This paper proposes a new cluster validity criterion designed for the fuzzy c-regression model (FCRM) clustering algorithm. The proposed cluster validity criterion is utilized to determine the appropriate number of clusters in the FCRM. A systematic procedure for the T-S fuzzy model identification is proposed based on the FCRM accompanied with the new cluster validity criterion. Simulation results show that for a given nonlinear system, the proposed algorithm can effectively and accurately obtain a T-S fuzzy model for it.
Published in:
Fuzzy Systems, 2004. Proceedings. 2004 IEEE International Conference on
(Volume:3
)
Date of Conference: 25-29 July 2004