Error Analysis and Design of Sliding-Mode-Observer-Based Sensorless PMSM Drives Under a Low Sampling Ratio

Sliding-mode observer (SMO) is widely used to achieve high-performance sensorless control of permanent-magnet-synchronous-motor drives. However, the conventional SMO presents increased position estimation error as the sampling ratio decreases even without parameter errors. It is shown in this article that the position error resulted from an SMO itself can be decomposed into three components, which are caused by model discretization, nonideal sliding-mode motion, and phase-locked loop (PLL), respectively. The first two error components are opposite in sign but the sum of them may be not zero, which causes steady-state estimation error, while the last one only exists during transient processes. To address the aforementioned issues, steady-state error components are theoretically analyzed and quantified. Then, calculation of back-electromotive force is redesigned with an improved SMO to eliminate the position estimation error under a low sampling ratio. After that, a flux-compensation-based PLL is proposed to suppress the position error during fast speed variations. Finally, simulation and experimental tests were carried out to validate the effectiveness of the proposed method. The obtained results, along with a video demonstration, justify theoretical analysis and satisfactory performance of the proposed method.