Adaptive Step-Size Predictive PLL Based Rotor Position Estimation Method for Sensorless IPMSM Drives

The fixed-gain position observer for sensorless interior permanent magnet synchronous motor (IPMSM) drives requires repeated trials for parameter tuning and has poor dynamic response capability. A novel adaptive step-size predictive phase-locked loop (ASS-PPLL) based rotor position estimation method is proposed to improve dynamic performance in this article. A cost function using position-tracking error decoupled from a high-frequency current response through a high-frequency square-wave injection is established. In addition, the step-size and direction are automatically adjusted by the predefined cost function to speed up the iterative search for an optimal rotor position estimate in a finite position set. Compared with the fixed-gain observers, the proposed ASS-PPLL effectively improves dynamic performance without a complex and time-consuming parameter tuning process. Compared with the conventional predictive PLL, the proposed method reduces the computational burden with fewer iterations, while ensuring the position estimation accuracy. Finally, the effectiveness of the proposed ASS-PPLL is comprehensively verified on a 2.2-kW IPMSM drive platform.