Title: Self-adapting control parameters with multi-parent crossover in differential evolution algorithm

Authors: Yuanyuan Fan; Qingzhong Liang; Chao Liu; Xuesong Yan

Addresses: School of Computer Science, China University of Geosciences, Wuhan, 0086-430074, China ' School of Computer Science, China University of Geosciences, Wuhan, 0086-430074, China ' School of Computer Science, China University of Geosciences, Wuhan, 0086-430074, China ' School of Computer Science, China University of Geosciences, Wuhan, 0086-430074, China

Abstract: The performance of differential evolution (DE) algorithm is influenced by the setting of control parameters, which is quite dependent on the problem and difficult to be determined. Therefore, the studies on parameter adaptation mechanisms have gradually become more popular. In this paper, we present a self-adaptive DE algorithm (GaDE), in which the adaptation of amplification factor and crossover rate is executed with a multi-parent crossover, while the adaptation timing is decided by the comparative result between the target vector and its offspring. The performance of GaDE algorithm is evaluated on a suite of bound-constrained numerical optimisation problems. The results show that our algorithm is better than, or at least comparable to, the canonical DE, and the two other adaptive DE algorithms.

Keywords: differential evolution; self-adaptive DE; multi-parent crossover; self-adapting control parameters; adaptation timing; bound-constrained numerical optimisation.

DOI: 10.1504/IJCSM.2015.067540

International Journal of Computing Science and Mathematics, 2015 Vol.6 No.1, pp.40 - 48

Received: 23 Jul 2014
Accepted: 24 Aug 2014
Published online: 19 Feb 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article