关键词: 叶端定时, 共振识别, 稀疏重构, 裂纹识别

Abstract: The blade tip timing (BTT) method facilitates non-contact vibration measurement for real-time online monitoring of rotating blade cracks. One of its primary challenges is the reconstruction of undersampled blade tip timing signals. This paper introduces the Cluster contraction Stage wise Orthogonal Matching Pursuit (CS-OMP) algorithm to address this issue, proposing a CS-OMP-based method for reconstructing undersampled vibration signals of compressor blades in aeroengines. The proposed method aims to improve the reconstruction accuracy and real-time performance of blade vibration signals. Additionally, it identifies blade cracks by detecting changes in the first-order modal frequen-cy of rotating compressor blades. The simulation analysis verified that this method can accurately identify the single-modal and multi-modal vibration parameters of blades. Furthermore, blade crack identification experiments were con-ducted using a scaled-down compressor test component of an aero-engine. The results demonstrated that this meth-od outperforms the traditional Orthogonal Matching Pursuit (OMP) algorithm in blade vibration reconstruction and achieves accurate identification of blade cracks.

Key words: blade tip timing, resonance identification, sparse reconstruction, Crack Identification