Reconstruction of blade tip-timing signals based on the MUSIC algorithm
Zhibo Liu,Fajie Duan,Guangyue Niu,Dechao Ye,Junnan Feng,Zhonghai Cheng,Xiao Fu,Jiajia Jiang,Jing Zhu,Meiru Liu
DOI: https://doi.org/10.1016/j.ymssp.2021.108137
IF: 8.4
2022-01-01
Mechanical Systems and Signal Processing
Abstract:<p>As a multi-frequency identification method, the MUSIC method has received more and more attention in the field of blade tip-timing (BTT) signal reconstruction. So far, an improved MUSIC method and a subspace dimension-reduced MUSIC (SDR-MUSIC) method were proposed to analyze BTT signals. However, they must rely on the expanded snapshot matrix to overcome the spectrum aliasing. The computational complexity of these methods was analyzed. The computation time of BTT signal reconstruction increased cubically as the size of the expanded matrix increased. This is unfavorable for real-time processing. In this paper, reconstruction conditions based on the MUSIC algorithm were proposed. This makes the traditional MUSIC method overcome spectrum aliasing without expanding the snapshot matrix. The computational complexity is greatly reduced. The feasibility of the reconstruction conditions was verified through simulations and aero-engine experiments. Under the reconstruction conditions, the frequency identification accuracy of the traditional MUSIC can reach that of the improved MUSIC method. Besides, an amplitude identification method was proposed based on the discrete Fourier transformation (DFT) and remainder theorem. It provides a feasible solution for the spectrum methods that cannot directly extract amplitude from the pseudo spectrum, such as the improved MUSIC, SDR-MUSIC, etc. The error of the amplitude identification was analyzed by simulations. And its feasibility was further verified by identifying all blade vibration amplitudes in the aero-engine experiments.</p>
engineering, mechanical