Improved auto-tuning niching genetic algorithm combined with grouping concept for inverse problem

Issue title: Special Volume: Proceedings of the eleventh International Symposium on Applied Electromagnetics and Mechanics ISEM-Versailles ISEM 03

Article type: Research Article

Authors: Kim, Jae-Kwang^{a; *} | Jung, Sang-Yong^a | Jung, Hyun-Kyo^a | Hahn, Song-Yop^a | Cho, Dong-Hyeok^b | Chung, Tae-Kyung^c

Affiliations: [a] Electromechanics Laboratory ENG420-040, School of Electrical Engineering, College of Engineering, Seoul National University, San 56-1, Shillim-Dong, Kwanak-Gu, Seoul, Korea | [b] Hyo-Sung Industry Company, Seoul, Korea | [c] Department of Electrical Engineering, Chung-Ang University, Seoul, Korea

Correspondence: [*] Corresponding author. Tel.: +82 2 880 7262; Fax: +82 2 878 1452; E-mail: light7@elecmech.snu.ac.kr

Abstract: Niching genetic algorithm has been applied to the optimisation of multimodal function. In this paper, improved niching genetic algorithm using auto-tuning concept and grouping factor is proposed. Population size and niche radii can be determined automatically and similar solutions unified by grouping factor. This method can be applied to the complex function such as inverse problem.

Keywords: niching genetic algorithm, auto-tuning concept, grouping factor

DOI: 10.3233/JAE-2004-627

Journal: International Journal of Applied Electromagnetics and Mechanics, vol. 19, no. 1-4, pp. 561-566, 2004