Designing fuzzy controllers for a class of MIMO systems using Hybrid Particle Swarm Optimization and Tabu Search

Article type: Research Article

Authors: Talbi, N.^{a; *} | Belarbi, K.^b

Affiliations: [a] Faculty of Science and Technology, Jijel University, Jijel, Algeria | [b] Faculty of Engineering, Mentoury University, Constantine, Algeria

Correspondence: [*] Corresponding author: N. Talbi, Faculty of Science and Technology, Jijel University, BP 98, Ouled Aissa, Jijel, Algeria. E-mail: [email protected]

Abstract: In this paper, Hybrid Particle Swarm Optimization and Tabu Search (HPSOTS) is proposed to automatically generate Fuzzy Controller for MIMO systems. The algorithm is used to simultaneously optimize the premise and consequent parameters of the fuzzy rules for the appropriate design of fuzzy controller for Takagi-Sugeno zero-order. Solution found after each step of PSO algorithm was introduced in Tabu search algorithm as initial solution to seek its best neighboring solution. This hybridization of global and local optimization algorithms minimizes the number of iterations and computation time while maintaining accuracy and minimum response time. The algorithm was tested to control two nonlinear dynamical MIMO systems. Results proved the effectiveness of the proposed algorithm.

Keywords: Tabu search, particle swarm optimization, fuzzy controller, MIMO system

DOI: 10.3233/HIS-120160

Journal: International Journal of Hybrid Intelligent Systems, vol. 10, no. 1, pp. 1-9, 2013