An adaptive quantum-behaved particle swarm optimizer for global optimization of inverse problem

Issue title: Proceedings from the 17th International Symposium on Applied Electromagnetics and Mechanics (ISEM 2015)

Guest editors: Fumio Kojima, Futoshi Kobayashi and Hiroyuki Nakamoto

Article type: Research Article

Authors: Wang, Luyu | Yang, Shiyou^* | Huang, Jin

Affiliations: College of Electrical Engineering, Zhejiang University, Hangzhou, Zhejiang, China

Correspondence: [*] Corresponding author: Shiyou Yang, College of Electrical Engineering, Zhejiang University, Zheda Rd. 38, Hangzhou 310027, Zhejiang, China. Fax: +86 571 8795 1625; E-mail:[email protected]

Abstract: An adaptive quantum-behaved particle swarm optimizer (AQPSO) is proposed to ensure a good balance between exploration and exploitation searches of the algorithm. In the proposed algorithm, some indicators to identify the searching states of a particular particle, the whole swarm, and the iterative process are proposed and used to design an adaptively tuning mechanism for the Contraction-Expansion Coefficient. Moreover, a partially regenerated swarm scheme is designed to select promising particles and to regenerate unpromising ones when the swarm is trapped into a local optimal point. The proposed AQPSO is used to solve the TEAM workshop benchmark problem 22, and compared with existing QPSO. The numerical results have demonstrated that the proposed algorithm outperforms the existing QPSO, especially in view of the global search ability.

Keywords: QPSO, global optimization, TEAM22

DOI: 10.3233/JAE-162058

Journal: International Journal of Applied Electromagnetics and Mechanics, vol. 52, no. 1-2, pp. 793-799, 2016