A stepwise parameter identification method of interior permanent magnet synchronous motor across full speed range based on digital twin

This paper proposes a stepwise identification method for the electrical parameters of an interior permanent magnet synchronous motor (IPMSM) system driven by a three-phase inverter, based on digital twin (DT) technology. This method achieves the identification of key internal parameters of the IPMSM without requiring signal injection, additional sensors, or other measurement equipment. Firstly, a model of the IPMSM driven by a three-phase inverter is established. The model is discretized and solved using the fourth-order Runge-Kutta method. Subsequently, the Sobol sensitivity analysis method is employed to analyze the influence degree of each key parameter under different objective functions and across different waveform segments. An optimized objective function design is proposed, which not only reduces the optimization pressure on the search algorithm and accelerates the optimization speed but also enhances the accuracy of parameter identification. Then, the particle swarm optimization (PSO) algorithm is utilized to iteratively optimize the parameters in the established DT model, enabling the estimation of multiple parameters. Finally, theoretical validation is conducted using full-speed-range Simulink simulation data of the IPMSM.