Authors: Qing Zhang; Sanyou Zeng; Changhe Li
Addresses: College of Mathematics and Computer Science, Huanggang Normal University, 438000 Huanggang, Hubei, China ' School of Computer Science, China University of Geosciences, 430074 Wuhan, Hubei, China ' School of Computer Science, China University of Geosciences, 430074 Wuhan, Hubei, China
Abstract: A dynamic and random differential evolution (DRDE) is designed in this paper to solve constrained optimisation problems (COPs). The dynamic means that the COP is transformed into a dynamic constrained optimisation problem (DCOP), then a DRDE is designed to solve the DCOP for solving the original COP. The key issue is to maintain a feasible population since a feasible population would vanish the difficulty in constraint handling. A deliberate setting of the environment can achieve the maintenance as much as possible. The 'random' in the DRDE means that the scaling factor of the DDE is randomly created in a range to increase search domain so as to enhance the performance of the DRDE. The DRDE is tested by a kit of widely used constraint benchmark problems. The experimental results suggest it outperforms or performs similarly to other state-of-the-art algorithms referred to in this paper.
Keywords: constrained optimisation; dynamic optimisation; differential evolution; feasible population.
International Journal of Computing Science and Mathematics, 2014 Vol.5 No.2, pp.137 - 150
Available online: 31 Jul 2014 *Full-text access for editors Access for subscribers Purchase this article Comment on this article