Sparse-assisted blade tip timing signal decomposition and automatic resonance region identification method

Abstract

Rotating blades are affected by various factors such as rotor–stator interaction, rotor vibration, unstable fluid excitation, and foreign object damage during operation, these factors can induce synchronous and asynchronous vibrations. Typically, the vibration states require offline analysis and diagnosis by experienced engineers. To achieve online automatic identification of synchronous and asynchronous resonance states of rotating blades, a blade tip timing (BTT) signal decomposition method based on sparse auxiliary optimization is proposed, which extracts the synchronous and asynchronous resonance signal components. In addition, automatic identification methods for synchronous and asynchronous resonance regions based on short-time energy (STE) and short-time zero-crossing rate (STZCR) are proposed separately. Finally, the effects of optimization parameters, transient response and vibration coupling effect on vibration signal decomposition and resonance region identification are revealed by means of numerical simulation and experimental verification. This study provides an effective approach for the online analysis of BTT signals and the automatic identification of resonance regions, especially provides effective theoretical support for blade fault diagnosis based on asynchronous vibration signal.