Sensorless Control of Sinusoidal Doubly Salient Electromagnetic Machine Using Synchronous Frequency Vector Resonant-Based Position Observer

Sliding mode observer (SMO) is widely used in medium- and high-speed sensorless control fields. However, existing SMOs suffer from significant position estimation deviation due to the use of a sliding mode function. In addition, an additional filter is required to suppress the harmonics in SMO, which increases the complexity and computational effort of the system. To address these problems, a synchronous frequency vector resonant (SFVR)-based position observer is proposed and applied to a sinusoidal doubly salient electromagnetic machine (SDSEM) drive in this article. The characteristic analysis for the amplitude- and phase-frequency of the SFVR reveals that the SFVR can replace the sliding mode function and filter used in existing SMOs, eliminating the position estimation deviation caused by the sliding mode function and solving the complexity issue caused by the filter. Moreover, the estimation performance and robustness of the proposed SFVR-based position observer are analyzed, and a comparison of the proposed method with the existing SMO methods is discussed. The proposed method is simple in structure, easy to implement, and improves the performance of both position and speed estimation under different operating conditions. Finally, the validity of the proposed method is verified on a SDSEM experimental platform.