摘要
蝙蝠算法(Bat algorithm,BA)是一种新型的、搜索全局最优解的元启发式算法.为解决蝙蝠算法局部搜索时易陷入局部极值的问题,提出一种基于速度越界处理与高斯扰动的改进蝙蝠算法(VGBA).该算法利用速度的越界处理控制蝙蝠位置更新的范围,利用高斯扰动增强蝙蝠算法的全局搜索能力.选取8个测试问题进行数值实验,实验结果表明,VGBA算法在收敛精度和稳定性上比BA算法有显著提升.
Bat algorithm is a new metaheuristic algorithm for solving global optimization problems.In order to overcome the shortage that the bat algorithm is prone to fall into local extremum in local search,we propose an improved bat algorithm(VGBA)based on velocity overstepping treatment and Gaussian perturbation.In the proposed algorithm,the update range of bat location is controlled by velocity overstepping treatment,and the global search of bat algorithm is enhanced by Gaussian perturbation.Eight test problems are used for numerical simulation experiments.The experimental results show that,compared to BA algorithm,VGBA algorithm has a significant improvement in convergence accuracy and stability.
作者
梁昔明
高超
龙文
LIANG Xi-ming;GAO Chao;LONG Wen(School of Science,Beijing University of Civil Engineering&Architecture,Beijing 102616,China;Guizhou Key Laboratory of Economics System Simulation,Guizhou University of Finance and Economics,Guiyang 550025,China)
出处
《数学的实践与认识》
北大核心
2019年第19期222-236,共15页
Mathematics in Practice and Theory
基金
国家自然科学基金(61463009)
北京自然科学基金(4122022)
中央支持地方科研创新团队项目(PXM2013-014210-000173)
贵州省科学技术基金(黔科合基础[2016]1022)
北京建筑大学市属高校科研业务费专项资金资助(X18193)
北京建筑大学研究生创新项目资助
贵州省高校科技拔尖人才支持计划(黔科合KY字[2017]070)
关键词
蝙蝠算法
VGBA
数值实验
bat algorithm
VGBA
Simulation experiments
