The active magnetic bearing system exhibits mass imbalance and sensor runout which cause the system to generate harmonic vibration force and moment. Repetitive control is an effective method to eliminate such harmonic vibration. Traditional repetitive control will eliminate all of the harmonic frequency components. However, in a practical system, the odd harmonic components usually dominate. Meanwhile, the existing method only suppresses the vibration force in the magnetic bearing system, and there is little research on the suppression of moment. Aiming at these problems, the harmonic vibration moment of the active magnetic bearing system is taken as the control object. This study investigates a hybrid control method that combines a second-order odd harmonic repetitive control with finite-dimensional repetitive control. And the virtual variable sampling is applied to construct any virtual sampling period in the proposed method, which effectively solves the problem of non-integer delay of digital repetitive control. The stability of the active magnetic bearing system is analyzed. The experimental results show that this method has faster response speed and better robustness when the frequency fluctuates.

主动式磁轴承系统表现出质量不平衡和传感器跳动，导致系统产生谐波振动力和力矩。重复控制是消除此类谐波振动的有效方法。传统的重复控制将消除所有谐波频率分量。但是，在实际系统中，奇次谐波分量通常占主导地位。同时，现有方法仅抑制电磁轴承系统中的振动力，而关于力矩抑制的研究很少。针对这些问题，以主动电磁轴承系统的谐波振动力矩为控制对象。本研究研究了一种混合控制方法，该方法将二阶奇次谐波重复控制与有限维重复控制相结合。该方法利用虚拟变量采样构造任意虚拟采样周期，有效解决了数字重复控制的非整数时延问题。分析了主动磁轴承系统的稳定性。实验结果表明，该方法在频率波动时具有更快的响应速度和更好的鲁棒性。