一种新型的柯西扰动黑猩猩优化算法  

Novel Chimp Optimization Algorithm with Cauchy Perturbation

作　　者：兰周新 何庆 LAN Zhou-xin;HE Qing(School of Big Data and Information Engineering,Guizhou University,Guiyang 550025,China;Guizhou Key Laboratory of Public Big Data,Guizhou University,Guiyang 550025,China)

机构地区：[1]贵州大学大数据与信息工程学院,贵阳550025 [2]贵州大学贵州省公共大数据重点实验室,贵阳550025

出　　处：《小型微型计算机系统》2023年第4期715-723,共9页Journal of Chinese Computer Systems

基　　金：贵州省科技计划项目重大专项项目(黔科合重大专项字[2018]3002,[2016]3022)资助;贵州省公共大数据重点实验室开放课题项目(2017BDKFJJ004)资助;贵州省教育厅青年科技人才成长项目(黔科合KY字[2016]124)资助;贵州大学培育项目(黔科合平台人才[2017]5788)资助。

摘　　要：针对黑猩猩优化算法(ChOA)在寻过程中求解精度低、收敛速度慢以及易陷入局部极值点等问题,提出一种新型的柯西扰动黑猩猩优化算法(CP-ChOA).首先采用佳点集映射初始化种群,增加算法在初始阶段的多样性;然后利用变异的柯西算子和反向学习策略,对当前最优位置进行扰动变异并产生新解,以提高算法的收敛速度,避免算法在迭代初期陷入局部极值;最后使用单纯形法策略改善最差个体的位置,增强算法的局部开发能力.选取8个基准函数和部分CEC2014测试函数进行试验仿真,结果表明CP-ChOA算法较标准ChOA算法、改进的ChOA算法以及其他元启发式算法具有更好的寻优性能,并通过优化2个工程设计问题,验证了CP-ChOA算法在工程上的可行性.Aiming at the problems of the chimp optimization algorithm(ChOA)in the search process,such as low accuracy,slow convergence speed and easy to fall into local extreme points,a novel chimp optimization algorithm with cauchy perturbation(CP-ChOA)is proposed.Firstly,the good point set mapping is used to initialize the population to increase the diversity of the algorithm in the initial stage.Then,the cauchy operator of mutationand reverse learning strategy are used to perturb the current optimal position and generate a new solution,so as to improve the convergence speed of the algorithm and avoid falling into local extremum at the beginning of iteration.Finally,the simplex method is used to improve the position of the worst individual and enhance the local development ability of the algorithm.Eight benchmark functions and part of CEC2014 test functions are selected for experimental simulation.The results show that the CP-ChOA algorithm has better optimization performance than the standard ChOA algorithm,the improved ChOA algorithm and other meta-heuristic algorithms,and the engineering feasibility of CP-CHOA algorithm is verified by optimizing two engineering design problems.

关 键 词：黑猩猩算法 柯西变异 反向学习 单纯形法 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]