Bulletin of the American Physical Society
66th Annual Meeting of the APS Division of Fluid Dynamics
Volume 58, Number 18
Sunday–Tuesday, November 24–26, 2013; Pittsburgh, Pennsylvania
Session A19: Turbulence Modeling I |
Hide Abstracts |
Chair: Naseem Ansari, University of Pittsburgh Room: 310/311 |
Sunday, November 24, 2013 8:00AM - 8:13AM |
A19.00001: A framework for Large Eddy Simulation (LES) based on spatiotemporal statistical information Prakash Vedula, Peter Attar, Allen LaBryer We present a computational framework that will have the potential to not only improve the efficiency of computational predictions based on LES but will also be able to address a major drawback of many existing constructs of LES, namely inaccurate predictions of the underlying spatiotemporal structure. The latter drawback could be especially critical in prediction of tornado paths and jet-noise intensities. In our proposed framework, the relevant sub-grid scale stress models are constructed based on information that is consistent with the underlying spatiotemporal statistics. Unlike in many existing constructs of LES, the proposed sub-grid scale stress models include non-Markovian or memory terms whose origins can be explained based on the theory of optimal prediction. These optimal models for LES are studied using a one-dimensional Burgers equation with and without forcing. Results indicate that the proposed framework performs better than most existing frameworks of LES, by virtue of accurate predictions of spatiotemporal structure. The presence of coarse-grained temporal information in our sub-grid scale models also allows for faster simulations by allowing for larger time steps. Implications of these findings to more complicated turbulent flows will also be discussed. [Preview Abstract] |
Sunday, November 24, 2013 8:13AM - 8:26AM |
A19.00002: A comparison of dynamic procedures for subgrid stresses in low- and high-speed channels Sungmin Ryu, Gianluca Iaccarino We present a novel dynamic procedure to determine the space- and time-dependent model coefficient of a subgrid scale (SGS) eddy-viscosity model. As a preceding step, an exact relationship between SGS Reynolds stresses and four different types of SGS closures is derived from the incompressible Navier-Stokes equation applying Reynolds decomposition and its filtered relation. To validate the proposed dynamic procedure, large eddy simulations (LES) of freely decaying isotropic turbulence and incompressible turbulent channel flow at $Re_\tau=395$ are performed with the Vreman model with the model coefficient dynamically determined by the SGS Reynolds stress based procedure. Moreover, LES with the dynamic Smagorinsky model and the Vreman model with the constant coefficient are also carried out to assess the performance of our dynamic model. Finally, LES with our dynamic model in compressible turbulent channel flows is evaluated to demonstrate that the relation of identity derived from the incompressible flow governing equations is also valid in compressible flows. [Preview Abstract] |
Sunday, November 24, 2013 8:26AM - 8:39AM |
A19.00003: Effects of filtering parameter value on simulation results Weiyun Liu, J.M. McDonough Aliasing is a fundamental issue in discrete solutions of the Navier--Stokes equations. It arises from under resolution of numerical approximations as occurs in large-eddy simulation and must be treated with a filter. Two approaches to filtering have been distinguished in the LES context: implicit and explicit. Implicit filtering is formally applied to governing equations without specification of a particular filter, and explicit filtering is performed on computed solutions via a prescribed filter, as in signal processing. With explicit filtering, since filtered velocities are used in subsequent time steps, the aliasing phenomenon can potentially be removed completely; we will focus on this form in the present work. Numerical filters, however, are constructed so as to allow control of the degree of aliasing via parameter values set by the user. We will demonstrate that poor choices of such parameters can result in completely non-physical, yet numerically stable, computed solutions for two widely-used filters, Pad\'{e} and Shuman, for a problem having abundant experimental data for comparisons. [Preview Abstract] |
Sunday, November 24, 2013 8:39AM - 8:52AM |
A19.00004: An improved dynamic non-equilibrium wall-model for large eddy simulation George Ilhwan Park, Parviz Moin A non-equilibrium wall-model based on unsteady 3D Reynolds-averaged Navier-Stokes (RANS) equations has been implemented in an unstructured mesh environment. The method is similar to that of the wall-model described by Wang and Moin [Phys. Fluids \textbf{14}, 2043--2051, (2002)], but is supplemented by a new dynamic eddy viscosity/conductivity model that corrects the effect of the resolved Reynolds stress (resolved turbulent heat flux) on the skin friction (wall heat flux). This correction is crucial for accurate prediction of the skin friction and wall heat flux. Unlike earlier models, this eddy viscosity/conductivity model does not have a stress-matching procedure or a tunable free parameter, and it shows consistent performance over a wide range of Reynolds numbers. The wall-model is validated against canonical (attached) transitional and fully turbulent flows at moderate to very high Reynolds number: a turbulent channel flow at $Re_\tau$ = 2000, an H-type transitional boundary layer up to $Re_\theta$ = 3300, and a high Reynolds number boundary layer at $Re_\theta$ = 31000. An application to the flow over NACA4412 airfoil is ongoing and hopefully will be presented. [Preview Abstract] |
Sunday, November 24, 2013 8:52AM - 9:05AM |
A19.00005: A dynamic two-level large-eddy simulation method for high Reynolds number flows Reetesh Ranjan, Suresh Menon We present a dynamic hybrid two-level large-eddy simulation method for high Reynolds number flows. The method combines the two level simulation model with a conventional large eddy simulation model through an additive scale separation operator in a dynamic manner. The two level simulation model performs the scale separation through a large-scale function instead of a spatial filtering used by a conventional large eddy simulation model and therefore, it does not from some of the limitations associated with the spatial filtering. The hybrid method ensures that the two level simulation model provides a dominant contribution in flow regions having sharp gradients and in other regions large-eddy simulation model assumes a dominant role leading to an overall efficient computational method for practical applications. The hybridization is achieved through a dynamic and a spatially smooth blending function based on a characteristic length scale. The dynamic evaluation of the blending function is essential for complex flows where a prior contribution of two-level and large-eddy simulation models can not be estimated accurately. Application of dynamic blending function will be demonstrated by simulating high Reynolds number separating/reattaching flow over a bump in a channel. [Preview Abstract] |
Sunday, November 24, 2013 9:05AM - 9:18AM |
A19.00006: A new Hybrid Filtered Favre average for compressible LES Massimo Germano, Antonella Abb\`a The statistical study of compressible or variable density turbulent flows is usually performed in terms of the Favre average weighted with the {\it statistical} density. The usual extension of this approach to the Large Eddy Simulation of compressible turbulent flows is in terms of the Filtered Favre average weighted with the {\it filtered} density. By consequence the Favre averages are not directly recovered by the statistical average of the Filtered Favre quantities. All that is inconvenient as regards the comparison between RANS and LES and as regards the development of some modeling approaches such as the dynamic modeling procedure and the hybrid RANS/LES strategies. In order to simplify the formulation of compressible LES and to relate more directly the filtered quantities to the statistical ones a new Hybrid Filtered Favre average weighted directly with the {\it statistical} density is proposed and compared with the usual one. [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