基于协同进化的多目标约束进化算法

吉林大学学报(信息科学版) ›› 2023, Vol. 41 ›› Issue (2): 321-328.

基于协同进化的多目标约束进化算法

刘仁云^1a, 张旭^1a , 姚亦飞^1b , 于繁华²

(1. 长春师范大学 a. 数学学院; b. 计算机科学与技术学院, 长春 130032; 2. 北华大学计算机科学与技术学院, 吉林吉林 132013)

收稿日期:2022-03-17 出版日期:2023-04-13 发布日期:2023-04-16
作者简介: 刘仁云(1968— ), 女, 辽宁大连人, 长春师范大学教授, 博士, 主要从事最优化理论及应用研究, (Tel)86-13174400345 (E-mail)liurenyun@ ccsfu. edu. cn
基金资助:
吉林省科技厅基金资助项目(20200201276JC); 吉林省教育厅基金资助项目(20200822KJ)

Multi-Objective Constrained Evolutionary Algorithm Based on Coevolution

LIU Renyun ^1a, ZHANG Xu ^1a , YAO Yifei ^1b , YU Fanhua ²

(1a. College of Mathematics; 1b. College of Computer Science and Technology, Changchun Normal University, Changchun 130032, China; 2. College of Computer Science and Technology, Beihua University, Jilin 132013, China)

Received:2022-03-17 Online:2023-04-13 Published:2023-04-16

摘要/Abstract

摘要： 针对约束多目标优化算法(COA: Constrained Optimization Algorithms)中存在的难以有效兼顾收敛性和多样性的问题, 提出了采用协同进化策略的多目标优化算法(CoMaC)。首先, 将一个 COA 转化为一个带动态约束处理的多目标进化算法。然后采用差分进化(DE: Differential Evolution)生成第 1 种群, 并将其中的已知可行解选入第 2 种群, 并与第 1 种群协同进化。第 1 种群通过保持原约束条件的全局搜索加快收敛。第 2 种群通过局部搜索进化, 保持并获得更多可行解。最后采用标准约束多目标测试函数进行实验, 以测试所提出算法的性能。实验结果表明, 与使用惩罚函数处理约束问题( PF: Penalty Function) 和使用动态处理约束边界方法 (DCMaOP: Dynamic Constrained Many Objective optimization Problem) 相比, 所提算法在反向世代距离( IGD: Inverted Generational Distance)和超体积(HV: Hypervolume)两个指标上均取得了良好的结果, 说明所提算法可以有效地兼顾收敛性和多样性。

关键词: 多目标进化算法, 动态约束处理, 协同进化, 全局搜索

Abstract: A CoMaCOA(Co-evolution Multi-Objective Constrained) optimization algorithm is proposed to deal with the problem that it cannot be combined convergence and diversity effectively in multi-objective COA (Constrained Optimization Algorithms). First, a COA is transformed into the multi-objective evolutionary algorithm with dynamic constraint processing. Then, DE(Differential Evolution) is used to generate the first population. The second population is generated by the known feasible solution in the first population and coevolved with the first. The first population accelerates convergence by global search that does not deal with constraints. The second population evolves through local search to maintain and obtain more feasible solutions. Finally, the standard constrained multi-objective test function is used for experiments in order to test the performance of the proposed algorithm. The experiment result shows that the proposed algorithm achieves good results on both IGD( Inverted Generational Distance) and HV( Hypervolume), comparing with PF ( Penalty Function) method and dynamic boundary processing to constrain problem DCMaOP(Dynamic Constrained Many Objective optimization Problem). It shows that the algorithm is both effective in convergence and diversity.

Key words: multi-objective evolutionary algorithm')">

multi-objective evolutionary algorithm, dynamic constraint processing, co-evolution, global search

中图分类号:

TP391

刘仁云 , 张旭 , 姚亦飞 , 于繁华. 基于协同进化的多目标约束进化算法[J]. 吉林大学学报(信息科学版), 2023, 41(2): 321-328.

LIU Renyun , ZHANG Xu , YAO Yifei , YU Fanhua . Multi-Objective Constrained Evolutionary Algorithm Based on Coevolution [J]. Journal of Jilin University (Information Science Edition), 2023, 41(2): 321-328.