Development of a Novel Pressure-Based Approach for Blade Synchronous Vibration Detection Including Mistuning

Abstract

Abstract. The monitoring of blade forced vibration is crucial for compressor operational safety and integrity. Conventional techniques, such as strain gauge and blade tip timing, are often expensive and susceptible to failure in harsh environments. Traditional unsteady pressure measurement presents a cost-effective alternative but suffers from a low signal-to-noise ratio due to spectral aliasing with the blade passing frequency. Thus, numerous probes are necessary to identify the signal related to blade vibration and obtain the vibration level. This study reveals that the blade mistuning breaks this aliasing by inducing the distinct spectral peaks of the unsteady pressure signal in the frequency domain. Leveraging this phenomenon, we develop a novel method to reconstruct the maximum vibration amplitude using a small quantity of unsteady pressure probes. A 1.5-stage compressor is chosen to validate the effectiveness of the proposed method. First, a mistuning model based on the component mistuning model and three-dimensional unsteady computational fluid dynamics simulations are applied. These simulations provide the mistuning amplification coefficients of blade amplitudes for different nodal diameters, along with the pressure sensitivity coefficients linking the vibration amplitude to unsteady pressure. The experimental data from casing-mounted unsteady pressure sensors are combined with the model to estimate the maximum blade amplitude. The results show a good agreement with blade tip timing measurements, demonstrating the method's reliability. This work advances online vibration monitoring techniques for turbomachinery rotor blades by exploiting the unsteady pressure characteristics. The proposed method offers a cost-effective solution for real-time measurement of synchronous blade vibration amplitudes.