Vibration suppression control for AMB system based on static learning process and feedforward current compensation.

Synchronous vibration caused by rotor eccentricity is an important obstacle to the development of an active magnetic bearing (AMB) system. To solve this problem, a novel autobalancing control strategy has been proposed, which includes a static learning process and dynamic feedforward current injection process. In this algorithm, a dynamic response process under unbalanced force is simulated by a static learning process. Then, a transfer function containing amplitude and phase information is obtained. According to the obtained transfer function information, a specific feedforward current is constructed to eliminate the synchronous component in the control current and realize the autobalancing of vibration displacement. Computational complexity compared with conventional control strategies is further investigated. Moreover, the velocity sensor is no longer needed in the proposed control strategy. The simulation and experimental results of the AMB system show that the synchronous component of the control current and the amplitude of the rotor vibration displacement can be significantly reduced. [ABSTRACT FROM AUTHOR]