解决高维INLP和MINLP问题的混沌差分进化算法  

Chaotic Differential Evolution Algorithm for High-dimensional INLP and MINLP Problems

作　　者：谭跃 赵政春 杨冰 肖湘 TAN Yue;ZHAO Zhengchun;YANG Bing;XIAO Xiang(College of Information and Electronic Engineering,Hunan City University,Yiyang,Hunan 413000,China)

机构地区：[1]湖南城市学院信息与电子工程学院

出　　处：《湖南城市学院学报（自然科学版）》2020年第1期53-59,共7页Journal of Hunan City University:Natural Science

基　　金：湖南省自然科学基金资助项目(2016JJ4014);湖南省教育厅科研项目(14B033,16C0301)

摘　　要：为改进差分进化(Differential Evolution,DE)算法的搜索能力,提出一种新的混沌差分进化算法(CGLSDE).首先,该算法利用混沌序列替换DE参数并采用混沌全局搜索算法来改进DE的全局搜索能力;其次,CGLSDE算法还采用了单维和多维的混沌局部搜索来改进DE的局部搜索能力.仿真结果表明:CGLSDE算法在解决高维整数非线性规划(INLP)问题和高维混合整数非线性(MINLP)问题上,其性能要好于其它3种混沌差分进化算法.To improve the search ability of differential evolution(DE), a new chaotic differential evolution algorithm(called CGLSDE) is proposed in this paper. Firstly, two methods(one is replacement of one of the parameters of DE using chaotic sequences, the other is chaotic global search) are used to improve the global search ability of DE. Secondly, in the CGLSDE, the single-dimensional and multi-dimensional chaotic local search ways are employed to enhance the local search capability of DE. The simulation results show that the CGLSDE is better than the other three chaotic differential evolution algorithms in solving high-dimensional integer nonlinear programming(INLP) and mixed integer nonlinear programming(MINLP) problems.

关 键 词：整数非线性规划(INLP) 混合整数非线性规划(MINLP) 差分进化(DE) 混沌局部搜索 混沌全局搜索 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]