Innovations in Intelligent Machines-5: Computational Intelligence in Control Systems Engineering

✍ Scribed by Balas, Valentina Emilia(Editor);Koprinkova-Hristova, Petia(Editor);Jain, Lakhmi C(Editor)

• Publisher: Springer
• Year: 2014
• Tongue: English
• Leaves: 261
• Category: Library

✦ Synopsis

This research monograph presents selected areas of applications in the field of control systems engineering using computational intelligence methodologies. A number of applications and case studies are introduced. These methodologies are increasing used in many applications of our daily lives. Approaches include, fuzzy-neural multi model for decentralized identification, model predictive control based on time dependent recurrent neural network development of cognitive systems, developments in the field of Intelligent Multiple Models based Adaptive Switching Control, designing military training simulators using modelling, simulation, and analysis for operational analyses and training, methods for modelling of systems based on the application of Gaussian processes, computational intelligence techniques for process control and image segmentation technique based on modified particle swarm optimized-fuzzy entropy.

✦ Table of Contents

Foreword......Page 6
Preface......Page 7
Contents......Page 11
1.1…Introduction......Page 16
1.2.1 Description of the RTNN Topology and Its Real-Time BP Learning......Page 18
1.2.2 Description of the Real-Time Second-Order Levenberg-Marquardt Learning......Page 22
1.3…Description of the Decentralized Direct I-Term Fuzzy-Neural Multi-Model Control System......Page 23
1.4…Description of the Decentralized Indirect (Sliding Mode) I-Term Fuzzy-Neural Multi-Model Control System......Page 27
1.4.1 Sliding Mode Control Systems Design......Page 29
1.5…Analytical Model of the Anaerobic Digestion Bioprocess Plant......Page 31
1.6.1 Simulation Results of the System Identification......Page 34
1.6.2 Simulation Results of the Direct HFNMM Control with and without I-Term......Page 38
1.6.3 Simulation Results of the Indirect HFNMM I-Term SMC......Page 42
1.7…Conclusion......Page 49
A.0. Appendix 1: Detailed Derivation of the Recursive Levenberg-Marquardt Optimal Learning Algorithm for the RTNN......Page 50
References......Page 56
2.1…Introduction......Page 59
2.2…Model Predictive Control Algorithms......Page 61
2.3…Neural Dynamical Process Model......Page 63
2.3.2 NN Auto Regressive with eXogenous Input (NNARX) Model......Page 64
2.3.4 NN Output Error (NNOE) Model......Page 65
2.3.4.3 Neural Model Structure for Training and MPC Implementation......Page 67
2.4.1 Classical (one test) Identification Experiment......Page 68
2.5.1 Problem Formulation......Page 69
2.5.3 Normalized ETMPC......Page 71
2.6.1 Process Description......Page 72
2.6.2 Crystallization Macro Model......Page 73
2.6.2.2 Energy Balance:......Page 74
2.7.1 Identification Experiments......Page 75
2.7.2.1 Lag Space Selection......Page 76
2.7.2.2 Number of Hidden Nodes......Page 77
2.7.2.3 NN Pruning......Page 78
2.8…NN ETMPC Control Tests......Page 79
2.8.1 Set-Point Tracking......Page 80
2.8.2 Computational Time Reduction......Page 82
2.8.3 Final Product Quality......Page 83
2.9…Conclusions......Page 84
References......Page 85
3.1…Introduction......Page 87
3.2…Single Model Adaptive Control and Multiple Models Switching Adaptive Control......Page 89
3.2.1 Adaptive Control Basics......Page 90
3.2.2 An Overview of Multiple Models Switching Control Methods......Page 91
3.2.2.1 Why Multiple Models? From Adaptive to Multiple Models Adaptive Switching Control......Page 92
3.2.2.2 Multiple Fixed Models Switching Control......Page 96
3.2.2.3 A Combination of Fixed and Adaptive Multiple Models......Page 97
3.2.2.4 Nonlinear Systems Control Using Multiple Models......Page 98
3.3.1 Multiple Models Adaptive Control for SISO and MIMO Discrete-Time Nonlinear Systems Using Neural Networks......Page 101
3.4…Multiple Models Switching Control and Fuzzy Systems......Page 104
3.4.2 Multiple T-S Fuzzy Models for More Reliable Control Systems......Page 105
3.4.3 Problem Statement and Single Identification Model......Page 106
3.4.4 Architecture......Page 107
3.4.5 Switching Rule and Cost Criterion......Page 108
3.4.6.1 T-S Identification Models......Page 109
3.4.6.2 Controller Design......Page 110
3.4.6.3 The Next Best Controller Logic......Page 111
3.4.7.1 Stability and Adaptive Laws......Page 112
3.4.7.2 Comments on Convergence and Computational Cost of the Switching Algorithm......Page 114
3.4.8 Another Approach with Hybrid T-S Multiple Models......Page 115
3.5…Numerical Example......Page 116
References......Page 119
Abstract......Page 123
4.1…Introduction......Page 124
4.2…A Computational Intelligence Based Approach Using Rough and Fuzzy Sets to Model the Component Retrieval Problem......Page 126
4.4…Fuzzy Sets......Page 127
4.5…Rough-Fuzzy Sets......Page 128
4.7…A Case Study of the Design and Development of Air Warfare Simulation Systems......Page 129
4.7.1 Use of Rough-Fuzzy Sets......Page 131
4.7.2 Use of Fuzzy-Rough Sets......Page 141
References......Page 144
Abstract......Page 147
5.2…An Agent-Based Architecture to Design Military Training Operations......Page 148
5.3…Mathematical Modelling of Pilot Behavior......Page 150
5.4.1 ANFIS......Page 155
5.4.2 Modeling the Human Factors and Situation Awareness of Pilot Agent......Page 157
5.5…Design of Conflict Situations and Discussion of Results......Page 160
5.6…Conclusions and Future Work......Page 165
References......Page 166
Abstract......Page 169
6.2…Systems Modelling with Gaussian Processes......Page 170
6.3.1 Inverse Dynamics Control......Page 174
6.3.2 Model-Based Predictive Control......Page 177
6.3.2.1 Internal Model Control......Page 180
6.3.2.2 Predictive Functional Control......Page 181
6.3.2.3 Approximate Explicit Stochastic Nonlinear Model Predictive Control......Page 182
6.3.3 Adaptive Control......Page 187
6.4.1 The Gas--Liquid Separator......Page 190
6.4.2 Gaussian Process Model of the Gas--Liquid Separator......Page 193
6.4.3 Design and Performance of Explicit GP-NMPC......Page 196
References......Page 201
7.1…Objectives and Conventional Automatic Control of Chemical Processes......Page 205
7.1.1 Objectives of Chemical Processes......Page 206
7.1.2 Conventional Automatic Control of Fractionating Processes......Page 210
7.1.3 Conventional Automatic Control of Heat Transfer Processes......Page 215
7.1.4 Conventional Automatic Control of Chemical Reactors......Page 218
7.2…Computational Intelligence Techniques for Process Control......Page 220
7.2.1 Computational Intelligence Techniques......Page 221
7.2.2.1 Fuzzy Systems Applications......Page 225
7.2.2.3 Genetic Algorithms Applications......Page 226
7.3…Case study: The Wastewater pH Neutralisation Process in a Wastewater Treatment Plant......Page 227
7.3.1 The Process Mathematical Model Development......Page 229
7.3.2 The R-ANFIS Controller Development......Page 231
7.3.3 pHACS Implementing in Matlab/Simulink......Page 233
7.4…Conclusion......Page 237
References......Page 238
Abstract......Page 241
8.1…Introduction......Page 242
8.2.1 Image as a Fuzzy Event......Page 243
8.2.2 Probability Partition Based Maximum Fuzzy Entropy......Page 244
8.2.3 Fuzzy Membership Functions for Dark and Bright Classes......Page 245
8.2.4 Modified Particle Swarm Optimization Algorithm......Page 247
8.3.1 Fuzzy Parameter Optimization Using MPSO......Page 249
8.4…Experimental Results......Page 251
8.4.1 Convergence Test......Page 256
References......Page 258
About the Editors......Page 260