Elliptic interval Type-2 intuitionistic fuzzy logic system for non-linear system identification

Abstract

An elliptic membership function has been proposed in the literature for interval type-2 fuzzy logic system. In this paper, elliptic non- membership function is incorporated into the conventional elliptic membership function model to obtain elliptic interval type-2 intuitionistic fuzzy sets for the first time. The elliptic interval type-2 intuitionistic fuzzy logic system so formed is applied for the prediction of two benchmark non-linear systems and results compared with Gaussian  interval type-2  intuitionistic fuzzy logic system. Experimental results show that the elliptic interval type-2 intuitionistic  fuzzy logic system outperforms the traditional Gaussian interval type-2 intuitionistic fuzzy logic system in the problem  instances investigated.