Search

How to control the vibration of many target positions other than the active magnetic bearings (AMBs) or corresponding AMB's sensors in noncollocated AMBs supporting flexible rotor systems, still is a key issue. This article focuses on the vibration control of multiple target positions of the noncollocated AMBs flexible rotor systems. First, in order to obtain the phase characteristics of noncollocated AMBs flexible rotor systems, the zero-pole distribution of noncollocated systems is studied by analytical solution of a Euler–Bernoulli beam, and by discrete numerical analysis of the flexible rotor system and sine sweeping tests. Then, a novel multiobjective iterative controller for vibration suppression is proposed, which can regulate the phase of target signals in real time by using phase characteristics of noncollocated systems. The controller's convergence is demonstrated for both single and multiobjective cases. To validate the effectiveness of the multiobjective iterative controller proposed under varied conditions such as single and multitarget, constant rotational speeds, and acceleration operations, experiments are carried out on an AMBs multidisk flexible rotor test rig. It is shown that the multiobjective iterative controller proposed can effectively realize real-time cooperative control of vibrations at multiple target positions, and control currents.

Active magnetic bearings (AMBs) are a type of mechatronics product widely used in high-speed rotating machinery. Considerable research has been conducted on the vibration rejection of the AMBs-rotor systems at the base standstill. However, little attention has been paid to the case where the rotor operates under base excitations, such as turbines mounted on ships with wave excitation. The increased rotor vibration caused by base excitations is a great challenge to operational safety. To address this issue, a base acceleration feedforward compensation (BAFC) algorithm was proposed in this article. Using the online base excitation data and the dynamic model of the rotor system, the proposed model-based algorithm determines the optimal compensation current in real time. Further, to reject the decreased compensation performance due to inaccurate modeling, a method for correcting the compensation error has been presented. Moreover, the effect of controller parameters on system stability was explored to avoid parametric instability during base excitations. By employing the BAFC algorithm in the experiments, the rotor vibration was reduced considerably under both the harmonic and random base excitations with a maximum vibration reduction of 85%, demonstrating the remarkable effectiveness of the BAFC algorithm.