A novel method for fuzzy self-tuning PID controllers

In this work, we are concerned with the use of the fuzzy inference mechanism in order to develop a novel on-line fuzzy self-tuning PID control scheme for improving the performance of the traditional PID controller. We first pre-tune a controlled system to determine its first order plus dead time (FOPDT) model and the steady state control signal (u_s). Then, the reference integral term can be set as (u_s/∫edt) since the reference integral control signal is equal to steady state control signal. Based on the control error, its first difference and some information of the controlled system (the coefficients of the FOPDT model), the fuzzy inference mechanism will be used to determine the proportional term, the increment of the integral term and the derivative term in the final step. Several numerical examples are given to demonstrate the feasibility of the proposed method in this paper