改进的多目标和声搜索算法  

Improved Harmony Search Algorithm for Multi-objective

作　　者：乔英[1] 高岳林[1] 江巧永[1] 

机构地区：[1]北方民族大学信息与系统科学研究所,银川750021

出　　处：《计算机工程》2012年第18期144-146,共3页Computer Engineering

基　　金：国家自然科学基金资助项目(60962006);北方民族大学校级基金资助项目(2011Y025)

摘　　要：针对和声搜索算法不能很好求解多目标优化问题的缺陷,引入邻域搜索算子,对和声记忆库内搜索到的分量进行扰动,对和声记忆库外进行Pareto邻域搜索,实现群体间信息交换,提高算法的全局搜索能力。数值实验选取4个常用测试函数并与NSGA-II、SPEA2、MOPSO 3个多目标算法进行比较,测试结果验证了改进算法的有效性。For solving the problems of basic harmony search algorithm can not be well used for multi-objective optimization,a kind of neighborhood search operator is introduced to improve the traditional harmony search algorithm.In the harmony memory of the proposed algorithm,every variable is disturbed by neighborhood search operator.Out of the harmony memory,every variable will be searched by Pareto neighborhood search operator.So it can exchange information between the populations during the proceeding of evolutionary and enhance the global searching ability of harmony search algorithm.Numerical experiments compare with NSGA-II,SPEA2 and MOPSO on four benchmark problems.The results show the effectiveness of the proposed algorithm.

关 键 词：多目标优化 和声搜索算法 邻域搜索算子 和声记忆 信息交换 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]