Sensorless speed control of a PM synchronous motor based on sliding mode observer and extended Kalman filter

In this paper a method for rotor speed and position detection in permanent magnet synchronous motors is presented, suitable for applications where low speed and standstill operations are not required. The approach is based on the estimation of the motor back-EMF through a sliding mode observer and a cascaded Kalman filtering. Due to its characteristics, the technique can be applied to motors having distorted and whatever shaped back-EMF waveform. The equation error for the rotor position estimation is derived and discussed. Test results are presented, which show the system performance including start-up capability, which is one of the most critical conditions in back-EMF based sensorless schemes.