Robust Model Predictive Control of Position Sensorless-Driven IPMSM Based on Cascaded EKF-LESO

This article proposed a robust model predictive control (MPC) for permanent magnet synchronous motors taking into account the effect of disturbance. A structure cascading extended Kalman filter (EKF) and linear extended state observer (LESO) are used to estimate the position of the motor without using encoders. First, the voltage vector reference containing information of the system speed and current is predicted using the deadbeat control. Meanwhile, the ultralocal model of the system is built considering the lumped disturbance of the motor. The disturbed part of the ultralocal model is updated online according to the information of voltage and current in the past two moments. Then, a cascaded EKF-LESO is designed to efficiently estimate position and speed, taking into account disturbances such as measurement noise and model parameter uncertainty. Experimental results verify the effectiveness and advantages of the proposed method.