland gradually developed an operational system to routinely monitor wildland fire by satellite observations. The Hazard Mapping System, as it is known today, allows a team of trained fire analysts to examine and integrate, on a daily basis, remote sensing data from Geostationary Operational Environmental Satellite (GOES), Advanced Very High Resolution Radiometer (AVHRR) and Moderate Resolution Imaging Spectroradiometer (MODIS) satellite sensors and generate a 24 hour fire product for the conterminous United States. Although assisted by automated fire detection algorithms, N O M has not been able to eliminate the human element from their fire detection procedures. As a consequence, the manually intensive effort has prevented NOAA from transitioning to a global fire product as urged particularly by climate modelers. NASA at Goddard Space Flight Center in Greenbelt, Maryland is helping N O M more fully automate the Hazard Mapping System by training neural networks to mimic the decision-making process of the frre analyst team as well as the automated algorithms.

  9. Detection of electrocardiographic 'left ventricular strain' using neural nets.

    PubMed

    Devine, B; Macfarlane, P W

    1993-07-01

    The use of artificial neural networks for classification of ST-T abnormalities of the electrocardiogram (ECG) was investigated. A training set of 356 lateral leads selected from 105 ECGs was visually classified as exhibiting one particular ST-T morphology (left ventricular (LV) strain) or not. Selected measurements, together with the classification, were fed as input to a three-layer software-based network during the learning process. The performance of the network was evaluated by comparing the results obtained from the network with conventional criteria, using two test sets. Set 1 comprised 63 lateral leads from 32 ECGs with ST-T changes showing atypical forms of LV strain. Set 2 consisted of 80 lateral leads from 20 ECGs containing normal and abnormal T-waves. For set 1, the network outperformed conventional criteria, having a higher sensitivity (96 per cent against 85 per cent) and specificity (67 per cent against 50 per cent). With test set 2, both network and conventional criteria were 100 per cent sensitive and 100 per cent specific. For sets 1 and 2 combined, the network had a higher overall sensitivity (97 per cent against 89 per cent) and specificity (88 per cent against 82 per cent). The results suggest that neural networks may be useful in selected areas of electrocardiography, but care is required when selecting patterns for use in the training process.

  10. Charged particle track reconstruction using artificial neural networks

    SciTech Connect

    Glover, C.; Fu, P.; Gabriel, T.; Handler, T.

    1992-12-31

    This paper summarizes the current state of our research in developing and applying artificial neural network (ANN) algorithm described here is based on a crude model of the retina. It takes as input the coordinates of each charged particle`s interaction point (``hit``) in the tracking chamber. The algorithm`s output is a set of vectors pointing to other hits that most likely to form a track.

  11. Charged particle track reconstruction using artificial neural networks

    SciTech Connect

    Glover, C.; Fu, P.; Gabriel, T. ); Handler, T. . Dept. of Physics)

    1992-01-01

    This paper summarizes the current state of our research in developing and applying artificial neural network (ANN) algorithm described here is based on a crude model of the retina. It takes as input the coordinates of each charged particle's interaction point ( hit'') in the tracking chamber. The algorithm's output is a set of vectors pointing to other hits that most likely to form a track.

  12. Electricity price short-term forecasting using artificial neural networks

    SciTech Connect

    Szkuta, B.R.; Sanabria, L.A.; Dillon, T.S.

    1999-08-01

    This paper presents the System Marginal Price (SMP) short-term forecasting implementation using the Artificial Neural Networks (ANN) computing technique. The described approach uses the three-layered ANN paradigm with back-propagation. The retrospective SMP real-world data, acquired from the deregulated Victorian power system, was used for training and testing the ANN. The results presented in this paper confirm considerable value of the ANN based approach in forecasting the SMP.

  13. Confirmation of artificial neural networks: Nuclear power plant fault diagnostics

    SciTech Connect

    Kim, K.; Aljundi, T.L.; Barlett, E.B. )

    1992-01-01

    A fault diagnostics adviser was developed by training a backpropagation artificial neural network (ANN) to diagnose the status of the San Onofre nuclear generating station using data obtained from the plant's training simulator. These data simulate the plant's conditions during ten distinct transients. Stacked generalization is then used to confirm the diagnosis of the ANN. The network is capable of diagnosing each of the ten transients in a timely manner.

  14. Artificial neural network modeling of p-cresol photodegradation

    PubMed Central

    2013-01-01

    Background The complexity of reactions and kinetic is the current problem of photodegradation processes. Recently, artificial neural networks have been widely used to solve the problems because of their reliable, robust, and salient characteristics in capturing the non-linear relationships between variables in complex systems. In this study, an artificial neural network was applied for modeling p-cresol photodegradation. To optimize the network, the independent variables including irradiation time, pH, photocatalyst amount and concentration of p-cresol were used as the input parameters, while the photodegradation% was selected as output. The photodegradation% was obtained from the performance of the experimental design of the variables under UV irradiation. The network was trained by Quick propagation (QP) and the other three algorithms as a model. To determine the number of hidden layer nodes in the model, the root mean squared error of testing set was minimized. After minimizing the error, the topologies of the algorithms were compared by coefficient of determination and absolute average deviation. Results The comparison indicated that the Quick propagation algorithm had minimum root mean squared error, 1.3995, absolute average deviation, 3.0478, and maximum coefficient of determination, 0.9752, for the testing data set. The validation test results of the artificial neural network based on QP indicated that the root mean squared error was 4.11, absolute average deviation was 8.071 and the maximum coefficient of determination was 0.97. Conclusion Artificial neural network based on Quick propagation algorithm with topology 4-10-1 gave the best performance in this study. PMID:23731706

  15. Using Artificial Neural Networks to Assess Changes in Microbial Communities

    SciTech Connect

    Brandt, C.C.; Macnaughton, S.; Palumbo, A.V.; Pfiffner, S.M.; Schryver, J.C.

    1999-04-19

    We evaluated artificial neural networks (ANNs) as a technique for assessing changes in soil microbial communities following exposure to metals. We analyzed signature lipid biomarker (SLB) data collected from two soil microcosm experiments using traditional statistical techniques and ANN. Two phases of data analysis were done; pattern recognition and prediction. In general, the ANNs were better able to detect patterns and relationships in the SLB data than were the traditional statistical techniques.

  16. Homology-based gene prediction using neural nets.

    PubMed

    Cai, Y; Bork, P

    1998-12-15

    We have developed and implemented a method for computational gene identification called GIN (gene identification using neural nets and homology information) that has been particularly designed to avoid false positive predictions. It thus predicts 55% of all genes tested correctly, has a specificity of 99%, but also has an overall accuracy of 92% on a benchmark set of 570 vertebrate genes constructed by Burset and Guigo. The method combines homology searches in protein and expressed sequence tag databases with several neural networks designed to recognize start codons, Poly(A) signals, stop codons, and splice sites. Predicted exons are assembled into genes using a homology-based scoring function. GIN is able to recognize multiple genes within genomic DNA as demonstrated by the identification of a globin gene (gamma-globin-1(G)) that has not been annotated as a coding region in the widely used the test set of Burset and Guigo. Furthermore, GIN identifies more than 107 other protein hits in noncoding regions and classifies them into possible pseudogenes or splice variants.

  17. The importance of artificial neural networks in biomedicine

    SciTech Connect

    Burke, H.B.

    1995-12-31

    The future explanatory power in biomedicine will be at the molecular-genetic level of analysis (rather than the epidemiologic-demographic or anatomic-cellular levels). This is the level of complex systems. Complex systems are characterized by nonlinearity and complex interactions. It is difficult for traditional statistical methods to capture complex systems because traditional methods attempt to find the model that best fits the statistician`s understanding of the phenomenon; complex systems are difficult to understand and therefore difficult to fit with a simple model. Artificial neural networks are nonparametric regression models. They can capture any phenomena, to any degree of accuracy (depending on the adequacy of the data and the power of the predictors), without prior knowledge of the phenomena. Further, artificial neural networks can be represented, not only as formulae, but also as graphical models. Graphical models can increase analytic power and flexibility. Artificial neural networks are a powerful method for capturing complex phenomena, but their use requires a paradigm shift, from exploratory analysis of the data to exploratory analysis of the model.

  18. A comparison of artificial neural networks and statistical analyses

    SciTech Connect

    Blough, D.K.; Anderson, K.K.

    1994-01-01

    Artificial neural networks have come to be used in a wide variety of data analytic applications, many of which were traditionally approached using statistical methods. It is the purpose of this paper to discuss the nature of the information obtained by each methodology, that of artificial neural networks and that of statistical analyses. Two aspects of the comparison will be considered: (1) what are the requirements needed for each approach in terms of model specification, data requirements, and computing power, and (2) what sort of information is contained in the results of each approach. Example analyses are presented characterizing the differences in the two approaches. A specific problem (hydrodynamic yield estimation) is presented with a corresponding data set. This data is then analyzed using statistical methods, and the results are compared with those obtained by using an artificial neural network. The requirements and results of the two approaches are then summarized as general guidelines an investigator can use in deciding which approach would be best for analyzing a given data set.

  19. Neural-Net Based Optical NDE Method for Structural Health Monitoring

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J.; Weiland, Kenneth E.

    2003-01-01

    This paper answers some performance and calibration questions about a non-destructive-evaluation (NDE) procedure that uses artificial neural networks to detect structural damage or other changes from sub-sampled characteristic patterns. The method shows increasing sensitivity as the number of sub-samples increases from 108 to 6912. The sensitivity of this robust NDE method is not affected by noisy excitations of the first vibration mode. A calibration procedure is proposed and demonstrated where the output of a trained net can be correlated with the outputs of the point sensors used for vibration testing. The calibration procedure is based on controlled changes of fastener torques. A heterodyne interferometer is used as a displacement sensor for a demonstration of the challenges to be handled in using standard point sensors for calibration.

  20. A TLD dose algorithm using artificial neural networks

    SciTech Connect

    Moscovitch, M.; Rotunda, J.E.; Tawil, R.A.; Rathbone, B.A.

    1995-12-31

    An artificial neural network was designed and used to develop a dose algorithm for a multi-element thermoluminescence dosimeter (TLD). The neural network architecture is based on the concept of functional links network (FLN). Neural network is an information processing method inspired by the biological nervous system. A dose algorithm based on neural networks is fundamentally different as compared to conventional algorithms, as it has the capability to learn from its own experience. The neural network algorithm is shown the expected dose values (output) associated with given responses of a multi-element dosimeter (input) many times. The algorithm, being trained that way, eventually is capable to produce its own unique solution to similar (but not exactly the same) dose calculation problems. For personal dosimetry, the output consists of the desired dose components: deep dose, shallow dose and eye dose. The input consists of the TL data obtained from the readout of a multi-element dosimeter. The neural network approach was applied to the Harshaw Type 8825 TLD, and was shown to significantly improve the performance of this dosimeter, well within the U.S. accreditation requirements for personnel dosimeters.

  1. Artificial neural network modeling of dissolved oxygen in reservoir.

    PubMed

    Chen, Wei-Bo; Liu, Wen-Cheng

    2014-02-01

    The water quality of reservoirs is one of the key factors in the operation and water quality management of reservoirs. Dissolved oxygen (DO) in water column is essential for microorganisms and a significant indicator of the state of aquatic ecosystems. In this study, two artificial neural network (ANN) models including back propagation neural network (BPNN) and adaptive neural-based fuzzy inference system (ANFIS) approaches and multilinear regression (MLR) model were developed to estimate the DO concentration in the Feitsui Reservoir of northern Taiwan. The input variables of the neural network are determined as water temperature, pH, conductivity, turbidity, suspended solids, total hardness, total alkalinity, and ammonium nitrogen. The performance of the ANN models and MLR model was assessed through the mean absolute error, root mean square error, and correlation coefficient computed from the measured and model-simulated DO values. The results reveal that ANN estimation performances were superior to those of MLR. Comparing to the BPNN and ANFIS models through the performance criteria, the ANFIS model is better than the BPNN model for predicting the DO values. Study results show that the neural network particularly using ANFIS model is able to predict the DO concentrations with reasonable accuracy, suggesting that the neural network is a valuable tool for reservoir management in Taiwan. PMID:24078053

  2. Science of artificial neural networks; Proceedings of the Meeting, Orlando, FL, Apr. 21-24, 1992

    SciTech Connect

    Ruck, D.W.

    1992-01-01

    The present conference discusses high-order neural networks with adaptive architecture, a parallel cascaded one-step learning machine, stretch and hammer neural networks, visual grammars for neural networks, the net pruning of a multilayer perceptron, neural correlates of the sensorial and cognitive control of behavior, neural nets for massively parallel optimization, parametric and additive perturbations for global optimization, design rules for multilayer perceptrons, the negative transfer problem in neural networks, and a vision-based neural multimap pattern recognition architecture. Also discussed are function prediction with recurrent neural networks, fuzzy neural computing systems, edge detection via fuzzy neural networks, modeling confusion for autonomous systems, self-organization by fuzzy clustering, neural nets in information retrieval, neighborhoods and trajectories in Kohonen maps, the random structure of error surfaces, and conceptual recognition by neural networks.

  3. Background considerations in the analysis of PIXE spectra by Artificial Neural Systems.

    NASA Astrophysics Data System (ADS)

    Correa, R.; Morales, J. R.; Requena, I.; Miranda, J.; Barrera, V. A.

    2016-05-01

    In order to study the importance of background in PIXE spectra to determine elemental concentrations in atmospheric aerosols using artificial neural systems ANS, two independently trained ANS were constructed, one which considered as input the net number of counts in the peak, and another which included the background. In the training and validation phases thirty eight spectra of aerosols collected in Santiago, Chile, were used. In both cases the elemental concentration values were similar. This fact was due to the intrinsic characteristic of ANS operating with normalized values of the net and total number of counts under the peaks, something that was verified in the analysis of 172 spectra obtained from aerosols collected in Mexico city. Therefore, networks operating under the mode which include background can reduce time and cost when dealing with large number of samples.

  4. Neural-net based coordinated stabilizing control for the exciter and governor loops of low head hydropower plants

    SciTech Connect

    Djukanovic, M.; Novicevic, M.; Dobrijevic, D.; Babic, B.; Sobajic, D.J.; Pao, Y.H. |

    1995-12-01

    This paper presents a design technique of a new adaptive optimal controller of the low head hydropower plant using artificial neural networks (ANN). The adaptive controller is to operate in real time to improve the generating unit transients through the exciter input, the guide vane position and the runner blade position. The new design procedure is based on self-organization and the predictive estimation capabilities of neural-nets implemented through the cluster-wise segmented associative memory scheme. The developed neural-net based controller (NNC) whose control signals are adjusted using the on-line measurements, can offer better damping effects for generator oscillations over a wide range of operating conditions than conventional controllers. Digital simulations of hydropower plant equipped with low head Kaplan turbine are performed and the comparisons of conventional excitation-governor control, state-space optimal control and neural-net based control are presented. Results obtained on the non-linear mathematical model demonstrate that the effects of the NNC closely agree with those obtained using the state-space multivariable discrete-time optimal controllers.

  5. Control of Wind Tunnel Operations Using Neural Net Interpretation of Flow Visualization Records

    NASA Technical Reports Server (NTRS)

    Buggele, Alvin E.; Decker, Arthur J.

    1994-01-01

    Neural net control of operations in a small subsonic/transonic/supersonic wind tunnel at Lewis Research Center is discussed. The tunnel and the layout for neural net control or control by other parallel processing techniques are described. The tunnel is an affordable, multiuser platform for testing instrumentation and components, as well as parallel processing and control strategies. Neural nets have already been tested on archival schlieren and holographic visualizations from this tunnel as well as recent supersonic and transonic shadowgraph. This paper discusses the performance of neural nets for interpreting shadowgraph images in connection with a recent exercise for tuning the tunnel in a subsonic/transonic cascade mode of operation. That mode was operated for performing wake surveys in connection with NASA's Advanced Subsonic Technology (AST) noise reduction program. The shadowgraph was presented to the neural nets as 60 by 60 pixel arrays. The outputs were tunnel parameters such as valve settings or tunnel state identifiers for selected tunnel operating points, conditions, or states. The neural nets were very sensitive, perhaps too sensitive, to shadowgraph pattern detail. However, the nets exhibited good immunity to variations in brightness, to noise, and to changes in contrast. The nets are fast enough so that ten or more can be combined per control operation to interpret flow visualization data, point sensor data, and model calculations. The pattern sensitivity of the nets will be utilized and tested to control wind tunnel operations at Mach 2.0 based on shock wave patterns.

  6. Predicting stream water quality using artificial neural networks (ANN)

    SciTech Connect

    Bowers, J.A.

    2000-05-17

    Predicting point and nonpoint source runoff of dissolved and suspended materials into their receiving streams is important to protecting water quality and traditionally has been modeled using deterministic or statistical methods. The purpose of this study was to predict water quality in small streams using an Artificial Neural Network (ANN). The selected input variables were local precipitation, stream flow rates and turbidity for the initial prediction of suspended solids in the stream. A single hidden-layer feedforward neural network using backpropagation learning algorithms was developed with a detailed analysis of model design of those factors affecting successful implementation of the model. All features of a feedforward neural model were investigated including training set creation, number and layers of neurons, neural activation functions, and backpropagation algorithms. Least-squares regression was used to compare model predictions with test data sets. Most of the model configurations offered excellent predictive capabilities. Using either the logistic or the hyperbolic tangent neural activation function did not significantly affect predicted results. This was also true for the two learning algorithms tested, the Levenberg-Marquardt and Polak-Ribiere conjugate-gradient descent methods. The most important step during model development and training was the representative selection of data records for training of the model.

  7. Evaluating the performance of two neutron spectrum unfolding codes based on iterative procedures and artificial neural networks

    NASA Astrophysics Data System (ADS)

    Ortiz-Rodríguez, J. M.; Reyes Alfaro, A.; Reyes Haro, A.; Solís Sánches, L. O.; Miranda, R. Castañeda; Cervantes Viramontes, J. M.; Vega-Carrillo, H. R.

    2013-07-01

    In this work the performance of two neutron spectrum unfolding codes based on iterative procedures and artificial neural networks is evaluated. The first one code based on traditional iterative procedures and called Neutron spectrometry and dosimetry from the Universidad Autonoma de Zacatecas (NSDUAZ) use the SPUNIT iterative algorithm and was designed to unfold neutron spectrum and calculate 15 dosimetric quantities and 7 IAEA survey meters. The main feature of this code is the automated selection of the initial guess spectrum trough a compendium of neutron spectrum compiled by the IAEA. The second one code known as Neutron spectrometry and dosimetry with artificial neural networks (NDSann) is a code designed using neural nets technology. The artificial intelligence approach of neural net does not solve mathematical equations. By using the knowledge stored at synaptic weights on a neural net properly trained, the code is capable to unfold neutron spectrum and to simultaneously calculate 15 dosimetric quantities, needing as entrance data, only the rate counts measured with a Bonner spheres system. Similarities of both NSDUAZ and NSDann codes are: they follow the same easy and intuitive user's philosophy and were designed in a graphical interface under the LabVIEW programming environment. Both codes unfold the neutron spectrum expressed in 60 energy bins, calculate 15 dosimetric quantities and generate a full report in HTML format. Differences of these codes are: NSDUAZ code was designed using classical iterative approaches and needs an initial guess spectrum in order to initiate the iterative procedure. In NSDUAZ, a programming routine was designed to calculate 7 IAEA instrument survey meters using the fluence-dose conversion coefficients. NSDann code use artificial neural networks for solving the ill-conditioned equation system of neutron spectrometry problem through synaptic weights of a properly trained neural network. Contrary to iterative procedures, in neural

  8. Evaluating the performance of two neutron spectrum unfolding codes based on iterative procedures and artificial neural networks

    SciTech Connect

    Ortiz-Rodriguez, J. M.; Reyes Alfaro, A.; Reyes Haro, A.; Solis Sanches, L. O.; Miranda, R. Castaneda; Cervantes Viramontes, J. M.; Vega-Carrillo, H. R.

    2013-07-03

    In this work the performance of two neutron spectrum unfolding codes based on iterative procedures and artificial neural networks is evaluated. The first one code based on traditional iterative procedures and called Neutron spectrometry and dosimetry from the Universidad Autonoma de Zacatecas (NSDUAZ) use the SPUNIT iterative algorithm and was designed to unfold neutron spectrum and calculate 15 dosimetric quantities and 7 IAEA survey meters. The main feature of this code is the automated selection of the initial guess spectrum trough a compendium of neutron spectrum compiled by the IAEA. The second one code known as Neutron spectrometry and dosimetry with artificial neural networks (NDSann) is a code designed using neural nets technology. The artificial intelligence approach of neural net does not solve mathematical equations. By using the knowledge stored at synaptic weights on a neural net properly trained, the code is capable to unfold neutron spectrum and to simultaneously calculate 15 dosimetric quantities, needing as entrance data, only the rate counts measured with a Bonner spheres system. Similarities of both NSDUAZ and NSDann codes are: they follow the same easy and intuitive user's philosophy and were designed in a graphical interface under the LabVIEW programming environment. Both codes unfold the neutron spectrum expressed in 60 energy bins, calculate 15 dosimetric quantities and generate a full report in HTML format. Differences of these codes are: NSDUAZ code was designed using classical iterative approaches and needs an initial guess spectrum in order to initiate the iterative procedure. In NSDUAZ, a programming routine was designed to calculate 7 IAEA instrument survey meters using the fluence-dose conversion coefficients. NSDann code use artificial neural networks for solving the ill-conditioned equation system of neutron spectrometry problem through synaptic weights of a properly trained neural network. Contrary to iterative procedures, in neural

  9. Lexical Tone Recognition with an Artificial Neural Network

    PubMed Central

    Zhou, Ning; Zhang, Wenle; Lee, Chao-Yang; Xu, Li

    2008-01-01

    Objectives Tone production is particularly important for communicating in tone languages such as Mandarin Chinese. In the present study, an artificial neural network was used to recognize tones produced by adult native speakers. The purposes of the study were (1) to test the sensitivity of the neural network to speaker variation typically in adult speaker groups, (2) to evaluate two normalization procedures to overcome the effects of speaker variation, and (3) to compare tone recognition performance of the neural network with that of the human listeners. Design A feedforward multilayer neural network was used. Twenty-nine adult native Mandarin Chinese speakers were recruited to record tone samples. The F0 contours of the vowel part of the 1044 monosyllabic words recorded were extracted using an autocorrelation method. Samples from the F0 contours were used as inputs to the neural network. The efficacy of the neural network was first tested by varying the number of inputs and the number of neurons in the hidden layer from 1 to 16. The sensitivity of the neural network to speaker variation was tested by (1) using the raw F0 data from speech tokens of a number of randomly drawn speakers that varied from 1 to 29, (2) using the raw F0 data from speech tokens of either male-only or female-only speakers, and (3) using two sets of normalized F0 data (i.e., tone 1-based normalization and first-order derivative) from speech tokens from a number of randomly drawn speakers that varied from 1 to 29. The recognition performance of the neural network under several experimental conditions was compared with the corresponding recognition performance of 10 normal-hearing, native Mandarin Chinese speaking adult listeners. Results Three inputs and four hidden neurons were found to be sufficient for the neural network to perform at about 85% correct using speech samples without normalization. The performance of the neural network was affected by variation across speakers particularly

  10. Development of a neural net paradigm that predicts simulator sickness

    SciTech Connect

    Allgood, G.O.

    1993-03-01

    A disease exists that affects pilots and aircrew members who use Navy Operational Flight Training Systems. This malady, commonly referred to as simulator sickness and whose symptomatology closely aligns with that of motion sickness, can compromise the use of these systems because of a reduced utilization factor, negative transfer of training, and reduction in combat readiness. A report is submitted that develops an artificial neural network (ANN) and behavioral model that predicts the onset and level of simulator sickness in the pilots and aircrews who sue these systems. It is proposed that the paradigm could be implemented in real time as a biofeedback monitor to reduce the risk to users of these systems. The model captures the neurophysiological impact of use (human-machine interaction) by developing a structure that maps the associative and nonassociative behavioral patterns (learned expectations) and vestibular (otolith and semicircular canals of the inner ear) and tactile interaction, derived from system acceleration profiles, onto an abstract space that predicts simulator sickness for a given training flight.

  11. The Development of Animal Behavior: From Lorenz to Neural Nets

    NASA Astrophysics Data System (ADS)

    Bolhuis, Johan J.

    In the study of behavioral development both causal and functional approaches have been used, and they often overlap. The concept of ontogenetic adaptations suggests that each developmental phase involves unique adaptations to the environment of the developing animal. The functional concept of optimal outbreeding has led to further experimental evidence and theoretical models concerning the role of sexual imprinting in the evolutionary process of sexual selection. From a causal perspective it has been proposed that behavioral ontogeny involves the development of various kinds of perceptual, motor, and central mechanisms and the formation of connections among them. This framework has been tested for a number of complex behavior systems such as hunger and dustbathing. Imprinting is often seen as a model system for behavioral development in general. Recent advances in imprinting research have been the result of an interdisciplinary effort involving ethology, neuroscience, and experimental psychology, with a continual interplay between these approaches. The imprinting results are consistent with Lorenz' early intuitive suggestions and are also reflected in the architecture of recent neural net models.

  12. The development of animal behavior: from Lorenz to neural nets.

    PubMed

    Bolhuis, J J

    1999-03-01

    In the study of behavioral development both causal and functional approaches have been used, and they often overlap. The concept of ontogenetic adaptations suggests that each developmental phase involves unique adaptations to the environment of the developing animal. The functional concept of optimal outbreeding has led to further experimental evidence and theoretical models concerning the role of sexual imprinting in the evolutionary process of sexual selection. From a causal perspective it has been proposed that behavioral ontogeny involves the development of various kinds of perceptual, motor, and central mechanisms and the formation of connections among them. This framework has been tested for a number of complex behavior systems such as hunger and dustbathing. Imprinting is often seen as a model system for behavioral development in general. Recent advances in imprinting research have been the result of an interdisciplinary effort involving ethology, neuroscience, and experimental psychology, with a continual interplay between these approaches. The imprinting results are consistent with Lorenz' early intuitive suggestions and are also reflected in the architecture of recent neural net models.

  13. Disparity tuning as simulated by a neural net.

    PubMed

    Lippert, J; Fleet, D J; Wagner, H

    2000-07-01

    Previous research has suggested that the processing of binocular disparity in complex cells may be described with an energy formalism. The energy formalism allows for a representation of disparity by differences in the position or in the phase of monocular receptive subfields of binocular cells, or by combination of these two types. We studied the coding of disparities with an approach complementary to previous algorithmic investigations. Since realization of these representations is probably not genetically determined but learned during ontogeny, we used backpropagation networks to study which of these three possibilities were realized within neural nets. Three types of networks were trained with noise patterns in analogy to the three types of energy models. The networks learned the task and generalized to untrained correlated noise pattern input. Outputs were broadly tuned to spatial frequency and did not respond to anti-correlated noise patterns. Although the energy model was not explicitly implemented, we could analyze the outputs of the networks using predictions of the energy formalism. After learning was completed, the model neurons preferred position shifts over phase shifts in representing disparity. We discuss the general meaning of these findings and the correspondences and deviations between the energy model, V1 neurons, and our networks.

  14. Design of Jetty Piles Using Artificial Neural Networks

    PubMed Central

    2014-01-01

    To overcome the complication of jetty pile design process, artificial neural networks (ANN) are adopted. To generate the training samples for training ANN, finite element (FE) analysis was performed 50 times for 50 different design cases. The trained ANN was verified with another FE analysis case and then used as a structural analyzer. The multilayer neural network (MBPNN) with two hidden layers was used for ANN. The framework of MBPNN was defined as the input with the lateral forces on the jetty structure and the type of piles and the output with the stress ratio of the piles. The results from the MBPNN agree well with those from FE analysis. Particularly for more complex modes with hundreds of different design cases, the MBPNN would possibly substitute parametric studies with FE analysis saving design time and cost. PMID:25177724

  15. Short-term load forecasting using an artificial neural network

    SciTech Connect

    Lee, K.Y.; Cha, Y.T. ); Park, J.H. )

    1992-02-01

    Artificial Neural Network (ANN) Method is applied to forecast the short-term load for a large power system. The load has two distinct patterns: weekday and weekend-day patterns. The weekend-day pattern include Saturday, Sunday, and Monday loads. In this paper a nonlinear load model is proposed and several structures of ANN for short-term load forecasting are tested. Inputs to the ANN are past loads and the output of the ANN is the load forecast for a given day. The network with one or two hidden layers are tested with various combination of neurons, and results are compared in terms of forecasting error. The neural network, when grouped into different load patterns, gives good load forecast.

  16. Fault Tolerant Characteristics of Artificial Neural Network Electronic Hardware

    NASA Technical Reports Server (NTRS)

    Zee, Frank

    1995-01-01

    The fault tolerant characteristics of analog-VLSI artificial neural network (with 32 neurons and 532 synapses) chips are studied by exposing them to high energy electrons, high energy protons, and gamma ionizing radiations under biased and unbiased conditions. The biased chips became nonfunctional after receiving a cumulative dose of less than 20 krads, while the unbiased chips only started to show degradation with a cumulative dose of over 100 krads. As the total radiation dose increased, all the components demonstrated graceful degradation. The analog sigmoidal function of the neuron became steeper (increase in gain), current leakage from the synapses progressively shifted the sigmoidal curve, and the digital memory of the synapses and the memory addressing circuits began to gradually fail. From these radiation experiments, we can learn how to modify certain designs of the neural network electronic hardware without using radiation-hardening techniques to increase its reliability and fault tolerance.

  17. Fuzzy logic -- artificial neural networks integration for transient identification

    SciTech Connect

    Ikonomopoulos, A.; Tsoukalas, L.H. . Dept. of Nuclear Engineering); Uhrig, R.E. . Dept. of Nuclear Engineering Oak Ridge National Lab., TN )

    1991-01-01

    A methodology is presented that integrates pretrained artificial neural networks (ANNs) with rule-based fuzzy logic systems, for the purpose of distinguishing different transients in a Nuclear Power Plant (NPP). In general this approach appears to provide timely, concise and task specific information about the status of a system under consideration. The pretrained neural network typifies different transient scenarios and derives membership functions which independently represent individual transients. The overall system successfully performs transient identification, in a time span faster or at least comparable to that of transient development. In order to examine the proposed methodology simulated accidents are used. The results obtained demonstrate the excellent noise tolerance of ANNs and suggest a new approach for transient identification.

  18. Fuzzy logic -- artificial neural networks integration for transient identification

    SciTech Connect

    Ikonomopoulos, A.; Tsoukalas, L.H.; Uhrig, R.E. |

    1991-12-31

    A methodology is presented that integrates pretrained artificial neural networks (ANNs) with rule-based fuzzy logic systems, for the purpose of distinguishing different transients in a Nuclear Power Plant (NPP). In general this approach appears to provide timely, concise and task specific information about the status of a system under consideration. The pretrained neural network typifies different transient scenarios and derives membership functions which independently represent individual transients. The overall system successfully performs transient identification, in a time span faster or at least comparable to that of transient development. In order to examine the proposed methodology simulated accidents are used. The results obtained demonstrate the excellent noise tolerance of ANNs and suggest a new approach for transient identification.

  19. Nuclear power plant fault-diagnosis using artificial neural networks

    SciTech Connect

    Kim, Keehoon; Aljundi, T.L.; Bartlett, E.B.

    1992-01-01

    Artificial neural networks (ANNs) have been applied to various fields due to their fault and noise tolerance and generalization characteristics. As an application to nuclear engineering, we apply neural networks to the early recognition of nuclear power plant operational transients. If a transient or accident occurs, the network will advise the plant operators in a timely manner. More importantly, we investigate the ability of the network to provide a measure of the confidence level in its diagnosis. In this research an ANN is trained to diagnose the status of the San Onofre Nuclear Generation Station using data obtained from the plant's training simulator. Stacked generalization is then applied to predict the error in the ANN diagnosis. The data used consisted of 10 scenarios that include typical design basis accidents as well as less severe transients. The results show that the trained network is capable of diagnosing all 10 instabilities as well as providing a measure of the level of confidence in its diagnoses.

  20. Artificial Neural Network for Location Estimation in Wireless Communication Systems

    PubMed Central

    Chen, Chien-Sheng

    2012-01-01

    In a wireless communication system, wireless location is the technique used to estimate the location of a mobile station (MS). To enhance the accuracy of MS location prediction, we propose a novel algorithm that utilizes time of arrival (TOA) measurements and the angle of arrival (AOA) information to locate MS when three base stations (BSs) are available. Artificial neural networks (ANN) are widely used techniques in various areas to overcome the problem of exclusive and nonlinear relationships. When the MS is heard by only three BSs, the proposed algorithm utilizes the intersections of three TOA circles (and the AOA line), based on various neural networks, to estimate the MS location in non-line-of-sight (NLOS) environments. Simulations were conducted to evaluate the performance of the algorithm for different NLOS error distributions. The numerical analysis and simulation results show that the proposed algorithms can obtain more precise location estimation under different NLOS environments. PMID:22736978

  1. Communication: Separable potential energy surfaces from multiplicative artificial neural networks

    SciTech Connect

    Koch, Werner Zhang, Dong H.

    2014-07-14

    We present a potential energy surface fitting scheme based on multiplicative artificial neural networks. It has the sum of products form required for efficient computation of the dynamics of multidimensional quantum systems with the multi configuration time dependent Hartree method. Moreover, it results in analytic potential energy matrix elements when combined with quantum dynamics methods using Gaussian basis functions, eliminating the need for a local harmonic approximation. Scaling behavior with respect to the complexity of the potential as well as the requested accuracy is discussed.

  2. Video data compression using artificial neural network differential vector quantization

    NASA Technical Reports Server (NTRS)

    Krishnamurthy, Ashok K.; Bibyk, Steven B.; Ahalt, Stanley C.

    1991-01-01

    An artificial neural network vector quantizer is developed for use in data compression applications such as Digital Video. Differential Vector Quantization is used to preserve edge features, and a new adaptive algorithm, known as Frequency-Sensitive Competitive Learning, is used to develop the vector quantizer codebook. To develop real time performance, a custom Very Large Scale Integration Application Specific Integrated Circuit (VLSI ASIC) is being developed to realize the associative memory functions needed in the vector quantization algorithm. By using vector quantization, the need for Huffman coding can be eliminated, resulting in superior performance against channel bit errors than methods that use variable length codes.

  3. Error bounds on the output of artificial neural networks

    SciTech Connect

    Bartlett, E.B.; Kim, H. )

    1993-01-01

    Resolving the uncertainties associated with solutions obtained from artificial neural networks (ANNs) is a major concern for ANN researchers. Error bounds on the solutions are important because they are an integral part of verification and validation. In this research, stacked generalization (SG) is applied to provide error bounds for novel solutions obtained from ANNS. An outline of SG and its use is given. The data used in this demonstration of SG are given. This work shows that SG can provide error bounds on ANN results. We have applied SG to nuclear power plant fault detection for verification of diagnoses provided by ANNs.

  4. Stress calculation of crankshaft using artificial neural network

    SciTech Connect

    Shiomi, Kazuyuki; Watanabe, Sei

    1995-12-31

    A system that calculates the stress concentration factor of the crankpin fillet from six characteristic dimensions of the crankshaft was developed using an artificial neural network. The learning database was constructed based on the finite element analysis, and an ``adaptive transfer function algorithm`` was used for the learning calculations. The calculation errors of the stress concentration factors applied to crankshafts of small utility engines and outboard motors were found to be within {minus}6.9 to +6.3% of the measured values. With this system, designers can calculate the stress concentrated at crankpin fillets precisely in a short time.

  5. Adaptive conventional power system stabilizer based on artificial neural network

    SciTech Connect

    Kothari, M.L.; Segal, R.; Ghodki, B.K.

    1995-12-31

    This paper deals with an artificial neural network (ANN) based adaptive conventional power system stabilizer (PSS). The ANN comprises an input layer, a hidden layer and an output layer. The input vector to the ANN comprises real power (P) and reactive power (Q), while the output vector comprises optimum PSS parameters. A systematic approach for generating training set covering wide range of operating conditions, is presented. The ANN has been trained using back-propagation training algorithm. Investigations reveal that the dynamic performance of ANN based adaptive conventional PSS is quite insensitive to wide variations in loading conditions.

  6. Simulation of nonlinear strutures with artificial neural networks

    SciTech Connect

    Paez, T.L.

    1996-03-01

    Structural system simulation is important in analysis, design, testing, control, and other areas, but it is particularly difficult when the system under consideration is nonlinear. Artificial neural networks offer a useful tool for the modeling of nonlinear systems, however, such modeling may be inefficient or insufficiently accurate when the system under consideration is complex. This paper shows that there are several transformations that can be used to uncouple and simplify the components of motion of a complex nonlinear system, thereby making its modeling and simulation a much simpler problem. A numerical example is also presented.

  7. Using Artificial Neural Networks to Assess Microbial Communities

    SciTech Connect

    Almeida, J.S.; Brand, C.C.; Palumbo, A.V.; Pfiffner, S.M.; Schryver, J.C.

    1998-09-08

    We are evaluating artificial neural networks (ANNs) as tools for assessing changes in soil microbial communities following exposure to metals. We analyzed signature lipid biomarker data collected from two soil microcosm experiments using an autoassociative ANN. In one experiment, the microcosms were exposed to O, 100, or 250 ppm of metals, and in the other experiment the microcosms were exposed to O or 500 ppm of metals. The ANNs were able to distinguish between microcosms exposed and not exposed to metals in both experiments.

  8. Analog neural nets with gaussian or other common noise distribution cannot recognize arbitrary regular languages.

    PubMed

    Maass, W; Sontag, E D

    1999-04-01

    We consider recurrent analog neural nets where the output of each gate is subject to gaussian noise or any other common noise distribution that is nonzero on a sufficiently large part of the state-space. We show that many regular languages cannot be recognized by networks of this type, and we give a precise characterization of languages that can be recognized. This result implies severe constraints on possibilities for constructing recurrent analog neural nets that are robust against realistic types of analog noise. On the other hand, we present a method for constructing feedforward analog neural nets that are robust with regard to analog noise of this type.

  9. Hybrid multiobjective evolutionary design for artificial neural networks.

    PubMed

    Goh, Chi-Keong; Teoh, Eu-Jin; Tan, Kay Chen

    2008-09-01

    Evolutionary algorithms are a class of stochastic search methods that attempts to emulate the biological process of evolution, incorporating concepts of selection, reproduction, and mutation. In recent years, there has been an increase in the use of evolutionary approaches in the training of artificial neural networks (ANNs). While evolutionary techniques for neural networks have shown to provide superior performance over conventional training approaches, the simultaneous optimization of network performance and architecture will almost always result in a slow training process due to the added algorithmic complexity. In this paper, we present a geometrical measure based on the singular value decomposition (SVD) to estimate the necessary number of neurons to be used in training a single-hidden-layer feedforward neural network (SLFN). In addition, we develop a new hybrid multiobjective evolutionary approach that includes the features of a variable length representation that allow for easy adaptation of neural networks structures, an architectural recombination procedure based on the geometrical measure that adapts the number of necessary hidden neurons and facilitates the exchange of neuronal information between candidate designs, and a microhybrid genetic algorithm ( microHGA) with an adaptive local search intensity scheme for local fine-tuning. In addition, the performances of well-known algorithms as well as the effectiveness and contributions of the proposed approach are analyzed and validated through a variety of data set types.

  10. Prediction aluminum corrosion inhibitor efficiency using artificial neural network (ANN)

    NASA Astrophysics Data System (ADS)

    Ebrahimi, Sh; Kalhor, E. G.; Nabavi, S. R.; Alamiparvin, L.; Pogaku, R.

    2016-06-01

    In this study, activity of some Schiff bases as aluminum corrosion inhibitor was investigated using artificial neural network (ANN). Hence, corrosion inhibition efficiency of Schiff bases (in any type) were gathered from different references. Then these molecules were drawn and optimized in Hyperchem software. Molecular descriptors generating and descriptors selection were fulfilled by Dragon software and principal component analysis (PCA) method, respectively. These structural descriptors along with environmental descriptors (ambient temperature, time of exposure, pH and the concentration of inhibitor) were used as input variables. Furthermore, aluminum corrosion inhibition efficiency was used as output variable. Experimental data were split into three sets: training set (for model building) and test set (for model validation) and simulation (for general model). Modeling was performed by Multiple linear regression (MLR) methods and artificial neural network (ANN). The results obtained in linear models showed poor correlation between experimental and theoretical data. However nonlinear model presented satisfactory results. Higher correlation coefficient of ANN (R > 0.9) revealed that ANN can be successfully applied for prediction of aluminum corrosion inhibitor efficiency of Schiff bases in different environmental conditions.

  11. Applications of artificial neural networks in medical science.

    PubMed

    Patel, Jigneshkumar L; Goyal, Ramesh K

    2007-09-01

    Computer technology has been advanced tremendously and the interest has been increased for the potential use of 'Artificial Intelligence (AI)' in medicine and biological research. One of the most interesting and extensively studied branches of AI is the 'Artificial Neural Networks (ANNs)'. Basically, ANNs are the mathematical algorithms, generated by computers. ANNs learn from standard data and capture the knowledge contained in the data. Trained ANNs approach the functionality of small biological neural cluster in a very fundamental manner. They are the digitized model of biological brain and can detect complex nonlinear relationships between dependent as well as independent variables in a data where human brain may fail to detect. Nowadays, ANNs are widely used for medical applications in various disciplines of medicine especially in cardiology. ANNs have been extensively applied in diagnosis, electronic signal analysis, medical image analysis and radiology. ANNs have been used by many authors for modeling in medicine and clinical research. Applications of ANNs are increasing in pharmacoepidemiology and medical data mining. In this paper, authors have summarized various applications of ANNs in medical science.

  12. Landslide susceptibility analysis using an artificial neural network model

    NASA Astrophysics Data System (ADS)

    Mansor, Shattri; Pradhan, Biswajeet; Daud, Mohamed; Jamaludin, Normalina; Khuzaimah, Zailani

    2007-10-01

    This paper deals with landslide susceptibility analysis using an artificial neural network model for Cameron Highland, Malaysia. Landslide locations were identified in the study area from interpretation of aerial photographs and field surveys. Topographical/geological data and satellite images were collected and processed using GIS and image processing tools. There are ten landslide inducing parameters which are considered for the landslide hazards. These parameters are topographic slope, aspect, curvature and distance from drainage, all derived from the topographic database; geology and distance from lineament, derived from the geologic database; landuse from Landsat satellite images; soil from the soil database; precipitation amount, derived from the rainfall database; and the vegetation index value from SPOT satellite images. Landslide hazard was analyzed using landslide occurrence factors employing the logistic regression model. The results of the analysis were verified using the landslide location data and compared with logistic regression model. The accuracy of hazard map observed was 85.73%. The qualitative landslide susceptibility analysis was carried out using an artificial neural network model by doing map overlay analysis in GIS environment. This information could be used to estimate the risk to population, property and existing infrastructure like transportation network.

  13. Adaptive evolutionary artificial neural networks for pattern classification.

    PubMed

    Oong, Tatt Hee; Isa, Nor Ashidi Mat

    2011-11-01

    This paper presents a new evolutionary approach called the hybrid evolutionary artificial neural network (HEANN) for simultaneously evolving an artificial neural networks (ANNs) topology and weights. Evolutionary algorithms (EAs) with strong global search capabilities are likely to provide the most promising region. However, they are less efficient in fine-tuning the search space locally. HEANN emphasizes the balancing of the global search and local search for the evolutionary process by adapting the mutation probability and the step size of the weight perturbation. This is distinguishable from most previous studies that incorporate EA to search for network topology and gradient learning for weight updating. Four benchmark functions were used to test the evolutionary framework of HEANN. In addition, HEANN was tested on seven classification benchmark problems from the UCI machine learning repository. Experimental results show the superior performance of HEANN in fine-tuning the network complexity within a small number of generations while preserving the generalization capability compared with other algorithms. PMID:21968733

  14. The use of neural nets for matching compressors with diesel engines

    SciTech Connect

    Nelson, S.A. II; Filipi, Z.S.; Assanis, D.N.

    1996-12-31

    A technique which uses trained neural nets to model the compressor in the context of a turbocharged diesel engine simulation is introduced. This technique replaces the usual interpolation of compressor maps with the evaluation of a smooth mathematical function, thus providing engine simulations with greater robustness and flexibility. Following presentation of the methodology, the proposed neural net technique is validated against data from a truck type, 6-cylinder, 14 liter diesel engine. Furthermore, with the introduction of an additional parameter, the proposed neural net can be trained to simulate an entire family of compressors. As a demonstration, five compressors of different sizes are represented with the neural net model, and used for matching calculations with intercooled and non-intercooled engine configurations at different speeds. This novel approach readily allows for evaluation of various options prior to prototype production, and is thus a powerful design tool for selection of the best compressor for a given diesel engine system.

  15. Practical approach to implementation of neural nets at the molecular level.

    PubMed

    Rambidi, N G

    1995-01-01

    Potentialities for implementing simple neural net information processing devices based on chemical and biochemical dynamic media are discussed. This approach gives an opportunity to construct efficient systems capable of performing some primitive operations important for imaging processing.

  16. Neural-Net Processed Characteristic Patterns for Measurement of Structural Integrity of Pressure Cycled Components

    NASA Technical Reports Server (NTRS)

    Decker, A. J.

    2001-01-01

    A neural-net inspection process has been combined with a bootstrap training procedure and electronic holography to detect changes or damage in a pressure-cycled International Space Station cold plate to be used for cooling instrumentation. The cold plate was excited to vibrate in a normal mode at low amplitude, and the neural net was trained by example to flag small changes in the mode shape. The NDE (nondestructive-evaluation) technique is straightforward but in its infancy; its applications are ad-hoc and uncalibrated. Nevertheless previous research has shown that the neural net can detect displacement changes to better than 1/100 the maximum displacement amplitude. Development efforts that support the NDE technique are mentioned briefly, followed by descriptions of electronic holography and neural-net processing. The bootstrap training procedure and its application to detection of damage in a pressure-cycled cold plate are discussed. Suggestions for calibrating and quantifying the NDE procedure are presented.

  17. NETS - A NEURAL NETWORK DEVELOPMENT TOOL, VERSION 3.0 (MACHINE INDEPENDENT VERSION)

    NASA Technical Reports Server (NTRS)

    Baffes, P. T.

    1994-01-01

    NETS, A Tool for the Development and Evaluation of Neural Networks, provides a simulation of Neural Network algorithms plus an environment for developing such algorithms. Neural Networks are a class of systems modeled after the human brain. Artificial Neural Networks are formed from hundreds or thousands of simulated neurons, connected to each other in a manner similar to brain neurons. Problems which involve pattern matching readily fit the class of problems which NETS is designed to solve. NETS uses the back propagation learning method for all of the networks which it creates. The nodes of a network are usually grouped together into clumps called layers. Generally, a network will have an input layer through which the various environment stimuli are presented to the network, and an output layer for determining the network's response. The number of nodes in these two layers is usually tied to some features of the problem being solved. Other layers, which form intermediate stops between the input and output layers, are called hidden layers. NETS allows the user to customize the patterns of connections between layers of a network. NETS also provides features for saving the weight values of a network during the learning process, which allows for more precise control over the learning process. NETS is an interpreter. Its method of execution is the familiar "read-evaluate-print" loop found in interpreted languages such as BASIC and LISP. The user is presented with a prompt which is the simulator's way of asking for input. After a command is issued, NETS will attempt to evaluate the command, which may produce more prompts requesting specific information or an error if the command is not understood. The typical process involved when using NETS consists of translating the problem into a format which uses input/output pairs, designing a network configuration for the problem, and finally training the network with input/output pairs until an acceptable error is reached. NETS

  18. NETS - A NEURAL NETWORK DEVELOPMENT TOOL, VERSION 3.0 (MACINTOSH VERSION)

    NASA Technical Reports Server (NTRS)

    Phillips, T. A.

    1994-01-01

    NETS, A Tool for the Development and Evaluation of Neural Networks, provides a simulation of Neural Network algorithms plus an environment for developing such algorithms. Neural Networks are a class of systems modeled after the human brain. Artificial Neural Networks are formed from hundreds or thousands of simulated neurons, connected to each other in a manner similar to brain neurons. Problems which involve pattern matching readily fit the class of problems which NETS is designed to solve. NETS uses the back propagation learning method for all of the networks which it creates. The nodes of a network are usually grouped together into clumps called layers. Generally, a network will have an input layer through which the various environment stimuli are presented to the network, and an output layer for determining the network's response. The number of nodes in these two layers is usually tied to some features of the problem being solved. Other layers, which form intermediate stops between the input and output layers, are called hidden layers. NETS allows the user to customize the patterns of connections between layers of a network. NETS also provides features for saving the weight values of a network during the learning process, which allows for more precise control over the learning process. NETS is an interpreter. Its method of execution is the familiar "read-evaluate-print" loop found in interpreted languages such as BASIC and LISP. The user is presented with a prompt which is the simulator's way of asking for input. After a command is issued, NETS will attempt to evaluate the command, which may produce more prompts requesting specific information or an error if the command is not understood. The typical process involved when using NETS consists of translating the problem into a format which uses input/output pairs, designing a network configuration for the problem, and finally training the network with input/output pairs until an acceptable error is reached. NETS

  19. Artificial Neural Network Analysis in Preclinical Breast Cancer

    PubMed Central

    Motalleb, Gholamreza

    2014-01-01

    Objective: In this study, artificial neural network (ANN) analysis of virotherapy in preclinical breast cancer was investigated. Materials and Methods: In this research article, a multilayer feed-forward neural network trained with an error back-propagation algorithm was incorporated in order to develop a predictive model. The input parameters of the model were virus dose, week and tamoxifen citrate, while tumor weight was included in the output parameter. Two different training algorithms, namely quick propagation (QP) and Levenberg-Marquardt (LM), were used to train ANN. Results: The results showed that the LM algorithm, with 3-9-1 arrangement is more efficient compared to QP. Using LM algorithm, the coefficient of determination (R2) between the actual and predicted values was determined as 0.897118 for all data. Conclusion: It can be concluded that this ANN model may provide good ability to predict the biometry information of tumor in preclinical breast cancer virotherapy. The results showed that the LM algorithm employed by Neural Power software gave the better performance compared with the QP and virus dose, and it is more important factor compared to tamoxifen and time (week). PMID:24381857

  20. Heterogeneous artificial neural network for short term electrical load forecasting

    SciTech Connect

    Piras, A.; Germond, A.; Buchenel, B.; Imhof, K.; Jaccard, Y.

    1995-12-31

    Short term electrical load forecasting is a topic of major interest for the planning of energy production and distribution. The use of artificial neural networks has been demonstrated as a valid alternative to classical statistical methods in terms of accuracy of results. However, a common architecture able to forecast the load in different geographical regions, showing different load shape and climate characteristics, is still missing. In this paper the authors discuss a heterogeneous neural network architecture composed of an unsupervised part, namely a neural gas, which is used to analyze the process in submodels finding local features in the data and suggesting regression variables, and a supervised one, a multilayer perceptron, which performs the approximation of the underlying function. The results outputs are then summed by a weighted fuzzy average, allowing a smooth transition between sub models. The effectiveness of the proposed architecture is demonstrated by two days ahead load forecasting of EOS power system sub areas, corresponding to five different geographical regions, and of its total electrical load.

  1. Heterogeneous artificial neural network for short term electrical load forecasting

    SciTech Connect

    Piras, A.; Germond, A.; Buchenel, B.; Imhof, K.; Jaccard, Y.

    1996-02-01

    Short term electrical load forecasting is a topic of major interest for the planning of energy production and distribution. The use of artificial neural networks has been demonstrated as a valid alternative to classical statistical methods in terms of accuracy of results. However a common architecture able to forecast the load in different geographical regions, showing different load shape and climate characteristics, is still missing. In this paper the authors discuss a heterogeneous neural network architecture composed of an unsupervised part, namely a neural gas, which is used to analyze the process in sub models finding local features in the data and suggesting regression variables, and a supervised one, a multilayer perceptron, which performs the approximation of the underlying function. The resulting outputs are then summed by a weighted fuzzy average, allowing a smooth transition between sub models. The effectiveness of the proposed architecture is demonstrated by two days ahead load forecasting of EOS power system sub areas, corresponding to five different geographical regions, and of its total electrical load.

  2. On the feasibility of using neural nets to derive hearing-aid prescriptive procedures.

    PubMed

    Kates, J M

    1995-07-01

    A neural net is a "black box" information processing system that can be used for pattern matching, optimal prediction, or functional approximation. A neural net requires a minimal amount of a priori knowledge about the problem to be solved, but can require large amounts of data to converge to a solution. For a hearing-aid fitting procedure, a multilayer perceptron net was trained to generate an optimum match between a set of input pure-tone audiograms and the corresponding best frequency response and gain for each subject. The feasibility of using neural nets to select hearing-aid response characteristics was tested using both simulated and real audiometric data. The simulation results indicate that a neural net can be successfully trained to reproduce a fitting rule such as the NAL-R procedure, and that a minimum of about 50 sets of audiometric response data are needed for the net to converge to a generalized solution. When used to predict, from the pure-tone audiograms, the best frequency response characteristics determined for subjects having severe-to-profound hearing losses, the neural net was more accurate than the NAL-R fitting procedure derived from the same data.

  3. NETS - A NEURAL NETWORK DEVELOPMENT TOOL, VERSION 3.0 (MACHINE INDEPENDENT VERSION)

    NASA Technical Reports Server (NTRS)

    Baffes, P. T.

    1994-01-01

    NETS, A Tool for the Development and Evaluation of Neural Networks, provides a simulation of Neural Network algorithms plus an environment for developing such algorithms. Neural Networks are a class of systems modeled after the human brain. Artificial Neural Networks are formed from hundreds or thousands of simulated neurons, connected to each other in a manner similar to brain neurons. Problems which involve pattern matching readily fit the class of problems which NETS is designed to solve. NETS uses the back propagation learning method for all of the networks which it creates. The nodes of a network are usually grouped together into clumps called layers. Generally, a network will have an input layer through which the various environment stimuli are presented to the network, and an output layer for determining the network's response. The number of nodes in these two layers is usually tied to some features of the problem being solved. Other layers, which form intermediate stops between the input and output layers, are called hidden layers. NETS allows the user to customize the patterns of connections between layers of a network. NETS also provides features for saving the weight values of a network during the learning process, which allows for more precise control over the learning process. NETS is an interpreter. Its method of execution is the familiar "read-evaluate-print" loop found in interpreted languages such as BASIC and LISP. The user is presented with a prompt which is the simulator's way of asking for input. After a command is issued, NETS will attempt to evaluate the command, which may produce more prompts requesting specific information or an error if the command is not understood. The typical process involved when using NETS consists of translating the problem into a format which uses input/output pairs, designing a network configuration for the problem, and finally training the network with input/output pairs until an acceptable error is reached. NETS

  4. NETS - A NEURAL NETWORK DEVELOPMENT TOOL, VERSION 3.0 (MACINTOSH VERSION)

    NASA Technical Reports Server (NTRS)

    Phillips, T. A.

    1994-01-01

    NETS, A Tool for the Development and Evaluation of Neural Networks, provides a simulation of Neural Network algorithms plus an environment for developing such algorithms. Neural Networks are a class of systems modeled after the human brain. Artificial Neural Networks are formed from hundreds or thousands of simulated neurons, connected to each other in a manner similar to brain neurons. Problems which involve pattern matching readily fit the class of problems which NETS is designed to solve. NETS uses the back propagation learning method for all of the networks which it creates. The nodes of a network are usually grouped together into clumps called layers. Generally, a network will have an input layer through which the various environment stimuli are presented to the network, and an output layer for determining the network's response. The number of nodes in these two layers is usually tied to some features of the problem being solved. Other layers, which form intermediate stops between the input and output layers, are called hidden layers. NETS allows the user to customize the patterns of connections between layers of a network. NETS also provides features for saving the weight values of a network during the learning process, which allows for more precise control over the learning process. NETS is an interpreter. Its method of execution is the familiar "read-evaluate-print" loop found in interpreted languages such as BASIC and LISP. The user is presented with a prompt which is the simulator's way of asking for input. After a command is issued, NETS will attempt to evaluate the command, which may produce more prompts requesting specific information or an error if the command is not understood. The typical process involved when using NETS consists of translating the problem into a format which uses input/output pairs, designing a network configuration for the problem, and finally training the network with input/output pairs until an acceptable error is reached. NETS

  5. Identifing Atmospheric Pollutant Sources Using Artificial Neural Networks

    NASA Astrophysics Data System (ADS)

    Paes, F. F.; Campos, H. F.; Luz, E. P.; Carvalho, A. R.

    2008-05-01

    The estimation of the area source pollutant strength is a relevant issue for atmospheric environment. This characterizes an inverse problem in the atmospheric pollution dispersion. In the inverse analysis, an area source domain is considered, where the strength of such area source term is assumed unknown. The inverse problem is solved by using a supervised artificial neural network: multi-layer perceptron. The conection weights of the neural network are computed from delta rule - learning process. The neural network inversion is compared with results from standard inverse analysis (regularized inverse solution). In the regularization method, the inverse problem is formulated as a non-linear optimization approach, whose the objective function is given by the square difference between the measured pollutant concentration and the mathematical models, associated with a regularization operator. In our numerical experiments, the forward problem is addressed by a source-receptor scheme, where a regressive Lagrangian model is applied to compute the transition matrix. The second order maximum entropy regularization is used, and the regularization parameter is calculated by the L-curve technique. The objective function is minimized employing a deterministic scheme (a quasi-Newton algorithm) [1] and a stochastic technique (PSO: particle swarm optimization) [2]. The inverse problem methodology is tested with synthetic observational data, from six measurement points in the physical domain. The best inverse solutions were obtained with neural networks. References: [1] D. R. Roberti, D. Anfossi, H. F. Campos Velho, G. A. Degrazia (2005): Estimating Emission Rate and Pollutant Source Location, Ciencia e Natura, p. 131-134. [2] E.F.P. da Luz, H.F. de Campos Velho, J.C. Becceneri, D.R. Roberti (2007): Estimating Atmospheric Area Source Strength Through Particle Swarm Optimization. Inverse Problems, Desing and Optimization Symposium IPDO-2007, April 16-18, Miami (FL), USA, vol 1, p

  6. Artificial Neural Network L* from different magnetospheric field models

    NASA Astrophysics Data System (ADS)

    Yu, Y.; Koller, J.; Zaharia, S. G.; Jordanova, V. K.

    2011-12-01

    The third adiabatic invariant L* plays an important role in modeling and understanding the radiation belt dynamics. The popular way to numerically obtain the L* value follows the recipe described by Roederer [1970], which is, however, slow and computational expensive. This work focuses on a new technique, which can compute the L* value in microseconds without losing much accuracy: artificial neural networks. Since L* is related to the magnetic flux enclosed by a particle drift shell, global magnetic field information needed to trace the drift shell is required. A series of currently popular empirical magnetic field models are applied to create the L* data pool using 1 million data samples which are randomly selected within a solar cycle and within the global magnetosphere. The networks, trained from the above L* data pool, can thereby be used for fairly efficient L* calculation given input parameters valid within the trained temporal and spatial range. Besides the empirical magnetospheric models, a physics-based self-consistent inner magnetosphere model (RAM-SCB) developed at LANL is also utilized to calculate L* values and then to train the L* neural network. This model better predicts the magnetospheric configuration and therefore can significantly improve the L*. The above neural network L* technique will enable, for the first time, comprehensive solar-cycle long studies of radiation belt processes. However, neural networks trained from different magnetic field models can result in different L* values, which could cause mis-interpretation of radiation belt dynamics, such as where the source of the radiation belt charged particle is and which mechanism is dominant in accelerating the particles. Such a fact calls for attention to cautiously choose a magnetospheric field model for the L* calculation.

  7. DeepNet: An Ultrafast Neural Learning Code for Seismic Imaging

    SciTech Connect

    Barhen, J.; Protopopescu, V.; Reister, D.

    1999-07-10

    A feed-forward multilayer neural net is trained to learn the correspondence between seismic data and well logs. The introduction of a virtual input layer, connected to the nominal input layer through a special nonlinear transfer function, enables ultrafast (single iteration), near-optimal training of the net using numerical algebraic techniques. A unique computer code, named DeepNet, has been developed, that has achieved, in actual field demonstrations, results unattainable to date with industry standard tools.

  8. Controlling chaotic convection using neural nets-theory and experiments.

    PubMed

    Bau, Haim H.; Yuen, Po Ki

    1998-04-01

    An exploratory study is conducted to assess the feasibility of using neural networks to control flow patterns and to evaluate the performance of these controllers. Neural networks were used to control (suppress) chaotic convection both in experiments and in a theoretical model of a thermal convection loop. It is demonstrated that the neural network controller can successfully cause the flow to behave in a desired way. The performance of the neural network controllers was compared with that of previously used conventional linear proportional controllers.

  9. Neuron-Glia Interactions in Neural Plasticity: Contributions of Neural Extracellular Matrix and Perineuronal Nets.

    PubMed

    Dzyubenko, Egor; Gottschling, Christine; Faissner, Andreas

    2016-01-01

    Synapses are specialized structures that mediate rapid and efficient signal transmission between neurons and are surrounded by glial cells. Astrocytes develop an intimate association with synapses in the central nervous system (CNS) and contribute to the regulation of ion and neurotransmitter concentrations. Together with neurons, they shape intercellular space to provide a stable milieu for neuronal activity. Extracellular matrix (ECM) components are synthesized by both neurons and astrocytes and play an important role in the formation, maintenance, and function of synapses in the CNS. The components of the ECM have been detected near glial processes, which abut onto the CNS synaptic unit, where they are part of the specialized macromolecular assemblies, termed perineuronal nets (PNNs). PNNs have originally been discovered by Golgi and represent a molecular scaffold deposited in the interface between the astrocyte and subsets of neurons in the vicinity of the synapse. Recent reports strongly suggest that PNNs are tightly involved in the regulation of synaptic plasticity. Moreover, several studies have implicated PNNs and the neural ECM in neuropsychiatric diseases. Here, we highlight current concepts relating to neural ECM and PNNs and describe an in vitro approach that allows for the investigation of ECM functions for synaptogenesis.

  10. Neuron-Glia Interactions in Neural Plasticity: Contributions of Neural Extracellular Matrix and Perineuronal Nets

    PubMed Central

    Dzyubenko, Egor; Gottschling, Christine

    2016-01-01

    Synapses are specialized structures that mediate rapid and efficient signal transmission between neurons and are surrounded by glial cells. Astrocytes develop an intimate association with synapses in the central nervous system (CNS) and contribute to the regulation of ion and neurotransmitter concentrations. Together with neurons, they shape intercellular space to provide a stable milieu for neuronal activity. Extracellular matrix (ECM) components are synthesized by both neurons and astrocytes and play an important role in the formation, maintenance, and function of synapses in the CNS. The components of the ECM have been detected near glial processes, which abut onto the CNS synaptic unit, where they are part of the specialized macromolecular assemblies, termed perineuronal nets (PNNs). PNNs have originally been discovered by Golgi and represent a molecular scaffold deposited in the interface between the astrocyte and subsets of neurons in the vicinity of the synapse. Recent reports strongly suggest that PNNs are tightly involved in the regulation of synaptic plasticity. Moreover, several studies have implicated PNNs and the neural ECM in neuropsychiatric diseases. Here, we highlight current concepts relating to neural ECM and PNNs and describe an in vitro approach that allows for the investigation of ECM functions for synaptogenesis. PMID:26881114

  11. [The Identification of the Origin of Chinese Wolfberry Based on Infrared Spectral Technology and the Artificial Neural Network].

    PubMed

    Li, Zhong; Liu, Ming-de; Ji, Shou-xiang

    2016-03-01

    The Fourier Transform Infrared Spectroscopy (FTIR) is established to find the geographic origins of Chinese wolfberry quickly. In the paper, the 45 samples of Chinese wolfberry from different places of Qinghai Province are to be surveyed by FTIR. The original data matrix of FTIR is pretreated with common preprocessing and wavelet transform. Compared with common windows shifting smoothing preprocessing, standard normal variation correction and multiplicative scatter correction, wavelet transform is an effective spectrum data preprocessing method. Before establishing model through the artificial neural networks, the spectra variables are compressed by means of the wavelet transformation so as to enhance the training speed of the artificial neural networks, and at the same time the related parameters of the artificial neural networks model are also discussed in detail. The survey shows even if the infrared spectroscopy data is compressed to 1/8 of its original data, the spectral information and analytical accuracy are not deteriorated. The compressed spectra variables are used for modeling parameters of the backpropagation artificial neural network (BP-ANN) model and the geographic origins of Chinese wolfberry are used for parameters of export. Three layers of neural network model are built to predict the 10 unknown samples by using the MATLAB neural network toolbox design error back propagation network. The number of hidden layer neurons is 5, and the number of output layer neuron is 1. The transfer function of hidden layer is tansig, while the transfer function of output layer is purelin. Network training function is trainl and the learning function of weights and thresholds is learngdm. net. trainParam. epochs=1 000, while net. trainParam. goal = 0.001. The recognition rate of 100% is to be achieved. It can be concluded that the method is quite suitable for the quick discrimination of producing areas of Chinese wolfberry. The infrared spectral analysis technology

  12. [The Identification of the Origin of Chinese Wolfberry Based on Infrared Spectral Technology and the Artificial Neural Network].

    PubMed

    Li, Zhong; Liu, Ming-de; Ji, Shou-xiang

    2016-03-01

    The Fourier Transform Infrared Spectroscopy (FTIR) is established to find the geographic origins of Chinese wolfberry quickly. In the paper, the 45 samples of Chinese wolfberry from different places of Qinghai Province are to be surveyed by FTIR. The original data matrix of FTIR is pretreated with common preprocessing and wavelet transform. Compared with common windows shifting smoothing preprocessing, standard normal variation correction and multiplicative scatter correction, wavelet transform is an effective spectrum data preprocessing method. Before establishing model through the artificial neural networks, the spectra variables are compressed by means of the wavelet transformation so as to enhance the training speed of the artificial neural networks, and at the same time the related parameters of the artificial neural networks model are also discussed in detail. The survey shows even if the infrared spectroscopy data is compressed to 1/8 of its original data, the spectral information and analytical accuracy are not deteriorated. The compressed spectra variables are used for modeling parameters of the backpropagation artificial neural network (BP-ANN) model and the geographic origins of Chinese wolfberry are used for parameters of export. Three layers of neural network model are built to predict the 10 unknown samples by using the MATLAB neural network toolbox design error back propagation network. The number of hidden layer neurons is 5, and the number of output layer neuron is 1. The transfer function of hidden layer is tansig, while the transfer function of output layer is purelin. Network training function is trainl and the learning function of weights and thresholds is learngdm. net. trainParam. epochs=1 000, while net. trainParam. goal = 0.001. The recognition rate of 100% is to be achieved. It can be concluded that the method is quite suitable for the quick discrimination of producing areas of Chinese wolfberry. The infrared spectral analysis technology

  13. Artificial metaplasticity neural network applied to credit scoring.

    PubMed

    Marcano-Cedeño, Alexis; Marin-de-la-Barcena, A; Jimenez-Trillo, J; Piñuela, J A; Andina, D

    2011-08-01

    The assessment of the risk of default on credit is important for financial institutions. Different Artificial Neural Networks (ANN) have been suggested to tackle the credit scoring problem, however, the obtained error rates are often high. In the search for the best ANN algorithm for credit scoring, this paper contributes with the application of an ANN Training Algorithm inspired by the neurons' biological property of metaplasticity. This algorithm is especially efficient when few patterns of a class are available, or when information inherent to low probability events is crucial for a successful application, as weight updating is overemphasized in the less frequent activations than in the more frequent ones. Two well-known and readily available such as: Australia and German data sets has been used to test the algorithm. The results obtained by AMMLP shown have been superior to state-of-the-art classification algorithms in credit scoring.

  14. ANNz: Estimating Photometric Redshifts Using Artificial Neural Networks

    NASA Astrophysics Data System (ADS)

    Collister, Adrian A.; Lahav, Ofer

    2004-04-01

    We introduce ANNz, a freely available software package for photometric redshift estimation using artificial neural networks. ANNz learns the relation between photometry and redshift from an appropriate training set of galaxies for which the redshift is already known. Where a large and representative training set is available, ANNz is a highly competitive tool when compared with traditional template-fitting methods. The ANNz package is demonstrated on the Sloan Digital Sky Survey Data Release 1, and for this particular data set the rms redshift error in the range 0<~z<~0.7 is σrms=0.023. Nonideal conditions (spectroscopic sets that are small or brighter than the photometric set for which redshifts are required) are simulated, and the impact on the photometric redshift accuracy is assessed.2

  15. Practical application of artificial neural networks in the neurosciences

    NASA Astrophysics Data System (ADS)

    Pinti, Antonio

    1995-04-01

    This article presents a practical application of artificial multi-layer perceptron (MLP) neural networks in neurosciences. The data that are processed are labeled data from the visual analysis of electrical signals of human sleep. The objective of this work is to automatically classify into sleep stages the electrophysiological signals recorded from electrodes placed on a sleeping patient. Two large data bases were designed by experts in order to realize this study. One data base was used to train the network and the other to test its generalization capacity. The classification results obtained with the MLP network were compared to a type K nearest neighbor Knn non-parametric classification method. The MLP network gave a better result in terms of classification than the Knn method. Both classification techniques were implemented on a transputer system. With both networks in their final configuration, the MLP network was 160 times faster than the Knn model in classifying a sleep period.

  16. Artificial neural networks to model and diagnose cardiovascular systems

    SciTech Connect

    Kangas, L.J.; Keller, P.E.; Allen, P.A.

    1995-12-31

    In this paper, a novel approach to modeling and diagnosing the cardiovascular system is introduced. A model exhibits a subset of the dynamics of the cardiovascular behavior of an individual by using a recurrent artificial neural network. Potentially, a model will be incorporated into a cardiovascular diagnostic system. This approach is unique in that each cardiovascular model is developed from physiological measurements of an individual. Any differences between the modeled variables and. the actual variables of an individual at a given time are used for diagnosis. This approach also exploits sensor fusion to optimize the utilization of biomedical sensors. The advantage of sensor fusion has been demonstrated in applications including control and diagnostics of mechanical and chemical processes.

  17. Artificial neural networks optimization method for radioactive source localization

    SciTech Connect

    Wacholder, E.; Elias, E.; Merlis, Y.

    1995-05-01

    An optimization artificial neural networks model is developed for solving the ill-posed inverse transport problem associated with localizing radioactive sources in a medium with known properties and dimensions. The model is based on the recurrent (or feedback) Hopfield network with fixed weights. The source distribution is determined based on the response of a limited number of external detectors of known spatial deployment in conjunction with a radiation transport model. The algorithm is tested and evaluated for a large number of simulated two-dimensional cases. Computations are carried out at different noise levels to account for statistical errors encountered in engineering applications. The sensitivity to noise is found to depend on the number of detectors and on their spatial deployment. A pretest empirical procedure is, therefore, suggested for determining an effective arrangement of detectors for a given problem.

  18. Applications of artificial neural networks (ANNs) in food science.

    PubMed

    Huang, Yiqun; Kangas, Lars J; Rasco, Barbara A

    2007-01-01

    Artificial neural networks (ANNs) have been applied in almost every aspect of food science over the past two decades, although most applications are in the development stage. ANNs are useful tools for food safety and quality analyses, which include modeling of microbial growth and from this predicting food safety, interpreting spectroscopic data, and predicting physical, chemical, functional and sensory properties of various food products during processing and distribution. ANNs hold a great deal of promise for modeling complex tasks in process control and simulation and in applications of machine perception including machine vision and electronic nose for food safety and quality control. This review discusses the basic theory of the ANN technology and its applications in food science, providing food scientists and the research community an overview of the current research and future trend of the applications of ANN technology in the field.

  19. Automatic segmentation of cerebral MR images using artificial neural networks

    SciTech Connect

    Alirezaie, J.; Jernigan, M.E.; Nahmias, C.

    1996-12-31

    In this paper we present an unsupervised clustering technique for multispectral segmentation of magnetic resonance (MR) images of the human brain. Our scheme utilizes the Self Organizing Feature Map (SOFM) artificial neural network for feature mapping and generates a set of codebook vectors. By extending the network with an additional layer the map will be classified and each tissue class will be labelled. An algorithm has been developed for extracting the cerebrum from the head scan prior to the segmentation. Extracting the cerebrum is performed by stripping away the skull pixels from the T2 image. Three tissue types of the brain: white matter, gray matter and cerebral spinal fluid (CSF) are segmented accurately. To compare the results with other conventional approaches we applied the c-means algorithm to the problem.

  20. Prediction of Dried Durian Moisture Content Using Artificial Neural Networks

    NASA Astrophysics Data System (ADS)

    Husna, Marati; Purqon, Acep

    2016-08-01

    Moisture content has a crucial issue in post-harvest processing since it plays main role to estimate a quality of dried product. However, estimating the moisture content is difficult since it shows mathematically nonlinear systems and complex physical processes. We investigate the prediction of moisture content of dried product by using Artificial Neural Networks (ANN). Our sample is a Bengkulu's local durian that is dried using a microwave oven. Our results show that ANN can predict the moisture content by performing with R2 value is 98.47%. Moreover, the RMSE values is 3.97% and MSE values is 0.16%. Our results indicate that ANN model have high capability for predicting moisture content and it is potentially applied in post-harvest product, especially in drying product quality control.

  1. Incomplete fuzzy data processing systems using artificial neural network

    NASA Technical Reports Server (NTRS)

    Patyra, Marek J.

    1992-01-01

    In this paper, the implementation of a fuzzy data processing system using an artificial neural network (ANN) is discussed. The binary representation of fuzzy data is assumed, where the universe of discourse is decartelized into n equal intervals. The value of a membership function is represented by a binary number. It is proposed that incomplete fuzzy data processing be performed in two stages. The first stage performs the 'retrieval' of incomplete fuzzy data, and the second stage performs the desired operation on the retrieval data. The method of incomplete fuzzy data retrieval is proposed based on the linear approximation of missing values of the membership function. The ANN implementation of the proposed system is presented. The system was computationally verified and showed a relatively small total error.

  2. Flood estimation at ungauged sites using artificial neural networks

    NASA Astrophysics Data System (ADS)

    Dawson, C. W.; Abrahart, R. J.; Shamseldin, A. Y.; Wilby, R. L.

    2006-03-01

    Artificial neural networks (ANNs) have been applied within the field of hydrological modelling for over a decade but relatively little attention has been paid to the use of these tools for flood estimation in ungauged catchments. This paper uses data from the Centre for Ecology and Hydrology's Flood Estimation Handbook (FEH) to predict T-year flood events and the index flood (the median of the annual maximum series) for 850 catchments across the UK. When compared with multiple regression models, ANNs provide improved flood estimates that can be used by engineers and hydrologists. Comparisons are also made with the empirical model presented in the FEH and a preliminary study is made of the spatial distribution of ANN residuals, highlighting the influence that geographical factors have on model performance.

  3. Modeling biodegradation and kinetics of glyphosate by artificial neural network.

    PubMed

    Nourouzi, Mohsen M; Chuah, Teong G; Choong, Thomas S Y; Rabiei, F

    2012-01-01

    An artificial neural network (ANN) model was developed to simulate the biodegradation of herbicide glyphosate [2-(Phosphonomethylamino) acetic acid] in a solution with varying parameters pH, inoculum size and initial glyphosate concentration. The predictive ability of ANN model was also compared with Monod model. The result showed that ANN model was able to accurately predict the experimental results. A low ratio of self-inhibition and half saturation constants of Haldane equations (< 8) exhibited the inhibitory effect of glyphosate on bacteria growth. The value of K(i)/K(s) increased when the mixed inoculum size was increased from 10(4) to 10(6) bacteria/mL. It was found that the percentage of glyphosate degradation reached a maximum value of 99% at an optimum pH 6-7 while for pH values higher than 9 or lower than 4, no degradation was observed. PMID:22424071

  4. Applications of Artificial Neural Networks (ANNs) in Food Science

    SciTech Connect

    HUang, Yiqun; Kangas, Lars J.; Rasco, Barbara A.

    2007-02-01

    Abstract Artificial neural networks (ANNs) have been applied in almost every aspect of food science over the past two decade, although most applications are in the development stage. ANNs are useful tools for food safety and quality analyses, which include modeling of microbial growth and from this predicting food safety, interpreting spectroscopic data, and predicting physical, chemical, functional and sensory properties of various food products during processing and distribution. ANNs have a great deal of promise for modeling complex tasks in process control and simulation, and in applications of machine perception including machine vision and the electronic nose for food safety and quality control. This review discusses the basic theory of the ANN technology and its applications in food science, providing food scientists and the research community an overview of the current research and future trend of the applications of ANN technology in this field.

  5. Artificial neural networks for plasma x-ray spectroscopic analysis

    SciTech Connect

    Larsen, J.T. ); Morgan, W.L. ); Goldstein, W.H. )

    1992-10-01

    Modern diagnostic instrumentation produces a vast amount of data that often requires substantial analysis efforts. New methods are needed to improve the efficiency of the analysis process. Artificial neural networks have been applied to a variety of signal processing and image recognition problems. The feed-forward, back-propagation technique is well suited for the analysis of scientific laboratory data, which is viewed as a pattern-matching problem. We summarize the concepts and algorithms as implemented on a personal computer, and illustrate the method using a nonlocal thermodynamic equilibrium theoretical atomic physics model for {ital k}-shell x-ray spectroscopy of a high density, high temperature aluminum plasma. Extensions to other types of spectroscopy data analysis are discussed.

  6. Inflow forecasting using Artificial Neural Networks for reservoir operation

    NASA Astrophysics Data System (ADS)

    Chiamsathit, Chuthamat; Adeloye, Adebayo J.; Bankaru-Swamy, Soundharajan

    2016-05-01

    In this study, multi-layer perceptron (MLP) artificial neural networks have been applied to forecast one-month-ahead inflow for the Ubonratana reservoir, Thailand. To assess how well the forecast inflows have performed in the operation of the reservoir, simulations were carried out guided by the systems rule curves. As basis of comparison, four inflow situations were considered: (1) inflow known and assumed to be the historic (Type A); (2) inflow known and assumed to be the forecast (Type F); (3) inflow known and assumed to be the historic mean for month (Type M); and (4) inflow is unknown with release decision only conditioned on the starting reservoir storage (Type N). Reservoir performance was summarised in terms of reliability, resilience, vulnerability and sustainability. It was found that Type F inflow situation produced the best performance while Type N was the worst performing. This clearly demonstrates the importance of good inflow information for effective reservoir operation.

  7. Magnetotelluric inversion for azimuthally anisotropic resistivities employing artificial neural networks

    NASA Astrophysics Data System (ADS)

    Montahaei, Mansoure; Oskooi, Behrooz

    2014-02-01

    An extension of an artificial neural network (ANN) approach to solve the magnetotelluric (MT) inverse problem for azimuthally anisotropic resistivities is presented and applied for a real dataset. Three different model classes, containing general 1-D and 2-D azimuthally anisotropic features, have been considered. For each model class, characteristics of three-layer feed forward ANNs trained through an error back propagation algorithm have been adjusted to approximate the inverse modeling function. It appears that, at least for synthetic models, reasonable results would be obtained by applying the amplitudes of the complex impedance tensor elements as inputs. Furthermore, the Levenberg-Marquart algorithm possesses optimal performance as a learning paradigm for this problem. The evaluation of applicability of the trained ANNs for unknown data sets excluded from the learning procedure reveals that the trained ANNs possess acceptable interpolation and extrapolation abilities to estimate model parameters accurately. This method was also successfully used for a field dataset wherein anisotropy had been previously recognized.

  8. The application of artificial neural networks in indirect cost estimation

    NASA Astrophysics Data System (ADS)

    Leśniak, Agnieszka

    2013-10-01

    Estimating of the costs of construction project is one of the most important task in the management of the project. The total costs can be divided into direct costs that are related to executing the works, and indirect costs that accompany delivery. A precise costs estimation is usually a highly labour and time-intensive task especially when using manual calculation methods. This paper presents Artificial Neural Network (ANN) approach to predicting index of indirect cost of construction projects in Poland. A quantitative study was undertaken on the factors conditioning indirect costs of polish construction projects and a determination was made of the actual costs incurred by enterprises during project implementation. As a result of these studies, a data set was assembled covering 72 real-life cases of building projects constructed in Poland.

  9. Representation and learning of nonlinear compliance using neural nets

    SciTech Connect

    Asada, Haruhiko . Dept. of Mechanical Engineering)

    1993-12-01

    A new approach to compliant motion control using neural networks is presented. In the paper, compliance is treated as a nonlinear mapping from a measured force to a corrected motion. The nonlinear mapping by a multilayer neural network is represented, which allows us to deal with complex control strategies that cannot be represented by linear compliance, such as in stiffness and damping control.

  10. The application of neural networks with artificial intelligence technique in the modeling of industrial processes

    SciTech Connect

    Saini, K. K.; Saini, Sanju

    2008-10-07

    Neural networks are a relatively new artificial intelligence technique that emulates the behavior of biological neural systems in digital software or hardware. These networks can 'learn', automatically, complex relationships among data. This feature makes the technique very useful in modeling processes for which mathematical modeling is difficult or impossible. The work described here outlines some examples of the application of neural networks with artificial intelligence technique in the modeling of industrial processes.

  11. An adaptive training method for optimal interpolative neural nets.

    PubMed

    Liu, T Z; Yen, C W

    1997-04-01

    In contrast to conventional multilayered feedforward networks which are typically trained by iterative gradient search methods, an optimal interpolative (OI) net can be trained by a noniterative least squares algorithm called RLS-OI. The basic idea of RLS-OI is to use a subset of the training set, whose inputs are called subprototypes, to constrain the OI net solution. A subset of these subprototypes, called prototypes, is then chosen as the parameter vectors of the activation functions of the OI net to satisfy the subprototype constraints in the least squares (LS) sense. By dynamically increasing the numbers of subprototypes and prototypes, RLS-OI evolves the OI net from scratch to the extent sufficient to solve a given classification problem. To improve the performance of RLS-OI, this paper addresses two important problems in OI net training: the selection of the subprototypes and the selection of the prototypes. By choosing subprototypes from poorly classified regions, this paper proposes a new subprototype selection method which is adaptive to the changing classification performance of the growing OI net. This paper also proposes a new prototype selection criterion to reduce the complexity of the OI net. For the same training accuracy, simulation results demonstrate that the proposed approach produces smaller OI net than the RLS-OI algorithm. Experimental results also show that the proposed approach is less sensitive to the variation of the training set than RLS-OI.

  12. Simulation of an array-based neural net model

    NASA Technical Reports Server (NTRS)

    Barnden, John A.

    1987-01-01

    Research in cognitive science suggests that much of cognition involves the rapid manipulation of complex data structures. However, it is very unclear how this could be realized in neural networks or connectionist systems. A core question is: how could the interconnectivity of items in an abstract-level data structure be neurally encoded? The answer appeals mainly to positional relationships between activity patterns within neural arrays, rather than directly to neural connections in the traditional way. The new method was initially devised to account for abstract symbolic data structures, but it also supports cognitively useful spatial analogue, image-like representations. As the neural model is based on massive, uniform, parallel computations over 2D arrays, the massively parallel processor is a convenient tool for simulation work, although there are complications in using the machine to the fullest advantage. An MPP Pascal simulation program for a small pilot version of the model is running.

  13. Hourly load forecasting using artificial neural networks. Final report

    SciTech Connect

    Khotanzad, A.

    1995-09-01

    An artificial neural network short-term load forecaster (ANNSTLF) and an artificial neural network (ANN) based temperature forecaster have been developed by Southern Methodist University under contracts RP2473-44 and RP3573-4. ANNSTLF can produce hourly load forecasts for one to 168 hours ahead (one to seven days ahead) with errors ranging from 2 to 4% depending on utility size and characteristics. Implementation of ANNSTLF requires an initial training with historical hourly load and weather data. Two weather parameters, temperature and relative humidity, from either one or multiple locations can be utilized. In the operational phase, the previous day`s load and weather data and hourly weather forecasts are needed. The temperature forecaster can generate hourly temperature forecasts from the predicted values for high and low temperatures of future days. Both forecasters run on a PC platform under the MS-DOS operating system. The development of ANNSTLF is based on decomposition of the load-weather relationship into three distinct trends: Weekly, daily, and hourly. Each trend is modeled by a separate module containing several multi-layer feed-forward ANNs trained by the back-propagation learning rule. The forecasts produced by each module are combined by adaptive filters to arrive at the final forecast. During the forecasting phase, the parameters of the ANNs within each module are adoptively changed according to the latest forecast accuracy. The temperature forecaster consists of a single ANN that requires the previous day`s hourly temperatures and the next day`s predicted high and low temperatures as inputs. The resulting hourly forecasts are adoptively scaled to assure that the high and low temperatures match their respective predictions. The system is capable of forecasting up to seven days ahead. ANNSTLF has been implemented at twenty utilities across the nation and is being used on-Ene by several of them.

  14. South America downscaling: using spatial artificial neural network

    NASA Astrophysics Data System (ADS)

    Mendes, David; Marengo, José

    2010-05-01

    The mathematical models used to simulate the present climate and project future climate with forcing by greenhouse gases and aerosols are generally referred to as General Circulation Models or Global Climate Models (GCMs). However, the spatial resolution of GCMs remains quite coarse, in the order of 300 x 300 km, and at scale, the regional and local details of the climate which are influenced by spatial heterogeneities in the regional physiography are lost. Therefore, there is the need to convert the GCM outputs into a reliable data set with higher spatial resolution, with daily rainfall and temperature time series at the scale of the watershed or a region to which the climate impact is going to be investigated. The methods used to convert GCM outputs into local meteorological variables required for reliable climate modeling are usually referred to as downscaling techniques. There are a variety of downscaling techniques in the literature, but two major approaches can be identified at the moment, namely, dynamic downscaling and empirical (statistical) downscaling. The most widely used empirical downscaling methods are the multiple linear regression and stochastic weather generation. However, the interest in nonlinear regression methods, namely, artificial neural network (ANN), is nowadays increasing because of their high potential for complex, nonlinear and time-varying input-output mapping. The main aim of this work is to develop and test a novel type of statistical downscaling technique based on the Artificial Neural Network (ANN), applied of the climate change. This work analyses the performance of the IPCC models in simulate the present and future climate using ANN. The ANN used here are based on a feed forward configuration of the multilayer perception that has been used by a growing number of authors. To carry out statistical downscaling for each meteorological date (grid point), the predictors and predictands were supplied to the models (ANN) and spatial

  15. Electric demand prediction using artificial neural network technology

    SciTech Connect

    Gibson, G.L.; Kraft, T.T. )

    1993-03-01

    As a means of promoting demand-side management (DSM) technologies, electric utilities have developed increasingly complex electric rate structures. Electric rates are typically based on both demand and energy use and, in some instances, can change on an hourly basis. The ability of a building's owner or operator to react to the variability of these rates would be greatly enhanced if a building's electric demand and energy use could be accurately predicted on a daily basis. This is especially true for buildings that are equipped with thermal energy storage (TES) systems for building cooling. TES systems are designed to shift the electric demand associated with building cooling to night-time hours when electric rates are usually lowest. TES systems are typically designed to provide the maximum benefit under design day weather and building usage conditions. As a result, TES systems are often under-utilized (with an associated reduction in savings) during time periods when less than design day conditions exist. To optimize TES system equipment operation, it is first necessary to predict building electric and cooling demand under non-design day conditions. A personal computer-based software package that operates in conjunction with a building's energy management and control system (EMCS) to automatically optimize TES system operation is currently installed and operating in an office building in the northeastern United States. This software package uses artificial neural network (ANN) technology to model several parameters related to building energy use and TES system operation. The purpose of this article is to report on the initial performance of the artificial neural network in its prediction of building electric load.

  16. Comparison of sonar discrimination: dolphin and an artificial neural network.

    PubMed

    Au, W W

    1994-05-01

    The capability of an echolocating dolphin to discriminate differences in the wall thickness of cylinders (3.81 cm o.d. and 12.7 cm length) was determined by Au and Pawloski [J. Comp. Physiol. A 170, 41-47 (1992)]. The dolphin was required to discriminate a standard target from comparison targets of differing wall thicknesses. Performance varied from 96% to 56% correct depending on the wall thickness of the comparison targets. The 75% correct threshold was determined to be wall thickness differences of -0.23 mm for comparison targets with thinner walls and +0.27 mm for comparison targets with thicker walls than the standard. The dolphin performance was unchanged in the presence of artificial broadband masking noise until the echo-energy-to-noise ratio fell below approximately 15 dB. A counterpropagation artificial neural network was used to examine broadband echo features from the same cylinders. Features of the echoes were determined by passing them through a filter bank of constant-Q filters. Echo features of the standard and each comparison target were analyzed in pairs by a neural network having two output nodes. Twenty echoes per target were used in the training set and 30 additional echoes per target were used in the test set. For the noise free condition, the network performed at a comparable level to the dolphin for Q values between 4 and 5. In the presence of noise, Q values between 7 and 8 were needed before the network could perform at a comparable level to the dolphin for echo-energy-to-noise ratios of 10 and 15 dB.(ABSTRACT TRUNCATED AT 250 WORDS)

  17. Comparison of sonar discrimination: dolphin and an artificial neural network.

    PubMed

    Au, W W

    1994-05-01

    The capability of an echolocating dolphin to discriminate differences in the wall thickness of cylinders (3.81 cm o.d. and 12.7 cm length) was determined by Au and Pawloski [J. Comp. Physiol. A 170, 41-47 (1992)]. The dolphin was required to discriminate a standard target from comparison targets of differing wall thicknesses. Performance varied from 96% to 56% correct depending on the wall thickness of the comparison targets. The 75% correct threshold was determined to be wall thickness differences of -0.23 mm for comparison targets with thinner walls and +0.27 mm for comparison targets with thicker walls than the standard. The dolphin performance was unchanged in the presence of artificial broadband masking noise until the echo-energy-to-noise ratio fell below approximately 15 dB. A counterpropagation artificial neural network was used to examine broadband echo features from the same cylinders. Features of the echoes were determined by passing them through a filter bank of constant-Q filters. Echo features of the standard and each comparison target were analyzed in pairs by a neural network having two output nodes. Twenty echoes per target were used in the training set and 30 additional echoes per target were used in the test set. For the noise free condition, the network performed at a comparable level to the dolphin for Q values between 4 and 5. In the presence of noise, Q values between 7 and 8 were needed before the network could perform at a comparable level to the dolphin for echo-energy-to-noise ratios of 10 and 15 dB.(ABSTRACT TRUNCATED AT 250 WORDS) PMID:8207144

  18. A novel technology for fabricating customizable VLSI artificial neural network chips

    SciTech Connect

    Fu, C.Y.; Law, B.; Chapline, G.; Swenson, D.

    1992-02-05

    This paper describes an implementation of hardware neural networks using highly linear thin-film resistor technology and an 8-bit binary weight circuit to produce customizable artificial neural network chips and systems. These neural networks are programmed using precision laser cutting and deposition. The fast turnaround of laser-based customization allows us to explore different neural network architectures and to rapidly program the synaptic weights. Our customizable chip allows us to expand an artificial network laterally and vertically. This flexibility permits us to build very large neural network systems.

  19. Robustness against S.E.U. of an artificial neural network space application

    SciTech Connect

    Assoum, A.; Radi, N.E.; Velazco, R.; Elie, F.; Ecoffet, R.

    1996-06-01

    The authors study the sensitivity of Artificial Neural Networks (ANN) to Single Event Upsets (SEU). A neural network designed to detect electronic and protonic whistlers has been implemented using a dedicated VLSI circuit: the LNeuro neural processor. Results of both SEU software simulations and heavy ion tests point out the fault tolerance properties of ANN hardware implementations.

  20. Bootstrapped neural nets versus regression kriging in the digital mapping of pedological attributes: the automatic and time-consuming perspectives

    NASA Astrophysics Data System (ADS)

    Langella, Giuliano; Basile, Angelo; Bonfante, Antonello; Manna, Piero; Terribile, Fabio

    2013-04-01

    Digital soil mapping procedures are widespread used to build two-dimensional continuous maps about several pedological attributes. Our work addressed a regression kriging (RK) technique and a bootstrapped artificial neural network approach in order to evaluate and compare (i) the accuracy of prediction, (ii) the susceptibility of being included in automatic engines (e.g. to constitute web processing services), and (iii) the time cost needed for calibrating models and for making predictions. Regression kriging is maybe the most widely used geostatistical technique in the digital soil mapping literature. Here we tried to apply the EBLUP regression kriging as it is deemed to be the most statistically sound RK flavor by pedometricians. An unusual multi-parametric and nonlinear machine learning approach was accomplished, called BAGAP (Bootstrap aggregating Artificial neural networks with Genetic Algorithms and Principal component regression). BAGAP combines a selected set of weighted neural nets having specified characteristics to yield an ensemble response. The purpose of applying these two particular models is to ascertain whether and how much a more cumbersome machine learning method could be much promising in making more accurate/precise predictions. Being aware of the difficulty to handle objects based on EBLUP-RK as well as BAGAP when they are embedded in environmental applications, we explore the susceptibility of them in being wrapped within Web Processing Services. Two further kinds of aspects are faced for an exhaustive evaluation and comparison: automaticity and time of calculation with/without high performance computing leverage.

  1. Training Spiking Neural Models Using Artificial Bee Colony

    PubMed Central

    Vazquez, Roberto A.; Garro, Beatriz A.

    2015-01-01

    Spiking neurons are models designed to simulate, in a realistic manner, the behavior of biological neurons. Recently, it has been proven that this type of neurons can be applied to solve pattern recognition problems with great efficiency. However, the lack of learning strategies for training these models do not allow to use them in several pattern recognition problems. On the other hand, several bioinspired algorithms have been proposed in the last years for solving a broad range of optimization problems, including those related to the field of artificial neural networks (ANNs). Artificial bee colony (ABC) is a novel algorithm based on the behavior of bees in the task of exploring their environment to find a food source. In this paper, we describe how the ABC algorithm can be used as a learning strategy to train a spiking neuron aiming to solve pattern recognition problems. Finally, the proposed approach is tested on several pattern recognition problems. It is important to remark that to realize the powerfulness of this type of model only one neuron will be used. In addition, we analyze how the performance of these models is improved using this kind of learning strategy. PMID:25709644

  2. Didactic Strategy Discussion Based on Artificial Neural Networks Results.

    NASA Astrophysics Data System (ADS)

    Andina, D.; Bermúdez-Valbuena, R.

    2009-04-01

    Artificial Neural Networks (ANNs) are a mathematical model of the main known characteristics of biological brian dynamics. ANNs inspired in biological reality have been useful to design machines that show some human-like behaviours. Based on them, many experimentes have been succesfully developed emulating several biologial neurons characteristics, as learning how to solve a given problem. Sometimes, experimentes on ANNs feedback to biology and allow advances in understanding the biological brian behaviour, allowing the proposal of new therapies for medical problems involving neurons performing. Following this line, the author present results on artificial learning on ANN, and interpret them aiming to reinforce one of this two didactic estrategies to learn how to solve a given difficult task: a) To train with clear, simple, representative examples and feel confidence in brian generalization capabilities to achieve succes in more complicated cases. b) To teach with a set of difficult cases of the problem feeling confidence that the brian will efficiently solve the rest of cases if it is able to solve the difficult ones. Results may contribute in the discussion of how to orientate the design innovative succesful teaching strategies in the education field.

  3. Training spiking neural models using artificial bee colony.

    PubMed

    Vazquez, Roberto A; Garro, Beatriz A

    2015-01-01

    Spiking neurons are models designed to simulate, in a realistic manner, the behavior of biological neurons. Recently, it has been proven that this type of neurons can be applied to solve pattern recognition problems with great efficiency. However, the lack of learning strategies for training these models do not allow to use them in several pattern recognition problems. On the other hand, several bioinspired algorithms have been proposed in the last years for solving a broad range of optimization problems, including those related to the field of artificial neural networks (ANNs). Artificial bee colony (ABC) is a novel algorithm based on the behavior of bees in the task of exploring their environment to find a food source. In this paper, we describe how the ABC algorithm can be used as a learning strategy to train a spiking neuron aiming to solve pattern recognition problems. Finally, the proposed approach is tested on several pattern recognition problems. It is important to remark that to realize the powerfulness of this type of model only one neuron will be used. In addition, we analyze how the performance of these models is improved using this kind of learning strategy. PMID:25709644

  4. Reliability and risk analysis using artificial neural networks

    SciTech Connect

    Robinson, D.G.

    1995-12-31

    This paper discusses preliminary research at Sandia National Laboratories into the application of artificial neural networks for reliability and risk analysis. The goal of this effort is to develop a reliability based methodology that captures the complex relationship between uncertainty in material properties and manufacturing processes and the resulting uncertainty in life prediction estimates. The inputs to the neural network model are probability density functions describing system characteristics and the output is a statistical description of system performance. The most recent application of this methodology involves the comparison of various low-residue, lead-free soldering processes with the desire to minimize the associated waste streams with no reduction in product reliability. Model inputs include statistical descriptions of various material properties such as the coefficients of thermal expansion of solder and substrate. Consideration is also given to stochastic variation in the operational environment to which the electronic components might be exposed. Model output includes a probabilistic characterization of the fatigue life of the surface mounted component.

  5. Gait quality assessment using self-organising artificial neural networks.

    PubMed

    Barton, Gabor; Lisboa, Paulo; Lees, Adrian; Attfield, Steve

    2007-03-01

    In this study, the challenge to maximise the potential of gait analysis by employing advanced methods was addressed by using self-organising neural networks to quantify the deviation of patients' gait from normal. Data including three-dimensional joint angles, moments and powers of the two lower limbs and the pelvis were used to train Kohonen artificial neural networks to learn an abstract definition of normal gait. Subsequently, data from patients with gait problems were presented to the network which quantified the quality of gait in the form of a single curve by calculating the quantisation error during the gait cycle. A sensitivity analysis involving the manipulation of gait variables' weighting was able to highlight specific causes of the deviation including the anatomical location and the timing of wrong gait patterns. Use of the quantisation error can be regarded as an extension of previously described gait indices because it measures the goodness of gait and additionally provides information related to the causes underlying gait deviations.

  6. Nuclear power plant fault-diagnosis using artificial neural networks

    SciTech Connect

    Kim, Keehoon; Aljundi, T.L.; Bartlett, E.B.

    1992-12-31

    Artificial neural networks (ANNs) have been applied to various fields due to their fault and noise tolerance and generalization characteristics. As an application to nuclear engineering, we apply neural networks to the early recognition of nuclear power plant operational transients. If a transient or accident occurs, the network will advise the plant operators in a timely manner. More importantly, we investigate the ability of the network to provide a measure of the confidence level in its diagnosis. In this research an ANN is trained to diagnose the status of the San Onofre Nuclear Generation Station using data obtained from the plant`s training simulator. Stacked generalization is then applied to predict the error in the ANN diagnosis. The data used consisted of 10 scenarios that include typical design basis accidents as well as less severe transients. The results show that the trained network is capable of diagnosing all 10 instabilities as well as providing a measure of the level of confidence in its diagnoses.

  7. Classification of Images Acquired with Colposcopy Using Artificial Neural Networks

    PubMed Central

    Simões, Priscyla W; Izumi, Narjara B; Casagrande, Ramon S; Venson, Ramon; Veronezi, Carlos D; Moretti, Gustavo P; da Rocha, Edroaldo L; Cechinel, Cristian; Ceretta, Luciane B; Comunello, Eros; Martins, Paulo J; Casagrande, Rogério A; Snoeyer, Maria L; Manenti, Sandra A

    2014-01-01

    OBJECTIVE To explore the advantages of using artificial neural networks (ANNs) to recognize patterns in colposcopy to classify images in colposcopy. PURPOSE Transversal, descriptive, and analytical study of a quantitative approach with an emphasis on diagnosis. The training test e validation set was composed of images collected from patients who underwent colposcopy. These images were provided by a gynecology clinic located in the city of Criciúma (Brazil). The image database (n = 170) was divided; 48 images were used for the training process, 58 images were used for the tests, and 64 images were used for the validation. A hybrid neural network based on Kohonen self-organizing maps and multilayer perceptron (MLP) networks was used. RESULTS After 126 cycles, the validation was performed. The best results reached an accuracy of 72.15%, a sensibility of 69.78%, and a specificity of 68%. CONCLUSION Although the preliminary results still exhibit an average efficiency, the present approach is an innovative and promising technique that should be deeply explored in the context of the present study. PMID:25374454

  8. Artificial neural network prediction of antisense oligodeoxynucleotide activity.

    PubMed

    Giddings, Michael C; Shah, Atul A; Freier, Sue; Atkins, John F; Gesteland, Raymond F; Matveeva, Olga V

    2002-10-01

    An mRNA transcript contains many potential antisense oligodeoxynucleotide target sites. Identification of the most efficacious targets remains an important and challenging problem. Building on separate work that revealed a strong correlation between the inclusion of short sequence motifs and the activity level of an oligo, we have developed a predictive artificial neural network system for mapping tetranucleotide motif content to antisense oligo activity. Trained for high-specificity prediction, the system has been cross-validated against a database of 348 oligos from the literature and a larger proprietary database of 908 oligos. In cross- validation tests the system identified effective oligos (i.e. oligos capable of reducing target mRNA expression to <25% that of the control) with 53% accuracy, in contrast to the <10% success rates commonly reported for trial-and-error oligo selection, suggesting a possible 5-fold reduction in the in vivo screening required to find an active oligo. We have implemented a web interface to a trained neural network. Given an RNA transcript as input, the system identifies the most likely oligo targets and provides estimates of the probabilities that oligos targeted against these sites will be effective. PMID:12364609

  9. A radial basis function neural network based on artificial immune systems for remote sensing image classification

    NASA Astrophysics Data System (ADS)

    Yan, Qin; Zhong, Yanfei

    2008-12-01

    The radial basis function (RBF) neural network is a powerful method for remote sensing image classification. It has a simple architecture and the learning algorithm corresponds to the solution of a linear regression problem, resulting in a fast training process. The main drawback of this strategy is the requirement of an efficient algorithm to determine the number, position, and dispersion of the RBF. Traditional methods to determine the centers are: randomly choose input vectors from the training data set; vectors obtained from unsupervised clustering algorithms, such as k-means, applied to the input data. These conduce that traditional RBF neural network is sensitive to the center initialization. In this paper, the artificial immune network (aiNet) model, a new computational intelligence based on artificial immune networks (AIN), is applied to obtain appropriate centers for remote sensing image classification. In the aiNet-RBF algorihtm, each input pattern corresonds to an antigenic stimulus, while each RBF candidate center is considered to be an element, or cell, of the immune network model. The steps are as follows: A set of candidate centers is initialized at random, where the initial number of candidates and their positions is not crucial to the performance. Then, the clonal selection principle will control which candidates will be selected and how they will be upadated. Note that the clonal selection principle will be responsible for how the centers will represent the training data set. Finally, the immune network will identify and eliminate or suppress self-recognizing individuals to control the number of candidate centers. After the above learning phase, the aiNet network centers represent internal images of the inuput patterns presented to it. The algorithm output is taken to be the matrix of memory cells' coordinates that represent the final centers to be adopted by the RBF network. The stopping criterion of the proposed algorithm is given by a pre

  10. The utilization of neural nets in populating an object-oriented database

    NASA Technical Reports Server (NTRS)

    Campbell, William J.; Hill, Scott E.; Cromp, Robert F.

    1989-01-01

    Existing NASA supported scientific data bases are usually developed, managed and populated in a tedious, error prone and self-limiting way in terms of what can be described in a relational Data Base Management System (DBMS). The next generation Earth remote sensing platforms (i.e., Earth Observation System, (EOS), will be capable of generating data at a rate of over 300 Mbs per second from a suite of instruments designed for different applications. What is needed is an innovative approach that creates object-oriented databases that segment, characterize, catalog and are manageable in a domain-specific context and whose contents are available interactively and in near-real-time to the user community. Described here is work in progress that utilizes an artificial neural net approach to characterize satellite imagery of undefined objects into high-level data objects. The characterized data is then dynamically allocated to an object-oriented data base where it can be reviewed and assessed by a user. The definition, development, and evolution of the overall data system model are steps in the creation of an application-driven knowledge-based scientific information system.

  11. Association by synaptic facilitation in highly damped neural nets.

    PubMed

    Harth, E M; Edgar, S L

    1967-11-01

    Cognitive functions are sought in a homogeneous, randomly connected net of neuron-like elements. Information is assumed to be contained in the instantaneous states of the system, which specify the firing states (off or on) of each neuron in the net. The hypothesis of synaptic facilitation is assumed to be the basis of learning and memory. Owing to the high degree of damping no reverberations occur in the net. However, close analogies can be found between the performance of the net and known association functions of the cerebral cortex, among them various types of conditioned reflexes. The data are obtained by a combination of mathematical analysis and computer simulation. It is emphasized that the biological entity simulated by this model is at best a limited component of the cerebral cortex.

  12. Larger bases and mixed analog/digital neural nets

    SciTech Connect

    Beiu, V.

    1998-12-31

    The paper overviews results dealing with the approximation capabilities of neural networks, and bounds on the size of threshold gate circuits. Based on an explicit numerical algorithm for Kolmogorov`s superpositions the authors show that minimum size neural networks--for implementing any Boolean function--have the identity function as the activation function. Conclusions and several comments on the required precision are ending the paper.

  13. Simple method for identification of skeletons of aporphine alkaloids from 13C NMR data using artificial neural networks.

    PubMed

    Rufino, Alessandra R; Brant, Antônio J C; Santos, João B O; Ferreira, Marcelo J P; Emerenciano, Vicente P

    2005-01-01

    This paper describes the use of artificial neural networks as a theoretical tool in the structural determination of alkaloids from (13)C NMR chemical shift data, aiming to identify skeletal types of those compounds. For that, 162 aporphine alkaloids belonging to 12 different skeletons were codified with their respective (13)C NMR chemical shifts. Each skeleton pertaining to aporphine alkaloid type was used as output, and the (13)C NMR chemical shifts were used as input data of the net. Analyzing the obtained results, one can then affirm the skeleton to which each one of these compounds belongs with high degree of confidence (over 97%). The relation between the correlation coefficient and the number of epochs and the architecture of net (3-layer MLP or 4-layer MLP) were analyzed, too. The analysis showed that the results predicted by the 3-layer MLP networks trained with a number of the epochs higher than 900 epochs are the best ones. The artificial neural nets were shown to be a simple and efficient tool to solve structural elucidation problems making use of (13)C NMR chemical shift data, even when a similarity between the searched skeletons occurs, offering fast and accurate results to identification of skeletons of organic compounds.

  14. Communications and control for electric power systems: Power system stability applications of artificial neural networks

    NASA Technical Reports Server (NTRS)

    Toomarian, N.; Kirkham, Harold

    1994-01-01

    This report investigates the application of artificial neural networks to the problem of power system stability. The field of artificial intelligence, expert systems, and neural networks is reviewed. Power system operation is discussed with emphasis on stability considerations. Real-time system control has only recently been considered as applicable to stability, using conventional control methods. The report considers the use of artificial neural networks to improve the stability of the power system. The networks are considered as adjuncts and as replacements for existing controllers. The optimal kind of network to use as an adjunct to a generator exciter is discussed.

  15. Communications and control for electric power systems: Power system stability applications of artificial neural networks

    SciTech Connect

    Toomarian, N.; Kirkham, H.

    1993-12-01

    This report investigates the application of artificial neural networks to the problem of power system stability. The field of artificial intelligence, expert systems and neural networks is reviewed. Power system operation is discussed with emphasis on stability considerations. Real-time system control has only recently been considered as applicable to stability, using conventional control methods. The report considers the use of artificial neural networks to improve the stability of the power system. The networks are considered as adjuncts and as replacements for existing controllers. The optimal kind of network to use as an adjunct to a generator exciter is discussed.

  16. Surrogate Modeling of Deformable Joint Contact using Artificial Neural Networks

    PubMed Central

    Eskinazi, Ilan; Fregly, Benjamin J.

    2016-01-01

    Deformable joint contact models can be used to estimate loading conditions for cartilage-cartilage, implant-implant, human-orthotic, and foot-ground interactions. However, contact evaluations are often so expensive computationally that they can be prohibitive for simulations or optimizations requiring thousands or even millions of contact evaluations. To overcome this limitation, we developed a novel surrogate contact modeling method based on artificial neural networks (ANNs). The method uses special sampling techniques to gather input-output data points from an original (slow) contact model in multiple domains of input space, where each domain represents a different physical situation likely to be encountered. For each contact force and torque output by the original contact model, a multi-layer feed-forward ANN is defined, trained, and incorporated into a surrogate contact model. As an evaluation problem, we created an ANN-based surrogate contact model of an artificial tibiofemoral joint using over 75,000 evaluations of a fine-grid elastic foundation (EF) contact model. The surrogate contact model computed contact forces and torques about 1000 times faster than a less accurate coarse grid EF contact model. Furthermore, the surrogate contact model was seven times more accurate than the coarse grid EF contact model within the input domain of a walking motion. For larger input domains, the surrogate contact model showed the expected trend of increasing error with increasing domain size. In addition, the surrogate contact model was able to identify out-of-contact situations with high accuracy. Computational contact models created using our proposed ANN approach may remove an important computational bottleneck from musculoskeletal simulations or optimizations incorporating deformable joint contact models. PMID:26220591

  17. Non-Lipschitzian dynamics for neural net modelling

    NASA Technical Reports Server (NTRS)

    Zak, Michail

    1989-01-01

    Failure of the Lipschitz condition in unstable equilibrium points of dynamical systems leads to a multiple-choice response to an initial deterministic input. The evolution of such systems is characterized by a special type of unpredictability measured by unbounded Liapunov exponents. Possible relation of these systems to future neural networks is discussed.

  18. Goal-seeking neural net for recall and recognition

    NASA Astrophysics Data System (ADS)

    Omidvar, Omid M.

    1990-07-01

    Neural networks have been used to mimic cognitive processes which take place in animal brains. The learning capability inherent in neural networks makes them suitable candidates for adaptive tasks such as recall and recognition. The synaptic reinforcements create a proper condition for adaptation, which results in memorization, formation of perception, and higher order information processing activities. In this research a model of a goal seeking neural network is studied and the operation of the network with regard to recall and recognition is analyzed. In these analyses recall is defined as retrieval of stored information where little or no matching is involved. On the other hand recognition is recall with matching; therefore it involves memorizing a piece of information with complete presentation. This research takes the generalized view of reinforcement in which all the signals are potential reinforcers. The neuronal response is considered to be the source of the reinforcement. This local approach to adaptation leads to the goal seeking nature of the neurons as network components. In the proposed model all the synaptic strengths are reinforced in parallel while the reinforcement among the layers is done in a distributed fashion and pipeline mode from the last layer inward. A model of complex neuron with varying threshold is developed to account for inhibitory and excitatory behavior of real neuron. A goal seeking model of a neural network is presented. This network is utilized to perform recall and recognition tasks. The performance of the model with regard to the assigned tasks is presented.

  19. Vector control of wind turbine on the basis of the fuzzy selective neural net*

    NASA Astrophysics Data System (ADS)

    Engel, E. A.; Kovalev, I. V.; Engel, N. E.

    2016-04-01

    An article describes vector control of wind turbine based on fuzzy selective neural net. Based on the wind turbine system’s state, the fuzzy selective neural net tracks an maximum power point under random perturbations. Numerical simulations are accomplished to clarify the applicability and advantages of the proposed vector wind turbine’s control on the basis of the fuzzy selective neuronet. The simulation results show that the proposed intelligent control of wind turbine achieves real-time control speed and competitive performance, as compared to a classical control model with PID controllers based on traditional maximum torque control strategy.

  20. Artificial neural-network based feeder reconfiguration for loss reduction in distribution systems

    SciTech Connect

    Hoyong Kim; Yunseok Ko; Kyunghee Jung . Dept. of Distribution System)

    1993-07-01

    Neural networks have the capability to map the complex and extremely non-linear relationship between the load levels of zone and system topologies, which is required for feeder reconfiguration in distribution systems. This study is intended to propose the strategies to reconfigure the feeder, by using artificial neural networks with mapping ability. Artificial neural networks determine the appropriate system topology that reduces the power loss according to the variation of load pattern. The control strategy can be easily obtained from the system topology which is provided by artificial neural networks. Artificial neural networks are in groups. The first group estimates the proper load level from the load data of each zone, and the second determines the appropriate system topology from the input load level. In addition, several programs with the training set builder are developed for the design, the training and the accuracy test of artificial neural networks. The authors also evaluate the performance of neural networks designed here, on the test distribution system. Neural networks are implemented in FORTRAN language, and trained on the personal computer COMPAQ 386.

  1. A neural net based architecture for the segmentation of mixed gray-level and binary pictures

    NASA Technical Reports Server (NTRS)

    Tabatabai, Ali; Troudet, Terry P.

    1991-01-01

    A neural-net-based architecture is proposed to perform segmentation in real time for mixed gray-level and binary pictures. In this approach, the composite picture is divided into 16 x 16 pixel blocks, which are identified as character blocks or image blocks on the basis of a dichotomy measure computed by an adaptive 16 x 16 neural net. For compression purposes, each image block is further divided into 4 x 4 subblocks; a one-bit nonparametric quantizer is used to encode 16 x 16 character and 4 x 4 image blocks; and the binary map and quantizer levels are obtained through a neural net segmentor over each block. The efficiency of the neural segmentation in terms of computational speed, data compression, and quality of the compressed picture is demonstrated. The effect of weight quantization is also discussed. VLSI implementations of such adaptive neural nets in CMOS technology are described and simulated in real time for a maximum block size of 256 pixels.

  2. Neural net identification of thumb movement using spectral characteristics of magnetic cortical rhythms.

    PubMed

    Portin, K; Kajola, M; Salmelin, R

    1996-04-01

    Neural nets have shown great promise as tools for reducing and examining multi-dimensional data. When carefully tuned with selected data sets of individual subjects neural nets have indisputable potential in identifying distinct stages of voluntary finger movements. However, robust, automatized data description methods would be needed to eventually extend the use of neural networks into visualization of brain activity during more complex, multimodal tasks where the cortical processes are not equally well understood. We explored the suitability of a self-organizing map (SOM) in the widely studied case of voluntary finger movements (left and right thumb), using as input such spectral characteristics that showed systematic task-dependent changes when averaged over repeated movements. SOMs constructed without individual fine-tuning and with generally chosen training parameters from these spectral features identified correctly 85% of the ongoing movements but, somewhat surprisingly, not the side of thumb movement. Even for this inclusive choice of input, the neural nets were sensitive to transient signals, but focused fine tuning, based on a priori known subgroups in the data, is clearly required for more detailed classification. Thus, a neural net visualization is likely not the most attractive first approach for characterization of cortical processing during complex multimodal tasks.

  3. Predicting concrete corrosion of sewers using artificial neural network.

    PubMed

    Jiang, Guangming; Keller, Jurg; Bond, Philip L; Yuan, Zhiguo

    2016-04-01

    Corrosion is often a major failure mechanism for concrete sewers and under such circumstances the sewer service life is largely determined by the progression of microbially induced concrete corrosion. The modelling of sewer processes has become possible due to the improved understanding of in-sewer transformation. Recent systematic studies about the correlation between the corrosion processes and sewer environment factors should be utilized to improve the prediction capability of service life by sewer models. This paper presents an artificial neural network (ANN)-based approach for modelling the concrete corrosion processes in sewers. The approach included predicting the time for the corrosion to initiate and then predicting the corrosion rate after the initiation period. The ANN model was trained and validated with long-term (4.5 years) corrosion data obtained in laboratory corrosion chambers, and further verified with field measurements in real sewers across Australia. The trained model estimated the corrosion initiation time and corrosion rates very close to those measured in Australian sewers. The ANN model performed better than a multiple regression model also developed on the same dataset. Additionally, the ANN model can serve as a prediction framework for sewer service life, which can be progressively improved and expanded by including corrosion rates measured in different sewer conditions. Furthermore, the proposed methodology holds promise to facilitate the construction of analytical models associated with corrosion processes of concrete sewers. PMID:26841228

  4. Artificial neural network Radon inversion for image reconstruction.

    PubMed

    Rodriguez, A F; Blass, W E; Missimer, J H; Leenders, K L

    2001-04-01

    Image reconstruction techniques are essential to computer tomography. Algorithms such as filtered backprojection (FBP) or algebraic techniques are most frequently used. This paper presents an attempt to apply a feed-forward back-propagation supervised artificial neural network (BPN) to tomographic image reconstruction, specifically to positron emission tomography (PET). The main result is that the network trained with Gaussian test images proved to be successful at reconstructing images from projection sets derived from arbitrary objects. Additional results relate to the design of the network and the full width at half maximum (FWHM) of the Gaussians in the training sets. First, the optimal number of nodes in the middle layer is about an order of magnitude less than the number of input or output nodes. Second, the number of iterations required to achieve a required training set tolerance appeared to decrease exponentially with the number of nodes in the middle layer. Finally, for training sets containing Gaussians of a single width, the optimal accuracy of reconstructing the control set is obtained with a FWHM of three pixels. Intended to explore feasibility, the BPN presented in the following does not provide reconstruction accuracy adequate for immediate application to PET. However, the trained network does reconstruct general images independent of the data with which it was trained. Proposed in the concluding section are several possible refinements that should permit the development of a network capable of fast reconstruction of three-dimensional images from the discrete, noisy projection data characteristic of PET.

  5. Artificial neural networks (ANNs) and modeling of powder flow.

    PubMed

    Kachrimanis, K; Karamyan, V; Malamataris, S

    2003-01-01

    Effects of micromeritic properties (bulk, tapped and particle density, particle size and shape) on the flow rate through circular orifices are investigated, for three pharmaceutical excipients (Lactose, Emcompress and Starch) separated in four sieve fractions, and are modeled with the help of artificial neural networks (ANNs). Eight variables were selected as inputs and correlated by applying the Spearman product-moment correlation matrix and the visual component planes of trained Self-Organizing Maps (SOMs). Back-propagation feed-forward ANN with six hidden units in a single hidden layer was selected for modeling experimental data and its predictions were compared with those of the flow equation proposed by. It was found that SOMs are efficient for the identification of co-linearity in the input variables and the ANN is superior to the flow equation since it does not require separate regression for each excipient and its predictive ability is higher. Besides the orifice diameter, most influential and important variable was the difference between tapped and bulk density. From the pruned ANN an approximate non-linear model was extracted, which describes powder flow rate in terms of the four network's input variables of the greatest predictive importance or saliency (difference between tapped and bulk density (x(2)), orifice diameter (x(3)), circle equivalent particle diameter (x(4)) and particle density [equation in text].

  6. Application of artificial neural networks for prediction of photocatalytic reactor.

    PubMed

    Delnavaz, Mohammad

    2015-02-01

    In this paper, forecasting of kinetic constant and efficiency of photocatalytic process of TiO2 nano powder immobilized on light expanded clay aggregates (LECA) was investigated. Synthetic phenolic wastewater, which is toxic and not easily biodegradable, was selected as the pollutant. The efficiency of the process in various operation conditions, including initial phenol concentration, pH, TiO2 concentration, retention time, and UV lamp intensity, was then measured. The TiO2 nano powder was immobilized on LECA using slurry and sol-gel methods. Kinetics of photocatalytic reactions has been proposed to follow the Langmuir-Hinshelwood model in different initial phenol concentration and pH. Several steps of training and testing of the models were used to determine the appropriate architecture of the artificial neural network models (ANNs). The ANN-based models were found to provide an efficient and robust tool in predicting photocatalytic reactor efficiency and kinetic constant for treating phenolic compounds. PMID:25790514

  7. Prediction of Soil Deformation in Tunnelling Using Artificial Neural Networks

    PubMed Central

    Lai, Jinxing

    2016-01-01

    In the past few decades, as a new tool for analysis of the tough geotechnical problems, artificial neural networks (ANNs) have been successfully applied to address a number of engineering problems, including deformation due to tunnelling in various types of rock mass. Unlike the classical regression methods in which a certain form for the approximation function must be presumed, ANNs do not require the complex constitutive models. Additionally, it is traced that the ANN prediction system is one of the most effective ways to predict the rock mass deformation. Furthermore, it could be envisaged that ANNs would be more feasible for the dynamic prediction of displacements in tunnelling in the future, especially if ANN models are combined with other research methods. In this paper, we summarized the state-of-the-art and future research challenges of ANNs on the tunnel deformation prediction. And the application cases as well as the improvement of ANN models were also presented. The presented ANN models can serve as a benchmark for effective prediction of the tunnel deformation with characters of nonlinearity, high parallelism, fault tolerance, learning, and generalization capability. PMID:26819587

  8. Statistical downscaling rainfall using artificial neural network: significantly wetter Bangkok?

    NASA Astrophysics Data System (ADS)

    Vu, Minh Tue; Aribarg, Thannob; Supratid, Siriporn; Raghavan, Srivatsan V.; Liong, Shie-Yui

    2015-08-01

    Artificial neural network (ANN) is an established technique with a flexible mathematical structure that is capable of identifying complex nonlinear relationships between input and output data. The present study utilizes ANN as a method of statistically downscaling global climate models (GCMs) during the rainy season at meteorological site locations in Bangkok, Thailand. The study illustrates the applications of the feed forward back propagation using large-scale predictor variables derived from both the ERA-Interim reanalyses data and present day/future GCM data. The predictors are first selected over different grid boxes surrounding Bangkok region and then screened by using principal component analysis (PCA) to filter the best correlated predictors for ANN training. The reanalyses downscaled results of the present day climate show good agreement against station precipitation with a correlation coefficient of 0.8 and a Nash-Sutcliffe efficiency of 0.65. The final downscaled results for four GCMs show an increasing trend of precipitation for rainy season over Bangkok by the end of the twenty-first century. The extreme values of precipitation determined using statistical indices show strong increases of wetness. These findings will be useful for policy makers in pondering adaptation measures due to flooding such as whether the current drainage network system is sufficient to meet the changing climate and to plan for a range of related adaptation/mitigation measures.

  9. Detection of spikes with artificial neural networks using raw EEG.

    PubMed

    Ozdamar, O; Kalayci, T

    1998-04-01

    Artificial neural networks (ANN) using raw electroencephalogram (EEG) data were developed and tested off-line to detect transient epileptiform discharges (spike and spike/wave) and EMG activity in an ongoing EEG. In the present study, a feedforward ANN with a variable number of input and hidden layer units and two output units was used to optimize the detection system. The ANN system was trained and tested with the backpropagation algorithm using a large data set of exemplars. The effects of different EEG time windows and the number of hidden layer neurons were examined using rigorous statistical tests for optimum detection sensitivity and selectivity. The best ANN configuration occurred with an input time window of 150 msec (30 input units) and six hidden layer neurons. This input interval contained information on the wave component of the epileptiform discharge which improved detection. Two-dimensional receiver operating curves were developed to define the optimum threshold parameters for best detection. Comparison with previous networks using raw EEG showed improvement in both sensitivity and selectivity. This study showed that raw EEG can be successfully used to train ANNs to detect epileptogenic discharges with a high success rate without resorting to experimenter-selected parameters which may limit the efficiency of the system.

  10. Spatiotemporal modeling of monthly soil temperature using artificial neural networks

    NASA Astrophysics Data System (ADS)

    Wu, Wei; Tang, Xiao-Ping; Guo, Nai-Jia; Yang, Chao; Liu, Hong-Bin; Shang, Yue-Feng

    2013-08-01

    Soil temperature data are critical for understanding land-atmosphere interactions. However, in many cases, they are limited at both spatial and temporal scales. In the current study, an attempt was made to predict monthly mean soil temperature at a depth of 10 cm using artificial neural networks (ANNs) over a large region with complex terrain. Gridded independent variables, including latitude, longitude, elevation, topographic wetness index, and normalized difference vegetation index, were derived from a digital elevation model and remote sensing images with a resolution of 1 km. The good performance and robustness of the proposed ANNs were demonstrated by comparisons with multiple linear regressions. On average, the developed ANNs presented a relative improvement of about 44 % in root mean square error, 70 % in mean absolute percentage error, and 18 % in coefficient of determination over classical linear models. The proposed ANN models were then applied to predict soil temperatures at unsampled locations across the study area. Spatiotemporal variability of soil temperature was investigated based on the obtained database. Future work will be needed to test the applicability of ANNs for estimating soil temperature at finer scales.

  11. Applying artificial neural networks to modeling the middle atmosphere

    NASA Astrophysics Data System (ADS)

    Xiao, Cunying; Hu, Xiong

    2010-07-01

    An artificial neural network (ANN) is used to model the middle atmosphere using a large number of TIMED/SABER limb sounding temperature profiles. A three-layer feed-forward network is chosen based on the back-propagation (BP) algorithm. Latitude, longitude, and height are chosen as the input vectors of the network while temperature is the output vector. The temperature observations during the period from 13 January through 16 March 2007, which are in the same satellite yaw, are taken as samples to train an ANN. Results suggest that the network has high quality for modeling spatial variations of temperature. Quantitative comparisons between the ANN outputs and those from the popular empirical NRLMSISE-00 model illustrate their generally consistent features and some specific differences. The NRLMSISE-00 model’s zonal mean temperatures are too high by ˜6 K-10 K near the stratopause, and the amplitude and phase of the planetary wave number 1 activity are different in some respects from the ANN simulations above 45-50 km, suggesting improvement is needed in the NRLMSISE-00 model for more accurate simulation near and above the stratopause.

  12. Use of artificial neural network for spatial rainfall analysis

    NASA Astrophysics Data System (ADS)

    Paraskevas, Tsangaratos; Dimitrios, Rozos; Andreas, Benardos

    2014-04-01

    In the present study, the precipitation data measured at 23 rain gauge stations over the Achaia County, Greece, were used to estimate the spatial distribution of the mean annual precipitation values over a specific catchment area. The objective of this work was achieved by programming an Artificial Neural Network (ANN) that uses the feed-forward back-propagation algorithm as an alternative interpolating technique. A Geographic Information System (GIS) was utilized to process the data derived by the ANN and to create a continuous surface that represented the spatial mean annual precipitation distribution. The ANN introduced an optimization procedure that was implemented during training, adjusting the hidden number of neurons and the convergence of the ANN in order to select the best network architecture. The performance of the ANN was evaluated using three standard statistical evaluation criteria applied to the study area and showed good performance. The outcomes were also compared with the results obtained from a previous study in the area of research which used a linear regression analysis for the estimation of the mean annual precipitation values giving more accurate results. The information and knowledge gained from the present study could improve the accuracy of analysis concerning hydrology and hydrogeological models, ground water studies, flood related applications and climate analysis studies.

  13. Design The Cervical Cancer Detector Use The Artificial Neural Network

    NASA Astrophysics Data System (ADS)

    Intan Af'idah, Dwi; Didik Widianto, Eko; Setyawan, Budi

    2013-06-01

    Cancer is one of the contagious diseases that become a public health issue, both in the world and in Indonesia. In the world, 12% of all deaths caused by cancer and is the second killer after cardiovascular disease. Early detection using the IVA is a practical and inexpensive (only requiring acetic acid). However, the accuracy of the method is quite low, as it can not detect the stage of the cancer. While other methods have a better sensitivity than the IVA method, is a method of PAP smear. However, this method is relatively expensive, and requires an experienced pathologist-cytologist. According to the case above, Considered important to make the cancer cervics detector that is used to detect the abnormality and cervical cancer stage and consists of a digital microscope, as well as a computer application based on artificial neural network. The use of cervical cancer detector software and hardware are integrated each other. After the specifications met, the steps to design the cervical cancer detection are: Modifying a conventional microscope by adding a lens, image recording, and the lights, Programming the tools, designing computer applications, Programming features abnormality detection and staging of cancer.

  14. Synoptic Classification and Establishment of Analogues with Artificial Neural Networks

    NASA Astrophysics Data System (ADS)

    Michaelides, S. C.; Liassidou, F.; Schizas, C. N.

    2007-06-01

    Weather charts depicting the spatial distribution of various meteorological parameters constitute an indispensable pictorial tool for meteorologists, in diagnosing and forecasting synoptic conditions and the associated weather. The purpose of the present research is to investigate whether training artificial neural networks can be employed in the objective identification of synoptic patterns on weather charts. In order to achieve this, the daily analyses at 0000UTC for 1996 were employed. The respective data consist of the grid-point values of the geopotential height of the 500 hPa isobaric level in the atmosphere. A uniform grid-point spacing of 2.5° × 2.5° is used and the geographical area covered by the investigation lies between 25°N and 65°N and between 20°W and 50°E, covering Europe, the Middle East and the Northern African Coast. An unsupervised learning self-organizing feature map algorithm, namely the Kohonen's algorithm, was employed. The input consists of the grid-point data described above and the output is the synoptic class which each day belongs to. The results referred to in this study employ the generation of 15 and 20 synoptic classes (more classes have been investigated but the results are not reported here). The results indicate that the present technique produced a satisfactory classification of the synoptic patterns over the geographical region mentioned above. Also, it is revealed that the classification performed in this study exhibits a strong seasonal relationship.

  15. Multiobjective analysis of a public wellfield using artificial neural networks

    USGS Publications Warehouse

    Coppola, E.A.; Szidarovszky, F.; Davis, D.; Spayd, S.; Poulton, M.M.; Roman, E.

    2007-01-01

    As competition for increasingly scarce ground water resources grows, many decision makers may come to rely upon rigorous multiobjective techniques to help identify appropriate and defensible policies, particularly when disparate stakeholder groups are involved. In this study, decision analysis was conducted on a public water supply wellfield to balance water supply needs with well vulnerability to contamination from a nearby ground water contaminant plume. With few alternative water sources, decision makers must balance the conflicting objectives of maximizing water supply volume from noncontaminated wells while minimizing their vulnerability to contamination from the plume. Artificial neural networks (ANNs) were developed with simulation data from a numerical ground water flow model developed for the study area. The ANN-derived state transition equations were embedded into a multiobjective optimization model, from which the Pareto frontier or trade-off curve between water supply and wellfield vulnerability was identified. Relative preference values and power factors were assigned to the three stakeholders, namely the company whose waste contaminated the aquifer, the community supplied by the wells, and the water utility company that owns and operates the wells. A compromise pumping policy that effectively balances the two conflicting objectives in accordance with the preferences of the three stakeholder groups was then identified using various distance-based methods. ?? 2006 National Ground Water Association.

  16. Prediction of problematic wine fermentations using artificial neural networks.

    PubMed

    Román, R César; Hernández, O Gonzalo; Urtubia, U Alejandra

    2011-11-01

    Artificial neural networks (ANNs) have been used for the recognition of non-linear patterns, a characteristic of bioprocesses like wine production. In this work, ANNs were tested to predict problems of wine fermentation. A database of about 20,000 data from industrial fermentations of Cabernet Sauvignon and 33 variables was used. Two different ways of inputting data into the model were studied, by points and by fermentation. Additionally, different sub-cases were studied by varying the predictor variables (total sugar, alcohol, glycerol, density, organic acids and nitrogen compounds) and the time of fermentation (72, 96 and 256 h). The input of data by fermentations gave better results than the input of data by points. In fact, it was possible to predict 100% of normal and problematic fermentations using three predictor variables: sugars, density and alcohol at 72 h (3 days). Overall, ANNs were capable of obtaining 80% of prediction using only one predictor variable at 72 h; however, it is recommended to add more fermentations to confirm this promising result.

  17. Atmospheric controls on Puerto Rico precipitation using artificial neural networks

    NASA Astrophysics Data System (ADS)

    Ramseyer, Craig A.; Mote, Thomas L.

    2016-01-01

    The growing need for local climate change scenarios has given rise to a wide range of empirical climate downscaling techniques. One of the most critical decisions in these methodologies is the selection of appropriate predictor variables for the downscaled surface predictand. A systematic approach to selecting predictor variables should be employed to ensure that the most important variables are utilized for the study site where the climate change scenarios are being developed. Tropical study areas have been far less examined than mid- and high-latitudes in the climate downscaling literature. As a result, studies analyzing optimal predictor variables for tropics are limited. The objectives of this study include developing artificial neural networks for six sites around Puerto Rico to develop nonlinear functions between 37 atmospheric predictor variables and local rainfall. The relative importance of each predictor is analyzed to determine the most important inputs in the network. Randomized ANNs are produced to determine the statistical significance of the relative importance of each predictor variable. Lower tropospheric moisture and winds are shown to be the most important variables at all sites. Results show inter-site variability in u- and v-wind importance depending on the unique geographic situation of the site. Lower tropospheric moisture and winds are physically linked to variability in sea surface temperatures (SSTs) and the strength and position of the North Atlantic High Pressure cell (NAHP). The changes forced by anthropogenic climate change in regional SSTs and the NAHP will impact rainfall variability in Puerto Rico.

  18. Unique applications for artificial neural networks. Phase 1. Final report

    SciTech Connect

    Not Available

    1991-08-08

    The investigation concerns the application of modular neural networks, working synergistically with genetic search, to provide a powerful means of intelligently controlling heuristic mathematical algorithms for large-scale vehicle routing and scheduling problems. The design lends itself naturally to parallel computing on loosely coupled networks of computers, and to implementation on parallel architectures such as MIMD machines. Extensive developmental work, coding and computational testing was carried on generic vehicle routing problems. The results are consistently superior to known alternatives, and provide strong motivation to extend the approach into more complex problem domains and military applications. The basic approach was also applied to routing problems with time constraints, a significant complication of considerable practical importance. Results of this problem are also consistently good, and there is potential to further investigate the use of the approach in this domain. Finally, very preliminary results are available for applying the methodology to routing and mission planning for remote autonomous military vehicles, such as Tomahawk cruise missiles or other smart weapons systems. In summary, the high performance achieved suggests that the multiparadigm approaches that utilize methods from artificial intelligence in conjunction with powerful and proven methods from mathematical combinatorial optimization can build upon the strengths of each constituent, and achieve performance that none of the methods can obtain in isolation.

  19. Artificial Neural Networks in Mammography Interpretation and Diagnostic Decision Making

    PubMed Central

    Burnside, Elizabeth S.

    2013-01-01

    Screening mammography is the most effective means for early detection of breast cancer. Although general rules for discriminating malignant and benign lesions exist, radiologists are unable to perfectly detect and classify all lesions as malignant and benign, for many reasons which include, but are not limited to, overlap of features that distinguish malignancy, difficulty in estimating disease risk, and variability in recommended management. When predictive variables are numerous and interact, ad hoc decision making strategies based on experience and memory may lead to systematic errors and variability in practice. The integration of computer models to help radiologists increase the accuracy of mammography examinations in diagnostic decision making has gained increasing attention in the last two decades. In this study, we provide an overview of one of the most commonly used models, artificial neural networks (ANNs), in mammography interpretation and diagnostic decision making and discuss important features in mammography interpretation. We conclude by discussing several common limitations of existing research on ANN-based detection and diagnostic models and provide possible future research directions. PMID:23781276

  20. Automatic classification of DMSA scans using an artificial neural network.

    PubMed

    Wright, J W; Duguid, R; McKiddie, F; Staff, R T

    2014-04-01

    DMSA imaging is carried out in nuclear medicine to assess the level of functional renal tissue in patients. This study investigated the use of an artificial neural network to perform diagnostic classification of these scans. Using the radiological report as the gold standard, the network was trained to classify DMSA scans as positive or negative for defects using a representative sample of 257 previously reported images. The trained network was then independently tested using a further 193 scans and achieved a binary classification accuracy of 95.9%. The performance of the network was compared with three qualified expert observers who were asked to grade each scan in the 193 image testing set on a six point defect scale, from 'definitely normal' to 'definitely abnormal'. A receiver operating characteristic analysis comparison between a consensus operator, generated from the scores of the three expert observers, and the network revealed a statistically significant increase (α < 0.05) in performance between the network and operators. A further result from this work was that when suitably optimized, a negative predictive value of 100% for renal defects was achieved by the network, while still managing to identify 93% of the negative cases in the dataset. These results are encouraging for application of such a network as a screening tool or quality assurance assistant in clinical practice.

  1. Automatic classification of DMSA scans using an artificial neural network

    NASA Astrophysics Data System (ADS)

    Wright, J. W.; Duguid, R.; Mckiddie, F.; Staff, R. T.

    2014-04-01

    DMSA imaging is carried out in nuclear medicine to assess the level of functional renal tissue in patients. This study investigated the use of an artificial neural network to perform diagnostic classification of these scans. Using the radiological report as the gold standard, the network was trained to classify DMSA scans as positive or negative for defects using a representative sample of 257 previously reported images. The trained network was then independently tested using a further 193 scans and achieved a binary classification accuracy of 95.9%. The performance of the network was compared with three qualified expert observers who were asked to grade each scan in the 193 image testing set on a six point defect scale, from ‘definitely normal’ to ‘definitely abnormal’. A receiver operating characteristic analysis comparison between a consensus operator, generated from the scores of the three expert observers, and the network revealed a statistically significant increase (α < 0.05) in performance between the network and operators. A further result from this work was that when suitably optimized, a negative predictive value of 100% for renal defects was achieved by the network, while still managing to identify 93% of the negative cases in the dataset. These results are encouraging for application of such a network as a screening tool or quality assurance assistant in clinical practice.

  2. Using artificial neural network tools to analyze microbial biomarker data

    SciTech Connect

    Brandt, C.C.; Schryver, J.C.; Almeida, J.S.; Pfiffner, S.M.; Palumbo, A.V.

    2004-03-17

    A major challenge in the successful implementation of bioremediation is understanding the structure of the indigenous microbial community and how this structure is affected by environmental conditions. Culture-independent approaches that use biomolecular markers have become the key to comparative microbial community analysis. However, the analysis of biomarkers from environmental samples typically generates a large number of measurements. The large number and complex nonlinear relationships among these measurements makes conventional linear statistical analysis of the data difficult. New data analysis tools are needed to help understand these data. We adapted artificial neural network (ANN) tools for relating changes in microbial biomarkers to geochemistry. ANNs are nonlinear pattern recognition methods that can learn from experience to improve their performance. We have successfully applied these techniques to the analysis of membrane lipids and nucleic acid biomarker data from both laboratory and field studies. Although ANNs typically outperform linear data analysis techniques, the user must be aware of several considerations and issues to ensure that analysis results are not misleading: (1) Overfitting, especially in small sample size data sets; (2) Model selection; (3) Interpretation of analysis results; and (4) Availability of tools (code). This poster summarizes approaches for addressing each of these issues. The objectives are: (1) Develop new nonlinear data analysis tools for relating microbial biomolecular markers to geochemical conditions; (2) Apply these nonlinear tools to field and laboratory studies relevant to the NABIR Program; and (3) Provide these tools and guidance in their use to other researchers.

  3. Atmospheric controls on Puerto Rico precipitation using artificial neural networks

    NASA Astrophysics Data System (ADS)

    Ramseyer, Craig A.; Mote, Thomas L.

    2016-10-01

    The growing need for local climate change scenarios has given rise to a wide range of empirical climate downscaling techniques. One of the most critical decisions in these methodologies is the selection of appropriate predictor variables for the downscaled surface predictand. A systematic approach to selecting predictor variables should be employed to ensure that the most important variables are utilized for the study site where the climate change scenarios are being developed. Tropical study areas have been far less examined than mid- and high-latitudes in the climate downscaling literature. As a result, studies analyzing optimal predictor variables for tropics are limited. The objectives of this study include developing artificial neural networks for six sites around Puerto Rico to develop nonlinear functions between 37 atmospheric predictor variables and local rainfall. The relative importance of each predictor is analyzed to determine the most important inputs in the network. Randomized ANNs are produced to determine the statistical significance of the relative importance of each predictor variable. Lower tropospheric moisture and winds are shown to be the most important variables at all sites. Results show inter-site variability in u- and v-wind importance depending on the unique geographic situation of the site. Lower tropospheric moisture and winds are physically linked to variability in sea surface temperatures (SSTs) and the strength and position of the North Atlantic High Pressure cell (NAHP). The changes forced by anthropogenic climate change in regional SSTs and the NAHP will impact rainfall variability in Puerto Rico.

  4. Predicting concrete corrosion of sewers using artificial neural network.

    PubMed

    Jiang, Guangming; Keller, Jurg; Bond, Philip L; Yuan, Zhiguo

    2016-04-01

    Corrosion is often a major failure mechanism for concrete sewers and under such circumstances the sewer service life is largely determined by the progression of microbially induced concrete corrosion. The modelling of sewer processes has become possible due to the improved understanding of in-sewer transformation. Recent systematic studies about the correlation between the corrosion processes and sewer environment factors should be utilized to improve the prediction capability of service life by sewer models. This paper presents an artificial neural network (ANN)-based approach for modelling the concrete corrosion processes in sewers. The approach included predicting the time for the corrosion to initiate and then predicting the corrosion rate after the initiation period. The ANN model was trained and validated with long-term (4.5 years) corrosion data obtained in laboratory corrosion chambers, and further verified with field measurements in real sewers across Australia. The trained model estimated the corrosion initiation time and corrosion rates very close to those measured in Australian sewers. The ANN model performed better than a multiple regression model also developed on the same dataset. Additionally, the ANN model can serve as a prediction framework for sewer service life, which can be progressively improved and expanded by including corrosion rates measured in different sewer conditions. Furthermore, the proposed methodology holds promise to facilitate the construction of analytical models associated with corrosion processes of concrete sewers.

  5. Designing Artificial Neural Networks Using Particle Swarm Optimization Algorithms.

    PubMed

    Garro, Beatriz A; Vázquez, Roberto A

    2015-01-01

    Artificial Neural Network (ANN) design is a complex task because its performance depends on the architecture, the selected transfer function, and the learning algorithm used to train the set of synaptic weights. In this paper we present a methodology that automatically designs an ANN using particle swarm optimization algorithms such as Basic Particle Swarm Optimization (PSO), Second Generation of Particle Swarm Optimization (SGPSO), and a New Model of PSO called NMPSO. The aim of these algorithms is to evolve, at the same time, the three principal components of an ANN: the set of synaptic weights, the connections or architecture, and the transfer functions for each neuron. Eight different fitness functions were proposed to evaluate the fitness of each solution and find the best design. These functions are based on the mean square error (MSE) and the classification error (CER) and implement a strategy to avoid overtraining and to reduce the number of connections in the ANN. In addition, the ANN designed with the proposed methodology is compared with those designed manually using the well-known Back-Propagation and Levenberg-Marquardt Learning Algorithms. Finally, the accuracy of the method is tested with different nonlinear pattern classification problems. PMID:26221132

  6. Prediction of Soil Deformation in Tunnelling Using Artificial Neural Networks.

    PubMed

    Lai, Jinxing; Qiu, Junling; Feng, Zhihua; Chen, Jianxun; Fan, Haobo

    2016-01-01

    In the past few decades, as a new tool for analysis of the tough geotechnical problems, artificial neural networks (ANNs) have been successfully applied to address a number of engineering problems, including deformation due to tunnelling in various types of rock mass. Unlike the classical regression methods in which a certain form for the approximation function must be presumed, ANNs do not require the complex constitutive models. Additionally, it is traced that the ANN prediction system is one of the most effective ways to predict the rock mass deformation. Furthermore, it could be envisaged that ANNs would be more feasible for the dynamic prediction of displacements in tunnelling in the future, especially if ANN models are combined with other research methods. In this paper, we summarized the state-of-the-art and future research challenges of ANNs on the tunnel deformation prediction. And the application cases as well as the improvement of ANN models were also presented. The presented ANN models can serve as a benchmark for effective prediction of the tunnel deformation with characters of nonlinearity, high parallelism, fault tolerance, learning, and generalization capability. PMID:26819587

  7. Designing Artificial Neural Networks Using Particle Swarm Optimization Algorithms

    PubMed Central

    Garro, Beatriz A.; Vázquez, Roberto A.

    2015-01-01

    Artificial Neural Network (ANN) design is a complex task because its performance depends on the architecture, the selected transfer function, and the learning algorithm used to train the set of synaptic weights. In this paper we present a methodology that automatically designs an ANN using particle swarm optimization algorithms such as Basic Particle Swarm Optimization (PSO), Second Generation of Particle Swarm Optimization (SGPSO), and a New Model of PSO called NMPSO. The aim of these algorithms is to evolve, at the same time, the three principal components of an ANN: the set of synaptic weights, the connections or architecture, and the transfer functions for each neuron. Eight different fitness functions were proposed to evaluate the fitness of each solution and find the best design. These functions are based on the mean square error (MSE) and the classification error (CER) and implement a strategy to avoid overtraining and to reduce the number of connections in the ANN. In addition, the ANN designed with the proposed methodology is compared with those designed manually using the well-known Back-Propagation and Levenberg-Marquardt Learning Algorithms. Finally, the accuracy of the method is tested with different nonlinear pattern classification problems. PMID:26221132

  8. Signal processing using artificial neural network for BOTDA sensor system.

    PubMed

    Azad, Abul Kalam; Wang, Liang; Guo, Nan; Tam, Hwa-Yaw; Lu, Chao

    2016-03-21

    We experimentally demonstrate the use of artificial neural network (ANN) to process sensing signals obtained from Brillouin optical time domain analyzer (BOTDA). The distributed temperature information is extracted directly from the local Brillouin gain spectra (BGSs) along the fiber under test without the process of determination of Brillouin frequency shift (BFS) and hence conversion from BFS to temperature. Unlike our previous work for short sensing distance where ANN is trained by measured BGSs, here we employ ideal BGSs with different linewidths to train the ANN in order to take the linewidth variation due to different conditions from the training and testing phases into account, making it feasible for long distance sensing. Moreover, the performance of ANN is compared with other two techniques, Lorentzian curve fitting and cross-correlation method, and our results show that ANN has higher accuracy and larger tolerance to measurement error, especially at large frequency scanning step. We also show that the temperature extraction from BOTDA measurements employing ANN is significantly faster than the other two approaches. Hence ANN can be an excellent alternative tool to process BGSs measured by BOTDA and obtain temperature distribution along the fiber, especially when large frequency scanning step is adopted to significantly reduce the measurement time but without sacrifice of sensing accuracy. PMID:27136863

  9. Signal processing using artificial neural network for BOTDA sensor system.

    PubMed

    Azad, Abul Kalam; Wang, Liang; Guo, Nan; Tam, Hwa-Yaw; Lu, Chao

    2016-03-21

    We experimentally demonstrate the use of artificial neural network (ANN) to process sensing signals obtained from Brillouin optical time domain analyzer (BOTDA). The distributed temperature information is extracted directly from the local Brillouin gain spectra (BGSs) along the fiber under test without the process of determination of Brillouin frequency shift (BFS) and hence conversion from BFS to temperature. Unlike our previous work for short sensing distance where ANN is trained by measured BGSs, here we employ ideal BGSs with different linewidths to train the ANN in order to take the linewidth variation due to different conditions from the training and testing phases into account, making it feasible for long distance sensing. Moreover, the performance of ANN is compared with other two techniques, Lorentzian curve fitting and cross-correlation method, and our results show that ANN has higher accuracy and larger tolerance to measurement error, especially at large frequency scanning step. We also show that the temperature extraction from BOTDA measurements employing ANN is significantly faster than the other two approaches. Hence ANN can be an excellent alternative tool to process BGSs measured by BOTDA and obtain temperature distribution along the fiber, especially when large frequency scanning step is adopted to significantly reduce the measurement time but without sacrifice of sensing accuracy.

  10. Improved Diagnostics Using Polarization Imaging and Artificial Neural Networks

    PubMed Central

    Xuan, Jianhua; Klimach, Uwe; Zhao, Hongzhi; Chen, Qiushui; Zou, Yingyin; Wang, Yue

    2007-01-01

    In recent years, there has been an increasing interest in studying the propagation of polarized light in biological cells and tissues. This paper presents a novel approach to cell or tissue imaging using a full Stokes imaging system with advanced polarization image analysis algorithms for improved diagnostics. The key component of the Stokes imaging system is the electrically tunable retarder, enabling high-speed operation of the system to acquire four intensity images sequentially. From the acquired intensity images, four Stokes vector images can be computed to obtain complete polarization information. Polarization image analysis algorithms are then developed to analyze Stokes polarization images for cell or tissue classification. Specifically, wavelet transforms are first applied to the Stokes components for initial feature analysis and extraction. Artificial neural networks (ANNs) are then used to extract diagnostic features for improved classification and prediction. In this study, phantom experiments have been conducted using a prototyped Stokes polarization imaging device. In particular, several types of phantoms, consisting of polystyrene latex spheres in various diameters, were prepared to simulate different conditions of epidermal layer of skin. The experimental results from phantom studies and a plant cell study show that the classification performance using Stokes images is significantly improved over that using the intensity image only. PMID:18274657

  11. Classification of breast abnormalities using artificial neural network

    NASA Astrophysics Data System (ADS)

    Zaman, Nur Atiqah Kamarul; Rahman, Wan Eny Zarina Wan Abdul; Jumaat, Abdul Kadir; Yasiran, Siti Salmah

    2015-05-01

    Classification is the process of recognition, differentiation and categorizing objects into groups. Breast abnormalities are calcifications which are tumor markers that indicate the presence of cancer in the breast. The aims of this research are to classify the types of breast abnormalities using artificial neural network (ANN) classifier and to evaluate the accuracy performance using receiver operating characteristics (ROC) curve. The methods used in this research are ANN for breast abnormalities classifications and Canny edge detector as a feature extraction method. Previously the ANN classifier provides only the number of benign and malignant cases without providing information for specific cases. However in this research, the type of abnormality for each image can be obtained. The existing MIAS MiniMammographic database classified the mammogram images into three features only namely characteristic of background tissues, class of abnormality and radius of abnormality. However, in this research three other features are added-in. These three features are number of spots, area and shape of abnormalities. Lastly the performance of the ANN classifier is evaluated using ROC curve. It is found that ANN has an accuracy of 97.9% which is considered acceptable.

  12. Nuclear Mass Systematics With Neural Nets And Astrophysical Nucleosynthesis

    SciTech Connect

    Athanassopoulos, S.; Mavrommatis, E.; Gernoth, K. A.; Clark, J. W.

    2006-04-26

    We construct a neural network model that predicts the differences between the experimental mass-excess values {delta}Mexp and the theoretical values {delta}MFRDM given by the Finite Range Droplet Model of Moeller et al. This difficult study reveals that subtle regularities of nuclear structure not yet embodied in the best microscopic/phenomenological models of atomic-mass systematics do actually exist. By combining the FRDM and the above neural network model we construct a Hybrid Model with improved predictive performance in the majority of the calculations of the systematics of nuclear mass excess and of related quantities. Such systematics is of current interest among others in such astrophysical problems as nucleosynthesis processes and the justification of the present abundances.

  13. Confidence intervals in Flow Forecasting by using artificial neural networks

    NASA Astrophysics Data System (ADS)

    Panagoulia, Dionysia; Tsekouras, George

    2014-05-01

    One of the major inadequacies in implementation of Artificial Neural Networks (ANNs) for flow forecasting is the development of confidence intervals, because the relevant estimation cannot be implemented directly, contrasted to the classical forecasting methods. The variation in the ANN output is a measure of uncertainty in the model predictions based on the training data set. Different methods for uncertainty analysis, such as bootstrap, Bayesian, Monte Carlo, have already proposed for hydrologic and geophysical models, while methods for confidence intervals, such as error output, re-sampling, multi-linear regression adapted to ANN have been used for power load forecasting [1-2]. The aim of this paper is to present the re-sampling method for ANN prediction models and to develop this for flow forecasting of the next day. The re-sampling method is based on the ascending sorting of the errors between real and predicted values for all input vectors. The cumulative sample distribution function of the prediction errors is calculated and the confidence intervals are estimated by keeping the intermediate value, rejecting the extreme values according to the desired confidence levels, and holding the intervals symmetrical in probability. For application of the confidence intervals issue, input vectors are used from the Mesochora catchment in western-central Greece. The ANN's training algorithm is the stochastic training back-propagation process with decreasing functions of learning rate and momentum term, for which an optimization process is conducted regarding the crucial parameters values, such as the number of neurons, the kind of activation functions, the initial values and time parameters of learning rate and momentum term etc. Input variables are historical data of previous days, such as flows, nonlinearly weather related temperatures and nonlinearly weather related rainfalls based on correlation analysis between the under prediction flow and each implicit input

  14. A neural net-based approach to software metrics

    NASA Technical Reports Server (NTRS)

    Boetticher, G.; Srinivas, Kankanahalli; Eichmann, David A.

    1992-01-01

    Software metrics provide an effective method for characterizing software. Metrics have traditionally been composed through the definition of an equation. This approach is limited by the fact that all the interrelationships among all the parameters be fully understood. This paper explores an alternative, neural network approach to modeling metrics. Experiments performed on two widely accepted metrics, McCabe and Halstead, indicate that the approach is sound, thus serving as the groundwork for further exploration into the analysis and design of software metrics.

  15. Artificial Neural Networks Applications: from Aircraft Design Optimization to Orbiting Spacecraft On-board Environment Monitoring

    NASA Technical Reports Server (NTRS)

    Jules, Kenol; Lin, Paul P.

    2002-01-01

    This paper reviews some of the recent applications of artificial neural networks taken from various works performed by the authors over the last four years at the NASA Glenn Research Center. This paper focuses mainly on two areas. First, artificial neural networks application in design and optimization of aircraft/engine propulsion systems to shorten the overall design cycle. Out of that specific application, a generic design tool was developed, which can be used for most design optimization process. Second, artificial neural networks application in monitoring the microgravity quality onboard the International Space Station, using on-board accelerometers for data acquisition. These two different applications are reviewed in this paper to show the broad applicability of artificial intelligence in various disciplines. The intent of this paper is not to give in-depth details of these two applications, but to show the need to combine different artificial intelligence techniques or algorithms in order to design an optimized or versatile system.

  16. Neural-Net Processing of Characteristic Patterns From Electronic Holograms of Vibrating Blades

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J.

    1999-01-01

    Finite-element-model-trained artificial neural networks can be used to process efficiently the characteristic patterns or mode shapes from electronic holograms of vibrating blades. The models used for routine design may not yet be sufficiently accurate for this application. This document discusses the creation of characteristic patterns; compares model generated and experimental characteristic patterns; and discusses the neural networks that transform the characteristic patterns into strain or damage information. The current potential to adapt electronic holography to spin rigs, wind tunnels and engines provides an incentive to have accurate finite element models lor training neural networks.

  17. An artificial neural network system for diagnosing gas turbine engine fuel faults

    SciTech Connect

    Illi, O.J. Jr.; Greitzer, F.L.; Kangas, L.J.; Reeve, T.

    1994-04-01

    The US Army Ordnance Center & School and Pacific Northwest Laboratories are developing a turbine engine diagnostic system for the M1A1 Abrams tank. This system employs Artificial Neural Network (AN) technology to perform diagnosis and prognosis of the tank`s AGT-1500 gas turbine engine. This paper describes the design and prototype development of the ANN component of the diagnostic system, which we refer to as ``TEDANN`` for Turbine Engine Diagnostic Artificial Neural Networks.

  18. Sensitivity and Calibration of Non-Destructive Evaluation Method That Uses Neural-Net Processing of Characteristic Fringe Patterns

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J.; Weiland, Kenneth E.

    2003-01-01

    This paper answers some performance and calibration questions about a non-destructive-evaluation (NDE) procedure that uses artificial neural networks to detect structural damage or other changes from sub-sampled characteristic patterns. The method shows increasing sensitivity as the number of sub-samples increases from 108 to 6912. The sensitivity of this robust NDE method is not affected by noisy excitations of the first vibration mode. A calibration procedure is proposed and demonstrated where the output of a trained net can be correlated with the outputs of the point sensors used for vibration testing. The calibration procedure is based on controlled changes of fastener torques. A heterodyne interferometer is used as a displacement sensor for a demonstration of the challenges to be handled in using standard point sensors for calibration.

  19. Polarography and artificial neural network for the simultaneous determination of nalidixic acid and its main metabolite (7-hydroxymethylnalidixic acid).

    PubMed

    Guiberteau, Agustina; Díaz, Teresa Galeano; Rodríguez Cáceres, María I; Ortiz Burguillos, Juan M; Merás, Isabel Durán; López, Francisco Salinas

    2004-02-01

    Nalidixic acid (NA) and its main metabolite, 7-hydroxymethylnalidixic acid (OH-NA), are simultaneously determined by applying artificial neural networks (ANNs), to their square wave voltammetric signals. The scores of a PCR model, built with the voltammetric data of a set of standard samples, recorded between -0.70 and -1.0V, are used as training set for the net for each compound. The trained nets (ANNs) are used for the simultaneous determination of NA and OH-NA in urine. The recovery values are comprised between 91 and 109% for NA and between 82 and 112% for OH-NA, being these results better than the results obtained by application of partial least squares (PLS) multivariate calibration.

  20. Modeling Career Counselor Decisions with Artificial Neural Networks: Predictions of Fit across a Comprehensive Occupational Map.

    ERIC Educational Resources Information Center

    Carson, Andrew D.; Bizot, Elizabeth B.; Hendershot, Peggy E.; Barton, Margaret G.; Garvin, Mary K.; Kraemer, Barbara

    1999-01-01

    Career recommendations were made based on aptitude scores of 335 high school freshmen. Artificial neural networks were used to map recommendations to 12 occupational clusters. Overall accuracy of neural networks (.80) approached that of discriminant function analysis (.84). The two methods had different strengths and weaknesses. (SK)

  1. Using Artificial Neural Networks in Educational Research: Some Comparisons with Linear Statistical Models.

    ERIC Educational Resources Information Center

    Everson, Howard T.; And Others

    This paper explores the feasibility of neural computing methods such as artificial neural networks (ANNs) and abductory induction mechanisms (AIM) for use in educational measurement. ANNs and AIMS methods are contrasted with more traditional statistical techniques, such as multiple regression and discriminant function analyses, for making…

  2. Neural nets for generalization and classification: comment on Staddon and Reid (1990)

    PubMed

    Shepard, R N

    1990-10-01

    The neural net model of Staddon and Reid (1990) explains exponential and Gaussian generalization gradients in the same way as the diffusion model of Shepard (1958). The "cognitive" generalization theory of Shepard (1987), which also has been implemented as a connectionist network, goes beyond both of these models in accounting for classification learning.

  3. The use of neural nets to combine equalization with decoding for severe intersymbol interference channels.

    PubMed

    Al-Mashouq, K A; Reed, I S

    1994-01-01

    This paper deals with the problem of combining equalization with decoding in channels which have severe intersymbol interference. A multilayer neural net structure is proposed to achieve the process of equalization and decoding simultaneously. Experimental examples show that this method results in a substantial improvement over the more conventional methods of performing equalization and decoding.

  4. Neural Nets for Generalization and Classification: Comment on Staddon and Reid (1990).

    ERIC Educational Resources Information Center

    Shepard, Roger N.

    1990-01-01

    The neural net model of J. E. R. Staddon and A. K. Reid (1990) explains exponential and Gaussian generalization gradients in the same way as the diffusion model of R. N. Shepard (1958). The cognitive generalization theory of Shepard (1987), also implemented as a connectionist network, goes beyond both models in accounting for classification…

  5. A comparison between criterion functions for linear classifiers, with an application to neural nets

    SciTech Connect

    Barnard, E.; Casasent, D. )

    1989-09-01

    A variety of criterion functions (or scalar performance measures) have been suggested for the design of nonparametric linear classifiers. The classification performance of the most important of these on a typical two-class problem are investigated. The results of the investigation are then applied to the analysis and synthesis of neural-net classifiers.classifiers.

  6. A hybrid architecture for the implementation of the Athena neural net model

    NASA Technical Reports Server (NTRS)

    Koutsougeras, C.; Papachristou, C.

    1989-01-01

    The implementation of an earlier introduced neural net model for pattern classification is considered. Data flow principles are employed in the development of a machine that efficiently implements the model and can be useful for real time classification tasks. Further enhancement with optical computing structures is also considered.

  7. ALINET: neural net automatic alignment of high-energy laser resonator optical elements

    NASA Astrophysics Data System (ADS)

    Hart, George A.; Bailey, Adam W.; Palumbo, Louis J.; Kuperstein, Michael

    1993-10-01

    A novel neural net approach has successfully solved the time consuming practical problem of aligning the many optical elements used in the resonator of high power chemical lasers. Moreover, because the neural net can achieve optimal performance in only 2 - 4 steps, as compared with 50 for other techniques, the important ability to effect real time control is gained. This represents a significant experimental breakthrough because of the difficulty previously associated with this alignment process. Use of either near or far field image information produces excellent performance. The method is very robust in the presence of noise. For cases where the initial misalignment falls outside the regime encompassed by the training set, a hybrid approach utilizing an advanced conventional method can bring the optical system within the capture range of the neural net. This reported use of a neural net to rapidly convert imagery information into high precision control information is of broad applicability to optical, acoustic, or electromagnetic alignment, positioning, and control problems.

  8. Neural net classification of x-ray pistachio nut data

    NASA Astrophysics Data System (ADS)

    Casasent, David P.; Sipe, Michael A.; Schatzki, Thomas F.; Keagy, Pamela M.; Le, Lan Chau

    1996-12-01

    Classification results for agricultural products are presented using a new neural network. This neural network inherently produces higher-order decision surfaces. It achieves this with fewer hidden layer neurons than other classifiers require. This gives better generalization. It uses new techniques to select the number of hidden layer neurons and adaptive algorithms that avoid other such ad hoc parameter selection problems; it allows selection of the best classifier parameters without the need to analyze the test set results. The agriculture case study considered is the inspection and classification of pistachio nuts using x- ray imagery. Present inspection techniques cannot provide good rejection of worm damaged nuts without rejecting too many good nuts. X-ray imagery has the potential to provide 100% inspection of such agricultural products in real time. Only preliminary results are presented, but these indicate the potential to reduce major defects to 2% of the crop with 1% of good nuts rejected. Future image processing techniques that should provide better features to improve performance and allow inspection of a larger variety of nuts are noted. These techniques and variations of them have uses in a number of other agricultural product inspection problems.

  9. Application of Artificial Neural Networks in the Heart Electrical Axis Position Conclusion Modeling

    NASA Astrophysics Data System (ADS)

    Bakanovskaya, L. N.

    2016-08-01

    The article touches upon building of a heart electrical axis position conclusion model using an artificial neural network. The input signals of the neural network are the values of deflections Q, R and S; and the output signal is the value of the heart electrical axis position. Training of the network is carried out by the error propagation method. The test results allow concluding that the created neural network makes a conclusion with a high degree of accuracy.

  10. Pattern recognition of respirable dust particles by a back-propagation artificial neural network.

    PubMed

    Wippel, R; Pichler-Semmelrock, F P; Köck, M; Kosmus, W

    2001-05-01

    A back-propagation neural network was used as a pattern recognition tool for LAMMA mass spectral data. Standard EPA source profiles were used as training and test data of the net. The elemental patterns (10 elements) of the sum of 100 mass spectra of fine dust particles were presented to the trained nets and satisfactory recognition (> 50%) was obtained.

  11. Artificial neural network to search for metal-poor galaxies

    NASA Astrophysics Data System (ADS)

    Shi, Fei; Liu, Yu-Yan; Kong, Xu; Chen, Yang

    2014-02-01

    Aims: To find a fast and reliable method for selecting metal-poor galaxies (MPGs), especially in large surveys and huge databases, an artificial neural network (ANN) method is applied to a sample of star-forming galaxies from the Sloan Digital Sky Survey (SDSS) data release 9 (DR9) provided by the Max Planck Institute and the Johns Hopkins University (MPA/JHU). Methods: A two-step approach is adopted: (i) The ANN network must be trained with a subset of objects that are known to be either MPGs or metal rich galaxies (MRGs), treating the strong emission line flux measurements as input feature vectors in n-dimensional space, where n is the number of strong emission line flux ratios. (ii) After the network is trained on a sample of star-forming galaxies, the remaining galaxies are classified in the automatic test analysis as either MPGs or MRGs. We consider several random divisions of the data into training and testing sets; for instance, for our sample, a total of 70 percent of the data are involved in training the algorithm, 15 percent are involved in validating the algorithm, and the remaining 15 percent are used for blind testing the resulting classifier. Results: For target selection, we have achieved an acquisition rate for MPGs of 96 percent and 92 percent for an MPGs threshold of 12 + log (O/H) = 8.00 and 12 + log (O/H) = 8.39, respectively. Running the code takes minutes in most cases under the Matlab 2013a software environment. The ANN method can easily be extended to any MPGs target selection task when the physical property of the target can be expressed as a quantitative variable. The code in the paper is available on the web (http://fshi5388.blog.163.com).

  12. A 2D CMAC neural net algorithm for a positioning system of automated agriculture vehicle

    NASA Astrophysics Data System (ADS)

    Zhang, Fangming; Ying, Yibin

    2006-10-01

    In a machine vision-based guidance system, a camera must be corrected precisely to calculate the position of vehicle, however, it is not easy to obtain the intrinsic and extrinsic parameters of the camera, while neural nets have the advantage to set up a mapping relationship for a nonlinear system. We intended to use the CMAC neural net to construct two map relationships: image coordinates and offsets of the vehicle, and image coordinates and the heading angle of the vehicle. The net inputs were the coordinates of top and bottom points in the detected guidance line in the image coordinate system. The outputs were offsets and heading angles. The verified results show that the RMS of inferred offset is 10.5 mm, and the STD is 11.3 mm; the RMS of inferred heading is 1.1°, and the STD is 0.99°.

  13. A new neural net approach to robot 3D perception and visuo-motor coordination

    NASA Technical Reports Server (NTRS)

    Lee, Sukhan

    1992-01-01

    A novel neural network approach to robot hand-eye coordination is presented. The approach provides a true sense of visual error servoing, redundant arm configuration control for collision avoidance, and invariant visuo-motor learning under gazing control. A 3-D perception network is introduced to represent the robot internal 3-D metric space in which visual error servoing and arm configuration control are performed. The arm kinematic network performs the bidirectional association between 3-D space arm configurations and joint angles, and enforces the legitimate arm configurations. The arm kinematic net is structured by a radial-based competitive and cooperative network with hierarchical self-organizing learning. The main goal of the present work is to demonstrate that the neural net representation of the robot 3-D perception net serves as an important intermediate functional block connecting robot eyes and arms.

  14. Use of artificial neural networks to analyze nuclear power plant performance

    SciTech Connect

    Guo, Z.; Uhrig, R.E. )

    1992-07-01

    This paper discusses a hybrid artificial neural network used to model the thermodynamic behavior of the Tennessee Valley Authority's Sequoyah nuclear power plant using data for heat rate measurements acquired over a 1-yr period. The modeling process involves the use of a self-organizing network to rearrange the original data into several classes by clustering. Then, the centroids of these clusters are used as the training patterns for an artificial neural network that utilizes backpropagation training to adjust the weights on the connections between artificial neurons. This procedure greatly reduces the training time and reduces the system error.

  15. Conditions for periodic and aperiodic behavior of formal neural nets.

    PubMed

    Labos, E

    1994-05-01

    Formal neural networks (FNN) can display dynamical behaviours, more or less different from each other depending on their units, the functions attributed to these units, interconnections, parameters, state spaces and initial states, etc. Whatever is 'chaos' - of which several practical and more exact definitions exist -, it used to be emerging at special conditions. Its prediction most often requires an individual analysis of the dynamical system (DS) in question. A study of such conditions is usually necessary in order to reach suitable control, which now seems to become a new trend in chaos theory. In chaos control tasks quick commands and at least short-term foresight of trends are required. It is a primary question also to define in advance what is regarded to be a controlled case of chaos. Possible importance of these general considerations at molecular scale is also discussed, avoiding not well-founded speculations.

  16. Intelligent control based on fuzzy logic and neural net theory

    NASA Technical Reports Server (NTRS)

    Lee, Chuen-Chien

    1991-01-01

    In the conception and design of intelligent systems, one promising direction involves the use of fuzzy logic and neural network theory to enhance such systems' capability to learn from experience and adapt to changes in an environment of uncertainty and imprecision. Here, an intelligent control scheme is explored by integrating these multidisciplinary techniques. A self-learning system is proposed as an intelligent controller for dynamical processes, employing a control policy which evolves and improves automatically. One key component of the intelligent system is a fuzzy logic-based system which emulates human decision making behavior. It is shown that the system can solve a fairly difficult control learning problem. Simulation results demonstrate that improved learning performance can be achieved in relation to previously described systems employing bang-bang control. The proposed system is relatively insensitive to variations in the parameters of the system environment.

  17. Calibration of a shock wave position sensor using artificial neural networks

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J.; Weiland, Kenneth E.

    1993-01-01

    This report discusses the calibration of a shock wave position sensor. The position sensor works by using artificial neural networks to map cropped CCD frames of the shadows of the shock wave into the value of the shock wave position. This project was done as a tutorial demonstration of method and feasibility. It used a laboratory shadowgraph, nozzle, and commercial neural network package. The results were quite good, indicating that artificial neural networks can be used efficiently to automate the semi-quantitative applications of flow visualization.

  18. Multidimensional interpolation using artificial neural networks: Application to an H I cloud in Perseus

    NASA Astrophysics Data System (ADS)

    Serra-Ricart, Miquel; Trapero, Joaquin; Beckman, John E.; Garrido, Lluis; Gaitan, Vicens

    1995-01-01

    In this paper we propose a method for interpolating multidimensional unbinned data, which could also be sparse, using artificial neural network techniques. An artificial example is first presented in order to show the reliability and potential of the neural network interpolator. A robust behavior is found. We apply the technique to the mapping of a cloud of interstellar atomic hydrogen. The cloud was mapped in H I at 21 cm and we find the neural network method ideal for interpolating the unevenly sampled data, yielding a map from which the global physical parameters of the cloud can be readily obtained.

  19. Elementary derivative tasks and neural net multiscale analysis of tasks.

    PubMed

    Giraud, B G; Touzeau, A

    2002-01-01

    Formal neurons implementing wavelets have been shown to build nets that are able to approximate any multidimensional task. In this paper, we use a finite number of formal neurons implementing elementary tasks such as "sombrero" responses or even simpler "window" responses, with adjustable widths. We show this to provide a reasonably efficient, practical and robust, multifrequency analysis of tasks. The translation degree of freedom of wavelets is shown to be unnecessary. A training algorithm, optimizing the output task with respect to the widths of the responses, reveals two distinct training modes. The first mode keeps the formal neurons distinct. The other mode induces some of the formal neurons to become identical, with output weights of equal strengths but opposite signs. Hence this latter mode promotes tasks that are derivatives of the elementary tasks with respect to the width parameter. Such results, obtained from optimizations with respect to a width parameter, can be generalized for any other parameters of the elementary tasks.

  20. Geometrical approach to neural net control of movements and posture

    NASA Technical Reports Server (NTRS)

    Pellionisz, A. J.; Ramos, C. F.

    1993-01-01

    In one approach to modeling brain function, sensorimotor integration is described as geometrical mapping among coordinates of non-orthogonal frames that are intrinsic to the system; in such a case sensors represent (covariant) afferents and motor effectors represent (contravariant) motor efferents. The neuronal networks that perform such a function are viewed as general tensor transformations among different expressions and metric tensors determining the geometry of neural functional spaces. Although the non-orthogonality of a coordinate system does not impose a specific geometry on the space, this "Tensor Network Theory of brain function" allows for the possibility that the geometry is non-Euclidean. It is suggested that investigation of the non-Euclidean nature of the geometry is the key to understanding brain function and to interpreting neuronal network function. This paper outlines three contemporary applications of such a theoretical modeling approach. The first is the analysis and interpretation of multi-electrode recordings. The internal geometries of neural networks controlling external behavior of the skeletomuscle system is experimentally determinable using such multi-unit recordings. The second application of this geometrical approach to brain theory is modeling the control of posture and movement. A preliminary simulation study has been conducted with the aim of understanding the control of balance in a standing human. The model appears to unify postural control strategies that have previously been considered to be independent of each other. Third, this paper emphasizes the importance of the geometrical approach for the design and fabrication of neurocomputers that could be used in functional neuromuscular stimulation (FNS) for replacing lost motor control.

  1. Geometrical approach to neural net control of movements and posture.

    PubMed

    Pellionisz, A J; Ramos, C F

    1993-01-01

    In one approach to modeling brain function, sensorimotor integration is described as geometrical mapping among coordinates of non-orthogonal frames that are intrinsic to the system; in such a case sensors represent (covariant) afferents and motor effectors represent (contravariant) motor efferents. The neuronal networks that perform such a function are viewed as general tensor transformations among different expressions and metric tensors determining the geometry of neural functional spaces. Although the non-orthogonality of a coordinate system does not impose a specific geometry on the space, this "Tensor Network Theory of brain function" allows for the possibility that the geometry is non-Euclidean. It is suggested that investigation of the non-Euclidean nature of the geometry is the key to understanding brain function and to interpreting neuronal network function. This paper outlines three contemporary applications of such a theoretical modeling approach. The first is the analysis and interpretation of multi-electrode recordings. The internal geometries of neural networks controlling external behavior of the skeletomuscle system is experimentally determinable using such multi-unit recordings. The second application of this geometrical approach to brain theory is modeling the control of posture and movement. A preliminary simulation study has been conducted with the aim of understanding the control of balance in a standing human. The model appears to unify postural control strategies that have previously been considered to be independent of each other. Third, this paper emphasizes the importance of the geometrical approach for the design and fabrication of neurocomputers that could be used in functional neuromuscular stimulation (FNS) for replacing lost motor control. PMID:8234751

  2. Identification of Propionibacteria to the species level using Fourier transform infrared spectroscopy and artificial neural networks.

    PubMed

    Dziuba, B

    2013-01-01

    Fourier transform infrared spectroscopy (FTIR) and artificial neural networks (ANN's) were used to identify species of Propionibacteria strains. The aim of the study was to improve the methodology to identify species of Propionibacteria strains, in which the differentiation index D, calculated based on Pearson's correlation and cluster analyses were used to describe the correlation between the Fourier transform infrared spectra and bacteria as molecular systems brought unsatisfactory results. More advanced statistical methods of identification of the FTIR spectra with application of artificial neural networks (ANN's) were used. In this experiment, the FTIR spectra of Propionibacteria strains stored in the library were used to develop artificial neural networks for their identification. Several multilayer perceptrons (MLP) and probabilistic neural networks (PNN) were tested. The practical value of selected artificial neural networks was assessed based on identification results of spectra of 9 reference strains and 28 isolates. To verify results of isolates identification, the PCR based method with the pairs of species-specific primers was used. The use of artificial neural networks in FTIR spectral analyses as the most advanced chemometric method supported correct identification of 93% bacteria of the genus Propionibacterium to the species level.

  3. Latent Heat and Sensible Heat Fluxes Simulation in Maize Using Artificial Neural Networks

    NASA Astrophysics Data System (ADS)

    Safa, B.

    2015-12-01

    flux net radiation, wind speed, air temperature, leaf area index and vapor pressure deficit. This study reveals that artificial neural networks are not only a powerful technique for estimation of LE and H fluxes, but also can identify the effectiveness of each input on the fluxes.

  4. Artificial neural networks for closed loop control of in silico and ad hoc type 1 diabetes.

    PubMed

    Fernandez de Canete, J; Gonzalez-Perez, S; Ramos-Diaz, J C

    2012-04-01

    The closed loop control of blood glucose levels might help to reduce many short- and long-term complications of type 1 diabetes. Continuous glucose monitoring and insulin pump systems have facilitated the development of the artificial pancreas. In this paper, artificial neural networks are used for both the identification of patient dynamics and the glycaemic regulation. A subcutaneous glucose measuring system together with a Lispro insulin subcutaneous pump were used to gather clinical data for each patient undergoing treatment, and a corresponding in silico and ad hoc neural network model was derived for each patient to represent their particular glucose-insulin relationship. Based on this nonlinear neural network model, an ad hoc neural network controller was designed to close the feedback loop for glycaemic regulation of the in silico patient. Both the neural network model and the controller were tested for each patient under simulation, and the results obtained show a good performance during food intake and variable exercise conditions.

  5. The application of artificial neural networks in astronomy

    NASA Astrophysics Data System (ADS)

    Li, Li-Li; Zhang, Yan-Xia; Zhao, Yong-Heng; Yang, Da-Wei

    2006-12-01

    Artificial Neural Networks (ANNs) are computer algorithms inspired from simple models of human central nervous system activity. They can be roughly divided into two main kinds: supervised and unsupervised. The supervised approach lays the stress on "teaching" a machine to do the work of a mention human expert, usually by showing examples for which the true answer is supplied by the expert. The unsupervised one is aimed at learning new things from the data, and most useful when the data cannot easily be plotted in a two or three dimensional space. ANNs have been used widely and successfully in various fields, for instance, pattern recognition, financial analysis, biology, engineering and so on, because they have many merits such as self-learning, self-adapting, good robustness and dynamically rapid response as well as strong capability of dealing with non-linear problems. In the last few years there has been an increasing interest toward the astronomical applications of ANNs. In this paper, the authors firstly introduce the fundamental principle of ANNs together with the architecture of the network and outline various kinds of learning algorithms and network toplogies. The specific aspects of the applications of ANNs in astronomical problems are also listed, which contain the strong capabilities of approximating to arbitrary accuracy, any nonlinear functional mapping, parallel and distributed storage, tolerance of faulty and generalization of results. They summarize the advantages and disadvantages of main ANN models available to the astronomical community. Furthermore, the application cases of ANNs in astronomy are mainly described in detail. Here, the focus is on some of the most interesting fields of its application, for example: object detection, star/galaxy classification, spectral classification, galaxy morphology classification, the estimation of photometric redshifts of galaxies and time series analysis. In addition, other kinds of applications have been

  6. A solution method of unit commitment by artificial neural networks

    SciTech Connect

    Yokoyama, R. )

    1992-08-01

    This paper explores the possibility of applying the Hopfield neural network to combinatorial optimization problems in power systems, in particular to unit commitment. A large number of inequality constraints included in unit commitment are handled by dedicated neural networks. As an exact mapping of the problem onto the neural network is impossible with the state of the art, the authors have developed a two step solution method: firstly, generators to start up at each period are determined by the network and then their outputs are adjusted by a conventional algorithm. The proposed neural network could solve a unit commitment of 30 units over 24 periods, and results obtained are very encouraging.

  7. Optimization of Training Sets For Neural-Net Processing of Characteristic Patterns From Vibrating Solids

    NASA Technical Reports Server (NTRS)

    Decker, Arthur J. (Inventor)

    2006-01-01

    An artificial neural network is disclosed that processes holography generated characteristic pattern of vibrating structures along with finite-element models. The present invention provides for a folding operation for conditioning training sets for optimally training forward-neural networks to process characteristic fringe pattern. The folding pattern increases the sensitivity of the feed-forward network for detecting changes in the characteristic pattern The folding routine manipulates input pixels so as to be scaled according to the location in an intensity range rather than the position in the characteristic pattern.

  8. Simulated annealing and stochastic learning in optical neural nets: An optical Boltzmann machine

    SciTech Connect

    Shae, Zonyin.

    1989-01-01

    This dissertation deals with the study of stochastic learning and neural computation in opto-electronic hardware. It presents the first demonstration of a fully operational optical learning machine. Learning in the machine is stochastic taking place in a self-organized multi-layered opto-electronic neural net with plastic connectivity weights that are formed in a programmable non-volatile spatial light modulator. Operation of the machine is made possible by two developments in this work: (a) Fast annealing by optically induced tremors in the energy landscape of the net. The objective of this scheme is to exploit the parallelism of the optical noise pattern so as to speed up the simulated annealing process. The procedure can be viewed as that of generating controlled gradually decreasing deformations or tremors in the energy landscape of the net that prevents entrapment in a local minimum energy state. Both the random drawing of neurons and the state update of the net are now done in parallel at the same time and without having to computer explicitly the change in the energy of the net and associated Boltzmann factor as required ordinarily in the Metropolis-Kirkpartrik simulated annealing algorithm. This leads to significant acceleration of the annealing process. (b) Stochastic learning with binary weights. Learning in opto-electronic neural nets can be simplified greatly if binary weights can be used. A third development, that is the development of schemes for driving and enhancing the frame rate of magneto-optic spatial light modulators, can make the machine learning speed potentially fast. Details of these developments together with the principle, architecture, structure, and performance evaluation of this machine are given.

  9. Neural Net-Based Redesign of Transonic Turbines for Improved Unsteady Aerodynamic Performance

    NASA Technical Reports Server (NTRS)

    Madavan, Nateri K.; Rai, Man Mohan; Huber, Frank W.

    1998-01-01

    A recently developed neural net-based aerodynamic design procedure is used in the redesign of a transonic turbine stage to improve its unsteady aerodynamic performance. The redesign procedure used incorporates the advantages of both traditional response surface methodology (RSM) and neural networks by employing a strategy called parameter-based partitioning of the design space. Starting from the reference design, a sequence of response surfaces based on both neural networks and polynomial fits are constructed to traverse the design space in search of an optimal solution that exhibits improved unsteady performance. The procedure combines the power of neural networks and the economy of low-order polynomials (in terms of number of simulations required and network training requirements). A time-accurate, two-dimensional, Navier-Stokes solver is used to evaluate the various intermediate designs and provide inputs to the optimization procedure. The optimization procedure yields a modified design that improves the aerodynamic performance through small changes to the reference design geometry. The computed results demonstrate the capabilities of the neural net-based design procedure, and also show the tremendous advantages that can be gained by including high-fidelity unsteady simulations that capture the relevant flow physics in the design optimization process.

  10. Neural net controller for inlet pressure control of rocket engine testing

    NASA Technical Reports Server (NTRS)

    Trevino, Luis C.

    1994-01-01

    Many dynamic systems operate in select operating regions, each exhibiting characteristic modes of behavior. It is traditional to employ standard adjustable gain proportional-integral-derivative (PID) loops in such systems where no apriori model information is available. However, for controlling inlet pressure for rocket engine testing, problems in fine tuning, disturbance accommodation, and control gains for new profile operating regions (for research and development) are typically encountered. Because of the capability of capturing I/O peculiarities, using NETS, a back propagation trained neural network is specified. For select operating regions, the neural network controller is simulated to be as robust as the PID controller. For a comparative analysis, the higher order moment neural array (HOMNA) method is used to specify a second neural controller by extracting critical exemplars from the I/O data set. Furthermore, using the critical exemplars from the HOMNA method, a third neural controller is developed using NETS back propagation algorithm. All controllers are benchmarked against each other.

  11. Detection of apnea using a short-window FFT technique and an artificial neural network

    NASA Astrophysics Data System (ADS)

    Waldemark, Karina E.; Agehed, Kenneth I.; Lindblad, Thomas; Waldemark, Joakim T. A.

    1998-03-01

    Sleep apnea is characterized by frequent prolonged interruptions of breathing during sleep. This syndrome causes severe sleep disorders and is often responsible for development of other diseases such as heart problems, high blood pressure and daytime fatigue, etc. After diagnosis, sleep apnea is often successfully treated by applying positive air pressure (CPAP) to the mouth and nose. Although effective, the (CPAP) equipment takes up a lot of space and the connected mask causes a lot of inconvenience for the patients. This raised interest in developing new techniques for treatment of sleep apnea syndrome. Several studies have indicated that electrical stimulation of the hypoglossal nerve and muscle in the tongue may be a useful method for treating patients with severe sleep apnea. In order to be able to successfully prevent the occurrence of apnea it is necessary to have some technique for early and fast on-line detection or prediction of the apnea events. This paper suggests using measurements of respiratory airflow (mouth temperature). The signal processing for this task includes the use of a short window FFT technique and uses an artificial back propagation neural net to model or predict the occurrence of apneas. The results show that early detection of respiratory interruption is possible and that the delay time for this is small.

  12. Artificial Neural Network Test Support Development for the Space Shuttle PRCS Thrusters

    NASA Technical Reports Server (NTRS)

    Lehr, Mark E.

    2005-01-01

    A significant anomaly, Fuel Valve Pilot Seal Extrusion, is affecting the Shuttle Primary Reaction Control System (PRCS) Thrusters, and has caused 79 to fail. To help address this problem, a Shuttle PRCS Thruster Process Evaluation Team (TPET) was formed. The White Sands Test Facility (WSTF) and Boeing members of the TPET have identified many discrete valve current trace characteristics that are predictive of the problem. However, these are difficult and time consuming to identify and trend by manual analysis. Based on this exhaustive analysis over months, 22 thrusters previously delivered by the Depot were identified as high risk for flight failures. Although these had only recently been installed, they had to be removed from Shuttles OV103 and OV104 for reprocessing, by directive of the Shuttle Project Office. The resulting impact of the thruster removal, replacement, and valve replacement was significant (months of work and hundreds of thousands of dollars). Much of this could have been saved had the proposed Neural Network (NN) tool described in this paper been in place. In addition to the significant benefits to the Shuttle indicated above, the development and implementation of this type of testing will be the genesis for potential Quality improvements across many areas of WSTF test data analysis and will be shared with other NASA centers. Future tests can be designed to incorporate engineering experience via Artificial Neural Nets (ANN) into depot level acceptance of hardware. Additionally, results were shared with a NASA Engineering and Safety Center (NESC) Super Problem Response Team (SPRT). There was extensive interest voiced among many different personnel from several centers. There are potential spin-offs of this effort that can be directly applied to other data acquisition systems as well as vehicle health management for current and future flight vehicles.

  13. On the relevance of using artificial neural networks for estimating soil moisture content

    NASA Astrophysics Data System (ADS)

    Elshorbagy, Amin; Parasuraman, K.

    2008-11-01

    SummarySoil moisture is a key variable that defines the land surface-atmosphere (boundary layer) interactions, by contributing directly to the surface energy balance and water balance. This paper investigates the utility of the widely adopted data-driven model, namely artificial neural networks (ANNs), for modeling the complex soil moisture dynamics. Datasets from three experimental soil covers (D1, D2, and D3), with thickness of 0.50 m, 0.35 m, and 1.0 m, comprising a thin layer of peat mineral mix over varying thickness of till, are considered in this study. Volumetric soil moisture contents at both the peat and the till layers were modeled as a function of precipitation, air temperature, net radiation, and ground temperature at different layers. Initial simulations illustrated that, in the absence of time-lagged meteorological variables, the ground temperature is the most influential state variable for characterizing the soil moisture, highlighting the strong link between the soil thermal properties and the corresponding moisture status. With the objective of extracting the maximum information from the most influential state variables (ground temperature), a higher-order neural networks (HONNs) model was developed to characterize the soil moisture dynamics. The HONNs resulted in relatively higher correlation coefficient, than traditional ANNs, for some of the soil moisture simulations. Time-lagged inputs were used to improve the model performance and obtain optimum results. The ANN models performed better than a previously developed conceptual model for estimating the depth-averaged soil moisture content. Results from the study indicate that modeling of soil moisture using ANNs is challenging but achievable, and its performance is largely influenced by the structure and formation of the soil covers, which in turn governs the dynamics of soil moisture variability.

  14. A Electric Load Forecasting Approach Using Expert Systems and Artificial Neural Networks.

    NASA Astrophysics Data System (ADS)

    Moharari, Nader Shariat

    The knowledge of accurate electric load demand is desirable for a variety of reasons. Smooth and economic operation of power systems is dependent upon reliable load forecasting. Large errors in load estimates could be costly. While accurate electric load forecasting will help in reducing operating costs by arranging to maintain and run the most economic generating plants to meet consumer demand at any time. In this dissertation a short-term load forecasting model is introduced (Rule-Based ANN model). The model makes use of Artificial Neural Networks (ANN) and Expert Systems (ES). In the proposed model an auxiliary network (sub net), driven by the ES has been utilized to adjust the biases for the main network. The Expert System is based on a set of rules which have been established according to an analysis of historical patterns. The role of ES is to tune the input components for the auxiliary net. The general forecasting process is as follows: the raw data files act as input for the Expert System. Then based on the rules and information available in the raw data files the ES goes through a reasoning process in order to prepare the processed data files for both auxiliary and main networks. These processed data files are then introduced to the ANN for training and prediction purposes. The model is capable of hourly load forecasting for the next 168 hours which is necessary for unit commitment. The model is also able to predict daily peak load for one week ahead. Evaluation tests have proven the viability of this approach. The results generated by this model have been compared with some other production grade packages in most cases the Rule-Based ANN model has performed superior.

  15. Optical implementation of the Hopfield neural network using multiple fiber nets.

    PubMed

    Ito, F; Kitayama, K

    1989-10-01

    Associative memory based on the Hopfield model utilizing interconnections by multiple optical fiber nets is proposed and demonstrated experimentally. The coupling ratio from fiber to fiber represents the synaptic weight of the connection between units. The remarkable feature of the optical fiber neural networks is that global connections between 2-D units can be easily achieved, resulting in simultaneous multiplication of the weight matrix and input vector. In the experiment, multiple fiber nets having 25(2) connections for 5 x 5 binary patterns are prepared, and three patterns are stored and successfully retrieved.

  16. Resource constrained design of artificial neural networks using comparator neural network

    NASA Technical Reports Server (NTRS)

    Wah, Benjamin W.; Karnik, Tanay S.

    1992-01-01

    We present a systematic design method executed under resource constraints for automating the design of artificial neural networks using the back error propagation algorithm. Our system aims at finding the best possible configuration for solving the given application with proper tradeoff between the training time and the network complexity. The design of such a system is hampered by three related problems. First, there are infinitely many possible network configurations, each may take an exceedingly long time to train; hence, it is impossible to enumerate and train all of them to completion within fixed time, space, and resource constraints. Second, expert knowledge on predicting good network configurations is heuristic in nature and is application dependent, rendering it difficult to characterize fully in the design process. A learning procedure that refines this knowledge based on examples on training neural networks for various applications is, therefore, essential. Third, the objective of the network to be designed is ill-defined, as it is based on a subjective tradeoff between the training time and the network cost. A design process that proposes alternate configurations under different cost-performance tradeoff is important. We have developed a Design System which schedules the available time, divided into quanta, for testing alternative network configurations. Its goal is to select/generate and test alternative network configurations in each quantum, and find the best network when time is expended. Since time is limited, a dynamic schedule that determines the network configuration to be tested in each quantum is developed. The schedule is based on relative comparison of predicted training times of alternative network configurations using comparator network paradigm. The comparator network has been trained to compare training times for a large variety of traces of TSSE-versus-time collected during back-propagation learning of various applications.

  17. Identification of power system load dynamics using artificial neural networks

    SciTech Connect

    Bostanci, M.; Koplowitz, J.; Taylor, C.W. |

    1997-11-01

    Power system loads are important for planning and operation of an electric power system. Load characteristics can significantly influence the results of synchronous stability and voltage stability studies. This paper presents a methodology for identification of power system load dynamics using neural networks. Input-output data of a power system dynamic load is used to design a neural network model which comprises delayed inputs and feedback connections. The developed neural network model can predict the future power system dynamic load behavior for arbitrary inputs. In particular, a third-order induction motor load neural network model is developed to verify the methodology. Neural network simulation results are illustrated and compared with the induction motor load response.

  18. Classification of land cover using optimized neural nets on SPOT data

    SciTech Connect

    Dreyer, P. )

    1993-05-01

    An optimized neural net was developed for land-cover classification in a multispectral SPOT satellite image covering 10 km x 10 km region which contains a mixture of densely built-up areas, suburbs, rural land, and waterbodies. In the technique, segments in the image are described by textural features calculated from gray-level difference statistics. The size of the input layer (i.e., the input variables to be used), as well as the size of the hidden layer in the neural net are determined using the optimization algorithm proposed by Mozer and Smolensky (1989). The textural features are calculated in segments generated by region growing in an image which has been processed iteratively with an edge enhancing adaptive filter. 15 refs.

  19. Power system distributed on-line fault section estimation using decision tree based neural nets approach

    SciTech Connect

    Yang, H.T.; Chang, W.Y.; Huang, C.L.

    1995-01-01

    This paper proposes a distributed neural nets decision approach to on-line estimation of the fault section of a transmission and distribution (T and D) system. The distributed processing alleviates the burden of communication between the control center and local substations, and increases the reliability and flexibility of the diagnosis system. Besides, by using the algorithms of data-driven decision tree induction and direct mapping from the decision tree into neural net, the proposed diagnosis system features parallel processing and easy implementation, overcoming the limitations of overly large and complex system. The approach has been practically tested on a typical Taiwan Power (Taipower) T and D system. The feasibility of such a diagnosis system is presented.

  20. Artificial neural networks in evaluation and optimization of modified release solid dosage forms.

    PubMed

    Ibrić, Svetlana; Djuriš, Jelena; Parojčić, Jelena; Djurić, Zorica

    2012-10-18

    Implementation of the Quality by Design (QbD) approach in pharmaceutical development has compelled researchers in the pharmaceutical industry to employ Design of Experiments (DoE) as a statistical tool, in product development. Among all DoE techniques, response surface methodology (RSM) is the one most frequently used. Progress of computer science has had an impact on pharmaceutical development as well. Simultaneous with the implementation of statistical methods, machine learning tools took an important place in drug formulation. Twenty years ago, the first papers describing application of artificial neural networks in optimization of modified release products appeared. Since then, a lot of work has been done towards implementation of new techniques, especially Artificial Neural Networks (ANN) in modeling of production, drug release and drug stability of modified release solid dosage forms. The aim of this paper is to review artificial neural networks in evaluation and optimization of modified release solid dosage forms.

  1. Application of artificial neural networks in nonlinear analysis of trusses

    NASA Technical Reports Server (NTRS)

    Alam, J.; Berke, L.

    1991-01-01

    A method is developed to incorporate neural network model based upon the Backpropagation algorithm for material response into nonlinear elastic truss analysis using the initial stiffness method. Different network configurations are developed to assess the accuracy of neural network modeling of nonlinear material response. In addition to this, a scheme based upon linear interpolation for material data, is also implemented for comparison purposes. It is found that neural network approach can yield very accurate results if used with care. For the type of problems under consideration, it offers a viable alternative to other material modeling methods.

  2. Beneficial role of noise in artificial neural networks

    SciTech Connect

    Monterola, Christopher; Saloma, Caesar; Zapotocky, Martin

    2008-06-18

    We demonstrate enhancement of neural networks efficacy to recognize frequency encoded signals and/or to categorize spatial patterns of neural activity as a result of noise addition. For temporal information recovery, noise directly added to the receiving neurons allow instantaneous improvement of signal-to-noise ratio [Monterola and Saloma, Phys. Rev. Lett. 2002]. For spatial patterns however, recurrence is necessary to extend and homogenize the operating range of a feed-forward neural network [Monterola and Zapotocky, Phys. Rev. E 2005]. Finally, using the size of the basin of attraction of the networks learned patterns (dynamical fixed points), a procedure for estimating the optimal noise is demonstrated.

  3. Can artificial neural networks provide an "expert's" view of medical students performances on computer based simulations?

    PubMed

    Stevens, R H; Najafi, K

    1992-01-01

    Artificial neural networks were trained to recognize the test selection patterns of students' successful solutions to seven immunology computer based simulations. When new student's test selections were presented to the trained neural network, their problem solutions were correctly classified as successful or non-successful > 90% of the time. Examination of the neural networks output weights after each test selection revealed a progressive increase for the relevant problem suggesting that a successful solution was represented by the neural network as the accumulation of relevant tests. Unsuccessful problem solutions revealed two patterns of students performances. The first pattern was characterized by low neural network output weights for all seven problems reflecting extensive searching and lack of recognition of relevant information. In the second pattern, the output weights from the neural network were biased towards one of the remaining six incorrect problems suggesting that the student mis-represented the current problem as an instance of a previous problem.

  4. The Application of Artificial Neural Networks to Astronomical Classification

    NASA Astrophysics Data System (ADS)

    Naim, A.

    1995-12-01

    Galaxies are fundamental to the understanding of the structure and evolution of the universe. They contain stars, gas and dust, and serve as an astrophysical laboratory in which physical processes can be examined. In the context of the large scale structure of the universe galaxies can be viewed as test particles. They are bright and therefore visible at very large distances, and also numerous and so can be used to provide reliable statistics. In previous decades the major obstacle to studying the large scale structure of the universe was the relatively sparse data samples, because obtaining large quantities of galaxian images and spectra requires a lot of observing time, and the accumulation of significant data bases was therefore a slow process. This obstacle is in the process of being removed today, with the advent of large-scale surveys (e.g., the APM galaxy survey, the Sloan Digital Sky Survey and the 2 degree Field survey). The new challenge with which the astronomical community is faced is the management and analysis of the forthcoming extragalactic data bases. On top of the obvious need for better hardware to give large storage volumes and quick access, one needs to devise automated tools for data analysis. The sheer volume of the data renders manual analysis impractical. It would be best if one could somehow transfer the knowledge and expertise accumulated over years of painstaking manual analysis to a machine. This thesis is part of an effort to achieve this goal. I borrowed techniques that have proved useful in other fields (e.g., engineering) and applied them to astronomical datasets. The major tool I used was Artificial Neural Networks (ANNs), which was originally conceived as a simplified computational model for the brain. The scope of methods and algorithms referred to as ANNs is quite wide. In particular, a distinction is made between Supervised Learning algorithms and Unsupervised methods. The former put the emphasis on ``teaching'' a machine to do

  5. Risk stratification in heart failure using artificial neural networks.

    PubMed Central

    Atienza, F.; Martinez-Alzamora, N.; De Velasco, J. A.; Dreiseitl, S.; Ohno-Machado, L.

    2000-01-01

    Accurate risk stratification of heart failure patients is critical to improve management and outcomes. Heart failure is a complex multisystem disease in which several predictors are categorical. Neural network models have successfully been applied to several medical classification problems. Using a simple neural network, we assessed one-year prognosis in 132 patients, consecutively admitted with heart failure, by classifying them in 3 groups: death, readmission and one-year event-free survival. Given the small number of cases, the neural network model was trained using a resampling method. We identified relevant predictors using the Automatic Relevance Determination (ARD) method, and estimated their mean effect on the 3 different outcomes. Only 9 individuals were misclassified. Neural networks have the potential to be a useful tool for making prognosis in the domain of heart failure. PMID:11079839

  6. Adaptive artificial neural network for autonomous robot control

    NASA Technical Reports Server (NTRS)

    Arras, Michael K.; Protzel, Peter W.; Palumbo, Daniel L.

    1992-01-01

    The topics are presented in viewgraph form and include: neural network controller for robot arm positioning with visual feedback; initial training of the arm; automatic recovery from cumulative fault scenarios; and error reduction by iterative fine movements.

  7. Deep neural nets as a method for quantitative structure-activity relationships.

    PubMed

    Ma, Junshui; Sheridan, Robert P; Liaw, Andy; Dahl, George E; Svetnik, Vladimir

    2015-02-23

    Neural networks were widely used for quantitative structure-activity relationships (QSAR) in the 1990s. Because of various practical issues (e.g., slow on large problems, difficult to train, prone to overfitting, etc.), they were superseded by more robust methods like support vector machine (SVM) and random forest (RF), which arose in the early 2000s. The last 10 years has witnessed a revival of neural networks in the machine learning community thanks to new methods for preventing overfitting, more efficient training algorithms, and advancements in computer hardware. In particular, deep neural nets (DNNs), i.e. neural nets with more than one hidden layer, have found great successes in many applications, such as computer vision and natural language processing. Here we show that DNNs can routinely make better prospective predictions than RF on a set of large diverse QSAR data sets that are taken from Merck's drug discovery effort. The number of adjustable parameters needed for DNNs is fairly large, but our results show that it is not necessary to optimize them for individual data sets, and a single set of recommended parameters can achieve better performance than RF for most of the data sets we studied. The usefulness of the parameters is demonstrated on additional data sets not used in the calibration. Although training DNNs is still computationally intensive, using graphical processing units (GPUs) can make this issue manageable.

  8. Deep neural nets as a method for quantitative structure-activity relationships.

    PubMed

    Ma, Junshui; Sheridan, Robert P; Liaw, Andy; Dahl, George E; Svetnik, Vladimir

    2015-02-23

    Neural networks were widely used for quantitative structure-activity relationships (QSAR) in the 1990s. Because of various practical issues (e.g., slow on large problems, difficult to train, prone to overfitting, etc.), they were superseded by more robust methods like support vector machine (SVM) and random forest (RF), which arose in the early 2000s. The last 10 years has witnessed a revival of neural networks in the machine learning community thanks to new methods for preventing overfitting, more efficient training algorithms, and advancements in computer hardware. In particular, deep neural nets (DNNs), i.e. neural nets with more than one hidden layer, have found great successes in many applications, such as computer vision and natural language processing. Here we show that DNNs can routinely make better prospective predictions than RF on a set of large diverse QSAR data sets that are taken from Merck's drug discovery effort. The number of adjustable parameters needed for DNNs is fairly large, but our results show that it is not necessary to optimize them for individual data sets, and a single set of recommended parameters can achieve better performance than RF for most of the data sets we studied. The usefulness of the parameters is demonstrated on additional data sets not used in the calibration. Although training DNNs is still computationally intensive, using graphical processing units (GPUs) can make this issue manageable. PMID:25635324

  9. Artificial neural network to predict degradation of non-metallic lining materials from laboratory tests

    SciTech Connect

    Silverman, D.C.

    1994-12-31

    Artificial neural networks are computer simulations that have the potential of ``finding`` the same patterns that corrosion practitioners recognize to relate experimental test results to lifetime predictions. This potential ability was utilized to construct an artificial neural network to recognize the pattern between results from a sequential immersion test for organic non-metallic lining materials and their ability to function as linings in actual applications. The network so constructed has been shown to predict field performance from this test. The network was incorporated within an Expert System to simplify data input and output, allow for simple consistency checks, and to make the final prediction.

  10. The Classification of a Simulation Data of a Servo System via Evolutionary Artificial Neural Networks

    NASA Astrophysics Data System (ADS)

    Alkaya, Asil; Bayhan, G. Miraç

    Evolutionary neural networks (EANNs) are the combination of artificial neural networks and evolutionary algorithms. This merge enabled these two methods to complement the disadvantages of the other methods. Traditional artificial neural networks based on backpropagation algorithms have some limitations. Contribution by artificial neural networks was the flexibility of nonlinear function approximation, which cannot be easily implemented with prototype evolutionary algorithm. On the other hand, evolutionary algorithm has freed artificial neural networks from simple gradient descent approaches of optimization. Classification is an important task in many domains and though there are several methods that can be used to find the relationship between the input and output space , among the different works, EAs and NNs stands out as one of the most promising methods. In this study, the data gathered from a simulation of a servo system involving a servo amplifier, a motor, a lead screw/nut, and a sliding carriage of some sort is classified by the application coded in Qt programming environment to predict the rise time of a servomechanism in terms of two (continuous) gain settings and two (discrete) choices of mechanical linkages.

  11. Artificial neural network for identification of a substance from a Mössbauer data bank

    NASA Astrophysics Data System (ADS)

    Salles, Evandro O. T.; de Souza, P. A.; Garg, V. K.

    1994-12-01

    Mössbauer data and references of the minerals reported in the literature have been stored in a computer. Artificial neutral networks (ANN) were taught with the average values of experimental data of isomer shift quadrupole splitting of known mineral systems (sulphate, sulphide and sulphites, and silicates). Artificial neural networks successfully identified the unknown substance when fed with the new values of isomer shift and quadrupole splitting.

  12. On Design and Implementation of Neural-Machine Interface for Artificial Legs

    PubMed Central

    Zhang, Xiaorong; Liu, Yuhong; Zhang, Fan; Ren, Jin; Sun, Yan (Lindsay); Yang, Qing

    2011-01-01

    The quality of life of leg amputees can be improved dramatically by using a cyber physical system (CPS) that controls artificial legs based on neural signals representing amputees’ intended movements. The key to the CPS is the neural-machine interface (NMI) that senses electromyographic (EMG) signals to make control decisions. This paper presents a design and implementation of a novel NMI using an embedded computer system to collect neural signals from a physical system - a leg amputee, provide adequate computational capability to interpret such signals, and make decisions to identify user’s intent for prostheses control in real time. A new deciphering algorithm, composed of an EMG pattern classifier and a post-processing scheme, was developed to identify the user’s intended lower limb movements. To deal with environmental uncertainty, a trust management mechanism was designed to handle unexpected sensor failures and signal disturbances. Integrating the neural deciphering algorithm with the trust management mechanism resulted in a highly accurate and reliable software system for neural control of artificial legs. The software was then embedded in a newly designed hardware platform based on an embedded microcontroller and a graphic processing unit (GPU) to form a complete NMI for real time testing. Real time experiments on a leg amputee subject and an able-bodied subject have been carried out to test the control accuracy of the new NMI. Our extensive experiments have shown promising results on both subjects, paving the way for clinical feasibility of neural controlled artificial legs. PMID:22389637

  13. On the Relationships between Generative Encodings, Regularity, and Learning Abilities when Evolving Plastic Artificial Neural Networks

    PubMed Central

    Tonelli, Paul; Mouret, Jean-Baptiste

    2013-01-01

    A major goal of bio-inspired artificial intelligence is to design artificial neural networks with abilities that resemble those of animal nervous systems. It is commonly believed that two keys for evolving nature-like artificial neural networks are (1) the developmental process that links genes to nervous systems, which enables the evolution of large, regular neural networks, and (2) synaptic plasticity, which allows neural networks to change during their lifetime. So far, these two topics have been mainly studied separately. The present paper shows that they are actually deeply connected. Using a simple operant conditioning task and a classic evolutionary algorithm, we compare three ways to encode plastic neural networks: a direct encoding, a developmental encoding inspired by computational neuroscience models, and a developmental encoding inspired by morphogen gradients (similar to HyperNEAT). Our results suggest that using a developmental encoding could improve the learning abilities of evolved, plastic neural networks. Complementary experiments reveal that this result is likely the consequence of the bias of developmental encodings towards regular structures: (1) in our experimental setup, encodings that tend to produce more regular networks yield networks with better general learning abilities; (2) whatever the encoding is, networks that are the more regular are statistically those that have the best learning abilities. PMID:24236099

  14. On the relationships between generative encodings, regularity, and learning abilities when evolving plastic artificial neural networks.

    PubMed

    Tonelli, Paul; Mouret, Jean-Baptiste

    2013-01-01

    A major goal of bio-inspired artificial intelligence is to design artificial neural networks with abilities that resemble those of animal nervous systems. It is commonly believed that two keys for evolving nature-like artificial neural networks are (1) the developmental process that links genes to nervous systems, which enables the evolution of large, regular neural networks, and (2) synaptic plasticity, which allows neural networks to change during their lifetime. So far, these two topics have been mainly studied separately. The present paper shows that they are actually deeply connected. Using a simple operant conditioning task and a classic evolutionary algorithm, we compare three ways to encode plastic neural networks: a direct encoding, a developmental encoding inspired by computational neuroscience models, and a developmental encoding inspired by morphogen gradients (similar to HyperNEAT). Our results suggest that using a developmental encoding could improve the learning abilities of evolved, plastic neural networks. Complementary experiments reveal that this result is likely the consequence of the bias of developmental encodings towards regular structures: (1) in our experimental setup, encodings that tend to produce more regular networks yield networks with better general learning abilities; (2) whatever the encoding is, networks that are the more regular are statistically those that have the best learning abilities.

  15. Quantitative analysis of cefalexin based on artificial neural networks combined with modified genetic algorithm using short near-infrared spectroscopy

    NASA Astrophysics Data System (ADS)

    Huan, Yanfu; Feng, Guodong; Wang, Bin; Ren, Yulin; Fei, Qiang

    2013-05-01

    In this paper, a novel chemometric method was developed for rapid, accurate, and quantitative analysis of cefalexin in samples. The experiments were carried out by using the short near-infrared spectroscopy coupled with artificial neural networks. In order to enhancing the predictive ability of artificial neural networks model, a modified genetic algorithm was used to select fixed number of wavelength.

  16. Modeling of mass transfer of Phospholipids in separation process with supercritical CO2 fluid by RBF artificial neural networks

    Technology Transfer Automated Retrieval System (TEKTRAN)

    An artificial Radial Basis Function (RBF) neural network model was developed for the prediction of mass transfer of the phospholipids from canola meal in supercritical CO2 fluid. The RBF kind of artificial neural networks (ANN) with orthogonal least squares (OLS) learning algorithm were used for mod...

  17. Total solar irradiance reconstruction using artificial neural networks

    NASA Astrophysics Data System (ADS)

    Tebabal Yirdaw, Ambelu; Damtie, Baylie; Nigussie, Melessew; Bires, Abiyot; Yizengaw, Endawoke

    2015-08-01

    A feed-forward neural network which can account for nonlinear relationships was used to reconstruct total solar irradiance (TSI). A single layer feed forward neural network with back-propagation algorithm have been implemented for reconstructing daily total solar irradiance from daily photometric sunspot index, and core to wing ratio of Mg II index data. The data year from 1978 to 2013 was used for the training, validation and testing purpose. In order to obtain the optimum neural network for TSI reconstruction, the root mean square error (RMSE), mean absolute error (MAE) and regression coefficient have been taken into account. We have carried out the analysis is made by comparing the reconstructed TSI from neural networks (NNs ) and TSI measurement from satellite. We have found out that the reconstructed TSI and the PMOD composite have the correlation coefficient of about R=0.9307 over the span of the recorded, 1978 to 2013. The NNs model output indicates that reconstructed TSI from solar proxies (photometric index and MgII ) can explain 86.6% of the variance of TSI. Neural network is able to recreate TSI observations on a time scale of a day. This reconstructed TSI using NNs further strengthens the view that surface magnetism indeed plays a dominant role in modulating solar irradiance.

  18. An Examination of Application of Artificial Neural Network in Cognitive Radios

    NASA Astrophysics Data System (ADS)

    Bello Salau, H.; Onwuka, E. N.; Aibinu, A. M.

    2013-12-01

    Recent advancement in software radio technology has led to the development of smart device known as cognitive radio. This type of radio fuses powerful techniques taken from artificial intelligence, game theory, wideband/multiple antenna techniques, information theory and statistical signal processing to create an outstanding dynamic behavior. This cognitive radio is utilized in achieving diverse set of applications such as spectrum sensing, radio parameter adaptation and signal classification. This paper contributes by reviewing different cognitive radio implementation that uses artificial intelligence such as the hidden markov models, metaheuristic algorithm and artificial neural networks (ANNs). Furthermore, different areas of application of ANNs and their performance metrics based approach are also examined.

  19. A new approach to training back-propagation artificial neural networks: empirical evaluation on ten data sets from clinical studies.

    PubMed

    Ciampi, Antonio; Zhang, Fulin

    2002-05-15

    We present a new approach to training back-propagation artificial neural nets (BP-ANN) based on regularization and cross-validation and on initialization by a logistic regression (LR) model. The new approach is expected to produce a BP-ANN predictor at least as good as the LR-based one. We have applied the approach to ten data sets of biomedical interest and systematically compared BP-ANN and LR. In all data sets, taking deviance as criterion, the BP-ANN predictor outperforms the LR predictor used in the initialization, and in six cases the improvement is statistically significant. The other evaluation criteria used (C-index, MSE and error rate) yield variable results, but, on the whole, confirm that, in practical situations of clinical interest, proper training may significantly improve the predictive performance of a BP-ANN.

  20. A Model for Improving the Learning Curves of Artificial Neural Networks

    PubMed Central

    2016-01-01

    In this article, the performance of a hybrid artificial neural network (i.e. scale-free and small-world) was analyzed and its learning curve compared to three other topologies: random, scale-free and small-world, as well as to the chemotaxis neural network of the nematode Caenorhabditis Elegans. One hundred equivalent networks (same number of vertices and average degree) for each topology were generated and each was trained for one thousand epochs. After comparing the mean learning curves of each network topology with the C. elegans neural network, we found that the networks that exhibited preferential attachment exhibited the best learning curves. PMID:26901646

  1. Application of Artificial Neural Networks to the Design of Turbomachinery Airfoils

    NASA Technical Reports Server (NTRS)

    Rai, Man Mohan; Madavan, Nateri

    1997-01-01

    Artificial neural networks are widely used in engineering applications, such as control, pattern recognition, plant modeling and condition monitoring to name just a few. In this seminar we will explore the possibility of applying neural networks to aerodynamic design, in particular, the design of turbomachinery airfoils. The principle idea behind this effort is to represent the design space using a neural network (within some parameter limits), and then to employ an optimization procedure to search this space for a solution that exhibits optimal performance characteristics. Results obtained for design problems in two spatial dimensions will be presented.

  2. Optically Connected Multiprocessors For Simulating Artificial Neural Networks

    NASA Astrophysics Data System (ADS)

    Ghosh, Joydeep; Hwang, Kai

    1988-05-01

    This paper investigates the architectural requirements in simulating large neural networks using a highly parallel multiprocessor with distributed memory and optical interconnects. First, we model the structure of a neural network and the functional behavior of individual cells. These models are used to estimate the volume of messages that need to be exchanged among physical processors to simulate the weighted connections of the neural network. The distributed processor/memory organization is tailored to an electronic implementation for greater versatility and flexibility. Optical interconnects are used to satisfy the interprocessor communication bandwidth demands. The hybrid implementation attempts to balance the processing, memory and bandwidth demands in simulating asynchronous, value-passing models for cooperative parallel computation with self-learning capabilities.

  3. Vein matching using artificial neural network in vein authentication systems

    NASA Astrophysics Data System (ADS)

    Noori Hoshyar, Azadeh; Sulaiman, Riza

    2011-10-01

    Personal identification technology as security systems is developing rapidly. Traditional authentication modes like key; password; card are not safe enough because they could be stolen or easily forgotten. Biometric as developed technology has been applied to a wide range of systems. According to different researchers, vein biometric is a good candidate among other biometric traits such as fingerprint, hand geometry, voice, DNA and etc for authentication systems. Vein authentication systems can be designed by different methodologies. All the methodologies consist of matching stage which is too important for final verification of the system. Neural Network is an effective methodology for matching and recognizing individuals in authentication systems. Therefore, this paper explains and implements the Neural Network methodology for finger vein authentication system. Neural Network is trained in Matlab to match the vein features of authentication system. The Network simulation shows the quality of matching as 95% which is a good performance for authentication system matching.

  4. Evolving artificial neural networks to control chaotic systems

    NASA Astrophysics Data System (ADS)

    Weeks, Eric R.; Burgess, John M.

    1997-08-01

    We develop a genetic algorithm that produces neural network feedback controllers for chaotic systems. The algorithm was tested on the logistic and Hénon maps, for which it stabilizes an unstable fixed point using small perturbations, even in the presence of significant noise. The network training method [D. E. Moriarty and R. Miikkulainen, Mach. Learn. 22, 11 (1996)] requires no previous knowledge about the system to be controlled, including the dimensionality of the system and the location of unstable fixed points. This is the first dimension-independent algorithm that produces neural network controllers using time-series data. A software implementation of this algorithm is available via the World Wide Web.

  5. Artificial Neural Networks: A New Approach for Predicting Application Behavior. AIR 2001 Annual Forum Paper.

    ERIC Educational Resources Information Center

    Gonzalez, Julie M. Byers; DesJardins, Stephen L.

    This paper examines how predictive modeling can be used to study application behavior. A relatively new technique, artificial neural networks (ANNs), was applied to help predict which students were likely to get into a large Research I university. Data were obtained from a university in Iowa. Two cohorts were used, each containing approximately…

  6. The Use of Artificial Neural Networks to Estimate Speech Intelligibility from Acoustic Variables: A Preliminary Analysis.

    ERIC Educational Resources Information Center

    Metz, Dale Evan; And Others

    1992-01-01

    A preliminary scheme for estimating the speech intelligibility of hearing-impaired speakers from acoustic parameters, using a computerized artificial neural network to process mathematically the acoustic input variables, is outlined. Tests with 60 hearing-impaired speakers found the scheme to be highly accurate in identifying speakers separated by…

  7. Automatic Keyword Identification by Artificial Neural Networks Compared to Manual Identification by Users of Filtering Systems.

    ERIC Educational Resources Information Center

    Boger, Zvi; Kuflik, Tsvi; Shoval, Peretz; Shapira, Bracha

    2001-01-01

    Discussion of information filtering (IF) and information retrieval focuses on the use of an artificial neural network (ANN) as an alternative method for both IF and term selection and compares its effectiveness to that of traditional methods. Results show that the ANN relevance prediction out-performs the prediction of an IF system. (Author/LRW)

  8. Statistical Classification for Cognitive Diagnostic Assessment: An Artificial Neural Network Approach

    ERIC Educational Resources Information Center

    Cui, Ying; Gierl, Mark; Guo, Qi

    2016-01-01

    The purpose of the current investigation was to describe how the artificial neural networks (ANNs) can be used to interpret student performance on cognitive diagnostic assessments (CDAs) and evaluate the performances of ANNs using simulation results. CDAs are designed to measure student performance on problem-solving tasks and provide useful…

  9. Predicting Item Difficulty in a Reading Comprehension Test with an Artificial Neural Network.

    ERIC Educational Resources Information Center

    Perkins, Kyle; And Others

    1995-01-01

    This article reports the results of using a three-layer back propagation artificial neural network to predict item difficulty in a reading comprehension test. Three classes of variables were examined: text structure, propositional analysis, and cognitive demand. Results demonstrate that the networks can consistently predict item difficulty. (JL)

  10. Predicting Item Difficulty in a Reading Comprehension Test with an Artificial Neural Network.

    ERIC Educational Resources Information Center

    Perkins, Kyle; And Others

    This paper reports the results of using a three-layer backpropagation artificial neural network to predict item difficulty in a reading comprehension test. Two network structures were developed, one with and one without a sigmoid function in the output processing unit. The data set, which consisted of a table of coded test items and corresponding…

  11. Artificial neural network estimation of soil erosion and nutrient concentrations in runoff from land application areas

    Technology Transfer Automated Retrieval System (TEKTRAN)

    The transport of sediment and nutrients from land application areas is an environmental concern. New methods are needed for estimating soil and nutrient concentrations of runoff from cropland areas on which manure is applied. Artificial Neural Networks (ANN) trained with a Backpropagation (BP) algor...

  12. Reconstructing missing daily precipitation data using regression trees and artificial neural networks

    Technology Transfer Automated Retrieval System (TEKTRAN)

    Incomplete meteorological data has been a problem in environmental modeling studies. The objective of this work was to develop a technique to reconstruct missing daily precipitation data in the central part of Chesapeake Bay Watershed using regression trees (RT) and artificial neural networks (ANN)....

  13. [Optimization of pellet formulation with the help of artificial neural networks].

    PubMed

    Kása, Péter; Sovány, Tamás; Hódi, Klára

    2007-01-01

    The authors demonstrate the essence and the application possibility of artificial neural networks in the formulation of pharmaceutical preparations. They draw attention to that the use of ANN the data processing will speed up and more accurate which will cause the decrease of the preliminary investigations and the amounts of the materials. PMID:17933271

  14. Predicting Final GPA of Graduate School Students: Comparing Artificial Neural Networking and Simultaneous Multiple Regression

    ERIC Educational Resources Information Center

    Anderson, Joan L.

    2006-01-01

    Data from graduate student applications at a large Western university were used to determine which factors were the best predictors of success in graduate school, as defined by cumulative graduate grade point average. Two statistical models were employed and compared: artificial neural networking and simultaneous multiple regression. Both models…

  15. Discovering Neural Nets with Low Kolmogorov Complexity and High Generalization Capability.

    PubMed

    Schmidhuber, Jurgen

    1997-07-01

    Many neural net learning algorithms aim at finding "simple" nets to explain training data. The expectation is that the "simpler" the networks, the better the generalization on test data (--> Occam's razor). Previous implementations, however, use measures for "simplicity" that lack the power, universality and elegance of those based on Kolmogorov complexity and Solomonoff's algorithmic probability. Likewise, most previous approaches (especially those of the "Bayesian" kind) suffer from the problem of choosing appropriate priors. This paper addresses both issues. It first reviews some basic concepts of algorithmic complexity theory relevant to machine learing, and how the Solomonoff-Levin distribution (or universal prior) deals with the prior problem. The universal prior leads to a probabilistic method for finding "algorithmically simple" problem solutions with high generalization capability. The method is based on Levin complexity (a time-bounded generalization of Kolmogorov complexity) and inspired by Levin's optimal universal search algorithm. For a given problem, solution candidates are computed by efficient "self-sizing" programs that influence their own runtime and storage size. The probabilistic search algorithm finds the "good" programs (the ones quickly computing algorithmically probable solutions fitting the training data). Simulations focus on the task of discovering "algorithmically simple" neural networks with low Kolmogorov complexity and high generalization capability. It is demonstrated that the method, at least with certain toy problems where it is computationally feasible, can lead to generalization results unmatchable by previous neural network algorithms. Much remains to be done, however, to make large scale applications and "incremental learning" feasible. Copyright 1997 Elsevier Science Ltd.

  16. A genetic system based on simulated crossover: stability analysis and relationships with neural nets.

    PubMed

    Carpentieri, Marco

    2009-01-01

    We consider a marginal distribution genetic model based on crossover of sequences of genes and provide relations between the associated infinite population genetic system and the neural networks. A lower bound on population size is exhibited stating that the behavior of the finite population system, in the case of sufficiently large sizes, can be approximated by the behavior of the corresponding infinite population system. Assumptions on fitness and individual chromosomes are provided implying that the behavior of the finite population genetic system remains consistent with the behavior of the associated infinite population genetic system for suitably long trajectories. The attractors (with binary components) of the infinite population genetic system are characterized as equilibrium points of a discrete (neural network) system that can be considered as a variant of a Hopfield's network; it is shown that the fitness is a Lyapunov function for the variant of the discrete Hopfield's net. Our main result can be summarized by stating that the relation between marginal distribution genetic systems and neural nets is much more general than that already shown elsewhere for other simpler models.

  17. Applications of Artificial Neural Networks in integrated water management: fiction or future?

    PubMed

    Schulze, F H; Wolf, H; Jansen, H W; van der Veer, P

    2005-01-01

    An Artificial Neural Network (ANN) is nowadays recognized as a very promising tool for relating input data to output data. It is said that the possibilities of artificial neural networks are unlimited. Here we focus on the potential role of neural networks in integrated water management. An Artificial Neural Network (ANN) is a mathematical methodology which describes relations between cause (input data) and effects (output data) irrespective of the process laying behind and without the need for making assumptions considering the nature of the relations. The applications are widespread and vary from optimization of measuring networks, operational water management, prediction of drinking water consumption, on-line steering of wastewater treatment plants and sewage systems, up to more specific applications such as establishing a relationship between the observed erosion of groyne field sediments and the characteristics of passing vessels on the river Rhine. Especially where processes are complex, neural networks can open new possibilities for understanding and modelling these kinds of complex processes. Besides explaining the method of ANN this paper shows different applications. Three examples have been worked out in more detail. An intelligent monitoring system is shown for the on-line prediction of water consumption, ANN are successfully used for sludge cost monitoring and optimizing wastewater treatment and the usage of ANN is shown in optimizing and monitoring water quality measuring networks. An ANN appears to be a multiuse and powerful tool for modelling complex processes.

  18. Artificial neural network classification using a minimal training set - Comparison to conventional supervised classification

    NASA Technical Reports Server (NTRS)

    Hepner, George F.; Logan, Thomas; Ritter, Niles; Bryant, Nevin

    1990-01-01

    Recent research has shown an artificial neural network (ANN) to be capable of pattern recognition and the classification of image data. This paper examines the potential for the application of neural network computing to satellite image processing. A second objective is to provide a preliminary comparison and ANN classification. An artificial neural network can be trained to do land-cover classification of satellite imagery using selected sites representative of each class in a manner similar to conventional supervised classification. One of the major problems associated with recognition and classifications of pattern from remotely sensed data is the time and cost of developing a set of training sites. This reseach compares the use of an ANN back propagation classification procedure with a conventional supervised maximum likelihood classification procedure using a minimal training set. When using a minimal training set, the neural network is able to provide a land-cover classification superior to the classification derived from the conventional classification procedure. This research is the foundation for developing application parameters for further prototyping of software and hardware implementations for artificial neural networks in satellite image and geographic information processing.

  19. Artificial frame filling using adaptive neural fuzzy inference system for particle image velocimetry dataset

    NASA Astrophysics Data System (ADS)

    Akdemir, Bayram; Doǧan, Sercan; Aksoy, Muharrem H.; Canli, Eyüp; Özgören, Muammer

    2015-03-01

    Liquid behaviors are very important for many areas especially for Mechanical Engineering. Fast camera is a way to observe and search the liquid behaviors. Camera traces the dust or colored markers travelling in the liquid and takes many pictures in a second as possible as. Every image has large data structure due to resolution. For fast liquid velocity, there is not easy to evaluate or make a fluent frame after the taken images. Artificial intelligence has much popularity in science to solve the nonlinear problems. Adaptive neural fuzzy inference system is a common artificial intelligence in literature. Any particle velocity in a liquid has two dimension speed and its derivatives. Adaptive Neural Fuzzy Inference System has been used to create an artificial frame between previous and post frames as offline. Adaptive neural fuzzy inference system uses velocities and vorticities to create a crossing point vector between previous and post points. In this study, Adaptive Neural Fuzzy Inference System has been used to fill virtual frames among the real frames in order to improve image continuity. So this evaluation makes the images much understandable at chaotic or vorticity points. After executed adaptive neural fuzzy inference system, the image dataset increase two times and has a sequence as virtual and real, respectively. The obtained success is evaluated using R2 testing and mean squared error. R2 testing has a statistical importance about similarity and 0.82, 0.81, 0.85 and 0.8 were obtained for velocities and derivatives, respectively.

  20. RRAM-based hardware implementations of artificial neural networks: progress update and challenges ahead

    NASA Astrophysics Data System (ADS)

    Prezioso, M.; Merrikh-Bayat, F.; Chakrabarti, B.; Strukov, D.

    2016-02-01

    Artificial neural networks have been receiving increasing attention due to their superior performance in many information processing tasks. Typically, scaling up the size of the network results in better performance and richer functionality. However, large neural networks are challenging to implement in software and customized hardware are generally required for their practical implementations. In this work, we will discuss our group's recent efforts on the development of such custom hardware circuits, based on hybrid CMOS/memristor circuits, in particular of CMOL variety. We will start by reviewing the basics of memristive devices and of CMOL circuits. We will then discuss our recent progress towards demonstration of hybrid circuits, focusing on the experimental and theoretical results for artificial neural networks based on crossbarintegrated metal oxide memristors. We will conclude presentation with the discussion of the remaining challenges and the most pressing research needs.

  1. Evaluation of the efficiency of artificial neural networks for genetic value prediction.

    PubMed

    Silva, G N; Tomaz, R S; Sant'Anna, I C; Carneiro, V Q; Cruz, C D; Nascimento, M

    2016-01-01

    Artificial neural networks have shown great potential when applied to breeding programs. In this study, we propose the use of artificial neural networks as a viable alternative to conventional prediction methods. We conduct a thorough evaluation of the efficiency of these networks with respect to the prediction of breeding values. Therefore, we considered eight simulated scenarios, and for the purpose of genetic value prediction, seven statistical parameters in addition to the phenotypic mean in a network designed as a multilayer perceptron. After an evaluation of different network configurations, the results demonstrated the superiority of neural networks compared to estimation procedures based on linear models, and indicated high predictive accuracy and network efficiency. PMID:27051007

  2. Optimization with artificial neural network systems - A mapping principle and a comparison to gradient based methods

    NASA Technical Reports Server (NTRS)

    Leong, Harrison Monfook

    1988-01-01

    General formulae for mapping optimization problems into systems of ordinary differential equations associated with artificial neural networks are presented. A comparison is made to optimization using gradient-search methods. The performance measure is the settling time from an initial state to a target state. A simple analytical example illustrates a situation where dynamical systems representing artificial neural network methods would settle faster than those representing gradient-search. Settling time was investigated for a more complicated optimization problem using computer simulations. The problem was a simplified version of a problem in medical imaging: determining loci of cerebral activity from electromagnetic measurements at the scalp. The simulations showed that gradient based systems typically settled 50 to 100 times faster than systems based on current neural network optimization methods.

  3. Fault diagnosis in nuclear power plants using an artificial neural network technique

    SciTech Connect

    Chou, H.P. ); Prock, J.; Bonfert, J.P. )

    1993-01-01

    Application of artificial intelligence (AI) computational techniques, such as expert systems, fuzzy logic, and neural networks in diverse areas has taken place extensively. In the nuclear industry, the intended goal for these AI techniques is to improve power plant operational safety and reliability. As a computerized operator support tool, the artificial neural network (ANN) approach is an emerging technology that currently attracts a large amount of interest. The ability of ANNs to extract the input/output relation of a complicated process and the superior execution speed of a trained ANN motivated this study. The goal was to develop neural networks for sensor and process faults diagnosis with the potential of implementing as a component of a real-time operator support system LYDIA, early sensor and process fault detection and diagnosis.

  4. Neural networks for aircraft control

    NASA Technical Reports Server (NTRS)

    Linse, Dennis

    1990-01-01

    Current research in Artificial Neural Networks indicates that networks offer some potential advantages in adaptation and fault tolerance. This research is directed at determining the possible applicability of neural networks to aircraft control. The first application will be to aircraft trim. Neural network node characteristics, network topology and operation, neural network learning and example histories using neighboring optimal control with a neural net are discussed.

  5. Recurrent neural nets as dynamical Boolean systems with application to associative memory.

    PubMed

    Watta, P B; Wang, K; Hassoun, M H

    1997-01-01

    Discrete-time/discrete-state recurrent neural networks are analyzed from a dynamical Boolean systems point of view in order to devise new analytic and design methods for the class of both single and multilayer recurrent artificial neural networks. With the proposed dynamical Boolean systems analysis, we are able to formulate necessary and sufficient conditions for network stability which are more general than the well-known but restrictive conditions for the class of single layer networks: (1) symmetric weight matrix with (2) positive diagonal and (3) asynchronous update. In terms of design, we use a dynamical Boolean systems analysis to construct a high performance associative memory. With this Boolean memory, we can guarantee that all fundamental memories are stored, and also guarantee the size of the basin of attraction for each fundamental memory.

  6. Eye tracking using artificial neural networks for human computer interaction.

    PubMed

    Demjén, E; Aboši, V; Tomori, Z

    2011-01-01

    This paper describes an ongoing project that has the aim to develop a low cost application to replace a computer mouse for people with physical impairment. The application is based on an eye tracking algorithm and assumes that the camera and the head position are fixed. Color tracking and template matching methods are used for pupil detection. Calibration is provided by neural networks as well as by parametric interpolation methods. Neural networks use back-propagation for learning and bipolar sigmoid function is chosen as the activation function. The user's eye is scanned with a simple web camera with backlight compensation which is attached to a head fixation device. Neural networks significantly outperform parametric interpolation techniques: 1) the calibration procedure is faster as they require less calibration marks and 2) cursor control is more precise. The system in its current stage of development is able to distinguish regions at least on the level of desktop icons. The main limitation of the proposed method is the lack of head-pose invariance and its relative sensitivity to illumination (especially to incidental pupil reflections).

  7. Opening up the black box of artificial neural networks

    SciTech Connect

    Spining, M.T.; Darsey, J.A.; Sumpter, B.G.; Noid, D.W.

    1994-05-01

    In this paper, neural networks are divided according to training methods--supervised and unsupervised. Supervised training is used when a training set consisting of inputs and outputs is available. The network uses the training set to determine an error and then adjusts itself with respect to that error. Unsupervised networks are used when training sets with known outputs are not available, for example, for realtime learning. These networks use the inputs to adjust themselves so that similar input gives similar output. Another classification that will be used is feedforward and feedback networks. In a feedforward network, information is propagated through the network in one direction until it emerges as the network`s output. However, in a feedback (recurrent) network, the input information is propagated through the network but can also cycle back into the network (the signal is recurrent). In the present paper, the authors give a fundamental overview of feedforward neural networks, present some applications using them in chemical physics, and comment on the potential for future uses in chemistry. They begin by discussing some specific types of neural networks that provide the generality needed to pursue applications in the chemical sciences.

  8. Search for predictive generic model of aqueous solubility using Bayesian neural nets.

    PubMed

    Bruneau, P

    2001-01-01

    Several predictive models of aqueous solubility have been published. They have good performances on the data sets which have been used for training the models, but usually these data sets do not contain many structures similar to the structures of interest to the drug research and their applicability in drug hunting is questionable. A very diverse data set has been gathered with compounds issued from literature reports and proprietary compounds. These compounds have been grouped in a so-called literature data set I, a proprietary data set II, and a mixed data set III formed by I and II. About 100 descriptors emphasizing surface properties were calculated for every compound. Bayesian learning of neural nets which cumulates the advantages of neural nets without having their weaknesses was used to select the most parsimonious models and train them, from I, II, and III. The models were established by either selecting the most efficient descriptors one by one using a modified Gram-Schmidt procedure (GS) or by simplifying a most complete model using automatic relevance procedure (ARD). The predictive ability of the models was accessed using validation data sets as much unrelated to the training sets as possible, using two new parameters: NDD(x,ref) the normalized smallest descriptor distance of a compound x to a reference data set and CD(x,mod) the combination of NDD(x,ref) with the dispersion of the Bayesian neural nets calculations. The results show that it is possible to obtain a generic predictive model from database I but that the diversity of database II is too restricted to give a model with good generalization ability and that the ARD method applied to the mixed database III gives the best predictive model.

  9. Reduced-Order Modeling for Flutter/LCO Using Recurrent Artificial Neural Network

    NASA Technical Reports Server (NTRS)

    Yao, Weigang; Liou, Meng-Sing

    2012-01-01

    The present study demonstrates the efficacy of a recurrent artificial neural network to provide a high fidelity time-dependent nonlinear reduced-order model (ROM) for flutter/limit-cycle oscillation (LCO) modeling. An artificial neural network is a relatively straightforward nonlinear method for modeling an input-output relationship from a set of known data, for which we use the radial basis function (RBF) with its parameters determined through a training process. The resulting RBF neural network, however, is only static and is not yet adequate for an application to problems of dynamic nature. The recurrent neural network method [1] is applied to construct a reduced order model resulting from a series of high-fidelity time-dependent data of aero-elastic simulations. Once the RBF neural network ROM is constructed properly, an accurate approximate solution can be obtained at a fraction of the cost of a full-order computation. The method derived during the study has been validated for predicting nonlinear aerodynamic forces in transonic flow and is capable of accurate flutter/LCO simulations. The obtained results indicate that the present recurrent RBF neural network is accurate and efficient for nonlinear aero-elastic system analysis

  10. Architecture and biological applications of artificial neural networks: a tuberculosis perspective.

    PubMed

    Darsey, Jerry A; Griffin, William O; Joginipelli, Sravanthi; Melapu, Venkata Kiran

    2015-01-01

    Advancement of science and technology has prompted researchers to develop new intelligent systems that can solve a variety of problems such as pattern recognition, prediction, and optimization. The ability of the human brain to learn in a fashion that tolerates noise and error has attracted many researchers and provided the starting point for the development of artificial neural networks: the intelligent systems. Intelligent systems can acclimatize to the environment or data and can maximize the chances of success or improve the efficiency of a search. Due to massive parallelism with large numbers of interconnected processers and their ability to learn from the data, neural networks can solve a variety of challenging computational problems. Neural networks have the ability to derive meaning from complicated and imprecise data; they are used in detecting patterns, and trends that are too complex for humans, or other computer systems. Solutions to the toughest problems will not be found through one narrow specialization; therefore we need to combine interdisciplinary approaches to discover the solutions to a variety of problems. Many researchers in different disciplines such as medicine, bioinformatics, molecular biology, and pharmacology have successfully applied artificial neural networks. This chapter helps the reader in understanding the basics of artificial neural networks, their applications, and methodology; it also outlines the network learning process and architecture. We present a brief outline of the application of neural networks to medical diagnosis, drug discovery, gene identification, and protein structure prediction. We conclude with a summary of the results from our study on tuberculosis data using neural networks, in diagnosing active tuberculosis, and predicting chronic vs. infiltrative forms of tuberculosis.

  11. Forecasting geomagnetic activity indices using the Boyle index through artificial neural networks

    NASA Astrophysics Data System (ADS)

    Balasubramanian, Ramkumar

    2010-11-01

    Adverse space weather conditions affect various sectors making both human lives and technologies highly susceptible. This dissertation introduces a new set of algorithms suitable for short term space weather forecasts with an enhanced lead-time and better accuracy in predicting Kp, Dst and the AE index over some leading models. Kp is a 3-hour averaged global geomagnetic activity index good for midlatitude regions. The Dst index, an hourly index calculated using four ground based magnetic field measurements near the equator, measures the energy of the Earth's ring current. The Auroral Electrojet indices or AE indices are hourly indices used to characterize the global geomagnetic activity in the auroral zone. Our algorithms can predict these indices purely from the solar wind data with lead times up to 6 hours. We have trained and tested an ANN (Artificial Neural Network) over a complete solar cycle to serve this purpose. Over the last couple of decades, ANNs have been successful for temporal prediction problems amongst other advanced non-linear techniques. Our ANN-based algorithms receive near-real-time inputs either from ACE (Advanced Composition Explorer), located at L1, and a handful of ground-based magnetometers or only from ACE. The Boyle potential, phi = 10-4 (vkm/sec)2+ 11.7BnT sin3 (theta/2) kV, or the Boyle Index (BI) is an empirically-derived formula that approximates the Earth's polar cap potential and is easily derivable in real time using the solar wind data from ACE. The logarithms of both 3-hour and 1-hour averages of the Boyle Index correlate well with the subsequent Kp, Dst and AE: Kp = 8.93 log 10 - 12.55. Dst = 0.355 - 6.48, and AE = 5.87 - 83.46. Inputs to our ANN models have greatly benefitted from the BI and its proven record as a forecasting parameter since its initiation in October, 2003. A preconditioning event tunes the magnetosphere to a specific state before an impending geomagnetic storm. The neural net not only improves the

  12. Temporal and Spatial prediction of groundwater levels using Artificial Neural Networks, Fuzzy logic and Kriging interpolation.

    NASA Astrophysics Data System (ADS)

    Tapoglou, Evdokia; Karatzas, George P.; Trichakis, Ioannis C.; Varouchakis, Emmanouil A.

    2014-05-01

    The purpose of this study is to examine the use of Artificial Neural Networks (ANN) combined with kriging interpolation method, in order to simulate the hydraulic head both spatially and temporally. Initially, ANNs are used for the temporal simulation of the hydraulic head change. The results of the most appropriate ANNs, determined through a fuzzy logic system, are used as an input for the kriging algorithm where the spatial simulation is conducted. The proposed algorithm is tested in an area located across Isar River in Bayern, Germany and covers an area of approximately 7800 km2. The available data extend to a time period from 1/11/2008 to 31/10/2012 (1460 days) and include the hydraulic head at 64 wells, temperature and rainfall at 7 weather stations and surface water elevation at 5 monitoring stations. One feedforward ANN was trained for each of the 64 wells, where hydraulic head data are available, using a backpropagation algorithm. The most appropriate input parameters for each wells' ANN are determined considering their proximity to the measuring station, as well as their statistical characteristics. For the rainfall, the data for two consecutive time lags for best correlated weather station, as well as a third and fourth input from the second best correlated weather station, are used as an input. The surface water monitoring stations with the three best correlations for each well are also used in every case. Finally, the temperature for the best correlated weather station is used. Two different architectures are considered and the one with the best results is used henceforward. The output of the ANNs corresponds to the hydraulic head change per time step. These predictions are used in the kriging interpolation algorithm. However, not all 64 simulated values should be used. The appropriate neighborhood for each prediction point is constructed based not only on the distance between known and prediction points, but also on the training and testing error of

  13. Architectures for optoelectronic analogs of self-organizing neural networks.

    PubMed

    Farhat, N H

    1987-06-01

    Architectures for partitioning optoelectronic analogs of neural nets into input-output and internal groups to form a multilayered net capable of self-organization, self-programming, and learning are described. The architectures and implementation ideas given describe a class of optoelectronic neural net modules that, when interfaced to a conventional computer controller, can impart to it artificial intelligence attributes.

  14. Forecasting SPEI and SPI Drought Indices Using the Integrated Artificial Neural Networks.

    PubMed

    Maca, Petr; Pech, Pavel

    2016-01-01

    The presented paper compares forecast of drought indices based on two different models of artificial neural networks. The first model is based on feedforward multilayer perceptron, sANN, and the second one is the integrated neural network model, hANN. The analyzed drought indices are the standardized precipitation index (SPI) and the standardized precipitation evaporation index (SPEI) and were derived for the period of 1948-2002 on two US catchments. The meteorological and hydrological data were obtained from MOPEX experiment. The training of both neural network models was made by the adaptive version of differential evolution, JADE. The comparison of models was based on six model performance measures. The results of drought indices forecast, explained by the values of four model performance indices, show that the integrated neural network model was superior to the feedforward multilayer perceptron with one hidden layer of neurons.

  15. Forecasting SPEI and SPI Drought Indices Using the Integrated Artificial Neural Networks

    PubMed Central

    Maca, Petr; Pech, Pavel

    2016-01-01

    The presented paper compares forecast of drought indices based on two different models of artificial neural networks. The first model is based on feedforward multilayer perceptron, sANN, and the second one is the integrated neural network model, hANN. The analyzed drought indices are the standardized precipitation index (SPI) and the standardized precipitation evaporation index (SPEI) and were derived for the period of 1948–2002 on two US catchments. The meteorological and hydrological data were obtained from MOPEX experiment. The training of both neural network models was made by the adaptive version of differential evolution, JADE. The comparison of models was based on six model performance measures. The results of drought indices forecast, explained by the values of four model performance indices, show that the integrated neural network model was superior to the feedforward multilayer perceptron with one hidden layer of neurons. PMID:26880875

  16. Forecasting SPEI and SPI Drought Indices Using the Integrated Artificial Neural Networks.

    PubMed

    Maca, Petr; Pech, Pavel

    2016-01-01

    The presented paper compares forecast of drought indices based on two different models of artificial neural networks. The first model is based on feedforward multilayer perceptron, sANN, and the second one is the integrated neural network model, hANN. The analyzed drought indices are the standardized precipitation index (SPI) and the standardized precipitation evaporation index (SPEI) and were derived for the period of 1948-2002 on two US catchments. The meteorological and hydrological data were obtained from MOPEX experiment. The training of both neural network models was made by the adaptive version of differential evolution, JADE. The comparison of models was based on six model performance measures. The results of drought indices forecast, explained by the values of four model performance indices, show that the integrated neural network model was superior to the feedforward multilayer perceptron with one hidden layer of neurons. PMID:26880875

  17. Tuning the stator resistance of induction motors using artificial neural network

    SciTech Connect

    Cabrera, L.A.; Elbuluk, M.E.; Husain, I.

    1997-09-01

    Tuning the stator resistance of induction motors is very important, especially when it is used to implement direct torque control (DTC) in which the stator resistance is a main parameter. In this paper, an artificial network (ANN) is used to accomplish tuning of the stator resistance of an induction motor. The parallel recursive prediction error and backpropagation training algorithms were used in training the neural network for the simulation and experimental results, respectively. The neural network used to tune the stator resistance was trained on-line, making the DTC strategy more robust and accurate. Simulation results are presented for three different neural-network configurations showing the efficiency of the tuning process. Experimental results were obtained for the one of the three neural-network configuration. Both simulation and experimental results showed that the ANN have tuned the stator resistance in the controller to track actual resistance of the machine.

  18. Intraportal infusion of ghrelin could inhibit glucose-stimulated GLP-1 secretion by enteric neural net in Wistar rat.

    PubMed

    Zhang, Xiyao; Li, Wensong; Li, Ping; Chang, Manli; Huang, Xu; Li, Qiang; Cui, Can

    2014-01-01

    As a regulator of food intake and energy metabolism, the role of ghrelin in glucose metabolism is still not fully understood. In this study, we determined the in vivo effect of ghrelin on incretin effect. We demonstrated that ghrelin inhibited the glucose-stimulated release of glucagon-like peptide-1 (GLP-1) when infused into the portal vein of Wistar rat. Hepatic vagotomy diminished the inhibitory effect of ghrelin on glucose-stimulated GLP-1 secretion. In addition, phentolamine, a nonselective α receptor antagonist, could recover the decrease of GLP-1 release induced by ghrelin infusion. Pralmorelin (an artificial growth hormone release peptide) infusion into the portal vein could also inhibit the glucose-stimulated release of GLP-1. And growth hormone secretagogue receptor antagonist, [D-lys3]-GHRP-6, infusion showed comparable increases of glucose stimulated GLP-1 release compared to ghrelin infusion into the portal vein. The data showed that intraportal infusion of ghrelin exerted an inhibitory effect on GLP-1 secretion through growth hormone secretagogue receptor 1α (GHS1α receptor), which indicated that the downregulation of ghrelin secretion after food intake was necessary for incretin effect. Furthermore, our results suggested that the enteric neural net involved hepatic vagal nerve and sympathetic nerve mediated inhibition effect of ghrelin on incretin effect. PMID:25247193

  19. Bias-free Shear Estimation Using Artificial Neural Networks

    NASA Astrophysics Data System (ADS)

    Gruen, D.; Seitz, S.; Koppenhoefer, J.; Riffeser, A.

    2010-09-01

    Bias due to imperfect shear calibration is the biggest obstacle when constraints on cosmological parameters are to be extracted from large area weak lensing surveys such as Pan-STARRS-3π, DES, or future satellite missions like EUCLID. We demonstrate that bias present in existing shear measurement pipelines (e.g., KSB) can be almost entirely removed by means of neural networks. In this way, bias correction can depend on the properties of the individual galaxy instead of being a single global value. We present a procedure to train neural networks for shear estimation and apply this to subsets of simulated GREAT08 RealNoise data. We also show that circularization of the point-spread function (PSF) before measuring the shear reduces the scatter related to the PSF anisotropy correction and thus leads to improved measurements, particularly on low and medium signal-to-noise data. Our results are competitive with the best performers in the GREAT08 competition, especially for the medium and higher signal-to-noise sets. Expressed in terms of the quality parameter defined by GREAT08, we achieve a Q≈ 40, 140, and 1300 without and 50, 200, and 1300 with circularization for low, medium, and high signal-to-noise data sets, respectively.

  20. BIAS-FREE SHEAR ESTIMATION USING ARTIFICIAL NEURAL NETWORKS

    SciTech Connect

    Gruen, D.; Seitz, S.; Koppenhoefer, J.; Riffeser, A.

    2010-09-01

    Bias due to imperfect shear calibration is the biggest obstacle when constraints on cosmological parameters are to be extracted from large area weak lensing surveys such as Pan-STARRS-3{pi}, DES, or future satellite missions like EUCLID. We demonstrate that bias present in existing shear measurement pipelines (e.g., KSB) can be almost entirely removed by means of neural networks. In this way, bias correction can depend on the properties of the individual galaxy instead of being a single global value. We present a procedure to train neural networks for shear estimation and apply this to subsets of simulated GREAT08 RealNoise data. We also show that circularization of the point-spread function (PSF) before measuring the shear reduces the scatter related to the PSF anisotropy correction and thus leads to improved measurements, particularly on low and medium signal-to-noise data. Our results are competitive with the best performers in the GREAT08 competition, especially for the medium and higher signal-to-noise sets. Expressed in terms of the quality parameter defined by GREAT08, we achieve a Q{approx} 40, 140, and 1300 without and 50, 200, and 1300 with circularization for low, medium, and high signal-to-noise data sets, respectively.

  1. Biological and bionic hands: natural neural coding and artificial perception.

    PubMed

    Bensmaia, Sliman J

    2015-09-19

    The first decade and a half of the twenty-first century brought about two major innovations in neuroprosthetics: the development of anthropomorphic robotic limbs that replicate much of the function of a native human arm and the refinement of algorithms that decode intended movements from brain activity. However, skilled manipulation of objects requires somatosensory feedback, for which vision is a poor substitute. For upper-limb neuroprostheses to be clinically viable, they must therefore provide for the restoration of touch and proprioception. In this review, I discuss efforts to elicit meaningful tactile sensations through stimulation of neurons in somatosensory cortex. I focus on biomimetic approaches to sensory restoration, which leverage our current understanding about how information about grasped objects is encoded in the brain of intact individuals. I argue that not only can sensory neuroscience inform the development of sensory neuroprostheses, but also that the converse is true: stimulating the brain offers an exceptional opportunity to causally interrogate neural circuits and test hypotheses about natural neural coding.

  2. Biological and bionic hands: natural neural coding and artificial perception

    PubMed Central

    Bensmaia, Sliman J.

    2015-01-01

    The first decade and a half of the twenty-first century brought about two major innovations in neuroprosthetics: the development of anthropomorphic robotic limbs that replicate much of the function of a native human arm and the refinement of algorithms that decode intended movements from brain activity. However, skilled manipulation of objects requires somatosensory feedback, for which vision is a poor substitute. For upper-limb neuroprostheses to be clinically viable, they must therefore provide for the restoration of touch and proprioception. In this review, I discuss efforts to elicit meaningful tactile sensations through stimulation of neurons in somatosensory cortex. I focus on biomimetic approaches to sensory restoration, which leverage our current understanding about how information about grasped objects is encoded in the brain of intact individuals. I argue that not only can sensory neuroscience inform the development of sensory neuroprostheses, but also that the converse is true: stimulating the brain offers an exceptional opportunity to causally interrogate neural circuits and test hypotheses about natural neural coding. PMID:26240424

  3. Biological and bionic hands: natural neural coding and artificial perception.

    PubMed

    Bensmaia, Sliman J

    2015-09-19

    The first decade and a half of the twenty-first century brought about two major innovations in neuroprosthetics: the development of anthropomorphic robotic limbs that replicate much of the function of a native human arm and the refinement of algorithms that decode intended movements from brain activity. However, skilled manipulation of objects requires somatosensory feedback, for which vision is a poor substitute. For upper-limb neuroprostheses to be clinically viable, they must therefore provide for the restoration of touch and proprioception. In this review, I discuss efforts to elicit meaningful tactile sensations through stimulation of neurons in somatosensory cortex. I focus on biomimetic approaches to sensory restoration, which leverage our current understanding about how information about grasped objects is encoded in the brain of intact individuals. I argue that not only can sensory neuroscience inform the development of sensory neuroprostheses, but also that the converse is true: stimulating the brain offers an exceptional opportunity to causally interrogate neural circuits and test hypotheses about natural neural coding. PMID:26240424

  4. Classification of astrocytomas and malignant astrocytomas by principal components analysis and a neural net.

    PubMed

    McKeown, M J; Ramsay, D A

    1996-12-01

    The classification of astrocytomas, astrocytomas with anaplastic foci and glioblastoma multiformes is not always straightforward because the tumors form a histological continuum. The use of principal component analysis (PCA) and neural nets in the classification of these tumors is explored. PCA was performed on 14 histological features recorded from 52 gliomas classified by the Radiation Therapy Oncology Group method (17 astrocytomas, 18 astrocytomas with anaplastic foci, 17 glioblastoma multiformes). Four of the 14 possible 'scores' derived from this analysis were selected to summarize the histological variability seen in all the tumors. These scores were mostly significantly different between tumor types and were thus used to successfully train a neural net to correctly classify these tumors. The first principal component (score) supported the use of increasing cellularity, mitoses, endothelial proliferation, and necrosis in differentiating between the tumor categories, but accounted for only 39% of the variability seen. Other histological features that were significant components of the other scores included the presence of multinucleated or giant cells, gemistocytes, atypical mitoses and changes in nuclear chromatin. Computer programs derived from the methodology described provide a way of standardizing glioma diagnosis and may be extended to assist with management decisions.

  5. Cadherins as regulators for the emergence of neural nets from embryonic divisions.

    PubMed

    Redies, Christoph; Treubert-Zimmermann, Ullrich; Luo, Jiankai

    2003-01-01

    Cadherins are a large family of cell adhesion molecules that are expressed in a spatially restricted fashion during vertebrate CNS development. Each cadherin shows a characteristic expression pattern that differs from that of other cadherins. Early in development, the cadherin expression domains relate to the neuromeric organization of the vertebrate CNS. Later, as functional structures (brain nuclei, cortical regions, fiber tracts and synapses) emerge, the expression patterns of each cadherin become restricted to subsets of these structures that form parts of specific neural nets. Cadherins thus represent a system of potentially adhesive cues that play a role in the emergence of neural nets from embryonic CNS divisions. We review descriptive and experimental evidence for such a role of cadherins in CNS development. It is argued that descriptive studies (i.e., the mapping of gene expression) and functional studies (i.e., experimental manipulation of gene expression) are equally important for generating specific and firm ideas on the function of genes in brain development.

  6. Application of artificial neural networks to predict the deflections of reinforced concrete beams

    NASA Astrophysics Data System (ADS)

    Kaczmarek, Mateusz; Szymańska, Agnieszka

    2016-06-01

    Nonlinear structural mechanics should be taken into account in the practical design of reinforced concrete structures. Cracking is one of the major sources of nonlinearity. Description of deflection of reinforced concrete elements is a computational problem, mainly because of the difficulties in modelling the nonlinear stress-strain relationship of concrete and steel. In design practise, in accordance with technical rules (e.g., Eurocode 2), a simplified approach for reinforced concrete is used, but the results of simplified calculations differ from the results of experimental studies. Artificial neural network is a versatile modelling tool capable of making predictions of values that are difficult to obtain in numerical analysis. This paper describes the creation and operation of a neural network for making predictions of deflections of reinforced concrete beams at different load levels. In order to obtain a database of results, that is necessary for training and testing the neural network, a research on measurement of deflections in reinforced concrete beams was conducted by the authors in the Certified Research Laboratory of the Building Engineering Institute at Wrocław University of Science and Technology. The use of artificial neural networks is an innovation and an alternative to traditional methods of solving the problem of calculating the deflections of reinforced concrete elements. The results show the effectiveness of using artificial neural network for predicting the deflection of reinforced concrete beams, compared with the results of calculations conducted in accordance with Eurocode 2. The neural network model presented in this paper can acquire new data and be used for further analysis, with availability of more research results.

  7. Modeling discharge-sediment relationship using neural networks with artificial bee colony algorithm

    NASA Astrophysics Data System (ADS)

    Kisi, Ozgur; Ozkan, Coskun; Akay, Bahriye

    2012-03-01

    SummaryEstimation of suspended sediment concentration carried by a river is very important for many water resources projects. The accuracy of artificial neural networks (ANN) with artificial bee colony (ABC) algorithm is investigated in this paper for modeling discharge-suspended sediment relationship. The ANN-ABC was compared with those of the neural differential evolution, adaptive neuro-fuzzy, neural networks and rating curve models. The daily stream flow and suspended sediment concentration data from two stations, Rio Valenciano Station and Quebrada Blanca Station, were used as case studies. For evaluating the ability of the models, mean square error and determination coefficient criteria were used. Comparison results showed that the ANN-ABC was able to produce better results than the neural differential evolution, neuro-fuzzy, neural networks and rating curve models. The logarithm transformed data were also used as input to the proposed ANN-ABC models. It was found that the logarithm transform significantly increased accuracy of the models in suspended sediment estimation.

  8. THE CHOICE OF OPTIMAL STRUCTURE OF ARTIFICIAL NEURAL NETWORK CLASSIFIER INTENDED FOR CLASSIFICATION OF WELDING FLAWS

    SciTech Connect

    Sikora, R.; Chady, T.; Baniukiewicz, P.; Caryk, M.; Piekarczyk, B.

    2010-02-22

    Nondestructive testing and evaluation are under continuous development. Currently researches are concentrated on three main topics: advancement of existing methods, introduction of novel methods and development of artificial intelligent systems for automatic defect recognition (ADR). Automatic defect classification algorithm comprises of two main tasks: creating a defect database and preparing a defect classifier. Here, the database was built using defect features that describe all geometrical and texture properties of the defect. Almost twenty carefully selected features calculated for flaws extracted from real radiograms were used. The radiograms were obtained from shipbuilding industry and they were verified by qualified operator. Two weld defect's classifiers based on artificial neural networks were proposed and compared. First model consisted of one neural network model, where each output neuron corresponded to different defect group. The second model contained five neural networks. Each neural network had one neuron on output and was responsible for detection of defects from one group. In order to evaluate the effectiveness of the neural networks classifiers, the mean square errors were calculated for test radiograms and compared.

  9. Comparison of artificial neural networks with logistic regression for detection of obesity.

    PubMed

    Heydari, Seyed Taghi; Ayatollahi, Seyed Mohammad Taghi; Zare, Najaf

    2012-08-01

    Obesity is a common problem in nutrition, both in the developed and developing countries. The aim of this study was to classify obesity by artificial neural networks and logistic regression. This cross-sectional study comprised of 414 healthy military personnel in southern Iran. All subjects completed questionnaires on their socio-economic status and their anthropometric measures were measured by a trained nurse. Classification of obesity was done by artificial neural networks and logistic regression. The mean age±SD of participants was 34.4 ± 7.5 years. A total of 187 (45.2%) were obese. In regard to logistic regression and neural networks the respective values were 80.2% and 81.2% when correctly classified, 80.2 and 79.7 for sensitivity and 81.9 and 83.7 for specificity; while the area under Receiver-Operating Characteristic (ROC) curve were 0.888 and 0.884 and the Kappa statistic were 0.600 and 0.629 for logistic regression and neural networks model respectively. We conclude that the neural networks and logistic regression both were good classifier for obesity detection but they were not significantly different in classification.

  10. Nuclear power plant status diagnostics using an artificial neural network

    SciTech Connect

    Bartlett, E.B.; Uhrig, R.E. )

    1992-03-01

    In this paper, nuclear power plant operating status recognition is investigated using a self-optimizing stochastic learning algorithm artificial neutral network (ANN) with dynamic node architecture learning. The objective is to train the ANN to classify selected nuclear power plant accident conditions and assess the potential for future success in this area. The network is trained on normal operating conditions as well as on potentially unsafe conditions based on nuclear power plant training simulator-generated accident scenarios. These scenarios include hot-and cold-leg loss of coolant, control rod ejection, total loss of off-site power, main streamline break, main feedwater line break, and steam generator tube leak accidents as well as the normal operating condition. Findings show that ANNs can be used to diagnose and classify nuclear power plant conditions with good results. continued research work indicated.

  11. Artificial neural network study on organ-targeting peptides.

    PubMed

    Jung, Eunkyoung; Kim, Junhyoung; Choi, Seung-Hoon; Kim, Minkyoung; Rhee, Hokyoung; Shin, Jae-Min; Choi, Kihang; Kang, Sang-Kee; Lee, Nam Kyung; Choi, Yun-Jaie; Jung, Dong Hyun

    2010-01-01

    We report a new approach to studying organ targeting of peptides on the basis of peptide sequence information. The positive control data sets consist of organ-targeting peptide sequences identified by the peroral phage-display technique for four organs, and the negative control data are prepared from random sequences. The capacity of our models to make appropriate predictions is validated by statistical indicators including sensitivity, specificity, enrichment curve, and the area under the receiver operating characteristic (ROC) curve (the ROC score). VHSE descriptor produces statistically significant training models and the models with simple neural network architectures show slightly greater predictive power than those with complex ones. The training and test set statistics indicate that our models could discriminate between organ-targeting and random sequences. We anticipate that our models will be applicable to the selection of organ-targeting peptides for generating peptide drugs or peptidomimetics.

  12. A Rapid Aerodynamic Design Procedure Based on Artificial Neural Networks

    NASA Technical Reports Server (NTRS)

    Rai, Man Mohan

    2001-01-01

    An aerodynamic design procedure that uses neural networks to model the functional behavior of the objective function in design space has been developed. This method incorporates several improvements to an earlier method that employed a strategy called parameter-based partitioning of the design space in order to reduce the computational costs associated with design optimization. As with the earlier method, the current method uses a sequence of response surfaces to traverse the design space in search of the optimal solution. The new method yields significant reductions in computational costs by using composite response surfaces with better generalization capabilities and by exploiting synergies between the optimization method and the simulation codes used to generate the training data. These reductions in design optimization costs are demonstrated for a turbine airfoil design study where a generic shape is evolved into an optimal airfoil.

  13. Application of artificial neural networks to thermal detection of disbonds

    NASA Technical Reports Server (NTRS)

    Prabhu, D. R.; Howell, P. A.; Syed, H. I.; Winfree, W. P.

    1992-01-01

    A novel technique for processing thermal data is presented and applied to simulation as well as experimental data. Using a neural network of thermal data classification, good classification accuracies are obtained, and the resulting images exhibit very good contrast between bonded and disbonded locations. In order to minimize the preprocessing required before using the network of classification, the temperature values were directly employed to train a network using data from an on-site testing run of a commercial aircraft. Training was extremely fast, and the resulting classification also agreed reasonably well with an ultrasonic characterization of the panel. The results obtained using one sample show the partially disbonded vertical doubler. The vertical lines along the doubler correspond to the original extent of the doubler obtained using blueprints of the aircraft.

  14. Simple artificial neural networks that match probability and exploit and explore when confronting a multiarmed bandit.

    PubMed

    Dawson, Michael R W; Dupuis, Brian; Spetch, Marcia L; Kelly, Debbie M

    2009-08-01

    The matching law (Herrnstein 1961) states that response rates become proportional to reinforcement rates; this is related to the empirical phenomenon called probability matching (Vulkan 2000). Here, we show that a simple artificial neural network generates responses consistent with probability matching. This behavior was then used to create an operant procedure for network learning. We use the multiarmed bandit (Gittins 1989), a classic problem of choice behavior, to illustrate that operant training balances exploiting the bandit arm expected to pay off most frequently with exploring other arms. Perceptrons provide a medium for relating results from neural networks, genetic algorithms, animal learning, contingency theory, reinforcement learning, and theories of choice.

  15. Fault detection and diagnosis of power converters using artificial neural networks

    SciTech Connect

    Swarup, K.S.; Chandrasekharaiah, H.S.

    1995-12-31

    Fault detection and diagnosis in real-time are areas of research interest in knowledge-based expert systems. Rule-based and model-based approaches have been successfully applied to some domains, but are too slow to be effectively applied in a real-time environment. This paper explores the suitability of using artificial neural networks for fault detection and diagnosis of power converter systems. The paper describes a neural network design and simulation environment for real-time fault diagnosis of thyristor converters used in HVDC power transmission systems.

  16. Transient stability assessment in longitudinal power systems using artificial neural networks

    SciTech Connect

    Aboytes, F.; Ramirez, R.

    1996-11-01

    Results of the application of Artificial Neural Networks to the problem of Transient Stability Assessment are presented. This technique is applied to a real Longitudinal Power System that includes discrete supplementary controls. Different representations of the training space patterns and neural networks architectures are investigated. Input variables include topological changes, load and generation levels and contingencies. A special organization of training patterns with a separation by type of contingency is proposed to reduce classification errors. A graphical presentation of results is suggested as an aid to help system operators to select preventive control actions.

  17. Simple artificial neural networks that match probability and exploit and explore when confronting a multiarmed bandit.

    PubMed

    Dawson, Michael R W; Dupuis, Brian; Spetch, Marcia L; Kelly, Debbie M

    2009-08-01

    The matching law (Herrnstein 1961) states that response rates become proportional to reinforcement rates; this is related to the empirical phenomenon called probability matching (Vulkan 2000). Here, we show that a simple artificial neural network generates responses consistent with probability matching. This behavior was then used to create an operant procedure for network learning. We use the multiarmed bandit (Gittins 1989), a classic problem of choice behavior, to illustrate that operant training balances exploiting the bandit arm expected to pay off most frequently with exploring other arms. Perceptrons provide a medium for relating results from neural networks, genetic algorithms, animal learning, contingency theory, reinforcement learning, and theories of choice. PMID:19596631

  18. Discriminant analysis and neural nets: Valuable tools to optimize completion practices

    SciTech Connect

    Nitters, G.; Davies, D.R.; Epping, W.J.M.

    1995-06-01

    This paper describes the application of multi-variate statistical techniques, discriminant analysis and neural networks in identifying drilling and other completion practices that impact on well productivity. Discriminant analysis determines whether a well can be assigned to a group of wells, on the basis of a number of common characteristics and using linear multivariate correlations. Neural nets enable the use of nonlinear correlations for such a classification. In this study, 47 gas wells from two fields were classified into three groups: Group 1 -- no production; Group 2 -- production below 5,900 std m{sup 3}/h (5 MMscf/D); Group 3 -- production over 5,900 std m{sup 3}/h (5 MMscf/D). The variables used in the discriminant analysis included parameters such as completion type, total height of the perforated interval, mud weight, drawdown during perforation, type of mud and perforation size. This study has identified and, to some extent, quantified those parameters that either adversely or favorable affect well productivity. The results can be used to adjust operational procedures to maximize well productivity. The parameters identified as increasing productivity reflect, for the most part, sound engineering practices. Application of neural nets enables further quantification of the effects of petroleum engineering parameters on well productivity and is being developed to make it possible for the most economical preventive and remedial measures to be selected. However, statistical techniques are applicable only when a sufficiently large data base is available, i.e., they are suitable for reasonably large and fairly mature fields and/or areas.

  19. Application of neural nets to system identification and bifurcation analysis of real world experimental data

    SciTech Connect

    Adomaitis, R.A.; Kevrekidis, I.G. . Dept. of Chemical Engineering); Farber, R.M.; Lapedes, A.S. ); Hudson, J.L.; Kube, M. . Dept. of Chemical Engineering)

    1990-02-01

    We report results on the use of neural nets, and the closely related radial basis nets'', to analyze experimental time series from electro-chemical systems. We show how the nets may be used to derive a map that describes the nonlinear system, and how reserving an extra input line'' of the network allows one to learn the system behavior dependent on a control variable. Pruning'' of the network after training appears to result in elimination of spurious connection weights and enhanced predictive accuracy. Subsequent analysis of the learned map using techniques of bifurcation theory allows both nonlinear system identification and accurate and efficient predictions of long-term system behavior. The electrochemical system that was used involved the electrodissolution of copper in phosphoric acid. This system exhibits interesting low dimensional dynamics such transitions from steady state to oscillatory behavior and from period-one to period-two oscillations. This analysis provides an example of methodology that can be fruitful in understanding systems for which no adequate phenomenological model exists, or for which predictions of system behavior given a large scale, complicated model is inherently impractical. 17 refs., 2 figs.

  20. A study on precursors leading to geomagnetic storms using artificial neural network

    NASA Astrophysics Data System (ADS)

    Singh, Gaurav; Singh, A. K.

    2016-07-01

    Space weather prediction involves advance forecasting of the magnitude and onset time of major geomagnetic storms on Earth. In this paper, we discuss the development of an artificial neural network-based model to study the precursor leading to intense and moderate geomagnetic storms, following halo coronal mass ejection (CME) and related interplanetary (IP) events. IP inputs were considered within a 5-day time window after the commencement of storm. The artificial neural network (ANN) model training, testing and validation datasets were constructed based on 110 halo CMEs (both full and partial halo and their properties) observed during the ascending phase of the 24th solar cycle between 2009 and 2014. The geomagnetic storm occurrence rate from halo CMEs is estimated at a probability of 79%, by this model.