Composite Disturbance Compensation for Pmsm Motorized Spindle Systems Using Proportional-Integral and Sliding Mode Observers

Abstract

Electric spindle systems based on permanent magnet synchronous motors are widely used in high-speed machining and precision manufacturing. However, their speed control performance can be affected by mechanical friction, parameter variations, load fluctuations and unmodelled disturbances, particularly under low-speed reverse operation and dynamic tracking conditions. To improve disturbance immunity and tracking accuracy, this paper proposes a composite disturbance compensation strategy based on frequency-domain decomposition. First, a proportional-integral observer is designed to estimate and compensate for low-frequency, slowly varying disturbances. Subsequently, a sliding-mode observer is introduced to suppress the remaining fast dynamic disturbance residuals, including friction lag, high-frequency current harmonics and measurement uncertainties. By integrating these two observers, the proposed method achieves hierarchical disturbance estimation and compensation across different frequency ranges. Experimental results under various tracking conditions demonstrate that the proposed strategy effectively reduces speed asymmetry, peak error and residual disturbances, thereby improving the dynamic response and steady-state accuracy of the electric spindle system.