Smooth Low-Speed Sensorless Control of Switched Reluctance Machine With Heavy Load

Abstract

For the switched reluctance machine (SRM) with heavy load, the inductance of the conduction phase is nonlinear, which makes the sensorless control method based on the linear region of the phase inductance no longer applicable. Meanwhile, in the case of light or no load, the estimated rotor position will abruptly change at the connection points between neighboring segments, which has an impact on the estimation of the speed. In this paper, a low-speed sensorless control method for the SRM with heavy load is proposed. The idle-phase inductance is modeled over a wide rotor position range for position estimation, and a proportional gain decreasing type second-order generalized integrator based on a frequency-locked loop (SOGI-FLL) is constructed to estimate the speed while performing rotor position filtering. The effectiveness of the proposed method is experimentally verified with a 3-phase 12/8-pole SRM prototype, and the results show that the estimated rotor position and speed are accurate and smooth, which is beneficial to performance improvement of the sensorless control.