Global-best Brain Storm Optimization Algorithm
No Thumbnail Available
Authors
El-Abd, Mohammed
Issue Date
Type
Journal Article
Peer-Reviewed
Peer-Reviewed
Language
Keywords
Alternative Title
Abstract
Brain storm optimization (BSO) is a population-based metaheuristic algorithm that was recently developed to mimic the brainstorming process in humans. It has been successfully applied to many real-world engineering applications involving non-linear continuous optimization. In this work, we propose improving the performance of BSO by introducing a global-best version combined with per-variable updates and fitness-based grouping. In addition, the proposed algorithm incorporates a re-initialization scheme that is triggered by the current state of the population. The introduced Global-best BSO (GBSO) is compared against other BSO variants on a wide range of benchmark functions. Comparisons are based on final solutions and convergence characteristics. In addition, GBSO is compared against global-best versions of other meta-heuristics on recent benchmark libraries. Results prove that the proposed GBSO outperform previous BSO variants on a wide range of classical functions and different problem sizes. Moreover, GBSO outperforms other global-best meta-heuristic algorithms on the well-known CEC05 and CEC14 benchmarks.
Description
Citation
Publisher
Swarm and Evolutionary Computation, Elsevier
License
Journal
Volume
Vol. 37