An improved normalized PLL-based high-order SMO for Sensorless Control of PMSM

When the permanent magnet synchronous motor (PMSM) is reversed, using the traditional sliding-mode observer (SMO), there is a 180° phase difference between the actual electrical angle and the estimated value. This paper proposes an improved normalized phase-locked loop-based high-order sliding-mode observer (HSMO-INPLL) method to solve this problem. First, HSMO is used to observe the current and back-EMF to reduce the chattering of SMO and the phase delay caused by the low-pass filter (LPF). Secondly, the 180° phase difference in the reverse rotation of traditional SMO is analyzed. Finally, the angle and speed information are extracted by INPLL to solve this problem. The simulation model and ARM experimental platform are built. Simulation and experimental results show that HSMO-INPLL can significantly improve the velocity observation range and signal-to-noise ratio (SNR) of SMO, and it is highly efficient and feasible.