Title: Digital redesign of uncertain interval systems via a hybrid particle swarm optimiser

Authors: Chen-Chien Hsu, Chun-Hwei Gao

Addresses: Department of Applied Electronic Technology, National Taiwan Normal University, 162 He-Ping East Road, Section 1, Taipei 10610, Taiwan, Republic of China. ' Department of Electrical Engineering, Tamkang University, 151 Ying-Chuan Rd., Tamsui, Taipei County, 25137, Taiwan

Abstract: In this paper, a hybrid optimiser incorporating particle swarm optimisation (PSO) and an enhanced NM simplex search method is proposed to derive an optimal digital controller for uncertain interval systems having an interval plant based on frequency-response resemblance. By combining the uncertain plant and controller, extremal systems for the redesigned digital system and its continuous counterpart can be obtained as the basis for comparison. The design problem is then formulated as an optimisation problem of an aggregated error function in terms of frequency-response deviation between the redesigned digital system and its continuous counterpart, and subsequently optimised by the proposed optimiser to obtain an optimal set of parameters for the digital controller. Thanks to the performance of the proposed hybrid optimiser, frequency envelope of the redesigned digital system using the digital controller evolutionarily derived by the proposed approach bare a better resemblance to its continuous-time counterpart in comparison to those obtained using existing open-loop discretisation methods.

Keywords: digital redesign; particle swarm optimisation; PSO; interval plants; uncertain systems; Nelder and Mead simplex research; NM simplex search; frequency response; discretisation; extremal systems; hybrid optimisation; digital controllers.

DOI: 10.1504/IJICA.2008.026966

International Journal of Innovative Computing and Applications, 2008 Vol.1 No.4, pp.232 - 243

Published online: 11 Jul 2009 *

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