A gravitational search algorithm with hierarchy and distributed framework

Yirui Wang,Shangce Gao,Yang Yu,Zonghui Cai,Ziqian Wang
DOI: https://doi.org/10.1016/j.knosys.2021.106877
2021-04-01
Abstract:<p>Gravitational search algorithm is an effective population-based algorithm. It simulates the law of gravity to implement the interaction among particles. Although it can effectively optimize many problems, it generally suffers from premature convergence and low search capability. To address these limitations, a gravitational search algorithm with hierarchy and distributed framework is proposed. A distributed framework randomly groups several subpopulations and a three-layered hierarchy manages them. Communication among subpopulations finally enhances the search performance. Experiments discuss parameters and strategies of the proposed algorithm. Comparison between it and sixteen state-of-the-art algorithms demonstrates its superior performance. It also shows the practicality for two real-world optimization problems.</p>
computer science, artificial intelligence
What problem does this paper attempt to address?