自适应步长和发现概率的布谷鸟搜索算法  

Cuckoo search algorithm based on adaptive step and discovery probability

作　　者：彭建新[1] 詹志辉[2] 陈宗淦[2] 王子佳[2] PENG Jianxin ZHAN Zhihui CHEN Zonggan WANG Zijia(Department of Computer Science, Guangdong Police College, Guangzhou 510230, China School of Computer Science and Engineering, South China University of Technology, Guangzhou 510006, China)

机构地区：[1]广东警官学院计算机系,广东广州510230 [2]华南理工大学计算机科学与工程学院,广东广州510006

出　　处：《济南大学学报（自然科学版）》2016年第5期328-333,共6页Journal of University of Jinan(Science and Technology)

基　　金：国家自然科学基金(61402545);公安部应用创新计划项目(2013YYCXGDST015)

摘　　要：针对因参数设置为常数、个体参数设置相同而导致布谷鸟算法求解精度降低的问题,提出一种基于适应值分配的自适应步长和发现概率的布谷鸟搜索算法,进行仿真实验,并与其他改进算法进行对比研究。结果表明:自适应步长和发现概率的布谷鸟搜索算法提高了算法的计算精度,计算结果优于原始的布谷鸟算法;与其他改进的布谷鸟算法相比,具有较强的竞争性。For solving the problem of accuracy reduction of the cuckoo search （CS）algorithm due to its constant parameter set and the same parameter set for all individuals, an adaptive cuckoo search （ACS） algorithm, which could adapt the step size and discovery probability was proposed. Simulation experiments show that this algorithm improves the computational precision, superior to the original, and is more competitive than the other improved cuckoo search algorithms.

关 键 词：布谷鸟搜索算法 函数优化 自适应步长 

分 类 号：TU7[建筑科学—建筑技术科学]