Extended T-S fuzzy model based on interval arithmetic and its application to interval nonlinear regression analysis

In this paper, a new fuzzy system model structure - interval T-S fuzzy model (ITSFM) is proposed. Inspired from interval regression analysis, the interval arithmetic is incorporated with classical T-S fuzzy model and the parameters in consequent part of the ITSFM model become to be interval numbers. Thus, the outputs of the proposed ITSFM are interval numbers. In our ITSFM model, the membership functions are the same as the ones of the classical type. Finally, the proposed ITSFM is applied to interval nonlinear regression analysis with crisp inputs and interval outputs. Experimental results are then presented that indicate the validity and applicability of the proposed ITSFM.