Shi / Lim Handbook of Swarm Intelligence

Concepts, Principles and Applications
1. Auflage 2011
ISBN: 978-3-642-17390-5
Verlag: Springer
Format: PDF
Kopierschutz: Wasserzeichen (»Systemvoraussetzungen)

E-Book, Englisch, 544 Seiten

Reihe: Adaptation, Learning, and Optimization

From nature, we observe swarming behavior in the form of ant colonies, bird flocking, animal herding, honey bees, swarming of bacteria, and many more.  It is only in recent years that researchers have taken notice of such natural swarming systems as culmination of some form of innate collective intelligence, albeit swarm intelligence (SI) - a metaphor that inspires a myriad of computational problem-solving techniques.  In computational intelligence, swarm-like algorithms have been successfully applied to solve many real-world problems in engineering and sciences. This handbook volume serves as a useful foundational as well as consolidatory state-of-art collection of articles in the field from various researchers around the globe.  It has a rich collection of contributions pertaining to the theoretical and empirical study of single and multi-objective variants of swarm intelligence based algorithms like particle swarm optimization (PSO), ant colony optimization (ACO), bacterial foraging optimization algorithm (BFOA), honey bee social foraging algorithms, and harmony search (HS).  With chapters describing various applications of SI techniques in real-world engineering problems, this handbook can be a valuable resource for researchers and practitioners, giving an in-depth flavor of what SI is capable of achieving.

Autoren/Hrsg.

Shi, Yuhui

Lim, Meng-Hiot

Inhaltsverzeichnis

1;Preface;6
2;Contents;9
3;Part A Particle Swarm Optimization;12
3.1;From Theory to Practice in Particle Swarm Optimization;13
3.1.1;Introduction;13
3.1.2;Standard PSO in Short;14
3.1.3;When Nearer Is Better;17
3.1.3.1;Motivation;17
3.1.3.2;Notations and Definitions;17
3.1.3.3;The Nearer Is Better (NisB) Class;18
3.1.4;NisBPSOs;21
3.1.4.1;Pure NisB PSOs;21
3.1.4.2;Gamma NisB PSO;22
3.1.5;The Mythical Balance, or When PSO Does Not Exploit;23
3.1.5.1;A Ritual Claim;23
3.1.5.2;A Bit of Theory;23
3.1.5.3;Checking the Exploitation Rate;24
3.1.6;Balanced PSO: A Tool to Design a Specific Algorithm5;24
3.1.6.1;My Algorithm Is Better than Yours;26
3.1.6.2;A Modular Tool;27
3.1.6.3;Step 1: A Small “Representative” Benchmark;29
3.1.6.4;Step 2: A Specific Improvement Method;29
3.1.6.5;Step 3: Some General Improvement Options;30
3.1.6.6;Success Rate vs. “Search Effort”;30
3.1.7;Conclusion;32
3.1.8;References;32
3.2;What Makes Particle Swarm Optimization a Very Interesting and Powerful Algorithm?;47
3.2.1;Introduction;48
3.2.2;How Physics Can Help Heuristics: The Continuous PSO Model;52
3.2.2.1;Stochatic Analysis of the PSO Continuous Model;53
3.2.2.2;The Oscillation Center Dynamics;55
3.2.3;Other Family Members;57
3.2.4;Methodology for Stochastic Analysis of the Discrete Trajectories;61
3.2.4.1;The GPSO Second Order Trajectories;63
3.2.5;Does GPSO Converge towards the Continuous PSO?;66
3.2.5.1;The Second Order Homogeneous Trajectories;67
3.2.5.2;The Transient Trajectories;67
3.2.6;Tuning of the PSO Parameters;69
3.2.7;How to Select the PSO Member: Application to Inverse Problems;71
3.2.8;Conclusions;72
3.2.9;References;73
3.3;Developing Niching Algorithms in Particle Swarm Optimization;76
3.3.1;Introduction;76
3.3.2;Niching Methods;77
3.3.2.1;Difficulties Facing Niching Methods;78
3.3.3;Particle Swarm Optimization;80
3.3.3.1;PSO Niching Methods;80
3.3.4;New Niching Methods Using a lbest PSO;86
3.3.4.1;Memory-Swarm vs. Explorer-Swarm;87
3.3.4.2;lbest PSO Using a Ring Topology;88
3.3.4.3;Numerical Examples;90
3.3.4.4;Results and Discussion;91
3.3.4.5;Effect of Varying Population Size;93
3.3.5;Conclusions;94
3.3.6;References;94
3.4;Test Function Generators for Assessing the Performance of PSO Algorithms in Multimodal Optimization;98
3.4.1;Introduction;98
3.4.2;The Particle Swarm Optimization Algorithm;99
3.4.3;Linear Transformations and Homogeneous Coordinates;101
3.4.4;Function Composition;103
3.4.5;Test Functions Commonly Adopted;104
3.4.6;Generating Test Functions;109
3.4.6.1;Using Linear Transformations;109
3.4.6.2;Using Function Composition;114
3.4.7;Overview of Test Function Generators;120
3.4.8;Guidelines to Build a Test Function Generator;123
3.4.9;Conclusions;124
3.4.10;Future Work;124
3.4.11;References;125
3.5;Linkage Sensitive Particle Swarm Optimization;127
3.5.1;Introduction;127
3.5.2;Determining Linkages;129
3.5.3;Linkage-Sensitive Particle Swarm Optimization;131
3.5.4;Results;133
3.5.4.1;Ten-Dimensional Functions;134
3.5.4.2;Thirty-Dimensional Functions;135
3.5.4.3;Complexity Analysis;136
3.5.4.4;Linkage Learning Error;137
3.5.5;Conclusions;138
3.5.6;References;139
3.6;Parallel Particle Swarm Optimization Algorithm Based on Graphic Processing Units;141
3.6.1;Introduction;141
3.6.2;Related Work;142
3.6.2.1;Original Particle Swarm Optimization;143
3.6.2.2;Standard Particle Swarm Optimization;143
3.6.3;GPU-Based Computing;144
3.6.3.1;ProgrammingModel for GPU;145
3.6.3.2;Applications of GPU Based Computing;145
3.6.3.3;Compute Unified Device Architecture (CUDA);146
3.6.4;Implementation of PSO on GPU;148
3.6.4.1;Data Organization;148
3.6.4.2;Variable Definition and Initialization;149
3.6.4.3;Random Number Generation;149
3.6.4.4;Algorithmic Flow for GPU-PSO;149
3.6.4.5;Methods of Parallelization;149
3.6.5;Experimental Results and Discussion;152
3.6.5.1;Running Time and Speedup versus Swarm Population;152
3.6.5.2;Running Time and Speedup versus Dimension;158
3.6.5.3;Other Characteristics of GPU-PSO;158
3.6.5.4;Comparison with RelatedWorks;160
3.6.6;Conclusions;160
3.6.7;References;161
3.7;Velocity Adaptation in Particle Swarm Optimization;163
3.7.1;Introduction;163
3.7.2;Particle Swarm Optimization;164
3.7.2.1;PSO for Bound-Constrained Problems;165
3.7.2.2;Adaptive Particle Swarm Optimization;166
3.7.3;Theoretical Observations;167
3.7.4;PSO with Velocity Adaptation;168
3.7.5;Experimental Results;170
3.7.5.1;Experiment 1: Comparison with a Standard PSO;171
3.7.5.2;Experiment 2: Success Probability;173
3.7.5.3;Experiment 3: Local Velocity Lengths;175
3.7.5.4;Experiment 4: Scaling Strategy;175
3.7.5.5;Experiment 5: Initialization of Velocity Length l;177
3.7.6;Conclusion;179
3.7.7;References;179
3.8;Integral-Controlled Particle Swarm Optimization;182
3.8.1;Introduction;182
3.8.2;Control Analysis for Standard Particle Swarm Optimization;184
3.8.3;Integral-Controlled Particle Swarm Optimization;184
3.8.3.1;Update Equations;185
3.8.3.2;Stability Analysis;186
3.8.3.3;Pseudo-codes of ICPSO;187
3.8.4;Integral Particle Swarm Optimization with Dispersed Accelerator Information;188
3.8.4.1;Predicted Accelerator Index;188
3.8.4.2;Phase Translation Principle of Cognitive Learning Factor;189
3.8.4.3;Time-Varying Social Learning Factor Adjustment Strategy;191
3.8.4.4;Mutation Strategy;191
3.8.4.5;Pseudo-codes of IPSO-DAI;191
3.8.4.6;Simulation Results;192
3.8.5;Conclusion;203
3.8.6;References;203
3.9;Particle Swarm Optimization for Markerless Full Body Motion Capture;207
3.9.1;Introduction;207
3.9.2;Background;209
3.9.2.1;The Problem;209
3.9.2.2;Related Work;210
3.9.3;Particle Swarm Optimization;211
3.9.3.1;Canonical PSO;211
3.9.3.2;PSO for Dynamic Optimization;212
3.9.4;Our Approach;213
3.9.4.1;Overview;213
3.9.4.2;The Articulated Body Model;213
3.9.4.3;Volume Reconstruction;214
3.9.4.4;PSO for Pose Initialization;216
3.9.4.5;PSO for Pose Estimation and Tracking;218
3.9.5;Experiment Results;219
3.9.5.1;Pose Initialization Results;221
3.9.5.2;Pose Estimation and Tracking Results;222
3.9.6;Conclusion;223
3.9.7;References;223
3.10;An Adaptive Multi-Objective Particle Swarm Optimization Algorithm with Constraint Handling;227
3.10.1;Introduction;227
3.10.2;Multi-Objective Optimization;229
3.10.3;Particle Swarm Optimization (PSO);229
3.10.4;Related Study;230
3.10.5;Adaptive Multi-Objective Particle Swarm Optimization: AMOPSO;232
3.10.6;CAMOPSO: Constraint Handling in AMOPSO;235
3.10.7;Experimental Results;236
3.10.7.1;Test Problems and Performance Measures;237
3.10.7.2;Results;237
3.10.8;Conclusions and Discussion;242
3.10.9;References;243
3.11;Multiobjective Particle Swarm Optimization for Optimal Power Flow Problem;246
3.11.1;Introduction;246
3.11.2;OPF Problem Formulation;249
3.11.2.1;Problem Objectives;249
3.11.2.2;Problem Constraints;251
3.11.2.3;Problem Formulation;252
3.11.3;Multiobjective Optimization;253
3.11.3.1;Principles and Definitions;253
3.11.3.2;Dominance and Pareto Optimal Solutions;253
3.11.4;Multiobjective Particle Swarm Optimization (Mopso);254
3.11.4.1;Overview;254
3.11.4.2;MOPSO Algorithm;255
3.11.5;MOPSO Implementation;257
3.11.5.1;Reducing Pareto Set by Clustering;257
3.11.5.2;Best Compromise Solution;257
3.11.5.3;MOPSO Computational Flow;258
3.11.5.4;Implementation;260
3.11.6;Results and Discussions;261
3.11.7;Conclusions;266
3.11.8;References;267
3.12;A Multi-objective Resource Assignment Problem in Product Driven Supply Chain Using Quantum Inspired Particle Swarm Algorithm;274
3.12.1;Introduction;274
3.12.2;Related Work;276
3.12.3;The Mathematical Model;277
3.12.3.1;Assignment of Operational Quality to Resources;278
3.12.3.2;List of Notations and Decision Variables;280
3.12.3.3;Proposed Multi-objective Optimization Model;283
3.12.4;Overview of Particle Swarm Methodology;286
3.12.5;The Proposed QPSO Metaheuristic;287
3.12.6;Results and Discussions;290
3.12.7;Conclusions and Next Steps;296
3.12.8;References;296
4;Part B Bee Colony Optimization;298
4.1;Honeybee Optimisation – An Overview and a New Bee Inspired Optimisation Scheme;299
4.1.1;Introduction;299
4.1.2;The Nature of the Honeybee;301
4.1.2.1;Mating;301
4.1.2.2;Foraging Behavior;301
4.1.2.3;Nest-Site Selection;302
4.1.3;Bee Inspired Algorithms;303
4.1.3.1;Mating-Based Optimisation Algorithms;303
4.1.3.2;Foraging Behaviour Based Approaches;305
4.1.4;The Optimisation Potential of Nest-Site Selection;312
4.1.4.1;A Spatial Nest-Site Selection Model;313
4.1.4.2;Experiments;318
4.1.5;The BNSSS Scheme;324
4.1.6;Conclusion;324
4.1.7;References;325
4.2;Parallel Approaches for the Artificial Bee Colony Algorithm;332
4.2.1;Introduction;332
4.2.2;Bee Foraging Behavior;334
4.2.3;Artificial Bee Colony Algorithm;334
4.2.4;Implementations of the ABC;335
4.2.4.1;Enhanced Sequential Model;336
4.2.4.2;Parallel Master-Slave Model;336
4.2.4.3;Parallel Multi-Hive Model;337
4.2.4.4;Parallel Hybrid Hierarchical Model;337
4.2.5;Experiments;338
4.2.6;Results and Analysis;341
4.2.7;Conclusions and Future Work;345
4.2.8;References;346
4.3;Bumble Bees Mating Optimization Algorithm for the Vehicle Routing Problem;349
4.3.1;Introduction;349
4.3.2;The Vehicle Routing Problem;351
4.3.3;Bumble Bees Mating Optimization Algorithm for the Vehicle Routing Problem;353
4.3.3.1;Bumble Bees Behavior;353
4.3.3.2;General Description of the Algorithm;354
4.3.3.3;Path Representation;357
4.3.3.4;Calculation of Fitness Function;358
4.3.3.5;Crossover Operator;358
4.3.3.6;Expanding Neighborhood Search;358
4.3.4;Computational Results;360
4.3.5;Conclusions;366
4.3.6;References;366
5;Part C Ant Colony Optimization;372
5.1;Ant Colony Optimization: Principle, Convergence and Application;373
5.1.1;Introduction;373
5.1.2;The Basic Principle of Ant Colony Optimization;374
5.1.3;Ant Colony Optimization Mathematical Model;375
5.1.4;Convergence Proof of Ant Colony Optimization;375
5.1.5;NLPID Controller;378
5.1.6;Grid-Based ACO Algorithm;380
5.1.7;Experimental Results;383
5.1.8;Conclusions;387
5.1.9;References;388
5.2;Optimization of Fuzzy Logic Controllers for Robotic Autonomous Systems with PSO and ACO;389
5.2.1;Introduction;389
5.2.2;S-ACO and gbest PSO Algorithms;390
5.2.2.1;S-ACO Algorithm;390
5.2.2.2;gbest PSO Algorithm;391
5.2.3;Problem Statement;393
5.2.4;Fuzzy Logic Control Design;394
5.2.5;Optimization of the Fuzzy Controllers;397
5.2.5.1;ACO Architecture for the FLC Optimization;397
5.2.5.2;PSO Architecture (gbest);399
5.2.5.3;PSO Parameters Handling;399
5.2.6;Simulation Results;400
5.2.6.1;S-ACO Results for the Optimization of the Type-1 FLC;400
5.2.6.2;S-ACO Results for the Optimization of the Type-2 FLC;402
5.2.6.3;gbest PSO Algorithm Results for Type-1 FLC Optimization;403
5.2.6.4;gbest PSO Algorithm Results for Type-2 FLC Optimization;407
5.2.7;Statistical Comparison;409
5.2.7.1;Statistical Comparison among Bio-inspired Methods for Type-1 FLCs;411
5.2.7.2;Statistical Comparison among Bio-inspired Methods for Type-2 FLCs;413
5.2.8;Conclusions;416
5.2.9;References;416
6;Part D Other Swarm Techniques;418
6.1;A New Framework for Optimization Based-On Hybrid Swarm Intelligence;419
6.1.1;Introduction;419
6.1.2;Optimization in Swarm Intelligence;420
6.1.2.1;Particle Swarm Optimization;420
6.1.2.2;Bacterial Foraging;423
6.1.2.3;Ant System and Ant Colony System;425
6.1.2.4;Artificial Bee Colony Optimization;428
6.1.2.5;Cat Swarm Optimization;430
6.1.2.6;Parallel Cat Swarm Optimization;432
6.1.3;The Hybrid Swarm Intelligence Based on PCSO and ABC;435
6.1.4;Experimental Results and Conclusions;436
6.1.5;References;445
6.2;Glowworm Swarm Optimization for Multimodal Search Spaces;448
6.2.1;Introduction;448
6.2.2;Multimodal Function Optimization;450
6.2.3;Basic Principle of GSO;451
6.2.4;Group Level Phases of GSO Algorithm;454
6.2.4.1;Splitting of the Agent-Swarm into Subgroups;454
6.2.4.2;Local Convergence of Agents in Each Subgroup to the Peak Locations;457
6.2.5;GSO Applications;459
6.2.5.1;Multimodal Function Optimization;459
6.2.5.2;Multiple Signal Source Localization;459
6.2.5.3;Pursuit of Mobile Signal Sources;460
6.2.6;Conclusions;462
6.2.7;References;463
6.3;Direct and Inverse Modeling of Plants Using Cat Swarm Optimization;465
6.3.1;Introduction;465
6.3.2;System Modeling;466
6.3.2.1;Direct Modeling;466
6.3.2.2;Inverse Modeling;467
6.3.3;Cat Swarm Optimization;468
6.3.3.1;Seeking Mode;468
6.3.3.2;Tracing Mode;469
6.3.3.3;Algorithm;469
6.3.4;Simulation Results;471
6.3.4.1;Case-1;471
6.3.4.2;Case-2;475
6.3.4.3;Case-3;477
6.3.4.4;Case-4;478
6.3.5;Conclusion;480
6.3.6;References;480
6.4;Parallel Bacterial Foraging Optimization;482
6.4.1;Introduction;482
6.4.2;Bacterial Foraging Optimization;483
6.4.2.1;Chemotaxis;483
6.4.2.2;Reproduction;484
6.4.2.3;Elimination and Dispersal;484
6.4.3;Parallel Bacterial Foraging Optimization (Pbfo);486
6.4.3.1;Chemotaxis;487
6.4.3.2;Mutation;487
6.4.3.3;Reproduction;488
6.4.4;Mutation by PSO Variants;490
6.4.4.1;Standard Particle Swarm Optimization;490
6.4.4.2;Modified Particle Swarm Optimization;491
6.4.4.3;Linearly Decreasing Weight PSO (LDW-PSO);491
6.4.4.4;Supervisor Student Model in Particle Swarm Optimization (SSM-PSO);492
6.4.4.5;Particle Swarm Optimization with Time Varying Acceleration Coefficients (PSOTVAC);492
6.4.4.6;Global Local Best Particle Swarm Optimization (GLBest PSO);492
6.4.4.7;Parameter Free PSO Algorithm (pf-PSO);493
6.4.4.8;Particle Swarm Optimization with Extrapolation Technique (e-PSO);493
6.4.5;Experimental Results and Discussions;493
6.4.6;Issues for Implementation of Parallel Bacterial Foraging Optimization;495
6.4.7;Conclusion;496
6.4.8;References;496
6.5;Reliability-Redundancy Optimization Using a Chaotic Differential Harmony Search Algorithm;498
6.5.1;Introduction;498
6.5.2;Fundamentals of Reliability-Redundancy Optimization;500
6.5.2.1;Example 1: Overspeed Protection System for a Gas Turbine;501
6.5.2.2;Example 2: Series-Parallel System;502
6.5.3;Optimization Algorithms;503
6.5.3.1;Harmony Search Algorithm;503
6.5.3.2;Proposed CHSDE Algorithm;505
6.5.4;Simulation Results;506
6.5.5;Conclusion and Future Research;509
6.5.6;References;510
6.6;Gene Regulatory Network Identification from Gene Expression Time Series Data Using Swarm Intelligence;512
6.6.1;Introduction;512
6.6.1.1;Preliminaries;513
6.6.1.2;The Problem;513
6.6.1.3;Review;514
6.6.1.4;Sectional Classification;515
6.6.2;The Proposed Model;515
6.6.3;Gene Expression Generation for Known Network;518
6.6.4;Formulation as an Optimization Problem;519
6.6.5;Solving the Optimisation Problem by Particle Swarm Optimization;520
6.6.5.1;Brief Overview on Particle Swarm Optimization;521
6.6.5.2;Solving the Gene Regulatory Network Identification by Particle Swarm Optimization;522
6.6.5.3;Topology Finding Particle Swarm Optimization;524
6.6.6;The Results;526
6.6.7;Identification of E.Coli. SOS DNA Repair Network;532
6.6.8;Conclusion;534
6.6.9;References;534
7;Author Index;538