A High-Frequency Injection Sensorless Method for SynRM with Online Inductance Identification

The effectiveness of sensorless control of synchronous reluctance motors (SynRM) is affected by the accuracy of the inductance. Due to the presence of severe magnetic saturation in SynRM, inaccurate inductance parameters can produce observation angle errors, which affect the performance of the sensorless control system of the motor. In order to improve the robustness of the sensorless control system against parameter variations, this paper proposes a four-vector high-frequency injection (FV-HFI) sensorless method for online inductance identification of SynRM. Based on the accurate magnetic saturation model of the d-axis inductance, the method achieves accurate angle estimation by online estimation of the q-axis self-inductance and dq-axis mutual inductance. The simulation model is constructed based on the test results of a 3kW SynRM motor. The validity of the proposed method in suppressing the angle estimation error is verified by this simulation platform.