Neural Networks for Identification, Prediction and Control

In recent years, there has been a growing interest in applying neural networks to dynamic systems identification (modelling), prediction and control. Neural networks are computing systems characterised by the ability to learn from examples rather than having to be programmed in a conventional sense. Their use enables the behaviour of complex systems to be modelled and predicted and accurate control to be achieved through training, without a priori information about the systems' structures or parameters. This book describes examples of applications of neural networks In modelling, prediction and control. The topics covered include identification of general linear and non-linear processes, forecasting of river levels, stock market prices and currency exchange rates, and control of a time-delayed plant and a two-joint robot. These applications employ the major types of neural networks and learning algorithms. The neural network types considered in detail are the muhilayer perceptron (MLP), the Elman and Jordan networks and the Group-Method-of-Data-Handling (GMDH) network. In addition, cerebellar-model-articulation-controller (CMAC) networks and neuromorphic fuzzy logic systems are also presented. The main learning algorithm adopted in the applications is the standard backpropagation (BP) algorithm. Widrow-Hoff learning, dynamic BP and evolutionary learning are also described.

Table of Contents:
1 Artificial Neural Networks.- l.1 Types of Neural Networks.- l.2 Example Neural Networks.- 1.3 Summary.- References.- 2 Dynamic System Identification Using Feedforward Neural Networks.- 2.1Dynamic System Descriptions.- 2.2 Identification Based on System Inputs and Outputs.- 2.3 Identification Based on Measurable System States.- 2.4 Input-output Model Identification.- 2.5 State-space Model Identification.- 2.6 Discussion.- 2.7 Analysis of the Hybrid Network.- 2.8 Summary.- References.- 3 Dynamic System Modelling Using Recurrent Neural Networks.- 3.1 Basic Ehnan Network.- 3.2 Modified Ehnan Network.- 3.3 Dynamic System Modelling.- 3.4 Further Analysis of Ehnan Networks.- 3.5 Summary.- References.- 4 Modelling and Prediction Using GMDH Networks.- 4.1 N-Adaline Networks and Widrow-Hoff Learning.- 4.2 GMDH Network Based on N-Adalines.- 4.3 Applications.- 4.4 Discussion.- 4.5 Summary.- References.- 5 Financial Prediction Using Neural Networks.- 5.1 Stock Market Prediction.- 5.2 Currency Exchange Rate Prediction.- 5.3 Data Sets Adopted for Simulation.- 5.4 Prediction Based on GMDH Networks.- 5.5 Prediction Based on Multilayer Perception Networks.- 5.6 Prediction Based on Recurrent Networks.- 5.7 Discussion.- 5.8 Summary.- References.- 6 Neural Network Controllers.- 6.1 Neural Network Controllers.- 6.2 Comparison of Neural Network Controllers.- 6.3 Summary.- References.- 7 Neuromorphic Fuzzy Controller Design.- 7.1 Integrating Neural Networks and FLCs.- 7.2 Results of Neuromorphic Fuzzy Controllers Design.- 7.3 Summary.- References.- 8 Robot Manipulator Control Using Neural Networks.- 8.1 Modelling of a Multi-joint Robot.- 8.2 Control System.- 8.3 Application to a Two-joint Robot Arm.- 8.4 Discussion.- 8.5 Summary.- References.- Appendix A Introduction to Some Conventional Techniques of Identification, Prediction and Control.- Appendix B Fuzzy Sets and Fuzzy Logic Control.- Appendix C Genetic Algorithms.- Appendix D Program: Feedforward Network for System Identification.- Appendix E Program: Modified Elman Network for Identification.- Appendix F Program: GMDH Network for Prediction.- Author Index.