A Study on Tuning Methods of Fuzzy PI Controllers
Constantin Voloşencu · IntechOpen eBooks · 2024
The question addressed is placed in the broad context of fuzzy control, based on fuzzy PI controller, of industrial processes, the equivalence of fuzzy PI controllers with linear PI controllers, tuned with given methods, for the improvement of performance of control systems. The purpose of the study is to analyze the possibility of these implementation and to find the behavior of the resulted fuzzy control systems to compare with the linear control systems. The tuning criteria chosen are the modulus criterion in Kessler’s variant for second-order processes with two time constants, the symmetric criterion in Kessler’s variant, for the second order processes with purely integrator character and Ziegler-Nichols criterion for slow processes. The equivalence of the fuzzy controllers with the linear controllers was made using the input-output transfer characteristics of the fuzzy block and its linearization in the origin to obtain the gain in the origin. The results consist of an equivalence formula for the controller, simulation characteristics for the control systems, comparisons and analyses. The main conclusions are that fuzzy regulators can be designed by equating linear regulators with the presented method and better regulation performances can be obtained.