Simulation analysis of hysteresis-based, PI-Based and fuzzy PI-Based torque controller for direct torque control (DTC) of induction machine drives

Abstract

This thesis presents the simulation analysis of hysteresis-based, Proportional Integral (PI)-based and fuzzy PI-based torque controller for Direct Torque Control (DTC) of Induction Machine drives. It was well established that the implementation of conventional DTC drive which consist of hysteresis-based torque controller suffers from the varying switching frequency and high torque ripple. One of the many approaches to tackle these problems is by replacing this hysteresis controller with the Proportional Integral (PI)-based torque controller. By comparing its generated control signal with the triangular carrier waveforms, the torque controller manages to produce a constant switching frequency and reduce the torque ripple. However, to determine the parameters of the PI controller, the torque loop of the drive need to be modelled and rotor parameters need to be determined. To simplify this designing process, a new torque controller called fuzzy PI-based torque controller is proposed. Similar strategy as the PIbased torque controller is used, but with the proposed controller, the function of PI controller is replaced by the fuzzy PI controller. In this thesis, the topology of the hysteresis-based and PI-based torque controller is reviewed, while the design of the fuzzy PI-based torque controller is discussed. The simulation of each torque controller scheme is performed, and then the comparison study of their performance is presented. With the proposed controller, the fixed switching frequency is obtained while the torque ripple, especially at low speed is significantly reduced. Compared to the PI-based controller, the performance of the proposed controller is maintained, if not better.