Torque Disturbance Observer Based Model Predictive Control for Electric Drives

Model predictive control (NIPC) based electric drive systems has faster dynamics and can achieve similar performance as field oriented control (FOC) and direct torque control (DTC) based systems though much smaller control frequency is applied. In this work, model inverse deadbeat based MPC is applied, which maintains the fast dynamics of the model forward finite control set MPC (FCS-MPC) and has even simpler structure. However, since it still keeps the outer speed proportional-integral (PI) controller, integration time for speed and torque response is required when load torque variations occurs. To improve system dynamics and stability by reducing response time and torque ripples against load disturbances, a torque disturbance observer (TDO) is designed. The effectiveness and good overall performance of the proposed system is verified through simulations.