Improved performance of sliding mode observer using parameter adaptation in sensorless IPMSM drive

This paper presents a method for improving the performance of sliding mode observer (SMO) in sensorless interior permanent magnet synchronous motor (IPMSM) drive. Precise motor parameters are necessary for position estimation. Especially, in the case of the magnetic saturation, motor parameter error influences of the right motor parameter. In this proposed method, the recursive least square (RLS) method is used to identify the inductance parameters on the discrete-time model. Furthermore, the identified inductances are adapted to improve the performance of the system. Simulation results mainly verify the proposed method.