Neural Networks Modeling and Control: Applications for Unknown Nonlinear Delayed Systems in Discrete Time

Cover
Neural Networks Modeling and
Control:
Applications for Unknown Nonlinear Delayed Systems
in Discrete Time
Copyright
Dedication
Contents
About the authors
Preface
Acknowledgments
1 Introduction
	1.1 Systems with delay
	1.2 System model
	1.3 Neural identification
	1.4 Neural state observers
	1.5 Neural block control
		1.5.1 Discrete sliding modes
		1.5.2 Inverse optimal control
	1.6 Problem statement
	1.7 Background
		1.7.1 Previous work on systems with time delay
		1.7.2 Advantages of our schemes
2 Mathematical preliminaries
	2.1 Time delay systems
		2.1.1 Delay
		2.1.2 System with time delay
		2.1.3 Nonlinear discrete system with time delays
	2.2 Recurrent high-order neural networks
		2.2.1 Discrete high-order recurrent neural networks
		2.2.2 Extended Kalman filter training
			2.2.2.1 RHONN training using EKF
3 Neural identification using recurrent high-order neural networks for discrete nonlinear systems with unknown time delays
	3.1 Identification of the system
	3.2 Neural identification
	3.3 Identifier design based on recurrent high-order neural networks for uncertain nonlinear systems with delay
	3.4 Results of RHONN identifier
		3.4.1 Simulation results: Van der Pol oscillator
			3.4.1.1 Observer based on high-order neural network
			3.4.1.2 RHONN-based identifier
			3.4.1.3 Simulation results
		3.4.2 Simulation results: differential robot
4 Identifier-controller scheme for uncertain nonlinear discrete systems with unknown time delays
	4.1 Identifier-controller scheme, sliding modes
		4.1.1 Block control with sliding modes
	4.2 Results of identifier-controller scheme, sliding modes
		4.2.1 Real-time results: linear induction motor with variant delays Test 4.1
		4.2.2 Real-time results: linear induction motor with variants delays Test 4.2
		4.2.3 Real-time results: linear induction motor with varying delays Test 4.3
	4.3 Identifier-controller scheme, inverse optimal control
		4.3.1 Inverse optimal control
	4.4 Results of identifier-controller scheme, inverse optimal control
		4.4.1 Application to a tank differential robot
			4.4.1.1 Inverse optimal control for a tank differential robot
		4.4.2 Real-time results: differential robot Test 4.4
		4.4.3 Real-time results: differential robot Test 4.5
5 Neural observer based on a RHONN for uncertain nonlinear discrete systems with unknown time delays
	5.1 Neural observer
	5.2 Full-order neural observer design based on a RHONN for discrete-time nonlinear systems with unknown delays
		5.2.1 Results of full-order RHONN observer
			5.2.1.1 Simulation results
			5.2.1.2 Experimental results
	5.3 Reduced-order neural observer design based on RHONNs for discrete-time nonlinear systems with unknown delays
	5.4 Results of reduced-order neural observer
		5.4.1 Simulation results
		5.4.2 Real-time results
6 Observer-controller scheme for uncertain nonlinear discrete systems with unknown delays
	6.1 RHONN observer-controller scheme for uncertain nonlinear discrete systems with unknown delays
		6.1.1 Simulation results: reduced-order RHONN observer-controller
		6.1.2 Real-time results: reduced RHONN observer-controller
7 Conclusions
	7.1 Conclusions
APPENDIX
A Artificial neural networks
	A.1 Biological neural networks
		A.1.1 Biological neuron
		A.1.2 Biological synapse
		A.1.3 Types of neurons
	A.2 Artificial neural networks
		A.2.0.1 Artificial neuron
	A.3 Activation functions
	A.4 Classification of neural networks
		A.4.1 Single-layer neural networks
		A.4.2 Multilayer neural networks
		A.4.3 Recurrent neural networks
	A.5 Neural network training
APPENDIX
B Linear induction motor prototype
	B.1 Linear induction motor
		B.1.1 How a LIM works
		B.1.2 Model of a LIM
		B.1.3 Flux observer
	B.2 Linear induction motor prototype
		B.2.1 Electric drive by induction motor
		B.2.2 LIM prototype
APPENDIX
C Differential tracked robot prototype
	C.1 Tracked robot
		C.1.1 Tracked robot model
	C.2 Prototype
Bibliography
Index
Back Cover