Research on the preprocessing of vibration signals of rotating blades and the identification of asynchronous vibration oriented to blade tip timing vibration measurement

Blade tip timing vibration measurement technology is gradually replacing the traditional contact measurement technology and becoming a new way for the dynamic monitoring of rotating blades. However, during the actual measurement process, the vibration signals of the blades are inevitably contaminated by noise. This paper proposes a simple yet effective preprocessing method for vibration signals of rotating blades by integrating Ensemble Empirical Mode Decomposition with the wavelet threshold technique. In this paper, the experiment of identifying the asynchronous vibration parameters of rotating blades at a constant speed is carried out on a high-speed rotating blade test bench. The results show that: after denoising and smoothing using this method, the removal effect of noise spikes is quite ideal, and the identification of the asynchronous vibration of rotating blades is successfully achieved.