Sliding-Mode Sensorless Control of PMSM With Inverter Nonlinearity Compensation

In this paper, a robust adaptive sliding-mode observer (SMO) is designed based on the surface permanent-magnet synchronous machine (SPMSM) model in rotor reference frame (γδ-axis), and an online inverter nonlinearity identification and compensation method is proposed. In order to reduce the chattering of the SMO, an adaptive law is presented to help estimate the back electromotive forces of an SPMSM; thus, smaller gains can be set for switching functions of the SMO. The small-signal model of the proposed sensorless scheme is derived for analyzing the steady-state and dynamic behavior of the sensorless scheme. Voltage distortion, caused by nonlinear characteristics of switching devices, not only causes (6k ± 1)th harmonics in phase currents but also leads to a rotor position estimation error. The equivalent amplitude of the voltage distortion can be identified based on the derived small-signal model of the proposed sensorless scheme. To improve the accuracy of the estimated voltage distortion, a recursive restricted total least squares is applied to obtain the estimated amplitude of the voltage distortion. Experimental results validate the proposed sensorless control scheme and its effectiveness.