Fuzzy linear parameter estimation algorithms: A new formulation

Abstract

In this paper, we present a new formulation of a fuzzy linear estimation problem as one of the linear programming where the objective is to minimize the spread of the data points, subject to constraints on each measurement point to guarantee that the original membership is included in the estimated membership. Different models are developed based on fuzzy triangular membership as well as fuzzy numbers of LR-type, and applied to different examples in fuzzy linear regression and finally we apply these models for estimating the electrical load on a substation busbar. It is demonstrated that since the nature of the load is characterized by uncertainty, the proposed technique is well suited to apply for electrical load estimation.