Application of Full-Order and Simplified EKFs to Sensorless PM Brushless AC Machines

This paper employs an extended Kalman filter (EKF) to estimate the rotor position and speed of a vector controlled surface-mounted permanent magnet (PM) brushless AC (BLAC) motor from measured terminal voltages and currents only. Both full-order and simplified EKFs are employed and their simulated performance capabilities are compared. Excellent agreement is achieved between estimated and commanded results. The EKF is also employed to identify the stator flux-linkage due to the PMs, which is influenced by temperature variation and magnetic saturation.