Differential Evolution: From Theory to Practice

This document was uploaded by one of our users. The uploader already confirmed that they had the permission to publish it. If you are author/publisher or own the copyright of this documents, please report to us by using this DMCA report form.

Simply click on the Download Book button.

Yes, Book downloads on Ebookily are 100% Free.

Sometimes the book is free on Amazon As well, so go ahead and hit "Search on Amazon"

This book addresses and disseminates state-of-the-art research and development of differential evolution (DE) and its recent advances, such as the development of adaptive, self-adaptive and hybrid techniques. Differential evolution is a population-based meta-heuristic technique for global optimization capable of handling non-differentiable, non-linear and multi-modal objective functions. Many advances have been made recently in differential evolution, from theory to applications. This book comprises contributions which include theoretical developments in DE, performance comparisons of DE, hybrid DE approaches, parallel and distributed DE for multi-objective optimization, software implementations, and real-world applications. The book is useful for researchers, practitioners, and students in disciplines such as optimization, heuristics, operations research and natural computing.

Author(s): B. Vinoth Kumar, Diego Oliva, P. N. Suganthan
Series: Studies in Computational Intelligence, 1009
Publisher: Springer
Year: 2022

Language: English
Pages: 395
City: Cham

Preface
Contents
Editors and Contributors
Analysis of Structural Bias in Differential Evolution Configurations
1 Introduction
2 Differential Evolution
2.1 Working Mechanism and Nomenclature
2.2 Considered Differential Mutations
2.3 The Bin and Exp crossover operators
3 Structural Bias
3.1 Types of Structural Bias
4 Methodology
4.1 Visual Analysis
4.2 Statistical Tests
4.3 BIAS Toolbox
4.4 Experimental Setup
5 Results
5.1 Structural Bias in Parameter Space
5.2 Structural Bias in Time
6 Conclusions and Recommendations
References
Spherical Model of Population Dynamics in Differential Evolution
1 Introduction
2 Differential Evolution
2.1 Differential Mutation
2.2 Crossover
2.3 Selection
2.4 Theoretical Results
3 Spherical Model of DE Dynamics
3.1 Spherical Objective Function
3.2 Population Dynamics Model
4 Discussion
4.1 Adequacy of the Assumptions and Limitations
4.2 Simulation-Based Example
4.3 Comparison with Gaussian Model ch2Zhang2007AnApproximate,ch2Zhang2009c
5 Conclusion
References
Reinforcement Learning-Based Differential Evolution for Global Optimization
1 Introduction
2 Basic Information
2.1 Concepts of Reinforcement Learning
2.2 Algorithms in the Study
3 The Proposed RL-SHADE
4 Experiments and Results
4.1 Algorithm Setup
4.2 Characteristics of the Benchmark Suites
4.3 Measures
4.4 Results
4.5 Discussion
5 Conclusion
References
Analytical Study on the Role of Scale Factor Parameter of Differential Evolution Algorithm on Its Convergence Nature
1 Introduction
2 Related Works
2.1 Review of Literature Surveys
2.2 Review of Parameter Control Mechanisms
2.3 Review of Parameter Analysis
2.4 Review of Parameter Analysis
3 Design and Implementation
4 Design of Experiments
5 Results and Discussions
5.1 Analyzing DE Variants with Varying F Values
5.2 Analyzing the Nature of Convergence of DE Variants
5.3 Categorical Analysis
5.4 Relational Analysis of F Values and NoC
6 Image Segmentation Using DE
6.1 General Working of DE
6.2 Candidate Representation
6.3 Mutation and Crossover
6.4 Fitness Evaluation and Selection
6.5 Results and Discussions
7 Conclusions
References
The Trap of Sisyphean Work in Differential Evolution and How to Avoid It
1 Introduction
1.1 Background
2 The Problem of Duplicates
2.1 Initial Experiment
3 Long-Term Memory Assistance for DE
4 Experiment
4.1 The Problem of Stopping Criterion
4.2 Benchmark Experiment
4.3 Static Economic Load Dispatch Problem
5 Discussion
5.1 How Common and How Big is It?
5.2 How to Detect It?
5.3 What are the Implications of Duplicates in DE?
5.4 How Can We Reduce the Impact of Duplicates on the Efficiency of DE?
6 Conclusion
References
Investigations on Distributed Differential Evolution Framework with Fault Tolerance Mechanisms
1 Introduction
2 Related Works
3 Differential Evolution (DE) Algorithm
4 Distributed Differential Evolution (dDE) Framework
5 Fault Tolerance
5.1 Single-node Failure
5.2 Communication Link Failure
5.3 Multi-node Failure
6 Design of Experiment
7 Results and Discussion
7.1 Case 1: Single-Node Failure
7.2 Case 2: Communication Link Failure
7.3 Case 3: Multi-node Failure
8 Conclusion
References
Differential Evolution for Water Management Problems
1 Introduction
2 Material and Method
2.1 Test Problem 1: Water Distribution Network Problem
2.2 Test Problem 2: Reservoir Operation Problem
2.3 Differential Evolution
3 Results and Discussion
3.1 Test Problem 1: Water Distribution Network Results
3.2 Test Problem 2: Reservoir Operation Problem
4 Conclusion
References
Sobol Sequence-based MOSaDE Algorithm for Multi-objective Design of Water Distribution Networks
1 Introduction
2 Materials and Methods
2.1 Optimization Model Formulation for Multi-objective Design of WDNs
2.2 Estimation of Mechanical Reliability of WDNs
2.3 MOSaDE Algorithm
2.4 Sobol Sequences
2.5 NSGA-II Algorithm
2.6 Performance Metrics
3 Application for Multi-objective Design of WDNs
3.1 Case Studies of WDNs
3.2 Performance Assessment of S-MOSaDE Algorithm
4 Summary and Concluding Remarks
References
A Comparative Study on Parameter Estimation of COVID Epidemiological Models Using Differential Evolution Algorithm
1 Introduction
2 Related Works
3 Proposed System
3.1 Model Description
4 Design of Experiments
5 Results and Discussions
5.1 Dataset
5.2 Training Time Comparison
5.3 Evaluating Model Performance
6 Conclusions
7 Future Works
References
Applications of Differential Evolution in Electric Power Systems
1 Introduction
2 Overview of DE Optimizer
2.1 Mutation
2.2 Crossover
2.3 Selection
3 Reactive Power Planning
3.1 Problem Objective
3.2 Constraints of Problem
3.3 DE-Based Reactive Power Planning
3.4 Applications
4 Congestion Management
4.1 Problem Objective
4.2 Constraints of Problem
4.3 DE-Based Congestion Management
4.4 Applications
5 Available Transfer Capability
5.1 Problem Objective
5.2 Constraints of Problem
5.3 DE-Based Available Transfer Capability
5.4 Applications
6 Economic Load Dispatch
6.1 Problem Objective
6.2 Constraints of Problem
6.3 DE-Based Economic Load Dispatch
6.4 Applications
7 Unit Commitment
7.1 Problem Objective
7.2 Constraints of Problem
7.3 DE-Based Unit Commitment
7.4 Applications
8 Optimal Power Flow
8.1 Problem Objective
8.2 Constraints of Problem
8.3 DE-Based Optimal Power Flow
8.4 Applications
9 Optimal Reactive Power Dispatch
9.1 Problem Objective
9.2 Constraints of Problem
9.3 DE-Based Optimal Reactive Power Dispatch
9.4 Applications
10 Discussion
11 Conclusion
12 Summary
References
Detection of Heavy Sandstorm Regions Using Composite Differential Evolution Algorithm
1 Introduction
1.1 Background Study
2 Classical Differential Evolution
3 Composite Differential Evolution
4 Experimental Settings
5 Image Segmentation Technique
6 Multi-level Thresholding with CODE Approach
7 Test Results on Image Segmentation
8 Conclusions
Appendix
References
A Hybrid Artificial Differential Evolution Gorilla Troops Optimizer for High-Dimensional Optimization Problems
1 Introduction
2 Theory and Techniques
2.1 Differential Evolution
2.2 Artificial Gorilla Troops Optimizer
2.3 Artificial Differential Evolution Gorilla Troops Optimizer
3 Experimental Setup
3.1 Benchmark Functions
3.2 Parameter Settings for Algorithms
4 Experimental Results
4.1 The F and CR Parameters Experiments of DE
4.2 The F and CR Parameters Experiments of ADEGTO
4.3 Random F Parameter and Constant CR Parameter Experiments of ADEGTO
4.4 Comparisons of ADEGTO with State-of-Art Algorithms
5 Conclusions
References
Multi-objective Adaptive Guided Differential Evolution for Multi-objective Optimal Power Flow Incorporating Wind-Solar-Small Hydro-Tidal Energy Sources
1 Introduction
1.1 Literature Review
2 Method
3 Definition of the MOOPF Problem Regarding Thermal Power Units and RESs
3.1 Cost Models of All Energy Generating Sources
3.2 Objective Functions
3.3 Constraints of the Problem
4 Stochastic Power Models of RESs
5 Application of the MOAGDE in Solving MOOPF Involving RESs
6 Conclusions
References
Applications and Performance of Fuzzy Differential Evolution (DEFIS) in CFD Modeling of Heat and Mass Transfer
1 Introduction
1.1 Background
1.2 Purpose and Goal
1.3 Current Solutions
2 Theory and Techniques
2.1 Differential Evolution
2.2 Population Structure
2.3 Initialization
2.4 Transmutation
2.5 Transmission
2.6 Selection
2.7 Fuzzy Inference System (FIS)
3 DEFIS Applications in CFD Modeling
4 Conclusions
References