期刊
INFORMATION SCIENCES
卷 394, 期 -, 页码 273-298出版社
ELSEVIER SCIENCE INC
DOI: 10.1016/j.ins.2017.01.038
关键词
Particle swarm optimization; Coevolving evolution; Vector partition; Centralized learning; Decentralized learning
资金
- National Natural Science Foundation of China [61572230, 61173078, 61573166]
In this paper, we propose a novel vector coevolving particle swarm optimization algorithm (VCPSO). In VCPSO, the full dimension of each particle is first randomly partitioned into several sub-dimensions. Then, we randomly assign either one of our newly designed scalar operators or learning operators to update the values in each sub-dimension. The scalar operators are designed to enhance the population diversity and avoid premature convergence. In addition, the learning operators are designed to enhance the global and local search ability. The proposed algorithm is compared with several other classical swarm optimizers on thirty-three benchmark functions. Comprehensive experimental results show that VCPSO displays a better or comparable performance compared to the other algorithms in terms of solution accuracy and statistical results. (C) 2017 Elsevier Inc. All rights reserved.
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