Fuzzy Control Using Piecewise Linear Membership Functions Based on Knowledge of Tuning PID Controller.

Abstract

Tuning of the control parameters of a fuzzy controller depends on trial and error. How this can be accomplished efficiently is an important subject that should be investigated in fuzzy control. We propose a method in which the membership functions of a fuzzy controller can be simply set up by utilizing the knowledge of tuning parameters in a conventional PID controller. In this method, fuzzy control is realized as follows: the piecewise linear membership functions that are determined from the knowledge of tuning parameters in a PID controller are set up in the antecedent parts of the four fuzzy control rules of a simple structure. Then, the simplified inference method that enables high-speed inference is applied to the fuzzy control rules.