Automatic balancing for magnetically suspended high-speed motor based on adaptive variable step-size LMS algorithm

Because of a large number of external imbalances in the magnetically suspended high-speed motor drive system, an adaptive variable step-size least mean square (LMS) algorithm is proposed to achieve real-time automatic balancing in working speed range. A control strategy based on switching the sign of the step-size factor in adaptive LMS algorithm is proposed to suppress synchronous current in the full speed range. By analyzing the effect of synchronous current suppression in different speeds with different step-size factor, a modified variable step-size factor with changing frequency is improved. So the rotor is in good condition with rotating around its axis of inertial before and after the cutoff frequency of the open-loop system. Experimental results indicate it can stably achieve real-time automatic balancing in the working speed range.