我所柳向斌老师课题组完成的论文《Adaptive Robust Sensorless Control for PMSM Based on a Wavelet Neural Network Back EMF Observer and Extended State Observer》于2026-03在《IEEE Transactions on Power Electronics》正式发表。论文摘要如下：
Abstract: Sliding mode observer (SMO) in the sensorless control system of permanent magnet synchronous motor (PMSM) results in chattering and high-frequency noise, reducing the accuracy of the back electromotive force (EMF) estimation. This article presents an adaptive wavelet neural network observer (AWWNO) design for the back EMF to overcome the indispensability or necessity of the SMO part. Through the adaptive wavelet neural network (AWNN), the AWNNO achieves chattering-free and high-precision EMF estimation based on the multiresolution analysis (MRA) approximation theory without relying on sign functions or low-pass filters. The speed controller comprises of the adaptive sliding mode control based on the integral sliding mode, and the extended state observer (ESO) in which the accurate estimates of the rotor speed and position are acquired. The proposed control scheme demonstrates strong robustness and self-correction capabilities in the presence of parameter uncertainties and external disturbances. Moreover, the convergence of the AWWNO and the stability of the speed controller are rigorously proven by using Lyapunov theory, ensuring the exponential convergence and the uniform ultimate bounded (UUB) stability. Experimental results validated the effectiveness and feasibility of the proposed robust sensorless control scheme, highlighting its nearly chattering-free, disturbance rejection and accurate estimation performance compared to the existing approaches.
引用信息：Sijie Li, Xiaoyu Zhang, Xiangbin Liu, Hongye Su. Adaptive Robust Sensorless Control for PMSM Based on a Wavelet Neural Network Back EMF Observer and Extended State Observer. IEEE Transactions on Power Electronics, 2026-03-23. DOI: 10.1109/tpel.2026.3676897