期刊
COMPUTERS & INDUSTRIAL ENGINEERING
卷 135, 期 -, 页码 299-313出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.cie.2019.06.015
关键词
Cuckoo search; Ensemble; Selection probability; External archive; Numerical optimization
资金
- National Natural Science Foundation of China [51669006, 61773314]
Cuckoo search is a simple yet effective evolutionary algorithm for solving numerical optimization problems. Recently, many variants of cuckoo search have been developed to further enhance the performance. These improved versions have different capabilities in tackling the optimization problems with different properties, so it is difficult to determine which algorithm is best for all problems. To address this issue, we present a new cuckoo search algorithm named the ensemble cuckoo search variant. In this developed version, a candidate pool consisting of three different cuckoo search algorithms is first constructed. According to the previous experiences in producing promising solutions, an adaptive scheme is then used to determine the probability that each algorithm can be assigned to distinct individuals in the current population. Also, an external archive is embedded to further discourage premature convergence. To assess the performance of this ensemble algorithm, 42 test problems derived from CEC 2005 and CEC 2013 are employed. Experimental results indicate that the proposed algorithm is a competitive method compared with seven well-established cuckoo search variants and several other well-known evolutionary algorithms.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据