Abstract
This paper presents a fuzzy modeling technique utilizing the so-called sector nonlinearity concept. To fully take advantage of the sector nonlinearity concept, we modify the Takagi-Sugeno fuzzy model and develop an algorithm to identify model parameters. The algorithm consists of two parts. In the first part, a sector candidate that satisfies a compatibility condition is found from input-output data of an original nonlinear system. The second part determines membership functions in premise parts so as to minimize identification error between an original nonlinear system and the fuzzy model. Identification examples illustrate the utility of our approach.
