摘要

This paper proposes a position estimation strategy for the sensorless control of permanent-magnet synchronous motors (PMSMs) based on an improved sliding-mode observer (SMO) and a tangent function based phase-locked loop (PLL) structure. The improved SMO adopts a rotor speed-related adaptive feedback gain, and is able to derive a flux model-based estimator that contains rotor position and speed direction information. To extract accurate rotor position from the proposed SMO and reduce position estimation errors in both forward and reverse rotation of the PMSM, a tangent function-based PLL structure is established. The proposed SMO together with the PLL structure realises a solution to position and speed estimation for sensorless PMSM drives. Compared with the conventional back electromotive force (EMF)-based position estimator, the proposed position estimation strategy has advantages of simple design and robust estimation performance at a wider speed range. Effectiveness of the proposed method has been validated with simulations on virtual platform.