Applications of Neural Networks in High Assurance Systems / Edition 1

Applications of Neural Networks in High Assurance Systems / Edition 1

by Johann M.Ph. Schumann
     
 

ISBN-10: 3642106897

ISBN-13: 9783642106897

Pub. Date: 02/01/2010

Publisher: Springer Berlin Heidelberg

The series Studies in Computational Intelligence (SCI) publishes new developments and advances in the various areas of computational intelligence-quickly and with a high quality. The intent is to cover the theory, applications, and design methods of computational intelligence, as embedded in the fields of engineering, computer science, physics and life science, as

Overview

The series Studies in Computational Intelligence (SCI) publishes new developments and advances in the various areas of computational intelligence-quickly and with a high quality. The intent is to cover the theory, applications, and design methods of computational intelligence, as embedded in the fields of engineering, computer science, physics and life science, as well as the methodologies behind them. The series contains monographs, lecture notes and edited volumes in computational intelligence spanning the areas of neural networks, connectionist systems, genetic algorithms, evolutionary computation, artificial intelligence, cellular automata, self-organizing systems, soft computing, fuzzy systems and hybrid intelligent systems. Critical to both contributors and readers are the short publication time and world-wide distribution-this permits a rapid and broad dissemination of research results.

'Applications of Neural Networks in High Assurance Systems' is the first book directly addressing a key part of neural network technology: methods used to pass the tough verification and validation (V&V) standards required in many safety-critical applications. This book presents relevant theoretical research and recent advances in industrial application and certification of neural network based learning systems in safety-related areas like

damage-adaptive aircraft control,

fault detection in automotive engines,

control of submarines, fuel cells, and oil-blending.

This book aims to provide a better understanding of the practical requirement for developing and deploying neuro-adaptive systems, and to inspire research on new methods and techniques for testing, certification, and standardization of neural network applications in high assurance systems.

Product Details

ISBN-13:
9783642106897
Publisher:
Springer Berlin Heidelberg
Publication date:
02/01/2010
Series:
Studies in Computational Intelligence Series, #268
Edition description:
2010
Pages:
248
Product dimensions:
6.10(w) x 9.25(h) x 0.04(d)

Table of Contents

Application of Neural Networks in High Assurance Systems: A Survey Johann Schumann Pramod Gupta Yan Liu 1

1 Introduction 1

2 Application Domains 3

2.1 Aircraft Control 4

2.2 Automotive 4

2.3 Power Systems 5

2.4 Medical Systems 6

2.5 Other Applications 7

3 Toward V&V of NNs in High Assurance Systems 8

3.1 V&V of Software Systems 8

3.2 V&V Issues and Gaps for NN-Based Applications 10

3.3 V&V Approaches for Neural Networks 11

4 Conclusions 15

References 16

Robust Adaptive Control Revisited: Semi-global Boundedness and Margins Anuradha M. Annaswamy Jinho Jang Eugene Lavretsky 21

1 Introduction 21

2 Problem Statement 22

3 Adaptive Controller 24

3.1 Reference Model 24

3.2 Adaptive Controller Design 24

3.3 Stability Analysis 25

4 Delay Margins 25

4.1 (1,1) Pade Approximation (η =Δ1 (s)u) 26

4.2 (2,2) Pade Approximation (η =Δ 2 (s)u) 30

5 Nonlinearity Margins 31

5.1 Interpretation of Theorem 3 34

5.2 Numerical Model: Hypersonic Vehicle 34

5.3 Relation between e(t0), ε(xp), and N 36

References 38

A Appendix 39

Network Complexity Analysis of Multilayer Feedforward Artificial Neural Networks Helen Yu 41

1 Introduction 41

2 Pruning Algorithms 44

3 Computer Simulation Results 50

4 Summary 53

References 53

Design and Flight Test of an Intelligent Flight Control System Tim Smith Jim Barhorst James M. Urnes 57

1 Introduction 57

2 IFCS Program 58

3 IFCS Experiment 59

4 Controller Architecture 61

5 Requirements Validation 63

5.1 System Stability 64

5.2 Aeroservoelastic Margin 64

5.3 Handling Qualities 66

5.4 Nonlinear Systems Requirements Validation 67

6 Flight Controls Software and System Verification 70

7 Flight Test 74

8 Conclusions 75

References 76

Stability, Convergence, and Verification and Validation Challenges of Neural Net Adaptive Flight Control Nhan T. Nguyen Stephen A. Jacklin 77

1 Introduction 77

2 Convergence and Stability of Neural Net Direct Adaptive Flight Control 79

2.1 Direct Adaptive Control Approach 80

2.2 Stability and Convergence 82

2.3 Unmodeled Dynamics 90

3 Potential Improvements 93

3.1 Direct Adaptive Control with Recursive Least Squares 93

3.2 Hybrid Direct-Indirect Adaptive Control with Recursive Least-Squares 96

4 Verification and Validation Challenges for Adaptive Systems 99

4.1 Simulation of Adaptive Control Systems 99

4.2 Approach for Adaptive System V&V 101

5 Future Research 103

5.1 Adaptive Control 103

5.2 Verification and Validation 105

6 Conclusions 107

References 107

Dynamic Allocation in Neural Networks for Adaptive Controllers Sampath Yerramalla Edgar Fuller Bojan Cukic 111

1 Introduction 111

1.1 Paper Overview 113

2 Dynamic Allocation in Neural Networks 113

2.1 Dynamic Cell Structures 114

2.2 Components of DCS Neural Netwrork 114

2.3 DCS Algorithm 117

3 Robustness Analysis of Dynamic Allocation 119

3.1 Node Insertion 119

3.2 Analysis for UC1 (Undesirable Condition 1) 120

3.3 Analysis for UC2 (Undesirable Condition 2) 122

4 Data-Driven Dynamic Allocation Algorithm 127

5 Case Study 131

6 Conclusion 138

References 138

Immune Systems Inspired Approach to Anomaly Detection, Fault Localization and Diagnosis in Automotive Engines Dragan Djurdjanovic Jianbo Liu Kenneth A. Marko Jun Ni 141

1 Introduction 141

2 Research Issues in Immune Systems Engineering 143

2.1 Anomaly Detection and Fault Localization 146

2.2 Fault Diagnosis 148

2.3 Automatic Control System Reconfiguration 149

3 Anomaly Detection, Fault Isolation and Diagnosis in an Automotive Electronic Throttle System 150

3.1 Anomaly Detection and Fault Isolation 150

3.2 Fault Diagnosis 152

3.3 Fever-Like Behavior in the Presence of an Unknown Fault 153

4 Anomaly Detection and Fault Isolation in Automotive Crankshaft Dynamics 156

5 Conclusions and Future Work 160

References 161

Pitch-Depth Control of Submarine Operating in Shallow Water via Neuro-adaptive Approach Y.D. Song Liguo Weng Medorian D. Gheorghiu 165

1 Introduction 165

2 Dynamics 166

2.1 Nonlinear Model 167

2.2 Fault Dynamics 168

3 Control Design 169

3.1 Nonlinear Model 169

3.2 Stability Analysis 171

4 Simulation Results 172

5 Conclusions 177

References 177

Stick-Slip Friction Compensation Using a General Purpose Neuro-adaptive Controller with Guaranteed Stability Ali Reza Mehrabian Mohammad Bagher Menhaj 179

1 Introduction 179

2 The Neural-Network-Based Control Strategy 182

2.1 Indirect Adaptive Neuro-Controller 182

2.2 Neural Network Scheme 183

2.3 Control Oriented On-Line Identification Method 183

2.4 Mathematical Description of the Control Scheme 184

2.5 Training Multilayer Neural Network (MLP) 186

2.6 Back-Propagation through the Model 186

3 Stability Analysis 187

4 Implementing the Proposed Adaptive-Neuro Control Method 188

4.1 NN Identifier Block 188

4.2 NN Controller Block 191

4.3 Controller Error Sensitivity Feedback Block 191

5 Simulation Studies 192

5.1 Example 1: A Non-linear System with a Second-Order Difference Equation and Variable Reference Model 192

5.2 Example 2: A Non-linear Plant Subjected to Uncertainty 194

6 Stick-Slip Friction Compensation Using the Introduced Neuro-Control Algorithm 195

6.1 Problem Statement 195

6.2 Simulation Results 196

7 Conclusions 200

References 201

Modeling of Crude Oil Blending via Discrete-Time Neural Networks Xiaoou Li Wen Yu 205

1 Introduction 205

2 Crude Oil Blending 206

3 Modeling of Crude Oil Blending via Discrete-Time Neural Networks 208

4 Application Study 213

5 Conclusion 218

References 219

Adaptive Self-Tuning Wavelet Neural Network Controller for a Proton Exchange Membrane Fuel Cell M. Sedighizadeh

1 Introduction 222

2 PEMFC System Model 223

3 Wavelet Neural Network and Identification Algorithm 226

3.1 Wavelet Neural Network 226

3.2 System Model Identification 228

4 Proposed Controller Design 230

4.1 Neural Network Controller Based on Wavelet 230

4.2 PID Neural Network Controller Based on Wavelets 230

5 Simulation Results 232

5.1 Identification of PEMFC 232

5.2 Control of PEMFC without Noise 233

5.3 Control of PEMFC with Input Noise 236

5.4 Control of PEMFC with Output Noise Problem 240

6 Conclusions 244

References 244

Customer Reviews

Average Review:

Write a Review

and post it to your social network

     

Most Helpful Customer Reviews

See all customer reviews >