摘要
针对人工蜂群算法收敛速度慢和易陷入局部最优的缺点,在雇佣蜂搜索阶段提出了一种基于多维搜索和一维搜索的混合搜索策略,能克服单一一维搜索下收敛速度慢的缺点,有效加快收敛速度;提出了新的跟随蜂蜜源选择策略,可保证种群多样性,增强算法全局搜索能力。通过对12个基准测试函数进行仿真实验并与原算法进行比较,其结果表明改进的算法在收敛速度和精度上均优于人工蜂群算法。
For the slow convergence and susceptibility to local minima of the artificial bee colony algorithm, ahybrid search strategy based on multi-dimensional search and linear search in the employment bee search is presen-ted, which improves convergence rate of the algorithm under single one-dimensional search strategy. A new selectionstrategy for following the bees is also proposed to enhance the diversity of population and strengthen global searchingability. Finally, the improved algorithm is compared with standard algorithms through simulation experiment ontwelve benchmark test functions, the results show that the improved algorithm outperform the standard algorithm inboth convergence rate and searching precision.
出处
《电子科技》
2015年第3期61-64,共4页
Electronic Science and Technology
关键词
人工蜂群算法
多维搜索
一维搜索
种群多样性
基准测试函数
artificial bee colony algorithm
multi-dimensional searching
one-dimensional searching
diversityof population
benchmark test function