scholarly journals Modeling unsteady turbulent flows over ripples: Reynolds-averaged Navier-Stokes equations (RANS) versus large-eddy simulation (LES)

2004 ◽  
Vol 109 (C9) ◽  
Author(s):  
Yeon S. Chang
Keyword(s):  
Large Eddy Simulation ◽  
Turbulent Flows ◽  
Stokes Equations ◽  
Navier Stokes ◽  
Navier Stokes Equations ◽  
Eddy Simulation ◽  
Large Eddy ◽  
Reynolds Averaged Navier Stokes

Partially Averaged Navier-Stokes Turbulence Modeling of Flow in 5x5 PWR Fuel Assembly With Spacer Grid

2018 ◽  
Author(s):  
Giacomo Busco ◽  
Yassin A. Hassan
Keyword(s):  
Large Eddy Simulation ◽  
Numerical Simulations ◽  
Stokes Equations ◽  
Direct Numerical Simulations ◽  
Navier Stokes ◽  
Spacer Grid ◽  
Navier Stokes Equations ◽  
Eddy Simulation ◽  
Large Eddy ◽  
Reynolds Averaged Navier Stokes

The highly turbulent flow inside a pressurized water reactor makes unpractical the use of scale resolving simulations, due to the large number of space and time turbulent structures. The high computational cost associated with typical large eddies simulations or direct numerical simulations techniques is unsuitable due to the large spatiotemporal resolution required. Partially averaged Navier-Stokes turbulence model is presented as bridging model between Reynolds averaged Navier-Stokes equations and direct numerical simulations. As filtered representation of the Navier-Stokes equations, the model is able to continuously shift its energy-based filter, inside the turbulence spectrum, being able to resolve the turbulent scales of interest. The choice of energy based cut-off filters gives the chance to directly impose the degree of needed resolution, where the most important large scales unsteadiness are resolved at minimal computational expenses. The partially averaged Navier-Stokes modelling approach has been tested for a Reynolds number of 14,000, inside a 5 × 5 fuel bundle, with a single spacer grid and split-type mixing vanes. Four different filters have been tested, whose resolution ranged from Reynolds averaged Navier-Stokes and large eddy simulation. A comparison with large eddy simulation will be presented. The results show that the partially averaged Navier-Stokes modeling produces results comparable to those of large eddy simulation when the appropriate cut-off energy filter is chosen. The turbulence models results will be compared with the available particle image velocimetry experimental data.

Large Eddy Simulation of Turbulent-Cavitation Interactions in a Venturi Nozzle

2010 ◽  
Vol 132 (12) ◽  
Author(s):  
Nagendra Dittakavi ◽  
Aditya Chunekar ◽  
Steven Frankel
Keyword(s):  
Large Eddy Simulation ◽  
Stokes Equations ◽  
Navier Stokes ◽  
Small Scale ◽  
Navier Stokes Equations ◽  
Vapor Cavity ◽  
Eddy Simulation ◽  
Downstream Region ◽  
Large Eddy ◽  
Venturi Nozzle

Large eddy simulation of turbulent cavitating flow in a venturi nozzle is conducted. The fully compressible Favre-filtered Navier–Stokes equations are coupled with a homogeneous equilibrium cavitation model. The dynamic Smagorinsky subgrid-scale turbulence model is employed to close the filtered nonlinear convection terms. The equations are numerically integrated in the context of a generalized curvilinear coordinate system to facilitate geometric complexities. A sixth-order compact finite difference scheme is employed for the Navier–Stokes equations with the AUSM+-up scheme to handle convective terms in the presence of large density gradients. The stiffness of the system due to the incompressibility of the liquid phase is addressed through an artificial increase in the Mach number. The simulation predicts the formation of a vapor cavity at the venturi throat with an irregular shedding of the small scale vapor structures near the turbulent cavity closure region. The vapor formation at the throat is observed to suppress the velocity fluctuations due to turbulence. The collapse of the vapor structures in the downstream region is a major source of vorticity production, resulting into formation of hair-pin vortices. A detailed analysis of the vorticity transport equation shows a decrease in the vortex-stretching term due to cavitation. A substantial increase in the baroclinic torque is observed in the regions where the vapor structures collapse. A spectra of the pressure fluctuations in the far-field downstream region show an increase in the acoustic noise at high frequencies due to cavitation.

Large Eddy Simulation of Flow Past Circular Cylinder Based on the Least Square Meshless Method

2013 ◽  
Vol 394 ◽  
pp. 128-133
Author(s):  
Yuan Ding Wang ◽  
Jun Jie Tan ◽  
Xiao Wei Cai ◽  
Deng Feng Ren
Keyword(s):  
Circular Cylinder ◽  
Large Eddy Simulation ◽  
Meshless Method ◽  
Stokes Equations ◽  
Least Square ◽  
Scale Model ◽  
Navier Stokes ◽  
Navier Stokes Equations ◽  
Eddy Simulation ◽  
Large Eddy

Large Eddy Simulation (LES) based on the least square meshless method was proposed in the present paper to simulate the classical turbulent flow around a stationary 2D circular cylinder. The subgrid scale model of Smagorinsky-Lily was employed to close the Navier-Stokes equations filtered by Favre filter. The Reynolds number is 3900 which means that the flow is subcritical and the wake is fully turbulent but the cylinder boundary is still laminar. Results obtained in this paper were evaluated by comparison with published experimental results and other numerical results. The results obtained in the present work show better agreement with the experimental values than other two-dimensional LES results .

LARGE EDDY SIMULATION OF SEDIMENT-LADEN TURBULENT FLOW IN AN OPEN CHANNEL

2008 ◽  
Vol 22 (16) ◽  
pp. 2517-2527 ◽  
Author(s):  
ZHANHONG WAN ◽  
ZHILIN SUN ◽  
ZHENJIANG YOU ◽  
QIYAN ZHANG
Keyword(s):  
Large Eddy Simulation ◽  
Open Channel ◽  
Stokes Equations ◽  
Sediment Concentration ◽  
Continuous Phase ◽  
Navier Stokes ◽  
Navier Stokes Equations ◽  
Eddy Simulation ◽  
Large Eddy ◽  
Concentration Equation

Sediment transport in fully developed turbulent open channel flow has been investigated using large eddy simulation (LES) of the incompressible Navier–Stokes equations. The scalar transport equation of the sediments concentration, which is based on the continuous-phase approach, is adopted. The settling process is taken into account with a modified settling velocity appearing in the sediment concentration equation. A Smagorinsky model allowing for the interaction between the fluid flow and the suspended sediment is used to simulate the unresolved, subgrid scale terms. The LES results are compared with the experimental data, and good general agreement is achieved.

Large-Eddy Simulation of Shallow Water Langmuir Turbulence Using Isogeometric Analysis and the Residual-Based Variational Multiscale Method

2011 ◽  
Vol 79 (1) ◽  
Author(s):  
Andrés E. Tejada-Martínez ◽  
Ido Akkerman ◽  
Yuri Bazilevs
Keyword(s):  
Large Eddy Simulation ◽  
Isogeometric Analysis ◽  
Stokes Equations ◽  
Navier Stokes ◽  
Navier Stokes Equations ◽  
Variational Multiscale ◽  
Eddy Simulation ◽  
Large Eddy ◽  
Coastal Shelf ◽  
Current Interaction

We develop a residual-based variational multiscale (RBVMS) method based on isogeometric analysis for large-eddy simulation (LES) of wind-driven shear flow with Langmuir circulation (LC). Isogeometric analysis refers to our use of NURBS (Non-Uniform Rational B-splines) basis functions which have been proven to be highly accurate in LES of turbulent flows (Bazilevs, Y., et al. 2007, Comput. Methods Appl. Mech. Eng., 197, pp. 173–201). LC consists of stream-wise vortices in the direction of the wind acting as a secondary flow structure to the primary, mean component of the flow driven by the wind. LC results from surface wave-current interaction and often occurs within the upper ocean mixed layer over deep water and in coastal shelf regions under wind speeds greater than 3 m s−1. Our LES of wind-driven shallow water flow with LC is representative of a coastal shelf flow where LC extends to the bottom and interacts with the sea bed boundary layer. The governing LES equations are the Craik-Leivobich equations (Tejada-Martínez, A. E., and Grosch, C. E., 2007, J. Fluid Mech., 576, pp. 63–108; Gargett, A. E., 2004, Science, 306, pp. 1925–1928), consisting of the time-filtered Navier-Stokes equations. These equations possess the same structure as the Navier-Stokes equations with an extra vortex force term accounting for wave-current interaction giving rise to LC. The RBVMS method with quadratic NURBS is shown to possess good convergence characteristics in wind-driven flow with LC. Furthermore, the method yields LC structures in good agreement with those computed with the spectral method in (Thorpe, S. A., 2004, Annu. Rev. Fluids Mech., 36, pp. 584 55–79) and measured during field observations in (D’Alessio, S. J., et al., 1998, J. Phys. Oceanogr., 28, pp. 1624–1641; Kantha, L., and Clayson, C. A., 2004, Ocean Modelling, 6, pp. 101–124).

Sign in / Sign up

Export Citation Format

Share Document