An Embedded Magnetic Encoder-Based Rotor Position Estimation for Switched Reluctance Machine

Abstract

This paper investigates the possibility of applying embedded magnetic encoder (EME) consisting of three linear Hall sensors (LHS) for rotor position estimation in SRM. The key concept of this scheme is to detect the flux density characteristics (FDC) inside of stator slots. However, the armature field is to be detected in SRM, which is rather different from permanent magnet (PM) machines, where the PM field is detected. In this paper, FDC in stator armature slot openings are firstly analyzed between different winding configurations. Thus, a selection principle of winding configuration is proposed. Secondly, the proper installation location of LHS is investigated, where the sensitivity of FDC to location deviation is discussed. Accordingly, a sensor holder is specifically designed and manufactured to reduce location deviation. Thirdly, based on the detected FDC, a simplified multiple complex-coefficient filter (MCCF) is utilized as a filter to extract the positive sequence of the fundamental component of LHS output signal. As a result, the angular frequency and rotor position can be tracked by the synchronous reference frame-based phase-locked loop (SRF-PLL). Finally, the proposed method is verified by a series of experiments under steady- and transient-state conditions, which confirm the feasibility of EME scheme.