Parameter estimation of nonlinear systems using a robust possibilistic c-regression model algorithm

This article studies the problem of inappropriate parameter estimation for nonlinear system when the dataset is contaminated by noise based on fuzzy c-regression models. In comparison to the existing algorithms in the literature, the proposed method uses a generalized objective function that reduces the errors of partitioning datasets contaminated by noise, and as a consequence an accurate model is obtained. Indeed, it combines a modified version of possibilistic c-means procedure with fuzzy c-regression models. The weighted least squares method is exploited to identify the parameters contained in the consequent (THEN part). The results of this study demonstrate the effectiveness of the proposed method compared with other extended versions of the fuzzy c-regression model algorithm such as modified fuzzy c-regression model algorithm, possibilistic c-regression model and interval type-2 fuzzy c-regression model algorithm as well as other techniques existing in the literature.