Adaptive Nonlinear ESO-Based Global Sliding-Mode Control of EV-PMSM With Unmeasurable Disturbance

As core traction components of electric vehicles (EVs), permanent magnet synchronous motors (PMSMs) directly determine EV efficiency, dynamic response, and driving safety. Thus, mitigating the adverse effects of unmeasurable lumped disturbances in PMSMs on control performance is crucial for enhancing overall EV performance. Previous methods suffer from insufficient disturbance rejection capability and low tracking accuracy, hindering EV driving smoothness. To address these issues, an improved global sliding-mode control method based on an adaptive nonlinear extended state observer is proposed. First, an adaptive nonlinear extended state observer (ANESO) is designed to estimate unmeasurable lumped disturbance, where its gain adaptively adjusted based on both state estimation error and tracking error. Then, an improved global sliding-mode control law (IGSMCL) is presented by employing a fused error that comprehensively integrates the tracking error and disturbance estimation error, aiming to ensure the global sliding-mode robustness and further improve tracking performance. Finally, the unmeasurable lumped disturbance estimated by the ANESO is adopted for compensation in the IGSMCL. Experimental results on a PMSM test suit with three kinds of disturbance validate the superiority of the proposed controller. More especially, the ANESO provides better disturbance rejection ability than the traditional extended state observers (ESOs) because its average disturbance estimation error is less than 0.16 % of the maximum disturbance amplitude and the accuracy improvement rate is larger than 69.4 %. Taking tracking accuracy into account, the proposed method achieves the less steady-state speed tracking error (< 0.00032 rad/s). Its average tracking error outperforms four competitors with the accuracy improvement rate more than 95 %. This confirms its suitability for high-performance EV-PMSM control.