基于骑手优化的动态布谷鸟搜索算法  

Dynamic cuckoo search algorithm based on rider optimization

作　　者：吴浩峻[1] 郝启润 聂群 王浩亮[1] 王丹 Wu Haojun;Hao Qirun;Nie Qun;Wang Haoliang;Wang Dan(College of Marine Engineering,Dalian Maritime University,Dalian Liaoning 116026,China;College of Marine Electrical Engineering,Dalian Maritime University,Dalian Liaoning 116026,China)

机构地区：[1]大连海事大学轮机工程学院,辽宁大连116026 [2]大连海事大学船舶电气工程学院,辽宁大连116026

出　　处：《计算机应用研究》2022年第7期2007-2011,2042,共6页Application Research of Computers

基　　金：辽宁省教育厅高等学校基本科研项目(面上项目)(LJKZ0044);大连市科技局高层次人才创新项目支持计划资助项目(2020RQ013);中央高校基本科研业务费专项资金资助项目(3132020197)。

摘　　要：针对传统布谷鸟搜索(cuckoo search,CS)算法对复杂问题收敛精度低、迭代步幅局限性大的特点,提出了基于骑手优化的动态布谷鸟搜索策略(rider optimization cuckoo search,ROCS)。结合骑手优化算法(rider optimization algorithm,ROA)思想,利用多种群在单周期内进行多策略寻优,动态使用最优策略进行加强搜索,提高算法对复杂问题的收敛效率;同时对Lévy飞行运动进行动态参数调节,改善算法搜索前期及末期表现。仿真测试结果显示,改进算法对复杂问题的优化表现优于对比算法,算法寻优效率得到显著改善。Aiming at the characteristics of the traditional cuckoo search algorithm with low convergence accuracy and large limitation of iterative step for complex problems,this paper proposed a dynamic cuckoo search strategy based on rider optimization.The algorithm combined with the idea of rider optimization algorithm,using multiple groups and optimal strategies in a single cycle.The optimal strategy could dynamically enhance the searching and improve the convergence efficiency of the algorithm for complex problems.At the same time,this algorithm adjusted the dynamic parameters of Lévy flight motion to improve the early and late performance of the algorithm search.The simulation test results show that the optimization performance of the improved algorithm ROCS is better than the comparison algorithms for complex problems,and the optimization efficiency of the algorithm has been significantly improved.

关 键 词：骑手优化算法 布谷鸟搜索算法 莱维飞行 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]