Prediction and Validation of the Transient Pressure Field in a Centrifugal Compressor With Vaned Diffuser for the Application in High Cycle Fatigue Stress Estimations

Abstract

Abstract In this paper, the transient pressure field in a centrifugal compressor is predicted by the Nonlinear Harmonic (NLH) method, as well as the unsteady Reynolds-Averaged Navier-Stokes (URANS) simulations and validated by transient pressure measurements. The accurate prediction of these pressure fluctuations is crucial, because they have a significant influence on the High-Cycle-Fatigue (HCF) behavior in turbomachinery applications. As the first step, excited and non-excited rotational speeds, caused by rotor-stator interactions, are identified by modal analysis performed on a single segment impeller assembly. In order to eliminate any additional pressure fluctuations caused by blade vibrations, three non-excited operating points at different rotational speed levels are selected for the transient flow simulations. The transient pressure fields predicted by these methods are validated by transient pressure measurements at three circumferentially different locations on the impeller shroud. A study of the modeling of these fast-response pressure probes in numerical calculations also falls within the scope of this work in order to identify its effect on the transient pressure amplitudes. The results of the numerical calculations and measurements are compared in the frequency domain by performing Fast-Fourier Transformations (FFT) and Short-Time Fourier Transformations (STFT) on the numerical and experimental data respectively. It is shown that the transient pressure field in a single staged centrifugal compressor is calculated accurately by both of the numerical methods in comparison to the transient flow measurements. This paper demonstrates that the numerical modeling of the fast-response pressure sensors has a significant impact on the unsteady pressure amplitudes, which needs to be taken into account for a reliable experimental validation process.