A Predictive Direct Torque Control of Induction Machines

In this paper, a predictive Direct Torque Control (DTC) of an Induction Motor (IM) drive is presented. Although the control uses stator flux and current vectors as state variables, a novel set of variables have been introduced in order to directly manage torque, stator flux and current magnitude variations in each sampling time interval. Then, the existence domain of the above mentioned variables in each sampling time interval is determined and depicted graphically, together with the inverter voltage saturation and current limitation constraints. As a result, the proposed predictive control algorithm allows a better tracking of the IM reference torque profile than that achievable by means of the traditional DTC technique, especially in terms of current and torque ripple. The effectiveness of the proposed predictive DTC algorithm is verified by simulating an IM drive by means of the Matlab Simulink tool. The simulation results, which also refer to the case of the traditional constant switching frequency DTC, highlight the good performance of the drive, especially referring to torque ripple and to low speed operation.