A Multifrequency Disturbances Identification and Suppression Method for the Self-Sensing AMB Rotor System

The self-sensing active magnetic bearing (SAMB) rotor systems own the benefits of reduced cost and better system integration, from the substitution of rotor displacement estimation for external sensors. However, the self-sensing errors between the real and estimated rotor displacement, together with the unbalance force and sensor runout, will act as disturbances that affect the SAMB rotor system, inducing current fluctuations and rotor vibrations. This paper presents a disturbance suppression method for the modulation-type SAMB rotor systems, in which both the amplitude and phase of disturbances are identified and real-time updated. A lumped multifrequency disturbances model is established. And a disturbances identification algorithm with object function value-based step vectors is proposed. The system stability and method convergence conditions are analyzed. The convergence rate, overshoot and antinoise capacity of the method are verified through simulations. In the end, the effectiveness and performance of the suppression method are validated through steady state and varying rotational speed experiments on a 4-DOF radial SAMB rigid rotor platform.