A Rotor Flux Observer with Adaptive Interfering Signal Elimination for SPMSM Sensorless Control

Abstract

In this paper, a rotor flux observer with adaptive interfering signal elimination for sensorless control of surface-mounted permanent magnet synchronous motor (SPMSM) is proposed. As we all known, the rotor flux of SPMSM can be obtained by integrator or closed-loop observer. But the interfering signals, i.e. ramp signal, dc offset and harmonics, always exist in obtained rotor flux due to unknown initial value, integral calculation, current detection errors and inverter nonlinearity. These interfering signals will greatly degrade the sensorless control performance. So an extended master-slave adaptive notch filter (EMS-ANF) with frequency adaptation is proposed to eliminate these interfering signals and extract the accurate fundamental wave of rotor flux. The theoretical analysis indicates that the EMS-ANF can eliminate the dc offset and specific harmonics in input, and avoid the ramp signal caused by integration. Then, based on EMS-ANF, a complete sensorless control scheme for SPMSM is designed. The experimental results under different conditions demonstrate the effectiveness and the superiority of the proposed method.