Enhanced Linear Active Disturbance Rejection Control With DC Offset Suppression for High-Accuracy Position Estimation in PMSM Sensorless Drives

In sensorless control systems of permanent magnet synchronous motor (PMSM), the back electromotive force (BEMF) estimated by conventional extended state observer (ESO) often suffers from phase lag and fails to effectively suppress dc bias and high-frequency noise. Furthermore, the phase-locked loop (PLL) inevitably introduces dynamic position tracking errors during acceleration and deceleration, while exhibiting insufficient robustness against ramp disturbances. To address these challenges, this article proposes a position estimation strategy combining a third-order model observer (THMO) and enhanced linear active disturbance rejection control (ELADRC). First, based on the mathematical model of the PMSM, a THMO is ingeniously designed by exploiting the differential of the current estimation error and the estimated BEMF, enabling more accurate extraction of specific periodic signals for improved BEMF extraction. Second, while retaining the advantages of the third-order observer in effectively tracking speed, a novel disturbance correction-based position observer (NDCPO) is proposed by introducing a disturbance correction factor $\alpha$, into the Linear Active Disturbance rejection Control (LADRC) framework. Finally, a linear error feedback controller is employed to replace the PI controller in the speed loop for compensating the total system disturbances. Through stability and estimation performance analysis, the proposed sensorless estimation strategy significantly improves position estimation accuracy during speed variations and enhances robustness against load disturbances. The effectiveness of the proposed method is experimentally verified on a DSP-based PMSM drive platform.