A blind estimation of speed and rotor position of permanent magnet synchronous motor from the measurements of stator currents using extended Kalman filter

A number of techniques have been developed for estimation of speed or rotor position in motor drives. In this paper, an extended Kalman filter (EKF) is developed for the estimation of speed and rotor position using only the measurements of three phase stator currents. Conventional mechanical sensors are not needed for this estimation. The paper proposes the modelling of the dynamic equations of the permanent magnet synchronous motor (PMSM) to describe the dynamics in terms of direct axis current and voltage, quadarture axis current and voltage, rotor position and synchronous speed. Unlike the earlier papers in this direction, the present work does not take any input like vd or vq into the Kalman filter, but uses only the sensed line currents as measurements. The performance evaluation of the extended Kalman filter with online estimation of the speed and rotor position of a PMSM is presented with simulation results, for a varying load torque. © 2013 Inderscience Enterprises Ltd.