A Pareto front estimation-based constrained multi-objective evolutionary algorithm

The balance of convergence, diversity, and feasibility plays a pivotal role in constrained multi-objective optimization problems. To address this issue, in this paper a novel method named PeCMOEA is proposed, in which the pivotal solutions, which are designed for estimating the constrained Pareto fr...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Applied intelligence (Dordrecht, Netherlands) Jg. 53; H. 9; S. 10380 - 10416
Hauptverfasser: Cao, Jie, Yan, Zesen, Chen, Zuohan, Zhang, Jianlin
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Springer US 01.05.2023
Springer Nature B.V
Schlagworte:
Artificial Intelligence
Computer Science
Convergence
Estimation
Evolutionary algorithms
Feasibility
Fitness
Genetic algorithms
Machines
Manufacturing
Mechanical Engineering
Methods
Multiple objective analysis
Optimization
Pareto optimization
Penalty function
Populations
Processes
Constrained multi-objective optimization
Multiple-populations
Constrained multi-objective evolutionary algorithms
Pareto front curvature
ISSN:0924-669X, 1573-7497
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The balance of convergence, diversity, and feasibility plays a pivotal role in constrained multi-objective optimization problems. To address this issue, in this paper a novel method named PeCMOEA is proposed, in which the pivotal solutions, which are designed for estimating the constrained Pareto front, are identified through an achievement scalarizing function. In addition, two different adaptive fitness functions are formulated to evaluate convergence- and diversity-oriented populations, respectively. Finally, the promising solutions from the two populations are reserved by their fitness values in the environmental selection while a self-adaptive penalty function is designed to repair infeasible solutions and ensure their feasibility. The performance of PeCMOEA is compared with five state-of-the-art constrained multi-objective evolutionary algorithms on five test suites. The experimental results illustrate that PeCMOEA exhibits competitive performance when utilised for this family of problems.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0924-669X
1573-7497
DOI:10.1007/s10489-022-03990-7