Muutke küpsiste eelistusi

Soft Computing In Systems And Control Technology [Kõva köide]

(National Technical Univ Of Athens, Greece)
Teised raamatud teemal:
Soft Computing In Systems And Control TechnologySuurem pilt
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9789810233815.html
Soft computing is a branch of computing which, unlike hard computing, can deal with uncertain, imprecise and inexact data. The three constituents of soft computing are fuzzy-logic-based computing, neurocomputing, and genetic algorithms. Fuzzy logic contributes the capability of approximate reasoning, neurocomputing offers function approximation and learning capabilities, and genetic algorithms provide a methodology for systematic random search and optimization. These three capabilities are combined in a complementary and synergetic fashion.This book presents a cohesive set of contributions dealing with important issues and applications of soft computing in systems and control technology. The contributions include state-of-the-art material, mathematical developments, fresh results, and how-to-do issues. Among the problems studied via neural, fuzzy, neurofuzzy and genetic methodologies are: data fusion, reinforcement learning, approximation properties, multichannel imaging, signal processing, system optimization, gaming, and several forms of control.The book can serve as a reference for researchers and practitioners in the field. Readers can find in it a large amount of useful and timely information, and thus save considerable effort in searching for other scattered literature.
Preface v
Contributors vii
Editorial xxi
PART I: NEURAL NETWORKS IN SYSTEM IDENTIFICATION AND CONTROL
Chapter 1 Supervised Learning in Multilayer Perceptrons: The Back-Propagation Algorithm
S.G. Tzafestas and Y. Anthopoulos
1. Introduction
3(1)
2. The supervised learning problem
3(5)
2.1 The main learning problems
7(1)
3. Multilayer perceptrons
8(5)
3.1 Approximating capabilities of multilayer perceptrons
9(4)
4. Supervised learning in multilayer perceptrons
13(6)
4.1 Generalization
16(3)
5. The back-propagation algorithm
19(6)
References
25(6)
Chapter 2 Identification of Two-Dimensional State Space Discrete Systems Using Neural Networks
D. Wang and A. Zilouchian
1. Introduction
31(2)
2. Problem Approach
33(3)
2.1 Motivation
33(2)
2.2 NN model representation
35(1)
3. A pattern mode training algorithm
36(6)
3.1 Training objective
37(1)
3.2 Weight adjustment
37(1)
3.3 Gradient calculation
38(3)
3.4 Algorithm summary
41(1)
4. Implementation issues
42(1)
4.1 Identifying a 2-D separable in denominator digital filter
42(1)
4.2 Order selection and order reduction
42(1)
4.3 Pattern versus batch mode training
42(1)
5. Simulation results
43(7)
5.1 Example 1: 2-D Gaussian filter design
43(2)
5.2 Example 2: 1-Q Gaussian SDDF filter design
45(3)
5.3 Example 3: A(2,2) order system
48(2)
6. Concluding remarks
50(1)
References
51(2)
Chapter 3 Neural Networks for Control
R.J. Mitchell
1. Introduction
53(4)
1.1 Neural network principles
53(1)
1.2 History of neural networks
54(1)
1.3 Properties of neural networks
55(1)
1.4 Applications of neural networks
56(1)
1.5 Types of neural networks
57(1)
2. Feedforward networks
57(8)
2.1 Multi-layer perception networks
57(6)
2.2 Radial basis function networks
63(2)
2.3 Comparison of MLP and RBF
65(1)
3. Recurrent networks
65(6)
3.1 Hopfield network
66(2)
3.2 Continuous version of the Hopfield network
68(1)
3.3 Real-time recurrent networks
69(2)
4. Weightless networks
71(2)
5. Conclusion
73(1)
References
74(3)
Chapter 4 Neuro-Based Adaptive Regulator
T. Tsuji, B.H. Xu and M. Kaneko
1. Introduction
77(1)
2. Neuro-based adaptive regulator
78(6)
2.1 System formulation
78(1)
2.2 Quadratic optimal regulator for linearized system
79(1)
2.3 Derivation of compensatory input
80(2)
2.4 Neuro-based adaptive regulator
82(2)
3. Multi-layer neural network and learning
84(4)
4. Computer simulation of double cart-spring system
88(7)
4.1 Control performance
90(2)
4.2 Ability of learning and identification
92(2)
4.3 Nonlinear uncertainties
94(1)
5. Conclusion
95(1)
Appendix
96(1)
References
96(3)
Chapter 5 Local Model Networks and Self Tuning Predictive Control
P.J. Gawthrop and E. Ronco
1. Introduction
99(3)
2. Continuous-time local model networks (LMN)
102(1)
3. Continuous-time generalized predictive control (GPC)
103(2)
4. LMNGPC
105(1)
5. Self tuning LMNGPC
106(1)
6. An example
107(6)
7. Conclusions
113(1)
References
113(6)
PART II: FUZZY AND NEURO-FUZZY SYSTEMS IN MODELLING, CONTROL AND ROBOT PATH PLANNING
Chapter 6 An On-Line Self Constructing Fuzzy Modeling Architecture Based on Neural and Fuzzy Concepts and Techniques
S.G. Tzafestas and K.C. Zikidis
1. Introduction
119(3)
2. General issues on the proposed architecture
122(3)
3. The basic Takagi-Sugeno-Kang.model
125(1)
4. The fuzzy ART algorithm
126(2)
5. Analytical presentation of the proposed architecture
128(10)
5.1 Adaptive parameters
128(1)
5.2 Input variable transformation for efficient output calculation
129(1)
5.3 Basic parameters and performance indices
130(1)
5.4 The core of the algorithm
130(6)
5.5 Discussion
136(2)
6. Membership functions
138(4)
6.1 First membership function
139(1)
6.2 Second membership function
139(1)
6.3 Third membership function
139(2)
6.4 Fourth membership function
141(1)
6.5 Fifth membership function
141(1)
7. Simulation testing
142(11)
7.1 First example: Approximation of the F6 function
142(3)
7.2 Second example: Modeling of a static three-variable function
145(4)
7.3 Third example: Modeling of the Box and Jenkins gas furnace process
149(1)
7.4 Fourth example: Prediction of the Mackey-Glass time series
150(3)
8. Conclusion
153(12)
Appendix A: Parameter identification using the rule
154(9)
Appendix B: List of variables and parameters of the proposed system
163(2)
References
165(4)
Chapter 7 Neuro-Fuzzy Model Based Control
D. Matko, K. Kavsek Biasizzo and J. Kocijan
1. Introduction
169(1)
2. The neuro-fuzzy models
170(5)
2.1 Dynamic models
171(1)
2.2 Experimental modelling
172(3)
3. Model based control
175(4)
3.1 Model based cancellation control
175(1)
3.2 Model based predictive control
176(3)
4. Robustness issues of neuro-fuzzy model based control
179(1)
5. Fuzzy versus classical robust control of a nonlinear process
180(4)
6. Neuro-fuzzy model based control of a laboratory scale heat exchanger
184(7)
7. Conclusion
191(1)
References
191(2)
Chapter 8 Fuzzy and Neurofuzzy Approaches to Mobile Robot Path and Motion Planning Under Uncertainty
C.S. Tzafestas and S.G. Tzafestas
1. Introduction
193(1)
2. General issues on fuzzy and neurofuzzy reasoning
194(4)
3. Fuzzy and neurofuzzy robot path planning and navigation
198(7)
3.1 Fuzzy obstacle avoidance in robot manipulators
198(2)
3.2 Fuzzy path planning in mobile robots
200(3)
3.3 Neurofuzzy mobile robot navigation
203(2)
4. Mobile robot motion planning and control
205(6)
5. Some illustrative examples
211(5)
5.1 Example 1: Manipulator obstacle avoidance
211(1)
5.2 Example 2: Mobile-robot obstacle avoidance
211(2)
5.3 Example 3: Reinforcement learning-based local path planning
213(1)
5.4 Example 4: Mobile robot path tracking
214(1)
5.5 Example 5: Robust neurofuzzy motion control
215(1)
6. Conclusions
216(7)
PART III: GENETIC-EVOLUTIONARY ALGORITHMS
Chapter 9 A Tutorial Overview of Genetic Algorithms and Their Applications
S.G. Tzafestas, M.-P. Saltouros and M. Markaki
1. Genetic algorithms: A tutorial introduction
223(16)
1.1 what are genetic algorithms?
223(1)
1.2 How are genetic algorithms different from traditional methods?
224(2)
1.3 Natural evolution: The initial inspiration
226(1)
1.4 A top-level view of the genetic algorithm
227(1)
1.5 A simple genetic algorithm
228(5)
1.6 Genetic algorithms at work
233(3)
1.7 How do genetic algorithms work?
236(3)
2. Modifications to the simple GA
239(12)
2.1 Selection mechanisms and scaling
239(2)
2.2 Crossover mechanisms
241(1)
2.3 Mutation mechanisms
242(1)
2.4 The inversion operator
243(1)
2.5 Control parameters
243(1)
2.6 Encoding
244(1)
2.7 Genetic algorithms with varying population size
245(2)
2.8 Hybrid genetic algorithms
247(4)
3. Applications of genetic algorithms
251(38)
3.1 Genetic synthesis of neural network architecture
251(15)
3.2 Representing trees in genetic algorithms
266(8)
3.3 A genetic algorithm applied to robot trajectory generation
274(15)
4. Conclusions
289(4)
Appendix: Application of genetic algorithms
289(4)
References
293(8)
Chapter 10 Results from a Variety of Genetic Algorithm Applications Showing the Robustness of the Approach
W.D. Potter, S.M. Bhandarkar and D.J. D'Angelo
1. Introduction
301(3)
2. Multiple fault diagnosis (MFD)
304(3)
2.1 The CAP-II prototype
305(1)
2.2 Experiment results
306(1)
3. Network configuration (IDA-NET)
307(3)
3.1 Mobile subscriber equipment
308(1)
3.2 IDA-NET: The network expert module
309(1)
3.3 IDA-NET results
310(1)
4. Edge detection
310(7)
4.1 Cost function for an edge image
311(2)
4.2 Computation of the cost function
313(1)
4.3 Cost factors in the cost function
313(1)
4.4 GA edge detection
314(1)
4.5 Meta-level GA operators
314(1)
4.6 Experimental results
315(2)
5. Fish distributions
317(2)
5.1 GA-P implementation
317(1)
5.2 Experimental results
318(1)
6. Route fording
319(5)
6.1 Snakes and coils
320(2)
6.2 GA experimental setup
322(1)
6.3 Results
323(1)
7. Conclusions
324(1)
References
324(3)
Chapter 11 Evolutionary Algorithms in Computer-Aided Design of Integrated Circuits
R. Drechsler, N. Drechsler, B. Becker and H. Esbensen
1. Introduction
327(3)
2. Evolutionary algorithms (EAs) in CAD: An overview
330(2)
3. Example applications of EAs in CAD
332(6)
3.1 Logic synthesis
332(1)
3.2 Mapping
333(3)
3.3 Testing
336(2)
4. Performance evaluation
338(1)
5. Learning heuristics by EAs
339(7)
5.1 The model
340(2)
5.2 Application of the model
342(4)
6. Conclusions
346(1)
References
347(8)
PART IV: SOFT COMPUTING APPLICATIONS
Chapter 12 Soft Data Fusion
C.G. Looney and Y. Varol
1. Introduction
355(3)
1.1 Datafusion
355(1)
1.2 Adaptive and intelligent data fusion
356(1)
1.3 An example of the need for data fusion
357(1)
2. An approach to intelligent data fusion
358(1)
2.1 The requirements for intelligent fusion
358(1)
2.2 Low level data fusion
358(1)
2.3 Higher level fusion
359(1)
3. An intelligent soft fusion methodology
359(8)
3.1 An overview of the process
359(2)
3.2 Fuzzy merging for low level fusion
361(2)
3.3 Neural fusing techniques
363(2)
3.4 High level fuzzy fusion
365(2)
4. An illustrative example
367(8)
4.1 Overview
367(2)
4.2 Tracking
369(2)
4.3 Indexing and retrieval
371(1)
4.4 Decision making
371(1)
4.5 Frames
372(1)
4.6 Reasoning from cases
373(2)
5. Conclusions
375(1)
References
376(3)
Chapter 13 Application of Neural Networks to Computer Gaming
N. Baba
1. Introduction
379(1)
2. Gaming
380(3)
2.1 History of gaming
380(1)
2.2 COMMONS game
380(1)
2.3 Personal Computer gaming system of the COMMONS game
381(2)
3. Neural networks
383(4)
4. An application of neural network technology to the COMMONS game
387(3)
4.1 Neural network model and teacher signal
387(1)
4.2 Training algorithm
388(2)
5. Computer simulation
390(3)
6. Concluding remarks
393(2)
References
395(2)
Chapter 14 Coherent Neural Networks and Their Applications to Control and Signal Processing
A. Hirose
1. Introduction
397(1)
2. Coherent neural networks as brain-type information processing systems in the future
397(2)
3. Fundamentals of coherent neural networks
399(15)
3.1 Complex-valued neural networks
399(8)
3.2 Coherent neural networks
407(7)
4. Stability of coherent neural networks for time-sequential signal processing
414(4)
5. Applications to control and signal processing
418(3)
5.1 Signal processing using coherent neural networks
418(1)
5.2 Control using coherent neural networks
419(1)
5.3 An example: Waveform synthesis
419(2)
6. Conclusions
421(1)
References
421(2)
Chapter 15 Neural, Fuzzy and Evolutionary Reinforcement Learning Systems: An Application Case Study
D.A. Linkeras and H.O. Nyongesa
1. Introduction
423(2)
2. Fuzzy logic systems
425(2)
3. Neural systems review
427(1)
4. Neural fuzzy systems
428(3)
4.1 Knowledge representation with radial basis functions
429(1)
4.2 Knowledge acquisition and modification
430(1)
5. Evolutionary and genetic algorithms
431(1)
6. Evolutionary reinforcement neural fuzzy systems
432(3)
7. A control system application study
435(3)
8. Conclusions and outlook
438(3)
References
441(4)
Chapter 16 Neural Networks in Industrial and Environmental Applications
G.C. Smith and C.L. Wrobel
1. Introduction
445(1)
2. Neural techniques for industrial air emissions
446(7)
2.1 Selected neural applications in industrial air emissions
447(1)
2.2 Modeling a kraft recovery boiler using neural networks
448(5)
3. Neural techniques for contaminants in ambient air
453(6)
3.1 Selected neural applications in ambient air quality
454(1)
3.2 Neural networks and honey bees for air biomonitoring
455(4)
4. Neural techniques for aqueous contaminants
459(5)
4.1 Selected neural applications for aqueous contaminants
459(1)
4.2 Identifying aromatic hydrocarbon sources in ground water
460(4)
5. Conclusion
464(1)
References
465(2)
Biographies of the Contributors 467(10)
Index 477