摘要
为解决高维多目标优化问题求解过程中存在的收敛性和多样性不平衡的问题,提出一种基于权值策略分类的高维多目标进化算法.首先,基于分解思想设计权值策略将种群分类,以保证解的多样性.其次,设计一种支配性排序方法,通过快速选出Pareto非支配个体,加快算法收敛速度.最后,采用WFG多目标测试函数集进行实验,并与MOEA/D、PeEA和Ma OEAIT算法进行比较.结果表明,所提出的算法在多样性和收敛性上有较为明显的改善.
In order to solve the imbalance between convergence and diversity in the solution process of many-objective optimization problems,a many-objective evolutionary algorithm based on classification of weighted strategy is proposed.First,a weighted strategy is designed to categorize the population based on the decomposition idea to ensure the diversity of solutions.Second,a dominance ordering method is designed to quickly select Pareto non-dominated individuals and speed up the convergence of the algorithm.Finally,the WFG multi-objective benchmark problem is used for experiments and compared with MOEA/D,PeEA and MaOEAIT algorithms,and the results show that the algorithm is able to improve the diversity and convergence significantly.
作者
李文彬
徐运昇
王培阳
LI Wenbin;XU Yunsheng;WANG Peiyang(School of Information Science and Engineering,Hunan Institute of Science and Technology,Yueyang 414006,China)
出处
《湖南理工学院学报(自然科学版)》
2025年第1期10-13,共4页
Journal of Hunan Institute of Science and Technology(Natural Sciences)
基金
湖南省自然科学基金项目(2024JJ7207,2024JJ7208,2024JJ7213)。
关键词
高维多目标优化
进化算法
权值策略
many-objective optimization
evolutionary algorithm
weighted strategy
作者简介
李文彬,男,博士,副教授.主要研究方向:进化算法、多目标优化;通信作者:徐运昇,男,硕士研究生.主要研究方向:进化算法、多目标优化。
