High-Precision Sensorless Control of High-Speed Permanent Magnet Synchronous Motor Based on the Prediction Methodology

In this paper, a high-precision sensorless control strategy is proposed for the high-speed permanent magnet synchronous motor (HSPMSM). The strategy is based on the prediction methodology and utilizes the HSPMSM mathematical model with filters. On the one hand, a continuous variable boundary-layer sliding mode observer (CVBSMO) is developed to mitigate the buffeting problem caused by rapid sliding mode motion. On the other hand, a position error feedforward compensation method is proposed to enhance the accuracy of HSPMSM position estimation, accounting for various delays and inverter nonlinearities. These two combinations perfectly overcome the influence of non-ideal factors on position estimation, effectively. It can obtain the total position estimation error without necessitating additional controllers and is highly practical. The control performance of the proposed strategy is verified on a 1.1kW, 30000r/min HSPMSM experimental platform, and compared to the traditional method. The experimental results demonstrate the feasibility and effectiveness of the proposed strategy.