子种群个体动态调整的人工蜂群算法  

Artificial bee colony algorithm for individual dynamical adjustment of sub-population

作　　者：龙文[1] 赵东泉[2] 徐松金[3] 唐明珠[4] 

机构地区：[1]贵州财经大学贵州省经济系统仿真重点实验室,贵州贵阳550004 [2]枣庄科技职业学院机械工程系,山东滕州277500 [3]铜仁学院数学科学学院,贵州铜仁554300 [4]长沙理工大学能源与动力工程学院,湖南长沙410114

出　　处：《兰州理工大学学报》2015年第6期93-98,共6页Journal of Lanzhou University of Technology

基　　金：国家自然科学基金(61463009);贵州省科学技术基金(黔科合J字[2013]2082号);贵州省高校优秀科技创新人才支持计划项目(黔教合KY字[2013]140);贵州省教育厅"125"重大科技专项资助项目(黔教合重大专项字[2012]011号);湖南省教育厅重点项目(12A007)

摘　　要：针对基本人工蜂群算法存在局部搜索能力差、收敛速度慢等缺点,提出一种动态调整子种群个体数目的改进人工蜂群算法用于求解无约束优化问题.该算法利用反向学习策略产生初始种群,以保证个体尽可能均匀分布在搜索空间中;基于个体适应度值,将种群分为两个子种群,分别采取不同的蜜源搜索公式,用于进行全局搜索和局部搜索.5个标准测试函数的仿真实验结果表明,改进算法具有较好的寻优性能.Aimed at the defects of poor global searching ability and slow convergence in artificial bee colony（ABC）algorithm,an improved artificial bee colony algorithm for dynamical adjustment of the number of individuals in sub-population is proposed to deal with the unconstrained optimization problem.In this algorithm,an initial population is generated based on reversed learning strategy to assure that the individuals are distributed in the search space as uniformly as possible.The population is divided into two sub-populations based on the fitness values of the individuals and different honey source searching formulae are selected to conduct global and local search,respectively.Experimental results of five benchmark function simulation show that the improved algorithm has better optimization performance.

关 键 词：人工蜂群 无约束优化问题 多种群 动态调整 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]