Nature-Inspired Algorithms and Applied Optimization 1st Edition Xin-She Yang (Eds.)

Computational Optimization: An Overview

Introduction

Computational Optimization

Optimization Procedure

Optimizer

Optimization Algorithms

Choice of Algorithms

Simulator

Numerical Solvers

Simulation Efficiency

Latest Developments

References

Optimization Algorithms

Introduction

Derivative-Based Algorithms

Newton’s Method and Hill-Climbing

Conjugate Gradient Method

Derivative-Free Algorithms

Pattern Search

Trust-Region Method

Metaheuristic Algorithms

Simulated Annealling

Genetic Algorithms and Differential Evolution

Particle Swarm Optimization

Harmony Search

Firefly Algorithm

Cuckoo Search

A Unified Approach to Metaheuristics

Characteristics of Metaheuristics

Generalized Evolutionary Walk Algorithm (GEWA)

To Be Inspired or Not to Be Inspired

References

Surrogate-Based Methods

Introduction

Surrogate-Based Optimization

Surrogate Models

Design of Experiments

Surrogate Modeling Techniques

Model Validation

Surrogate Correction

Surrogate-Based Optimization Techniques

Approximation Model Management Optimization

Space Mapping

Manifold Mapping

Surrogate Management Framework

Exploitation versus Exploration

Final Remarks

References

Derivative-Free Optimization

Introduction

Derivative-Free Optimization

Local Optimization

Pattern Search Methods

Derivative-Free Optimization with Interpolation and Approximation Models

Global Optimization

Evolutionary Algorithms

Estimation of Distribution Algorithms

Particle Swarm Optimization

Differential Evolution

Guidelines for Generally Constrained Optimization

Penalty Functions

Augmented Lagrangian Method

Filter Method

Other Approaches

Concluding Remarks

References

Maximum Simulated Likelihood Estimation: Techniques and Applications in Economics

Introduction

Copula Model

Estimation Methodology

The CRT Method

Optimization Technique

Application

Concluding Remarks

References

Optimizing Complex Multi-location Inventory Models Using Particle Swarm Optimization

Introduction

RelatedWork

Simulation Optimization

Multi-Location Inventory Models with Lateral Transshipments

Features of a General Model

Features of the Simulation Model

Particle Swarm Optimization

Experimentation

System Setup

Results and Discussion

Conclusion and Future Work

References

Traditional and Hybrid Derivative-Free Optimization Approaches for Black Box Functions

Introduction and Motivation

A Motivating Example

Some Traditional Derivative-Free Optimization Methods

Genetic Algorithms (GAs)

Deterministic Sampling Methods

Statistical Emulation

Some DFO Hybrids

APPS-TGP

EAGLS

DIRECT-IFFCO

DIRECT-TGP

Summary and Conclusion

References

Simulation-Driven Design in Microwave Engineering: Methods

Introduction

Direct Approaches

Surrogate-Based Design Optimization

Surrogate Models for Microwave Engineering

Microwave Simulation-Driven Design Exploiting Physically-Based Surrogates

Space Mapping

Simulation-Based Tuning and Tuning Space Mapping

Shape-Preserving Response Prediction

Multi-fidelity Optimization Using Coarse-Discretization EM Models

Optimization Using Adaptively Adjusted Design Specifications

Summary

References

Variable-Fidelity Aerodynamic Shape Optimization

Introduction

Problem Formulation

Computational Fluid Dynamic Modeling

Governing Equations

Numerical Modeling

Direct Optimization

Gradient-Based Methods

Derivative-Free Methods

Surrogate-Based Optimization

The Concept

Surrogate Modeling

Optimization Techniques

Summary

References

Evolutionary Algorithms Applied to Multi-Objective Aerodynamic Shape Optimization

Introduction

Basic Concepts

Pareto Dominance

Pareto Optimality

Pareto Front

Multi-Objective Aerodynamic Shape Optimization

Problem Definition

Surrogate-Based Optimization

Hybrid MOEA Optimization

Robust Design Optimization

Multi-Disciplinary Design Optimization

Data Mining and Knowledge Extraction

A Case Study

Objective Functions

Geometry Parameterization

Constraints

Evolutionary Algorithm

Results

Conclusions and Final Remarks

References

An Enhanced Support Vector Machines Model for Classification and Rule Generation

Basic Concept of Classification and Support Vector Machines

Data Preprocessing

Data Cleaning

Data Transformation

Data Reduction

Parameter Determination of Support Vector Machines by Meta-heuristics

Genetic Algorithm

Immune Algorithm

Particle Swarm Optimization

Rule Extraction Form Support Vector Machines

The Proposed Enhanced SVM Model

A Numerical Example and Empirical Results

Conclusion

References

Benchmark Problems in Structural Optimization

Introduction to Benchmark Structural Design

Structural Engineering Design and Optimization

Classifications of Benchmarks

Design Benchmarks

Truss Design Problems

Non-truss Design Problems

Discussions and Further Research