Permanent magnet synchronous motor active disturbance rejection control based on ultra-local modeling and improved extended state observer

Abstract

In this paper, we propose an active disturbance rejection control (ADRC) method—termed ADRC-IESO-ULM—that integrates an improved extended state observer (IESO) with an ultra-local modeling (ULM) adaptive quadratic sliding mode control strategy. This approach is designed to accelerate system response and effectively counter both external and internal disturbances in a permanent magnet synchronous motor (PMSM). First, we enhance the conventional extended state observer to better capture time-varying perturbations affecting the control system. Moreover, we employ a polynomial approximation to approximate a new fal function, addressing the derivative discontinuity issues at the segmentation points of the traditional fal function and thereby improving the overall control performance of the PMSM system. Second, the ULM technique eliminates the need for explicit state observer construction, offering robust performance while avoiding the instability problems associated with improperly designed observers. Finally, simulation and experimental results demonstrate that the IESO can more accurately track system state variables and errors and that the ADRC-IESO-ULM method achieves faster response and superior disturbance rejection compared to both the proportional-integral (PI) method and the traditional ADRC approach.