Frequency identification method based on active aliasing of biprobes blade tip timing without once-per-revolution

Abstract

Rotary blades are the core components of aeroengines, gas turbines and centrifugal compressors. However, due to the complexity of their work environment, monitoring their health is necessary. Blade tip timing (BTT) is an effective non-contact monitoring method for rotating blades. However, conventional BTT rely heavily on once-per-revolution (OPR) signals, which cannot be accurately obtained in complex environments. At the same time, there is a serious under-sampled problem in the BTT signal. Consequently, the extraction of blade vibration parameters from under-sampled BTT signals without relying on OPR signals remains a research hotspot. In this paper, an active aliasing method based on the vibration difference of the blade is proposed to extract the vibration frequency of the blade. In the proposed method, two BTT probes are used to calculate the blade vibration difference, and time delay estimation is used to estimate the approximate value of the natural frequency at small angle delays. Then, a series of aliasing frequencies are generated under various large angle delays. Next, the estimated natural frequency and aliasing frequency are combined to solve the exact natural frequency value. In addition, numerical simulations demonstrate the effectiveness and robustness of the method. Finally, experimental research was conducted on the BTT experimental platform. As a result, this method can accurately identify the inherent frequency of asynchronous vibration of the blades.