Constrained multi-objective evolutionary algorithm based on the correlation between objectives and constraints

doi:10.1016/j.swevo.2025.101903

Constrained multi-objective evolutionary algorithm based on the correlation between objectives and constraints

2025-03-01

0

PRE

AI

DOI：10.1016/j.swevo.2025.101903

原文链接

原文求助

摘要

En 中文

Many engineering optimization problems require simultaneous optimization of multiple objective functions under certain constraints, which are collectively referred to as constrained multi-objective problems (CMOPs). The crucial issue in solving CMOPs is to balance constraints and objectives. This paper proposes a constrained multi-objective evolutionary algorithm based on the correlation between objectives and constraints, termed CORCMO. CORCMO mainly comprises two stages: the learning stage and the evolving stage. The learning stage focuses on analyzing the correlation between each objective and constraints. In the evolving stage, the CMOP is decomposed into M constraint single-objective problems, which are optimized by M subpopulations cooperatively. For each subproblem, the corresponding fitness function, computed based on the correlation, is adopted to guide the evolution. Subsequently, CORCMO employs archive population update strategy to find the optimal solutions of the given CMOP. Experiments conducted on a series of benchmark problems demonstrate that CORCMO is promising to solve CMOPs.

Keyword:

Correlation

Constrained multi-objective optimization

Decomposition

Archive population

Constrained multi-objective evolutionary algorithm based on the correlation between objectives and constraints

摘要

期刊

机构

引用论文