A DC Error Suppression Adaptive Second-Order Backstepping Observer for Sensorless Control of PMSM

In the sensorless control system of permanent magnet synchronous motor, there are inevitable dc bias errors in phase currents and phase voltages due to factors such as measurement errors and inverter nonlinearity, which reduce the reliability of the system and the accuracy of rotor speed and position estimations. This article presents a dc error suppression adaptive second-order backstepping observer (ASOBO) to estimate the back electromotive force (BEMF). First, the ASOBO is inversely designed by constructing the second-order transfer function of the BEMF, which can completely eliminate the dc error and effectively suppress the high-frequency harmonics, and its harmonic suppression performance is not affected by the speed. Then, in order to reduce the estimation errors of rotor speed and position during the input as a frequency ramp signal, a position estimation model is constructed and an adaptive law for speed estimation is derived. Meanwhile, the parameters of this position estimation model are configured to further improve the filtering capability of the system. Finally, the proposed observer is compared with existing observers on an experimental platform, and the experimental results verify the effectiveness of the proposed scheme.