Online Parameter Identification of PMSM for Hybrid Locomotive Based on FFRLS

Aiming at the problems that the operating conditions of permanent magnet synchronous motors (PMSM) in the traction system of hybrid shunting locomotives are complex and the parameters are prone to change with temperature rise and load disturbance, this paper proposes an online parameter identification method based on forgetting-factor recursive least squares (FFRLS). The method identifies stator resistance, stator inductance and permanent magnet flux linkage in real time during motor operation. To verify the method under a locomotive-oriented power level, simulation and experiments are conducted on a 7.5 kW PMSM platform. The platform is used as a 40-times power-and-torque reduced-scale object of a 300 kW traction PMSM. The results show that the FFRLS method has fast convergence and practical online implementation capability on a DSP28335-based controller. The research provides a feasible parameter-tracking scheme for improving the stability and energy-efficiency optimization of hybrid shunting locomotive traction control.