所属单位：计算机科学与工程学院

发表刊物：2020 International Conference on Data Processing Algorithms and Models, ICDPAM 2020

项目来源：自选课题

摘要：In this paper, we propose an enhancement algorithm of colliding bodies optimization based on rotation learning, named by RBL-CBO. Firstly, we use the rotation-based learning to search for any point in the rotating space by adjusting the rotation angle, thereby improving the ability of the proposed algorithm jump out of the local optimum. Next, we leverage the sinusoid-based nonlinear adjustment strategy to modify the control parameters to improve the calculation accuracy of the proposed algorithm. Finally, we process the cross-boundary object by the mirroring strategy. We conduct extensive experiments to test the performance of the proposed algorithm. In simulation-based experiments, 23 benchmark functions are used to compare RBL-CBO algorithm with the CBO, DE, BBO, PSO and GSA algorithm. The experimental results demonstrate that the proposed RBL-CBO algorithm is superior to the other comparison algorithms, while the RBL-CBO algorithm is at least 20% higher than the CBO algorithm in terms of the accuracy of solving function optimization problem.

合写作者：刘慧,李勇,崔世琦

第一作者：刘冰

论文类型：期刊论文

卷号：1774

页面范围：2

ISSN号：1742-6588

是否译文：否

发表时间：2021-02-03