Affiliations: [a] Laboratory of Control and Computational Intelligence, Federal Institute of Technology of Maranhão, Imperatriz, MA, Brazil
| [b] Laboratory of Computational Intelligence Applied to Technology, Federal Institute of Technology of Maranhão, São Luís, MA, Brazil
Correspondence:
[*]
Corresponding authors. Edson B.M. Costa, IFMA, Laboratory of Control and Computational Intelligence, Federal Institute of Technology of Maranhão, Av. Newton Belo, s/n, Vila Maria, CEP 65919-050, Imperatriz, MA, Brazil. E-mail: edson.costa@ifma.edu.br and Ginalber L.O. Serra, IFMA, Laboratory of Computational Intelligence Applied to Technology, Federal Institute of Technology of Maranhão, Av. Getúlio Vargas, 04, Monte Castelo, CEP 65030-005, São Luís, MA, Brazil. Tel.: +55 98 3218 9088; Fax: +55 98 3218 9000; E-mail: ginalber@ifma.edu.br.
Note: [1] This classification is neither unique nor exhaustive, and many other different classifications can also be employed. However, this classification is suitable to the scope of this work.
Abstract: In this paper, an adaptive fuzzy controller design methodology via MultiObjective Particle Swarm Optimization (MOPSO) based on robust stability criterion, is proposed. The plant to be controlled is modeled from its input-output experimental data considering a Takagi-Sugeno (TS) fuzzy NARX model, by using the fuzzy C-Means clustering algorithm (antecedent parameters estimation) and Weighted Recursive Least Squares (WRLS) algorithm (consequent parameters estimation). An adaptation mechanism as MOPSO problem for online tuning of a fuzzy model based digital PID controller parameters, based on the gain and phase margins specifications, is formulated. Experimental results for adaptive fuzzy digital PID control of a thermal plant with time varying delay is presented to illustrate the efficiency and applicability of the proposed methodology.
Keywords: Adaptive fuzzy control, robust stability, PID controller, multiobjective optimization, gain and phase margins, particle swarm optimization
