Affiliations: [a] School of Electric Power Engineering, South China University of Technology, Guangzhou, Guangdong, China | [b] College of Electrical and Electronic Engineering, Shandong University of Technology, Zibo, Shandong, China | [c] Institute of Advanced Technology, Chinese Academy of Sciences, Institute of Biomedical and Health Engineering, Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen, Guangdong, China
Correspondence:
[*]
Corresponding author: Mengshi Li, School of Electric Power Engineering, South China University of Technology, Guangzhou 510641, Guangdong, China. E-mail:[email protected]
Abstract: Group search optimizer (GSO) is a stochastic, population-based optimization technique that has shown better performance as for global searching when optimizing multimodal benchmarks. However, it suffers from poor convergence because of its producer-scrounger model, which makes it unable to get a reliable estimiator for evolution path. In order to enhance the local search ability without impairing the global seach ability of GSO, this paper proposes a variant of group search optimizer (VGSO), which develops a producer-organizer model whereby group members are assumed to search for opportunities of either `finding' (producing) or `learning' (organization). This model enables VGSO to achieve a good trade-off between its exploration and exploitation abilities. A comprehensive experimental study is conducted on a set of benchmark functions and a practical optimization problem. Comparison results show that VGSO obtains a promising performance on these optimization problems.
Keywords: Evolutionary algorithm, group search optimizer, adaptive learning, differential vector, global optimization
