Parameter Identification of Recurrent Fuzzy Systems with Fuzzy Finite-State Automata Representation

Carlos A. Gama (1) , Alexandre Evsukoff (1) , Philippe Weber (2) , Nelson F. F. Ebecken (1)

1 COPPE-UFRJ - Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia
2 CRAN - Centre de Recherche en Automatique de Nancy

This paper presents the identification of non-linear dynamical systems by recurrent fuzzy system models. Two types of recurrent fuzzy systems (RFS) models are discussed, the Takagi-Sugeno-Kang (TSK) type and the linguistic or Mamdani type. Both models are equivalent and the latter model may be represented by a fuzzy finite-state automaton. An identification procedure is proposed based on a standard general purpose genetic algorithm. First, the TSK rule parameters are estimated and, in a second step, the TSK model is converted into an equivalent linguistic model. The parameter identification is evaluated in some benchmark problems for non-linear system identification described in literature. The results show that RFS models achieve good numerical performance while keeping the interpretability of the actual system dynamics.

Mots clés: Fuzzy finite-state automata; genetic algorithms; nonlinear systems; system identification; recurrent fuzzy systems

Domaines: Automatique

Carlos A. Gama, Alexandre Evsukoff, Philippe Weber, Nelson F. F. Ebecken. Parameter Identification of Recurrent Fuzzy Systems with Fuzzy Finite-State Automata Representation. IEEE Transactions on Fuzzy Systems, 2008, 16 (1), pp.213-224. ⟨10.1109/TFUZZ.2007.902015⟩. ⟨hal-00139493⟩