Enhanced PMSM Current Controller via DOB-Based Load Torque Forecasting

The current control of the D and Q axes of Permanent Magnet Synchronous Motors (PMSMs) is crucial in drives, especially when saliency is present in the system. The stability and performance of any outer loop control algorithm are significantly influenced by the performance of the D and Q axis current controllers. The outer loop typically functions as a speed or torque controller, often incorporating an MTPA. This paper proposes an inner loop controller that observes changes in load torque and adjusts the current components accordingly. To determine the load torque in the proposed model, a disturbance observer (DOB) approach is utilized, which eliminates the need for external load torque sensors. Unlike usual DOB approaches, this method considers disturbance torque for both Iq and Id inner loops. The predicted torque is calculated based on the references of D and Q axis currents, and the difference between dynamic torques is used to forecast the load torque. The forecasted load torque is then used to calculate the dq axis currents, which are injected back into the system. The PMSM and drive system were simulated under various torque inputs, comparing results for scenarios with and without the proposed controller, while keeping all other parameters constant. As the results are derived from a simulation environment, validating the proposed method through hardware implementation and analyzing the variations will be a critical aspect of future research. Simulation results demonstrate that the current controller incorporating the proposed method exhibits improved performance compared to the scenario without the proposed controller.