基于混沌局部搜索的多目标遗传算法  

Multi-objective genetic algorithm based on chaotic local search

作　　者：梁敬东[1] 郭建军[1] 梁海洪[2] 郭凤仪[3] 

机构地区：[1]南京农业大学信息科学技术学院,江苏南京210095 [2]辽宁省电力有限公司葫芦岛供电公司,辽宁葫芦岛125000 [3]辽宁工程技术大学电气与控制工程学院,辽宁葫芦岛125105

出　　处：《辽宁工程技术大学学报（自然科学版）》2010年第4期630-633,共4页Journal of Liaoning Technical University (Natural Science)

基　　金：国家自然科学基金资助项目(50677027);辽宁省优秀人才计划项目(2007R23)

摘　　要：针对多目标遗传算法存在的局部搜索能力弱和易早熟的问题,采取理论分析方法,提出了一种新的基于混沌局部搜索的多目标遗传算法(MOGA-CLS)。对按支配关系形成的第1、2层Pareto解进行混沌搜索获得更优解,并采用基于聚集度的聚类方法保持种群和Pareto解集的多样性和分布性。仿真实验结果表明:该算法能有效地提高搜索效率。In view of overcome the deficiencies of the multi-objective genetic algorithm, such as poor local search capability and premature convergence, a new multi-objective genetic algorithm （MOGA-CLS） was developed based on chaotic local search. A better solution was obtained around the 1st and 2nd rank of Pareto solutions with MOGA-CLS. Also, the distribution and diversity of population and Pareto solution set were maintained by adopting the cluster method based on crowding degree. The simulation results show that the new algorithm MOGA-CLS can greatly improve the searching efficiency.

关 键 词：混沌 局部搜索 目标 传算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]