APPLICATIONS OF SWARM INTELLIGENCE

APPLICATIONS OF SWARM INTELLIGENCE

CONTENTS

PREFACE

Chapter 1 SWARM INTELLIGENCE AND FUZZY SYSTEMS

Abstract

1. Optimizing the Parameters of Fuzzy Systems Using Swarm Intelligence Algorithms

1.1. Fuzzy Systems

1.1.1. Membership Functions

1.1.2. Fuzzy Rules

1.2. Designing a Fuzzy Classifier Using Particle Swarm Optimization Algorithm (PSO)

1.2.1. Integer-Valued Particle Swarm Optimization with Constriction Coefficient

1.2.2. Particle Representation

1.2.3. Fitness Function Definition

1.3. Experimental Results

1.4. Other Related Researches

2- Intelligently Controlling the Multi-objective Swarm Intelligence Parameters Using Fuzzy Systems

2.1. A Review on the Past Researches on Multi-objective PSO

2.2. Fuzzy-MOPSO Algorithm

2.2.1. Integer-Valued MOPSO with Constriction Coefficient

2.2.2. Designing Fuzzy-Controller for MOPSO

2.2.2.1. Metrics of Performance

a) Minimal spacing

b) Aggregation factor

2.2.2.2. Fuzzy Parameters

a) Inputs of fuzzy controller

b) Outputs of fuzzy controller

c) Fuzzy rules

Linguistic description on the effect of structural parameters of MOPSO on its search process

2.3. Space Allocation (Problem Description and Formulation)

2.4. Implementation and Experimental Results

2.4.1. Application on Well-Known Benchmarks

2.4.2. Application of Fuzzy-MOPSO on Space Allocation

a) Particle Representation

b) Experimental and Comparative Results

3. Conclusion

References

Chapter 2 EVOLUTIONARY STRATEGIES TO FIND PARETO FRONTS IN MULTIOBJECTIVE PROBLEMS

Abstract

1. Introduction

2. Pareto Optimality

3. Multi-objective Optimization with PSO

A1. Algorithm for MOPSO

4. Movement Strategies

4.1. Ms1: Pick a Global Guidance Located in the Least Crowded Areas

A2. Algorithm for Ms1

4.2. Ms2: Create the Perturbation with Differential Evolution Concept

A3. Algorithm for Ms2

4.3. Ms3: Search the Unexplored Space in the Non-Dominated Front

A4. Algorithm for Ms3

4.4. Ms4: Combination of Ms1 and Ms2

4.5. Ms5: Explore Solution Space with Mixed Particles

4.6. Ms6: Adaptive Weight Approach

5. Design of Experiments

6. Results and Discussions

7. Conclusions

Acknowledgment

References

Chapter 3 PARTICLE SWARM OPTIMIZATION APPLIED TO REAL-WORLD COMBINATORIAL PROBLEMS: THE CASE STUDY OF THE IN-CORE FUEL MANAGEMENT OPTIMIZATION

Abstract

1. Introduction

2. Particle Swarm Optimization

3. Models of Particle Swarm Optimization for Combinatorial Problems

4. Particle Swarm Optimization with Random Keys

4.1. Random Keys

4.2. Particle Swarm Optimization with Random Keys

5. Optimization of Real-World Problems: The Case Study of the in-Core Fuel Management Optimization

5.1. The Traveling Salesman Problem

5.2. The In-Core Fuel Management Optimization

5.2.1. A General Description of the ICFMO

5.2.2. Mathematical Formulation of the in-Core Fuel Management Optimization

5.2.2. Simulation of Angra 1 NPP with the Reactor Physics Code Recnod

5.2.2. PSORK Model for the ICFMO

6. Computational Experimental Results

6.1. Traveling Salesman Problem

6.2. In-Core Fuel Management Optimization

7. Discussion

7.1. Traveling Salesman Problem

7.2. In-Core Fuel Management Optimization

8. Conclusions

Acknowledgments

References

Chapter 4 SWARM INTELLIGENCE AND ARTIFICIAL NEURAL NETWORKS

Summary

1. Using Swarm Intelligence for Artificial Neural Networks Training and Structure Optimization

1.1. Artificial Neural Networks

1.2. Using Particle Swarm Optimization for Artificial Neural Networks Training and Structure Optimization

1.3. Using Ant Colony Optimization for Artificial Neural Networks Training

References

Chapter5SWARMINTELLIGENCEFORTHESELF-ASSEMBLYOFNEURALNETWORKS

Abstract

1.Introduction

2.Methods

2.1.Agents

2.2.Rules

2.3.Forces

2.3.1.ForcesGoverningCollective(“Swarm”)Movements

2.3.2.EnvironmentandRule-BasedForces

2.4.ExperimentalMethods

2.4.1.ComputationalExperiments

2.4.2.ImplementationDetails

2.4.3.ConnectivityMeasures

3.Results

3.1.SomatosensoryCortexModel

3.2.VisualCortexModel

3.3.RobustnessExperiments

4.Discussion

5.ConclusionsandFutureWork

Acknowledgment

References

Chapter 6 APPLICATION OF PARTICLE SWARM OPTIMIZATION METHOD TO INVERSE HEAT RADIATION PROBLEM

Abstract

1. Introduction

2. Principle of Algorithm

2.1. Hybrid Genetic Algorithm (HGA)

2.2. Particle Swarm Optimization (PSO)

3. Mathematical Formulation

3.1. Physical Model

3.2. Direct Problem

4. Results and Discussion

4.1. Inverse Analysis Procedure

4.2. Estimation of Wall Emissivities (Case 1, 2)

4.3. Simultaneous Estimation of an Absorption and a Scattering Coefficients (Case 3, 4)

4.4. Simultaneous Estimation of Emissivities, Absorption & Scattering Coefficients (Case 5, 6)

5. Conclusions

Acknowledgment

References

Chapter 7 ANT COLONY OPTIMIZATION FOR FUZZY SYSTEM PARAMETER OPTIMIZATION: FROM DISCRETE TO CONTINUOUS SPACE

Abstract

1. Introduction

2. Fuzzy System Parameter Optimization in Discrete and Continuous Spaces

3. Discrete Aco for Fuzzy System Parameter Optimization in Discrete Spcace

3.1. Basic Concept of Discrete Ant Colony Optimization (ACO)

3.2. Discrete ACO for FS Parameter Optimization

4. Continuous ACO for FS Parameter Optimization in Continuous Space

5. Simulations

6. Conclusion

References

Chapter 8 PARTICLE SWARM OPTIMIZATION: A SURVEY

Abstract

1. Introduction

2. Particle Swarm Optimization (PSO)

2.1. Conventional PSO (Original PSO)

2.2. Basic PSO

2.3. Parameter Adjustment

2.3.1. Acceleration Coefficients

2.3.2. Maximum and Minimum Velocity (vmax and vmin)

2.3.3. Inertia Weight (w)

2.3.4. Personal-Best and Global-Best

2.4. Neighbourhood Topology

3. PSO vs. GA

4. Conventional Weaknesses of PSO

5. Solutions and Proposed Modifications

6. Discussion and Conclusion

References

Chapter 9 APPLICATION OF PSO TO ELECTROMAGNETIC AND RADAR-RELATED PROBLEMS IN NON COOPERATIVE TARGET IDENTIFICATION

Abstract

1. Introduction

2. PSO Applied to Direction of Arrival Estimation

2.1. Problem Formulation

2.2. Application of PSO to DOA Estimation

2.3. The Fitness Function and the Solution Space Limits

2.4. Performance Analysis

2.4.1. Convergence Study

2.4.2. Angular Errors, Fitness Errors and Number of Iterations

2.4.3. Robustness against Noise

2.4.4. Resolution

2.4.5. Dependence of the Incoming Angle

3. Complex Dielectric Constant Estimation by PSO

3.1. Context

3.2. Introduction

3.3. Problem Formulation

3.4. Results

4. Conclusion

Formulas

References

Chapter 10 ANT COLONY OPTIMIZATION: A POWERFUL STRATEGY FOR BIOMARKER FEATURE SELECTION

Abstract

Introduction

Conclusion

Acknowledgments

References

Chapter 11 SWARM INTELLIGENCE BASED ANONYMOUS AUTHENTICATION PROTOCOL FOR DYNAMIC GROUP MANAGEMENT IN EHRM SYSTEM

Abstract

1. Introduction

2. Model of the System

3. BXAAP (Boolean Expression Anonymous Authentication Protocol)

BXAAP Protocol Model

Registration

Key Distribution

Verification

4. Ant Colony Optimization

5. Ant Colony Optimization Based Boolean Function Minimization

5.1. Model of an Ant System

6. ABXE Algorithm-Construction and Design

6.1. Pheromone Deposition of the Ant Agent

6.2. Assignment of Energy Value

6.3. Computation of Energy Value for a Large Number of Users in a Group

6.4. Algorithm: Ant Colony Optimized Boolean Expression Evolver

7. Experimental Results

8. Analysis

8.1. Security Analysis

8.2. Protocol Analysis

9. Comparison with Existing Group Rekeying Methods

9.1. Comparison of BXE Algorithm with ABXE Algorithm

Conclusion

Acknowledgment

References

INDEX

Blank Page