Vector coevolving particle swarm optimization algorithm

Publication Type:
Journal Article
Information Sciences, 2017, 394-395, pp. 273-298
Issue Date:
Filename Description Size
1-s2.0-S0020025517303717-main.pdf5.09 MB
Adobe PDF
Full metadata record
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.
Please use this identifier to cite or link to this item: