Research of Improved MRAS Based Sensorless Control of Permanent Magnet Synchronous Motor Considering Parameter Sensitivity

PMSM sensorless control performance is largely affected by the accuracy of rotor position estimation. In this paper, MRAS method is used in the PMSM sensorless control, the position estimation of traditional MRAS sensorless method is sensitive to parameter variation during the operation of PMSM. To solve the problem, simulation has been done at first to study the parameter sensitivity of traditional MRAS. Base on the simulation results that variations in inductance and flux linkage have a greater impact on MRAS position estimation, an improved MRAS sensorless method is proposed in this paper. The MRAS adjustable model is updated by the FFRLS parameter identification results in each control cycle. The simulation results show that the improved MRAS position estimator can effectively reduce the influence of parameter variation on rotor position estimation.