基于自适应浸入和不变理论的转子多频振动靶向抑制

Targeting Suppression of Multi-frequency Rotor Vibration Based on Adaptive Immersion and Invariance Theory

旋转机械转子在运行过程中受到周期性干扰力的激励,振动频率多与转子转速相关,呈现出同频、整数倍频和分频的特点。针对旋转机械所受的多频激励,将转子振动的各频率成分作为抑制靶点,将电磁作动器作为执行机构,提出一种基于自适应浸入和不变理论的转子多频振动靶向抑制方法。该方法在笛卡尔转换坐标系下对多频激励进行矢量分解,将高阶非线性系统浸入到低阶目标系统,且设计的控制算法使目标系统的像具有不变流形吸引性。设计陷波滤波器,提取振动信号的频率成分,该滤波器可根据实时转速自调整陷波频率,以应对转速波动的影响,满足多转速抑制的需要。建立转子-轴承-电磁作动器系统的有限元模型以及实时控制试验台,仿真和试验结果表明,所提方法可以有效地抑制转子多频振动,且鲁棒性较好。

Rotor is excited by periodic disturbances in the rotating machine during operation.The frequencies of vibration of rotor are mainly related to the speed of the rotor.The excitation frequencies present the characteristics of integral multiple or integral fraction of rotate speed.Aiming at the multi-frequency excitations in the rotating machinery,the multiple frequency components of the rotor vibration are set as the suppression target in this work.A multi-frequency rotor vibration targeting suppression approach is proposed based on the adaptive immersion and invariance theory using the active magnetic actuator.A Cartesian transformation coordinate method is used for the disturbing force vector decomposition,and the higher-order nonlinear system is immersed into a lower-order target system.The designed control law makes the image of the target system become an invariant and attractive manifold.Notch filters are designed from the view of the frequency domain to extract the frequency components of vibration signal,this filter can self-adjust its notch frequency based on the real-time rotate speed to satisfy the need for multi-speed control and avoid the effect of speed fluctuation.The model of the rotor-bearing-active magnetic actuator finite element model and real-time control test rig are established respectively.The results simulations and experiments show that the proposed method can suppress the multi-frequency rotor vibration effectively,and the algorithm has better robustness.