Survey on the Recent Design and Tuning Approaches for model-free Fuzzy PID/PI Controllers

Abstract

There is good experience in applying classical linear proportional, integral and derivative (PID) controllers to industrial plants due to its simple structure and easy tuning property. However, considering the fact that these linear controllers have fixed parameters, the uncertainties and nonlinearities in the controlled system can degrade the control performance. To tackle the mentioned problems in the application of PID controllers, one of the ways is to incorporate fuzzy logic systems (FLSs) into the PIDs and fine-adjust them. Like the PID controllers, one of the advantages of the FLS-based PID controllers is that they do not require a system mathematical model for control problems, thus these controllers have shown to be a viable control solution for various complicated nonlinear systems. This study presents a survey of development and analysis of model-free PIDs incorporated with FLSs. We focus especially on the widely used fuzzy gain scheduling PID (FGPID) and fuzzy PID (FPID) controllers which are two important combinations of the FLSs and PID controllers. In this study, different structures of FGPID and FPID controllers in which type-1 FLS (T1FLS), interval type-2 FLS (IT2FLS), and general type-2 FLS (GT2FLS) are used as the inference realization part, have also been reviewed. Moreover, the tuning methods of FGPID and FPID controllers’ parameters and their closed-loop stability problems are discussed.