A Rotor Position Estimation Scheme Using Inherent Excitation Source for Finite-Control Set Model Predictive Control in Low-Speed Region

The finite-control-set model predictive control (FCS-MPC) is a promising control method for permanent magnet synchronous motor (PMSM) drives. However, its discrete nature poses problems in sensorless control for low-speed operation. Since the cost function of FCS-MPC directly selects the switching states, the conventional injection methods relying on modulators are difficult to operate in FCS-MPC. To address this problem, a rotor position estimation scheme for low-speed operation is designed by the inherent excitation source of FCS-MPC. First, the estimated rotor position is extracted from the error between the predicted and sampled current slopes. Second, the compensation of parameter perturbation is conducted on the proposed position extraction method, which can improve the parameter immunity. In addition, the cost function is modified to match the position extraction method in the low-speed and light-load regions. Experiments are implemented on the FPGA1 ARM-based PMSM test rig to validate the dynamic, steady-state, load capacity, and parameter immunity of the proposed method in the low-speed region.