LMS-based Notch Filter for AMB Unbalance Compensation Control

The stability and other properties of the active magnetic bearing system influenced by periodic vibration impact with the same frequency of the rotor speed are analyzed. Firstly, on the basis of the periodic displacement signal, the filtering theory of least mean square (LMS) algorithm and the deficiency of LMS algorithm in the application of AMB system, such as frequency mismatch, the relation of step-size and the frequency of displacement signal, and the mutual influence among the step-size length and frequency of the rotor and the PM controller coefficient, etc are studied. And then, a new strategy of variable step-size and real-time switching control with LMS and PID combination to filter the sinusoidal displacement signal to achieve adaptive force balance compensation is proposed. Finally, the feasibility of the new method for the adaptive compensation of the vibration of AMB system is tested by the experimental results.