An artificial bee colony algorithm for inverse problems

Article type: Research Article

Authors: Ho, S.L.^{a; *} | Yang, Shiyou^b

Affiliations: [a] Department of Electrical Engineering, The Hong Kong Polytechnic University, Hong Kong, China | [b] College of Electrical Engineering, Zhejiang University, Hangzhou, China

Correspondence: [*] Corresponding author. Tel.: +852 27666170; Fax: +852 23301544; E-mail: [email protected]

Abstract: Based on the metaphor of the foraging mechanism of honey bee swarms as well as on available works, an artificial bee colony algorithm is proposed for solving difficult inverse electromagnetic problems. In the proposed algorithm, the entire searching process is divided into an intensification and a diversification phase. For intensification searches, some novel formulas are proposed for the employed bees and onlookers to carry out exploiting searches around specific memorized food sources; and scouts are used to generate new food sources to guarantee the diversity of the algorithm in diversification searches. Also, an age variable is introduced to measure the "exhausted" level of a food source and then decide when to abandon a memorized food source. Three numerical examples are reported to validate the robustness and to demonstrate the advantages of the proposed algorithm.

Keywords: Bee colony algorithm, evolutionary programming, inverse problem, swarm intelligence

DOI: 10.3233/JAE-2009-1056

Journal: International Journal of Applied Electromagnetics and Mechanics, vol. 31, no. 3, pp. 181-192, 2009