Bulletin of the American Physical Society
65th Annual Meeting of the APS Division of Fluid Dynamics
Volume 57, Number 17
Sunday–Tuesday, November 18–20, 2012; San Diego, California
Session G21: Turbulence Simulation: Wall Bounded |
Hide Abstracts |
Chair: Oscar Flores, Universidad Carlos III de Madrid Room: 30B |
Monday, November 19, 2012 8:00AM - 8:13AM |
G21.00001: Slow Growth Formulation for DNS of Chemically Reacting Temporal Boundary Layers with Forcing Victor Topalian, Todd Oliver, Robert Moser Extensions to a previously developed formulation for DNS of temporally evolving boundary layers are presented. The original formulation, which allows characterization of turbulence in a temporal boundary layer at a chosen stage of the development, uses a multiscale approach where the fast evolution of the turbulent fluctuations is simulated directly while the slow evolution of averaged quantities is modeled. Specifically, the source terms from slow evolution are modeled assuming self-similarity in the evolution of mean and RMS quantities. Here, the formulation is extended to enable DNS of chemically reacting boundary layers with forcing. These extensions are used to obtain DNS data for conditions similar to those observed in the boundary layer during athmospheric reentry of the NASA CEV. Data from this simulation will be used to inform turbulence model calibration and UQ. This work is supported by the Department of Energy [National Nuclear Security Administration] under Award Number [DE-FC52-08NA28615]. [Preview Abstract] |
Monday, November 19, 2012 8:13AM - 8:26AM |
G21.00002: Direct Numerical Simulation of two superposed viscous fluids in a channel with cavities on the wall Stefano Leonardi, Paolo Orlandi Parallel shear flow of two viscous fluids has received much attention in the past. An instability associated with the jump in viscosity at the interface between two fluids has been observed and it depends on the wavenumber of the flow disturbance, on the depth ratio and the viscosity ratio of the fluids. In the present paper, we extend previous findings by performing Direct Numerical Simulations of two superposed viscous fluids over rectangular cavities in a channel. The interface between the two fluids is slightly above the crests plane of the cavities. This configuration represents a simplified and preliminary model of a Slippery Liquid-Infused Porous Surface (SLIPS) promising for drag reduction. Periodic boundary conditions apply in the streamwise and spanwise direction. A parametric study has been carried out varying the viscous ratio, the Reynolds number of the inner flow, the aspect ratio of the cavities and the depth of the near wall fluid. A finite difference code, based on a Runge Kutta and fractional step, has been used. Roughness on the walls has been modeled using the immersed boundary method. Frictional and form drag and transition to turbulence depend strongly on the cavity shape and on the viscous ratio. [Preview Abstract] |
Monday, November 19, 2012 8:26AM - 8:39AM |
G21.00003: Aspect ratio effects in turbulent duct flows studied with DNS R. Vinuesa, A. Noorani, A. Lozano-Dur\'an, P. Schlatter, P. Fischer, H. Nagib Three-dimensional effects present in turbulent duct flows, i.e., side-wall boundary layers and secondary motions, are studied by means of direct numerical simulations (DNS). The spectral element code Nek5000, developed by Fischer {\it et. al.} (2008), is used to compute turbulent duct flows with aspect ratios 1 and 3 in streamwise-periodic boxes of length $25h$ (long enough to capture the longest streamwise structures). The total number of grid points is 28 and 62 million respectively, and the inflow conditions were adjusted iteratively in order to keep the same bulk Reynolds number at the centerplane $\left (Re_{b,c}=2800 \right )$ in both cases. Spanwise variations in wall shear, mean-flow profiles and turbulence statistics were analyzed with aspect ratio, and also compared with the 2D channel. The simulations were started from a laminar duct profile, and transition to turbulence was triggered by means of trip-forcing in the wall-normal direction, applied at the two horizontal walls. In addition, we developed a convergence criterion aimed at assessing the necessary averaging time $T_{A}$ for converged statistics. We find that econdary motions present in duct flows require longer averaging times and the total shear-stress profile is not necessarily linear. [Preview Abstract] |
Monday, November 19, 2012 8:39AM - 8:52AM |
G21.00004: Direct numerical simulation for turbulent channel flow at high Reynolds number Myoungkyu Lee, Nicholas Malaya, Robert D. Moser Direct numerical simulation (DNS) is a powerful tool in the study of wall-bounded turbulent flows. Of particular focus is the scaling of flow statistics with respect to Reynolds number. Investigations of these scalings require data at higher Reynolds numbers, which is limited by available computational power. We have developed a new codebase, optimized for Petascale machines, in order to perform a DNS at higher Reynolds number ($Re_\tau \approx 5000$) than previously performed. We simulate a canonical channel flow, with two infinite parallel plates driven by a constant pressure gradient. The numerical scheme is a Fourier spectral representation in the streamwise and spanwise directions, with B-Splines in the inhomogeneous direction. We demonstrate agreement between this code and previous DNS results at lower Reynolds numbers. Finally, we present some preliminary statistics including the mean velocity profile and the intensity of the fluctuations. [Preview Abstract] |
Monday, November 19, 2012 8:52AM - 9:05AM |
G21.00005: Application of mean wall shear stress boundary condition to complex turbulent flows using a wall-modeled large eddy simulation Minjeong Cho, Jungil Lee, Haecheon Choi The mean wall shear stress boundary condition was successfully applied to turbulent channel and boundary flows using large eddy simulation without resolving near-wall region (see Lee, Cho \& Choi in this book of abstracts). In the present study, we apply this boundary condition to more complex flows where flow separation and redeveloping flow exist. As a test problem, we consider flow over a backward-facing step at $Re_h = 22860$ based on the step height. Turbulent boundary layer flow at the inlet ($Re_\theta = 1050$) is obtained using inflow generation technique by Lund et al. (1998) but with wall shear stress boundary condition. First, we prescribe the mean wall shear stress distribution obtained from DNS (Kim, 2011, Ph.D. Thesis, Stanford U.) as the boundary condition of present simulation. Here we give no-slip boundary condition at flow-reversal region. The present results are in good agreements with the flow statistics by DNS. Currently, a dynamic approach of obtaining mean wall shear stress based on the log-law is being applied to the flow having flow separation and its results will be shown in the presentation. [Preview Abstract] |
Monday, November 19, 2012 9:05AM - 9:18AM |
G21.00006: Off-wall boundary conditions for turbulent flows obtained from buffer-layer minimal flow units Ricardo Garcia-Mayoral, Brian Pierce, James Wallace There is strong evidence that the transport processes in the buffer region of wall-bounded turbulence are common across various flow configurations, even in the embryonic turbulence in transition (Park et al., \textit{Phys.~Fl.}~\textbf{24}). We use this premise to develop off-wall boundary conditions for turbulent simulations. Boundary conditions are constructed from DNS databases using periodic minimal flow units and reduced order modeling. The DNS data was taken from a channel at $Re_{\tau} = 400$ and a zero-pressure gradient transitional boundary layer (Sayadi et al., submitted to \em{J.~Fluid~Mech.}). Both types of boundary conditions were first tested on a DNS of the core of the channel flow with the aim of extending their application to LES and to spatially evolving flows. [Preview Abstract] |
Monday, November 19, 2012 9:18AM - 9:31AM |
G21.00007: Effect of thermal boundary condition on wall-bounded, stably-stratified turbulence Oscar Flores, Manuel Garcia-Villalba The dynamics of stably stratified wall-bounded turbulent flows are of great importance for many engineering and geophysical problems. In some cases, like the stably stratified atmospheric boundary layer, it is unclear which is the most appropriate thermal boundary condition, i.e. constant temperature or constant flux at the ground. Here, we analyze the effect that this boundary condition has on the dynamics of turbulent motions in the near-wall region in the case of strong stable stratification. Two Direct Numerical Simulations of turbulent channels will be used, at $Re_\tau=u_\tau h/\nu = 560$ and $Ri_\tau=\Delta \rho gh/\rho_0u_\tau^2=600-900$, which are described in detail in Flores \& Riley (2011, Boundary-Layer Meteorol) and Garcia-Villalba \& del Alamo (2011, Phys.Fluids). For this range of Reynolds and Richardson numbers, the near-wall region is intermittent, with patches of laminar flow embedded in the otherwise turbulent flow. It is in this regime where the differences between the constant temperature and the constant flux boundary conditions are expected to be larger, with the thermal boundary condition affecting how the local relaminarization of the flow takes place. This research has been supported by ARO, NSF and the German Research Foundation. [Preview Abstract] |
Monday, November 19, 2012 9:31AM - 9:44AM |
G21.00008: The influence of viscosity stratification on boundary-layer turbulence Jin Lee, Seo Yoon Jung, Hyung Jin Sung, Tamer A. Zaki Direct numerical simulations of turbulent flows over isothermally-heated walls were performed to investigate the influence of viscosity stratification on boundary-layer turbulence and drag. The adopted model for temperature-dependent viscosity was typical of water. The free-stream temperature was set to 30$^{\circ}$C, and two wall temperatures, 70$^{\circ}$C and 99$^{\circ}$C, were simulated. In the heated flows, the mean shear-rate is enhanced near the wall and reduced in the buffer region, which induces a reduction in turbulence production. On the other hand, the turbulence dissipation is enhanced near the wall, despite the the reduction in fluid viscosity. The higher dissipation is attributed to a decrease in the smallest length scales and near-wall fine-scale motions. The combined effect of the reduced production and enhanced dissipation leads to lower Reynolds shear stresses and, as a result, reduction of the skin-friction coefficient. [Preview Abstract] |
Monday, November 19, 2012 9:44AM - 9:57AM |
G21.00009: Statistics of Stagnation Points in Turbulent Channel Flows with Wavy Walls Fabian Hennig, Michael Gauding, Jens Henrik Goebbert, Norbert Peters We investigate the turbulent velocity field my means of instantaneous streamlines. The streamlines are partitioned into segments and decompose the velocity field in a non-arbitrary way. The segments are defined by extreme points based on the velocity magnitude. The boundaries of all streamline segments define a surface in space where the gradient of the projected velocity in streamline direction $\partial u /\partial s$ vanishes. This surface contains all local extreme points of the velocity magnitude. Such points also include stagnation points of the flow field, which are absolute minimum points of the turbulent velocity field. The properties of the $\partial u /\partial s = 0$ surface (and thus of stagnation points) are affected by local pressure gradients. Therefore, direct numerical simulations (DNS) of a turbulent flow with wavy walls, which induce complex pressure effects, are conducted. For the DNS a spectral element code is employed. The results have been validated against DNS and experimental data from literature. Based on the DNS the surface $\partial u /\partial s = 0$ is investigated in detail and its interaction with streamlines is visualized. The location and statistics of stagnation points with respect to the specific flow geometry are examined. [Preview Abstract] |
Monday, November 19, 2012 9:57AM - 10:10AM |
G21.00010: Turbulent wall jets over rough surfaces Rayhaneh Banyassady, Ugo Piomelli The effects of surface roughness on plane and radial turbulent wall jets have been studied using large-eddy simulation. The Reynolds number is $18,800$ (based on the bulk velocity and diameter of the impinging jet) for the radial case, and $7,500$ (based on the jet velocity and height) for the plane one. To represent the random roughness elements a virtual sandpaper model and an immersed-boundary method (IBM) based on the volume-of-fluid (VOF) approach are used. The roughness Reynolds numbers in both simulations are in the transitionally rough regime, $5 < k^+ < 70$. The grid refinement study showed that the plane and radial wall jet simulations require $15$ and $22$ million grid points, respectively. The results are validated with available literature. The mean flow shows that surface roughness decreases maximum velocity of the wall jet and increases the wall jet characteristics length (distance from the wall where the velocity decreases to half of the maximum velocity). The rate of decay of maximum velocity and growth rate of the wall jet are not significantly affected by the surface roughness. An analysis of the effect of surface roughness on the magnitude and shape of the Reynolds stresses profiles is being carried out. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700