Date of this Version
IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 28, NO. 12, DECEMBER 2013
Advantages such as parameter insensitivity and high robustness to system structure uncertainty make the sliding-mode observer (SMO) a promising solution for sensorless control of interior permanentmagnet synchronousmachines (IPMSMs). In practical industry applications, in order to utilize digital controllers and achieve comparable performance under a lower sampling frequency, a discrete-time or quasi-SMO (QSMO) is commonly used. However, because of the saliency of an IPMSM, the magnitude of the extended electromotive force (EMF) will change with load (torque and/or speed) variations, whichmakes it challenging for the QSMO to estimate the extended EMF accurately. Without proper observer parameters, a phase shift will be observed in the QSMOestimated rotor position when the load changes. In order to overcome this problem, an adaptive QSMO using an online parameter adaption scheme is proposed to estimate the extended EMF components in an IPMSM, which are then used to estimate the rotor position of the IPMSM. The resulting position estimation has zero phase lags and is highly robust to load variations. The proposed adaptive QSMO is implemented on a 150-kW IPMSM drive system used in heavy-duty, off-road, hybrid electric vehicles. Testing results for ramp torque changes, four-quadrant operations, and complete torque reversals between full motoring and full braking modes are presented to verify the effectiveness of the proposed sensorless control algorithm.