Authors: Zhuanghua Zhu
Addresses: Economic Information Department, Shanxi Finance and Taxation College, WanBailin District, Taiyuan 030024, China
Abstract: As a new swarm intelligent algorithm, group search optimiser (GSO) attracts many scholars' attention. However, its performance is not good. To overcome this shortcoming, a new group search optimiser based on quadratic interpolation method (QIGSO) is proposed by Yao et al. (2011) in which one local optimum is estimated. In this paper, a new strategy, steepest gradient descent strategy is incorporated into the methodology of QIGSO to enhance the exploitation capability. This new variant of QIGSO (QIGSO-SDO) provides little estimation error, and obtains a better performance near the local optima. In this paper, QIGSO-SDO is employed to solve the directing orbits of chaotic systems, simulation results show this new variant increases the performance significantly when compared with the standard version of group search optimiser.
Keywords: group search optimiser; optimisation; GSO; steepest gradient descent; orbits of chaotic systems; swarm intelligence; quadratic interpolation; simulation; chaos.
International Journal of Computational Science and Engineering, 2012 Vol.7 No.2, pp.133 - 138
Received: 11 Jul 2011
Accepted: 30 Oct 2011
Published online: 22 Sep 2014 *