Optimum voltage vector selection in Direct Torque Controlled PMSM using intelligent controller

Direct Torque Controlled Permanent Magnet Synchronous Motor (DTCPMSM) has become popular because of its high speed, accuracy and absence of coordinate transformation. In DTC technique, the torque and flux are directly controlled by selecting the optimum voltage vector. But the conventional DTC method has some disadvantages such as high torque and flux ripple, difficulties in controlling torque and flux at low speed. To overcome the difficulties in the conventional DTC and to select optimum voltage vector, artificial intelligent controllers: Neural Network, Fuzzy Logic and ANFIS controllers are proposed. The proposed neural network controller is used as a switching vector selector, based on torque error, flux error and sector angle neural network selects the optimum voltage vector. The proposed fuzzy logic controller is used to control the upper and lower limits of the torque hysteresis bandwidth to minimize the torque ripple. To improve the speed response of DTC, the conventional PI controller is replaced with a Self-Tuned ANFIS controller. The proposed works are simulated in MATLAB/SIMULINK platform and its result proves that the proposed methods have less torque ripple than the conventional DTC.