Adaptive direct torque control using Luenberger-sliding mode observer for online stator resistance estimation for five-phase induction motor drives

Direct torque control is considered as one of the variable structure control techniques which are characterized by fast response, simplicity and provide direct control of both electromagnetic torque and stator flux by adequate selection of the inverter switches in each sampling period. In case of multiphase motor drives, the increase in voltage vectors offers flexibility to optimize the selection of the inverter switching states, thereby achieving more precise control of the torque and flux. Nevertheless, the criterion for the selection of the inverter states becomes more complex. This aspect is not considered an issue in the traditional three-phase motor drives but needs to be considered in designing the switching table of the direct torque control of five-phase induction motor. Due to the auxiliary vector plane, the low-frequency harmonics need to be eliminated and full utilization of the dc link voltage is desired. The effects of parameter variation (particularly, stator resistance) on the performance of the direct torque control. It is necessary the addition of parameter adaptation algorithm to compensate this effects. A novel direct torque control of five-phase induction motor using a new switching table combined with an adaptive variable structure observer to avoid the effects of stator resistance variations is presented in this paper. Simulation results provided show the effectiveness of the proposed control strategy.