Fuzzy tuned PI controllers for series connected DC motor drives

A new approach for tuning a PI controller for the speed adjustment of a series-connected DC motor drive system is presented. Using the classical Ziegler-Nichols technique, the nominal feedback gains of a PI speed controller are obtained. However, the model of the system is nonlinear and, in order to achieve an improved performance, it is proven that the PI gains must be corrected in accordance to the changes from the nominal operation. Taking into account this last ascertainment, a suitable fuzzy logic controller is constructed involving time variant feedback gains. Extensive simulation studies are conducted to compare the response of the system with the fuzzy tuned controller to the response given with the classical static Ziegler-Nichols controller. In all cases, a significantly improved performance of the transient behaviour of the closed-loop system is verified.