Online unbalance compensation of a maglev rotor with two active magnetic bearings based on the LMS algorithm and the influence coefficient method
Zhou Jian,Wu Huachun,Wang Weiyu,Yang Kezhen,Hu Yefa,Guo Xinhua,Song Chunsheng
DOI: https://doi.org/10.1016/j.ymssp.2021.108460
IF: 8.4
2022-03-01
Mechanical Systems and Signal Processing
Abstract:Traditional rotor dynamic balancing requires equipment shutdown to add or subtract a counterweight mass so that a rotor can reach a certain level of dynamic balancing. However, a rotor may still have some residual unbalance. As the unbalance condition changes due to factors such as load changes and rust, the original dynamic balancing will be compromised. To solve this problem, an online unbalance compensation algorithm with active magnetic bearings (AMBs) based on the least mean squares (LMS) method and the influence coefficient method (ICM) was proposed. This algorithm first applies the LMS to extract the rotor unbalance vibration signal in real time. By implementing the AMB, the trial signal with the same frequency and phase with the rotor unbalance vibration signal is added to each AMB of the balancing plane, instead of mass addition or subtraction in the traditional dynamic balancing process. An eddy current displacement sensor is used to measure the rotor displacement response in each measuring plane, and the rotor unbalance vibration compensation current can be calculated to realize the online unbalance compensation during the normal operation. The online unbalance compensation converged through the iterative algorithm. The principle of the algorithm is simple and does not depend on the model of the controller. The effectiveness of the algorithm was verified by the online unbalance compensation experiment of a maglev rotor. The experimental results revealed that the vibration amplitude of the rotor decreased by 83.6%, 84.5%, 68.6%, and 84.3% after three iterations of the algorithm in each direction, and the amplitude of the main harmonic declined by 94.2%, 96.4%, 90%, and 96.1%. The vibration amplitude of the main harmonic in the Y direction at the left AMB specifically decreased by 28.8 dB. Moreover, the adaptability of the proposed algorithm to different unbalance masses and the phase of the rotor was verified through experiment validation.
engineering, mechanical