Compressible effects modeling for turbulent cavitating flow in a small venturi channel: An empirical turbulent eddy viscosity correction

2021 ◽  
Vol 33 (3) ◽  
pp. 035148
Author(s):  
Xin-Lei Zhang ◽  
Ming-Ming Ge ◽  
Guang-Jian Zhang ◽  
Olivier Coutier-Delgosha
Keyword(s):  
Eddy Viscosity ◽  
Cavitating Flow ◽  
Turbulent Eddy

Flow Analyses in a Single-Stage Propulsion Pump

1996 ◽  
Vol 118 (2) ◽  
pp. 240-248 ◽  
Author(s):  
Y. T. Lee ◽  
C. Hah ◽  
J. Loellbach
Keyword(s):  
Boundary Layer ◽  
Numerical Method ◽  
Turbulence Model ◽  
Eddy Viscosity ◽  
Single Stage ◽  
Acoustic Properties ◽  
Turbulent Eddy ◽  
Trailing Edge ◽  
Local Flow ◽  
Blade Row

Steady-state analyses of the incompressible flow past a single-stage stator/rotor propulsion pump are presented and compared to experimental data. The purpose of the current study is to validate a numerical method for the design application of a typical propulsion pump and for the acoustic analysis based on predicted flowfields. A steady multiple-blade-row approach is used to calculate the flowfields of the stator and the rotor. The numerical method is based on a fully conservative control-volume technique. The Reynolds-averaged Navier–Stokes equations are solved along with the standard two-equation k–ε turbulence model. Numerical results for both mean flow and acoustic properties compare well with measurements in the wake of each blade row. The rotor blade has a thick boundary layer in the last quarter of the chord and the flow separates near the trailing edge. These features invalidate many Euler prediction results. Due to the dramatic reduction of the turbulent eddy viscosity in the thick boundary layer, the standard k–ε model cannot predict the correct local flow characteristics near the rotor trailing edge and in its near wake. Thus, a modification of the turbulence length scale in the turbulence model is applied in the thick boundary layer in response to the reduction of the turbulent eddy viscosity.

Reynolds Stress Anisotropy Based Turbulent Eddy Viscosity Model Applied to Numerical Ocean Models

2011 ◽  
Vol 133 (6) ◽  
Author(s):  
Subhendu Maity ◽  
Hari Warrior
Keyword(s):  
Reynolds Stress ◽  
Eddy Viscosity ◽  
State Of The Art ◽  
Test Case ◽  
Turbulent Eddy ◽  
Ocean Turbulence ◽  
Stress Anisotropy ◽  
Eddy Viscosity Model ◽  
Degree Of Anisotropy ◽  
Ocean Models

The present state-of-the-art ocean models use an eddy viscosity that depends on structure parameter (Cμ). In this paper we use a Reynolds stress anisotropy based formulation for the eddy viscosity because in addition to the value of turbulent kinetic energy, it also depends on the degree of anisotropy. The formulation is incorporated into the General Ocean Turbulence Model (GOTM) and simulated using the famous test case of Ocean Weather Station (OWS) Papa experiment. Even if there is not much of an improvement in terms of results with this model, it can be very easily incorporated into the ocean models removing cumbersome equations for structure parameters.

Low-reynolds number approximation for turbulent eddy viscosity

1994 ◽  
Vol 9 (3) ◽  
pp. 283-292 ◽  
Author(s):  
A. Yakhot ◽  
S. Rakib ◽  
W. S. Flannery
Keyword(s):  
Reynolds Number ◽  
Eddy Viscosity ◽  
Low Reynolds Number ◽  
Turbulent Eddy ◽  
Low Reynolds

Improvement in Numerical Prediction of Cavitating Flows Over Various 2D Geometries Using Modification to the Turbulence Model

2007 ◽  
Author(s):  
Arvind Jayaprakash ◽  
Kartikeya Mahalatkar ◽  
Urmila Ghia ◽  
Karman Ghia
Keyword(s):  
Turbulence Model ◽  
Eddy Viscosity ◽  
Cavity Length ◽  
Cavitating Flow ◽  
Turbulence Modification ◽  
Vapor Cloud ◽  
Shedding Frequency ◽  
Naca 0015 ◽  
Vapor Region ◽  
Liquid Vapor

Cavitation often causes performance breakdown and damage. So, it is very essential to accurately predict and control this phenomenon. In the present study, the unsteady effects associated with cavitation are investigated for various geometries including a NACA 0015 hydrofoil, a convergent-divergent nozzle, and a wedge, using the flow solver FLUENT. The turbulent viscosity and/or the turbulence dissipation in the k-epsilon turbulence model are modified. The cavitation phenomenon is represented based on the full cavitation model developed by Singhal et al. (2002), and it considers the liquid-vapor mixture as a homogeneous fluid whose density varies with respect to the static pressure and whose mass fraction is known in advance. Also, this model takes into account the formation and collapse of the vapor bubbles. The k-epsilon model was originally developed for fully incompressible fluids, and does not account for highly compressible two-phase mixtures. Hence, it has been found to be unsatisfactory for predicting cavitating flow in presence of high compressibility in the vapor region. Coutier-Delgosha et al. (2001) attributed this to the over-prediction of eddy viscosity in regions of flow with high vapor concentration, and suggested a modification for the calculation of eddy viscosity. Though the modification works in capturing the dynamic behavior of the cavitation sheet, the accuracy of cavity length and frequency are not accurately predicted for high cavitation numbers. This is due to inability of Coutier-Delgosha’s turbulence modification to completely account for all the complex flow features present in the cavity closure region. Thus, a further modification based on geometry and cavitation type is introduced in the turbulence modification of Coutier-Delgosha. Better results were obtained for moderate cavitation numbers, but this modification failed to accurately predict the frequency of vapor-cloud shedding at high cavitation numbers. This discrepancy is attributed to the large (40000:1) variation of density in the liquid-vapor region. Hence, a new modification is suggested in the present work where the closure coefficients of dissipation production (C1epsilon) and dissipation (C2epsilon) in the turbulent dissipation equation are dynamically varied in the liquid-vapor region. A User-Defined Function (UDF) is implemented in FLUENT to achieve this dynamic variation of the above mentioned closure coefficients. This modification is being tested to predict the time-averaged cavity length and vapor-cloud shedding frequency of cavitating flow over a NACA 0015 airfoil. The poster will present comparisons of cavity length and vapor-cloud shedding frequency over a wide range of cavitation numbers as predicted by the present and previous turbulence modifications and those observed in experimental studies.

NUMERICAL UNDERSTANDING OF THE SELF-INDUCED OSCILLATION OF FLOW PAST THE OPEN CAVITY WITH A STORE

2010 ◽  
Vol 24 (27) ◽  
pp. 5295-5307
Author(s):  
HAI-QING SI ◽  
TONG-GUANG WANG
Keyword(s):  
Experimental Data ◽  
Eddy Viscosity ◽  
Cavity Flow ◽  
Open Cavity ◽  
The Self ◽  
Turbulent Model ◽  
Free Shear Layer ◽  
Turbulent Eddy ◽  
Fully Implicit ◽  
Power Spectral

The flows around a typical open cavity with a store are numerically simulated in this paper. A fully implicit unfactored method is employed in the solver of RANS equations, where the S-A turbulent model is implemented to calculate the turbulent eddy viscosity. The flow-induced oscillation of the free shear layer around the cavity lip is captured and analyzed. Comparisons for the different cases of the cavity flow are made. The calculated results show that the pressure fluctuation is pertinent to the different positions of the store inside the cavity. The power spectral density (PSD) of the pressure by fast Fourier transformations (FFT) is also displayed in the paper. For the Rossiter's oscillatory mode, the numerical results compare well with the experimental data.

Sign in / Sign up

Export Citation Format

Share Document