Sensorless control for IPMSMs based on a multilayer discrete-time sliding-mode observer

Abstract

Because of the high robustness to system structure and parameter uncertainties, discrete-time sliding-mode observer (DSMO) has become a promising candidate for sensorless digital control of interior permanent magnet synchronous machines (IPMSMs). However, it is challenging to achieve fully sliding mode under a low sampling frequency, especially for high speed and heavy load applications. In order to overcome this problem, a multilayer DSMO is proposed to estimate the rotor position from the extended back electromagnetic force (EMF). To guarantee a fast response to speed and torque variations, the width of the boundary layer of the DSMO is designed to be adaptive to speed and torque variations. A parameter adaption scheme based on control law and sliding-mode dynamics is proposed for selecting the width of the boundary layer and the observer gain. The proposed multilayer DSMO is validated by simulations and experimental results on a 150 kW IPMSM drive system used for hybrid electric vehicles.