A Low-Speed Sensorless Control Strategy for IPMSM Under Heavy-Load Conditions Based on Flux Linkage Model Prediction

The high-frequency (HF) signal injection method suffers from increased position estimation error with load and reduced stability under heavy-load conditions. To address these issues, an improved sensorless control scheme for interior permanent magnet synchronous motor (IPMSM) based on flux linkage model prediction is proposed. By fully leveraging HF current information, an improved position error signal based on flux linkage response is constructed. Then, model prediction algorithm is employed to achieve accurate rotor position estimation. In addition, an Nth-order polynomial model is designed to describe the nonlinear properties of the motor. Simultaneously, a hybrid optimization algorithm combining a genetic algorithm and least squares method (GA-LS) is constructed to achieve high-accuracy flux linkage modeling. Furthermore, the impact of flux linkage model deviations on position estimation accuracy is quantitatively analyzed. Finally, the feasibility and effectiveness of the proposed method are verified by experiments on a 6 kW IPMSM drive platform.