Bulletin of the American Physical Society
62nd Annual Meeting of the APS Division of Fluid Dynamics
Volume 54, Number 19
Sunday–Tuesday, November 22–24, 2009; Minneapolis, Minnesota
Session BC: Turbulence Simulations II |
Hide Abstracts |
Chair: Meng Wang, University of Notre Dame Room: 101C |
Sunday, November 22, 2009 10:30AM - 10:43AM |
BC.00001: Subgrid scale physics in a turbulent boundary layer flow under varying convective stability conditions: an a-priori study Elie Bou-Zeid, Nikki Vercauteren, Chad Higgins, Hendrik Huwald, Marc B. Parlange, Charles Meneveau Using data sets collected during the Lake Atmosphere Turbulent Exchanges (LATEX, convectively unstable conditions) and the Snow Horizontal Array Turbulence Study (SnoHATS, convectively stable conditions) field experimental campaigns, we study the impact of this convective stability on the physics of small scale turbulence in the atmospheric boundary layer flow and the implications for modeling the subgrid scales stresses and fluxes (of heat and moisture) in large eddy simulation. Results indicate that the subgrid scale turbulent Prandtl number increases significantly as the flow transitions from unstable to stable. Under all stabilities, the TKE and scalar variance dissipation estimated based on the structure functions are very good estimates of the flux of energy to the subgrid scales; however, under stable conditions, a significant fraction of the TKE flux is destroyed by buoyancy rather than by viscous dissipation. Finally, the effect of stability on the coefficients of 2 SGS models is shown to be better explained by the Ozmidov scale under stable conditions. Overall, these results indicate that subgrid scale modeling is not drastically affected by atmospheric stability and hence a unified approach is possible. [Preview Abstract] |
Sunday, November 22, 2009 10:43AM - 10:56AM |
BC.00002: A Subgrid Scale Estimation Model for Large Eddy Simulation Rajes Sau, Krishnan Mahesh We discuss a novel estimation procedure to model the subgrid velocity for Large Eddy Simulation. The subgrid stress is obtained directly from the estimated subgrid velocity. The subgrid velocity is modeled as a function of resolved velocity ($\bar{u}_i$) and resolved strain--rate tensor ($\bar{S}_{ij}$). Using tensor invariants, we obtain an expression for subgrid velocity involving $\bar{u}_i$, that is quadratic in $\bar{S}_{ij}$ with three undetermined coefficients. These three coefficients are obtained by imposing the following constraints: (i) Galilean invariance, (ii) ensemble-averaged subgrid dissipation and (iii) local subgrid kinetic energy. Subgrid dissipation is obtained through a new dynamic procedure which uses two scalar level identities without least squares minimization, as opposed to the tensor level Germano identity. Subgrid kinetic energy is obtained either from the dynamic Yoshizawa model or a transport equation for subgrid kinetic energy. The estimation model is applied to isotropic turbulence and good results are obtained. Realistic backscatter is also predicted using this model. [Preview Abstract] |
Sunday, November 22, 2009 10:56AM - 11:09AM |
BC.00003: Grid-independent LES Sanjeeb Bose, Parviz Moin Grid independent turbulent statistics are obtained in a planar channel flow at $Re_{\tau}= 640$ by explicit filtering the governing equations for LES. Three dimensional filters (Vasilyev et al., JCP, 1998) are utilized such that commutation error is the same order as the truncation error of the fourth-order, conservative finite difference scheme (Morinishi et al., JCP, 1998). Several calculations are performed with a fixed filter width, but with varying grid resolutions. The grid-independent mean velocity profile is in good agreement with the experimental data of Hussain \& Reynolds (1970). The rms velocity profiles and one-dimensional energy spectra are compared with previous LES results and the unfiltered DNS of Abe et al. (2001), and show convergence toward a grid-independent profile. Ensemble averaged contributions of the dynamic Smagorinsky subgrid model to the Reynolds shear stress have also converged to a grid-independent profile across all grid resolutions. The effect of effective filter shape on the convergence of turbulence statistics will be discussed. Progress in the development of grid-independent LES for complex geometries with unstructured meshes will be presented. [Preview Abstract] |
Sunday, November 22, 2009 11:09AM - 11:22AM |
BC.00004: A dynamic wall model constrained by RANS Reynolds stress Aman Verma, Noma Park, Krishnan Mahesh We discuss a dynamic wall model obtained by incorporating RANS constraints into a dynamic SGS model. Unlike conventional approaches, Reynolds stresses are used as constraints on the mean SGS stress so that the constraining Reynolds stress closely matches the computed stress only in the mean sense. We use the Germano-identity error as an indicator of LES quality so that the RANS constraints are activated only where the Germano-identity error exceeds a certain threshold. The proposed model is applied to LES of turbulent channel flow at various Reynolds numbers and grid resolutions to obtain significant improvement over the dynamic Smagorinsky model, especially at coarse resolutions. The model has been implemented in spectral and structured finite volume solvers and is being extended to an unstructured solver. These developments will be discussed. [Preview Abstract] |
Sunday, November 22, 2009 11:22AM - 11:35AM |
BC.00005: LES of Supersonic Turbulent Flows with the Scalar FMDF Araz Banaeizadeh, Zhaorui Li, Farhad Jaberi The scalar filtered mass density function (FMDF) subgrid-scale model is further developed and tested for large eddy simulation (LES) of supersonic turbulent mixing and reacting flows in complex geometries. The LES/FMDF is implemented via a hybrid numerical method. In this method, the filtered compressible Navier-Stokes equations in curvilinear coordinate systems are solved with a generalized, high-order, multi-block, compact differencing scheme. To reduce the numerical oscillations of the compact scheme in shock regions, a localized high order artificial viscosity is added. The compressible scalar FMDF equation is solved with a stochastic Lagrangian Monte Carlo method. The results obtained with the LES/FMDF for shock tube and other compressible flows indicate that the pressure effects on the scalar field are well captured by the extended compressible FMDF model. The consistency of the filtered temperature and density fields as obtained from the Eulerian (finite difference) and Lagrangian (Monte Carlo) components of the LES/FMDF also indicate the reliability and the accuracy of the model in high speed flows. [Preview Abstract] |
Sunday, November 22, 2009 11:35AM - 11:48AM |
BC.00006: Dynamic k-Equation Model for LES of Compressible Flows Xiaochuan Chai, Krishnan Mahesh The sub-grid scale (SGS) kinetic energy (KE) has to be modeled in LES of compressible flows. Standard compressible versions of the dynamic Smagorinsky model (DSM) use Yoshizawa's expression for SGS KE. However, it is well known that Yoshizawa's Model tends to under-predict the magnitude of SGS KE. Obtaining the SGS KE from its transport equation,has shown improved performance for incompressible flows (e.g. Ghosal {\it et al.} 1995, Kim \& Menon 1996). We develop a compressible version of the DSM model with SGS KE equation. The SGS KE transport equation for compressible flow is derived, and the unclosed terms in the compressible KE equation are modeled and dynamically closed using the Germano identity. The proposed model is applied to decaying isotropic turbulence and normal shock/isotropic turbulence interaction. [Preview Abstract] |
Sunday, November 22, 2009 11:48AM - 12:01PM |
BC.00007: Subgrid scale modeling approaches for LES of low Mach number Jets Gregory Rodebaugh, Lester Su Large-eddy simulation of the low Mach number equations provides a computationally attractive option for simulating low speed, complex flows with large density and temperature gradients, particularly for reacting flow applications. Accurately predicting the scalar concentration fields is essential for the development of precise combustion simulations. In this work, we seek to elucidate what effects different subgrid scale (SGS) stress and scalar flux models have on both the mixing properties and turbulent statistics of the flow. Additionally, we aim to understand the coupling between the SGS stress and scalar flux models; therefore, we restrict the study to an isothermal flow with two species of different densities. There have been previous investigations of SGS models in compressible flows, but these focused primarily on higher Mach number regimes and SGS models for the energy equation. We study a canonical axisymmetric turbulent jet in this work, with the low Mach number equations being discretized in cylindrical coordinates. A predictor-corrector scheme is employed for time evolution of mass species fraction and momentum. [Preview Abstract] |
Sunday, November 22, 2009 12:01PM - 12:14PM |
BC.00008: The role of subgrid-scale models near the turbulent/nonturbulent interface in free shear flows Carlos B. da Silva This work addresses a challenging new problem for large-eddy simulations (LES) that exists in free shear flows where there are two distinct regions: the outer region where the flow is irrotational and the inner region where the flow is turbulent. The two regions are separated by a sharp interface: the turbulent/nonturbulent (T/NT) interface. It has been shown that important Reynolds stresses exist near the T/NT interface and that these stresses determine in part the mixing and combustion rates in jets. In the present work the role of several subgrid-scale models near the T/NT is analyzed in detail by using direct numerical simulations (DNS) and LES. The subgrid scales of motion near the T/NT interface are far from equilibrium and contain an important fraction of the total kinetic energy. Model constants used in several subgrid-scale models such as the Smagorinsky and the gradient models need to be corrected near the jet edge. The procedure used to obtain the dynamic Smagorinsky constant is not able to cope with the intermittent nature of this region. Both a-priori tests and LES show that near the jet edge the Smagorinsky model is superior both to the dynamic Smagorinsky and to the gradient models. [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