摘要
布谷鸟搜索算法利用Lévy Flights随机走动和Biased随机走动过程完成全局搜索和局部开发.针对原始的Lévy Flights随机走动仅采用固定的常数步长因子,介绍了一种使用每一代中个体的全局和局部最优适应值动态设置步长因子的方法,并提出了一种带全局-局部最优步长比例因子的布谷鸟搜索算法.在测试函数上的运行结果证明,该方法是可行的,且能够全面有效地加强布谷鸟搜索算法的收敛速度和求精能力,其性能总体上比采用固定因子、基于均匀分布随机数或基于贝塔分布随机数比例因子的布谷鸟搜索算法更优.
Lévy Flights random walk and Biased/selective random walk is employed in cuckoo search algorithm to search for new solutions.Instead of a fixed scaling factor in standard Lévy Flights,in our study,global and local best fitness of individuals in each generation are utilized to dynamically define the scaling factor.And a modified cuckoo search algorithm,called GlbestCS is proposed.Comprehensive experiments demonstrate that this strategy is feasible,and it can effectively strengthen the convergence speed and improve accuracy of cuckoo search algorithm.In addition,the performance of the proposed algorithm is generally better than that of cuckoo search algorithm that uses a constant scaling factor,either based on uniformly distributed random numbers or based on a beta distribution random numbers.
作者
梁毛毛
肖文
王李进
钟一文
Liang Maomao;Xiao Wen;Wang Lijin;Zhong Yiwen(College of Computer and Information Sciences,Fujian Agriculture and Forestry University,Fuzhou 350002,China;Key Laboratory of Smart Agriculture and Forestry,Fujian Agriculture and Forestry University,Fuzhou 350002,China)
出处
《南京师范大学学报（工程技术版）》
CAS
2022年第2期56-62,72,共8页
Journal of Nanjing Normal University(Engineering and Technology Edition)
基金
福建省自然科学基金项目(2021J01127)
福建农林大学科技创新专项基金项目(CXZX2020148C、CXZX2020150C)
数字福建旅游大数据研究所开放基金项目(DFJTBDRI2020103).
