A new self-tuning design for PID-type fuzzy logic controllers

In this study, a new method is proposed for tuning the coefficients of PID type fuzzy logic controllers. The new method adjusts the input scaling factor corresponding to the derivative coefficient and the output scaling factor corresponding to the integral coefficient of the PID type fuzzy logic controller using a fuzzy inference mechanism in an on-line manner. The fuzzy inference mechanism that adjusts the related coefficients has two inputs one of which is called "normalized acceleration" and the other one is the classical "error". The "normalized acceleration" gives the "relative rate" information about the fastness or slowness of the system response. An appropriate rule-base is generated for the adaptation of the derivative coefficient of the PID type fuzzy logic controller using these two input variables. The robustness and effectiveness of the new self-tuning algorithm have been compared with the other related tuning methods proposed in the literature through simulations. The simulations are done on a second order system with varying parameters and time delay.