一种求解多目标优化问题的进化算法混合框架  

A hybrid framework of evolutionary algorithm for solving multi-objective optimization problems

作　　者：田红军[1] 汪镭[1] 吴启迪[1] TIAN Hong-jun WANG Lei WU Qi-di(College of Electronics and Information Engineering, Tongji University, Shanghai 201804, China)

机构地区：[1]同济大学电子与信息工程学院,上海201804

出　　处：《控制与决策》2017年第10期1729-1738,共10页Control and Decision

基　　金：国家自然科学基金项目(61075064;61034004;61005090);教育部新世纪人才计划项目(NCET-10-0633)

摘　　要：为了提高多目标优化算法的求解性能,提出一种启发式的基于种群的全局搜索与局部搜索相结合的多目标进化算法混合框架.该框架采用模块化、系统化的设计思想,不同模块可以采用不同策略构成不同的算法.采用经典的改进非支配排序遗传算法(NSGA-Ⅱ)和基于分解的多目标进化算法(MOEA/D)作为进化算法的模块算法来验证所提混合框架的有效性.数值实验表明,所提混合框架具有良好性能,可以兼顾算法求解的多样性和收敛性,有效提升现有多目标进化算法的求解性能.In order to improve the solve performance of multi-objective evolutionary algorithms, a heuristic hybrid framework for evolutionary multi-objective optimization combining the global search and local search based on populations is proposed. The framework adopts modular, systematic design thoughts and different strategies can be used in different modules to form different algorithms. The classic non-dominated sorting genetic algorithm Ⅱ（NSGA-Ⅱ） and multiobjective evolutionary algorithm based on decomposition（MOEA/D） algorithm are used as the algorithm of evolutionary algorithm module to verify the effectiveness of the proposed hybrid framework. The numerical experiments show that,the proposed hybrid framework has good performance, which can achieve a balance between diversity and convergence of algorithm and effectively improve the solve performance of existing multi-objective evolutionary algorithms.

关 键 词：多目标优化 进化算法 混合框架 NSGA-Ⅱ MOEA/D 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]