Design of a robust interval-valued type-2 fuzzy c-regression model for a nonlinear system with noise and outliers

This paper presents the development of a novel interval-valued type-2 robust fuzzy c-regression model (IVT2RFCRM) clustering algorithm for identification of nonlinear systems taking into account the presence of noise and outliers in the associated dataset. On the one hand, the proposed method allows for the handling of the uncertainties of the FCRM due to its fixed fuzzier parameter m. In the other hand, the dataset is subject to various sources of uncertainty such as measurement uncertainty, fuzziness of information and environmental noise. As a result, obtaining a high-quality approximation of real processes is often a difficult task. In this paper, the structure of the proposed clustering algorithm is given and its parameter update rule is derived. First, the modified objective functions use a kernel measure of error to deal with the noisy data. Then, a credibility function is integrated into the clustering process in order to reduce the effect of outliers. Finally, the effectiveness of the proposed algorithm is evaluated by comparing the obtained results with others reported in the literature and also through the simulation results of a real liquid level process.