Genetic Algorithms in Engineering Systems

Genetic Algorithms in Engineering Systems

by A. M. Zalzala
     
 

ISBN-10: 0852969023

ISBN-13: 9780852969021

Pub. Date: 02/01/1999

Publisher: Institution of Engineering and Technology (IET)

The contributions presented in this book are extended version of commissioned papers from some of the highest quality contributions to the conference. Chosen for their experience in the field, the authors are drawn from academia and industry worldwide. The chapters cover the main fields of work as well as presenting tutorial material in this important subject, which

Overview

The contributions presented in this book are extended version of commissioned papers from some of the highest quality contributions to the conference. Chosen for their experience in the field, the authors are drawn from academia and industry worldwide. The chapters cover the main fields of work as well as presenting tutorial material in this important subject, which is currently receiving considerable attention from engineers.

Product Details

ISBN-13:
9780852969021
Publisher:
Institution of Engineering and Technology (IET)
Publication date:
02/01/1999
Series:
Control Engineering Series
Pages:
280
Product dimensions:
6.12(w) x 9.25(h) x (d)

Table of Contents

1Introduction to genetic algorithms1
1.1What are genetic algorithms?3
1.1.1Overview of GAs3
1.1.2GAs versus traditional methods5
1.2Major elements of the GA5
1.2.1Population representation and initialisation6
1.2.2The objective and fitness functions8
1.2.3Selection9
1.2.3.1Roulette wheel selection methods10
1.2.3.2Stochastic universal sampling12
1.2.4Crossover (recombination)12
1.2.4.1Multipoint crossover13
1.2.4.2Uniform crossover13
1.2.4.3Other crossover operators14
1.2.4.4Intermediate recombination14
1.2.4.5Line recombination15
1.2.4.6Discussion15
1.2.5Mutation16
1.2.6Reinsertion17
1.2.7Termination of the GA18
1.3Other evolutionary algorithms19
1.4Parallel GAs20
1.4.1Global GAs21
1.4.2Migration GAs22
1.4.3Diffusion GAs26
1.5GAs for engineering systems30
1.6Example application: gas turbine engine control33
1.6.1Problem specification34
1.6.2EA implementation36
1.6.3Results37
1.6.4Discussion40
1.7Concluding remarks40
1.8References41
2Levels of evolution for control systems46
2.1Introduction46
2.1.1Evolutionary algorithms46
2.1.2Control system applications48
2.1.3Overview49
2.2Evolutionary learning: parameters49
2.3Evolutionary learning: data structures51
2.4Evolutionary learning: program level52
2.4.1Knowledge representation54
2.4.2Rule strength54
2.4.3Mutation operators55
2.4.4Crossover in SAMUEL55
2.4.5Control applications of SAMUEL56
2.5Evolutionary algorithms for testing intelligent control systems57
2.6Summary60
2.7Acknowledgment60
2.8References60
3Multiobjective genetic algorithms63
3.1Multiobjective optimisation and preference articulation64
3.2How do MOGAs differ from simple GAs?64
3.2.1Scale-independent decision strategies65
3.2.2Cost to fitness mapping and selection67
3.2.3Sharing67
3.2.4Mating restriction70
3.2.5Interactive optimisation and changing environments70
3.3Putting it all together70
3.4Experimental results73
3.5Concluding remarks74
3.6Acknowledgment76
3.7References76
4Constraint resolutions in genetic algorithms79
4.1Introduction79
4.2Constraint resolution in genetic algorithms79
4.3Problems in encoding of constraints82
4.4Fuzzy encoding of contraints83
4.5Fuzzy logic84
4.5.1Membership84
4.5.2Rules86
4.5.3Defuzzification87
4.5.4Example87
4.5.5Advantages of fuzzy logic89
4.5.6Uses of fuzzy logic90
4.6Fuzzy logic to resolve constraints in genetic algorithms90
4.7Engineering applications of the technique [9]95
4.8Discussion97
4.9Acknowledgments98
4.10References98
5Towards the evolution of scaleable neural architectures99
5.1Introduction99
5.2Encoding neural networks in chromosomes100
5.3Evolutionary algorithms103
5.4Active weights and the simulation of neural networks105
5.5A set based chromosome structure107
5.5.1Set interconnections108
5.5.2Example chromosome108
5.5.3Results111
5.5.4Scaleability112
5.6Conclusions113
5.7Acknowledgment114
5.8References114
6Chaotic systems identification118
6.1Background119
6.1.1Chua's oscillator119
6.1.2Synchronisation of nonlinear systems121
6.1.3Genetic algorithms123
6.2Synchronisation-based identification124
6.2.1Description of the algorithm124
6.2.2Identification of Chua's oscillator126
6.3Experimental examples127
6.4Conclusions131
6.5References132
7Job shop scheduling134
7.1Introduction134
7.2Disjunctive graph135
7.2.1Active schedules137
7.3Binary representation138
7.3.1Local harmonisation139
7.3.2Global harmonisation140
7.3.3Forcing140
7.4Permutaion representation141
7.4.1Subsequence exchange crossover141
7.4.2Permuation with repetition142
7.5Heuristic crossover143
7.5.1GT crossover144
7.6Genetic enumeration145
7.6.1Priority rule based GA145
7.6.2Shifting bottleneck based GA146
7.7Genetic local search147
7.7.1Neighbourhood search147
7.7.2Multistep crossover fusion148
7.7.3Neighbourhood structures for the JSSP150
7.7.4Scheduling in the reversed order152
7.7.5MSXF-GA for the job shop scheduling154
7.8Benchmark problems155
7.8.1Muth and Thompson benchmark155
7.8.2The ten tough benchmark problems156
7.9Other heuristic methods158
7.10Conclusions158
7.11References158
Evolutionary algorithms for robotic systems: principles and implementations161
8.1Optimal motion of industrial robot arms162
8.1.1Formulation of the problem163
8.1.2Simulation of case studies165
8.1.2.1A two DOF arm165
8.1.2.2A six DOF arm167
8.1.3Parallel genetic algorithms169
8.2A comparative study of the optimisation of cubic polynomial robot motion170
8.2.1Background170
8.2.2Motion based on cubic splines171
8.2.3The genetic formulations171
8.2.4The objective functions172
8.2.4.1Pareto-based GA172
8.2.4.2Weighted-sum GA172
8.2.5Parameter initialisation173
8.2.6Evaluating the population174
8.2.6.1Ranking174
8.2.6.2Fitness assignment174
8.2.6.3Sharing scheme175
8.2.7Selection scheme175
8.2.8Shuffling175
8.2.9Recombination mechanisms175
8.2.10Modified feasable solution converter176
8.2.11Time intervals mutation177
8.2.12Simulation results177
8.2.12.1Case 1: Pareto-based GA178
8.2.12.2Case 2: Pareto-GA versus flexible polyhedron search180
8.2.12.3Case 3: weighted-sum GA180
8.3Multiple manipulator systems182
8.3.1Problem formulation183
8.3.2Encoding of paths as strings184
8.3.3Fitness function184
8.3.4The GA operators186
8.3.5Simulation results for two 3DOF arms187
8.4Mobile manipulator system with nonholonomic constraints190
8.4.1Multicriteria cost function191
8.4.2Parameter encoding using polynomials192
8.4.3Fitness function193
8.4.4Genetic evolution193
8.4.5Simulation results194
8.5Discussions and conclusions195
8.6Acknowledgment197
8.7References198
8.8Appendix199
8.8.1Motion based on cubic splines199
8.8.2Physical limits201
8.8.3The feasable solution converter (time scaling)202
9Aerodynamic inverse optimisation problems203
9.1Direct optimisation of airfoil206
9.1.1Approximation concept206
9.1.2Results of direct optimisation206
9.2Inverse optimisation of the airfoil210
9.2.1Coding210
9.2.2Simple GA with real number coding212
9.2.3Fitness evaluation: objective and constraints213
9.2.4Construction of fitness function214
9.2.5Inverse design cycle215
9.2.6Results of airfoil design217
9.3Inverse optimisation of the wing218
9.3.1Pressure distribution for the wing219
9.3.2MOGA220
9.3.4Results of wing design221
9.4Summary225
9.5References226
10Genetic design of VLSI layouts229
10.1Introduction229
10.2Physical VLSI design230
10.2.1Macro cell layouts231
10.2.2Placement233
10.2.3Routing233
10.2.4Previous genetic approaches235
10.3A GA for combined placement and routing236
10.3.1The genotype representation237
10.3.2Floorplanning238
10.3.3Integration of routing239
10.3.4Computation of the global routes239
10.3.5Hybrid creation of the initial population241
10.3.6Crossover242
10.3.7Mutation242
10.3.8Selection245
10.4Results245
10.5Conclusions249
10.6Acknowledgments251
10.7References252
Index254

Customer Reviews

Average Review:

Write a Review

and post it to your social network

     

Most Helpful Customer Reviews

See all customer reviews >