Current Compensation Based Sliding Mode Observer for Sensorless Control and Online Parameter Estimation of PMSM

For better performance of permanent magnetic synchronous motors (PMSM), a sensorless sliding mode observer (SMO) and an adaptive online parameter identification algorithm are proposed. An effective current compensation method is introduced to realize accurate current filtering with minimum delay. With the filtered current, an iterative SMO can precisely estimate the electromotive force (EMF). A modified phase locked loop algorithm detects the rotor position and speed from the estimated EMF. The output of SMO can be used to estimate motor parameter adaptively. The algorithm is proved to be robust and effective through a Lyapunov function and numerical simulations.