求解约束优化问题的加速CMODE算法  

Accelerated CMODE algorithm for solving constrained optimization problems

作　　者：閤大海[1,2] 李元香[1,2] 刘伟[3] 

机构地区：[1]武汉大学软件工程国家重点实验室,湖北武汉430072 [2]武汉大学计算机学院,湖北武汉430072 [3]南阳理工学院软件学院,河南南阳473000

出　　处：《华中科技大学学报（自然科学版）》2016年第4期48-52,共5页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家重大仪器专项资助项目(2011YQ170065.4)

摘　　要：为了加快CMODE(多目标与差分进化结合)算法的收敛速度,提出一种基于适应排序的分组选择方法将种群分为精英组与普通组,对精英组个体使用随机选择方式,对普通组个体使用适应排序选择方式,通过分组选择系数控制种群的选择压力.引入一种个体更新辅助策略对于淘汰个体给予一次机会与相似个体比较并保留优胜个体,提高优秀个体的生存几率并进一步加快收敛速度.改进后的算法在cec2006的测试函数上进行了测试,结果表明该算法使大多数函数的收敛速度明显提高.In order to speed up the convergence of CMODE,an adaptive ranking based grouping selection method was proposed.The method divided the population into elite group and common group at first,applying the random selection to the elite group and applying the adaptive ranking selection to the common group,which could control the selection pressure by the grouping selection coefficient.In addition,an individual update assistant strategy was introduced into the algorithm to give one more chance to the failed individual to survive in the population.The failed individual was compared with the most similar individual and would survive if it won.The strategy could improve the living odds of good individuals and speed up the convergence rate further.The improved algorithm was tested on the test functions of cec2006,and the results show that the improved algorithm can improve the convergence rate of most functions.

关 键 词：约束优化 多目标与差分进化结合算法 分组选择 个体更新辅助策略 收敛速度 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]