An improvement decomposition-based multi-objective evolutionary algorithm using multi-search strategy

The main goal of multi-objective optimization evolutionary algorithms (MOEAs) is to obtain a set of solutions with good diversity and convergence. However, how to concurrently improve the diversity and convergence is a difficult work. To address this problem, an updated strategy based on decompositi...

Full description

Saved in:
Bibliographic Details
Published in:Knowledge-based systems Vol. 163; pp. 572 - 580
Main Authors: Dong, Ning, Dai, Cai
Format: Journal Article
Language:English
Published: Amsterdam Elsevier B.V 01.01.2019
Elsevier Science Ltd
Subjects:
Algorithms
Convergence
Crossovers
Decomposition
Evolutionary algorithms
Experiments
Exploitation
Genetic algorithms
Global local relationship
Multi-objective optimization
Multi-search strategy
Multiple objective analysis
Objectives
Operators
Optimization
Search methods
Search strategies
State of the art
Strategies
Strategy
Multi-search strategy
Multi-objective optimization
Decomposition
ISSN:0950-7051, 1872-7409
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The main goal of multi-objective optimization evolutionary algorithms (MOEAs) is to obtain a set of solutions with good diversity and convergence. However, how to concurrently improve the diversity and convergence is a difficult work. To address this problem, an updated strategy based on decomposition is used to maintain the diversity, and the convergence is enhanced by improving the search efficiency. In this paper, an improvement decomposition-based multi-objective evolutionary algorithm using multi-search strategy is aimed at improving the search efficiency. In this work, three search strategies are used to help crossover operators to carry out the local search and global search. This multi-search strategy selects sparse non-dominated solutions to carry out the exploration, and selects convergent solutions to implement the exploitation. In the experiments, the proposed algorithm is compared with seven efficient state-of-the-art algorithms, e.g., NSGAII, MOEA/D, MOEA-DVA, MOEA-IGD-NS, MOEA/D-PaS, RVEA and MPSOD, on twenty-two benchmark functions. Empirical results show that the proposed algorithm can find a set of solutions with better diversity and convergence than six compared algorithms.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0950-7051
1872-7409
DOI:10.1016/j.knosys.2018.09.018