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 R22: Turbulence Modeling |
Hide Abstracts |
Chair: Karan Venayagamoorthy, Colorado State University Room: 30C |
Tuesday, November 20, 2012 1:00PM - 1:13PM |
R22.00001: In Marriage of Model and Numerics, Glimpses of the Future AliReza Nejadmalayeri, Oleg V. Vasilyev, Alexei Vezolainen A newly defined concept of \emph{m}-refinement (model-refinement), which provides two-way coupling of physical models and numerical methods, is employed to study the Reynolds scaling of SCALES with constant levels of fidelity. Within the context of wavelet-based methods, this new hybrid methodology provides a hierarchical space/time dynamically adaptive automatic smooth transition from resolving the Kolmogorov length-scale (WDNS) to decomposing deterministic-coherent/stochastic-incoherent modes (CVS) to capturing more/less energetic structures (SCALES). This variable fidelity turbulence modeling approach utilizes a unified single solver framework by means of a Lagrangian spatially varying thresholding technique. The fundamental findings of this computational complexity study are summarized as follows: 1) SCALES can achieve the objective of ``controlling the captured flow-physics as desired'' by profoundly small number of spatial modes; 2) Reynolds scaling of constant-dissipation SCALES is the same regardless of fidelity of the simulations; 3) the number of energy containing structures at a fixed level of resolved turbulent kinetic energy scales linearly with \emph{Re}; and 4) the fractal dimension of coherent energy containing structures is close to unity. [Preview Abstract] |
Tuesday, November 20, 2012 1:13PM - 1:26PM |
R22.00002: Detached-Eddy Simulation Based on the $v^2$-$f$ Model SolKeun Jee, Karim Shariff Detached-eddy simulation (DES) based on the $v^2$-$f$ Reynolds-averaged Navier-Stokes (RANS) model is developed and tested. The $v^2$-$f$ model incorporates anisotropy of near-wall turbulence, which is absent in other RANS models commonly used in the DES community. Here, we present preliminary but encouraging results for the proposed model. The constant, $C_{\rm{DES}}$, required in the DES formulation was calibrated by simulating both decaying and statistically-stationary isotropic turbulence. Both cases provide the same value of $C_{\rm{DES}}$, indicating that the forced case is an alternative way to determine the coefficient. After $C_{\rm{DES}}$ is calibrated, the $v^2$-$f$ DES formulation is tested for flow around a circular cylinder at a Reynolds number of 3900, in which case turbulence develops after separation. Simulations indicate that this model represents the turbulent wake nearly as accurately as the dynamic Smagorinsky model. For comparison, Spalart-Allmaras-based DES is also included in the cylinder flow simulation. [Preview Abstract] |
Tuesday, November 20, 2012 1:26PM - 1:39PM |
R22.00003: Finite-dimensional Asymptotics and Degrees-of-Freedom Estimation for Turbulence Models Incorporating Spectral Subgrid-scale Viscosity Joel Avrin We study the finite-dimensional large-time behavior of three-dimensional forced turbulence as modeled by a modified Navier-Stokes equation. Subgrid-scale viscous effects are modeled by adding a hyperviscous term, but only to the high frequencies past a cutoff wavenumber $m$. We theoretically establish for arbitrarily large Reynolds numbers that the asymptotic (i.e. large-time) behavior of the system is finite-dimensional with an estimate on the number of degrees of freedom well within the Landau-Lifschitz estimates. We also verify in the case that $m$ is large enough that the overall large-time dynamics are controlled by the large-time dynamics of the inertial range. Given these promising results, we now would like to explore the physicality of the model by modifying the arguments underlying the Chapman-Enskog expansion. [Preview Abstract] |
Tuesday, November 20, 2012 1:39PM - 1:52PM |
R22.00004: Invariant turbulence models Alexander Bihlo, Elsa Maria Dos Santos Cardoso-Bihlo, Jean-Christophe Nave, Roman Popovych Various subgrid-scale closure models break the invariance of the Euler or Navier--Stokes equations and thus violate the geometric structure of these equations. A method is shown which allows one to systematically derive invariant turbulence models starting from non-invariant turbulence models and thus to correct artificial symmetry-breaking. The method is illustrated by finding invariant hyperdiffusion schemes to be applied in the two-dimensional turbulence problem. [Preview Abstract] |
Tuesday, November 20, 2012 1:52PM - 2:05PM |
R22.00005: Kolmogorov hypotheses for variable-resolution turbulence simulations Dasia Reyes, Sharath Girimaji Variable-resolution (VR) turbulence computation approaches such as detached-eddy simulations (DES), hybrid RANS-LES, partially-averaged Navier-Stokes (PANS) methods and partially-integrated turbulence model (PITM) are gaining popularity in engineering applications. Justifiably, these methods can be considered direct numerical simulations (DNS) of a variable-viscosity (non-Newtonian) fluid. Subject to this paradigm, we extend Kolmogorov's first and second similarity hypotheses for VR calculations. The resulting scaling laws can be invaluable in assessing the physical validity of spatio-temporal fluctuations of VR methods. Investigation of PANS decaying isotropic turbulence shows that the resolved field Kolmogorov scales vary with resolution as expected. [Preview Abstract] |
Tuesday, November 20, 2012 2:05PM - 2:18PM |
R22.00006: Differential filtering on unstructured grids with application to grid adaptation Sanjeeb Bose, Parviz Moin, Frank Ham Extension of explicitly filtered LES methods and their corresponding SGS models require a filtering operator that is low-pass on arbitrary meshes and can be decoupled from the underlying grid topology. Previously, we have utilized the differential filters proposed by Germano (1986) to perform explicitly filtered LES on unstructured grids. This framework is now extended to extract an estimate of the mean SGS kinetic energy to determine regions where mesh refinement is required. This procedure is automated using a local, anisotropic mesh refinement tool, adapt. This approach has been applied to large eddy simulation of a three-dimensional diffuser at Re=50,000, experimentally characterized by Kolade (2010). Results from two different mesh resolutions will be presented; an initially coarse mesh and a mesh refined using the SGS kinetic energy estimates. The adapted mesh has increased resolution in the separated shear layers originating from the bottom and side expanding walls. The accuracy of the SGS model will also be assessed through comparison of the LES predictions with experimental measurements. Other recent applications to flow over a cylinder with heat transfer and to flow over a turbine blade will be presented. [Preview Abstract] |
Tuesday, November 20, 2012 2:18PM - 2:31PM |
R22.00007: ABSTRACT WITHDRAWN |
Tuesday, November 20, 2012 2:31PM - 2:44PM |
R22.00008: ABSTRACT MOVED TO R20.00010 |
Tuesday, November 20, 2012 2:44PM - 2:57PM |
R22.00009: Computational modeling of scalar transport and buoyancy effects in turbulent flows using ODTLES Alan Kerstein, Christoph Glawe, Heiko Schmidt, Rupert Klein, Esteban Gonzalez-Juez, Rodney Schmidt ODTLES is a stochastic model for turbulent flow simulation consisting of a lattice-work of instantiations of the one-dimensional-turbulence (ODT) model, each of which is time advanced on a 1D domain with full spatial and temporal resolution. Collectively they form a 3D coarse mesh on which 3D flow is captured by coupling the 1D domains so as to obtain a formulation that reduces to direct numerical simulation (DNS) and conventional large-eddy simulation in the appropriate limits. The advantage of ODTLES relative to the latter is the built-in resolution of small scales where needed (near walls, across buoyancy jumps, etc.) at lower cost than resolving them using 3D DNS. A recent formulation targeting confined flow [1] is generalized to incorporate scalar fields and buoyancy effects. The generalized formulation, illustrative applications, and planned future development are described. \\[4pt] [1] E. D. Gonzalez-Juez, R. C. Schmidt, A. R. Kerstein, Phys. Fluids \textbf{23}, 125102 (2011). [Preview Abstract] |
Tuesday, November 20, 2012 2:57PM - 3:10PM |
R22.00010: Commutative Recursive Filters for Explicit-Filter Large-Eddy Simulation of Turbulent Flows Myeongkyun Kim, Daegeun Yoon, Donghyun You One of the most notable drawbacks associated with the implicit-filter LES is that the simulation result is dependent on the numerical grid employed due to the inherent dependence of the filtering operation on the numerical discretization. Alternatively, commutative explicit filters can be applied to distinguish the filtering operation from the underlying mesh distribution, thereby eliminating grid sensitivities. The efficacy of explicit-filtering to obtain grid-independent solutions of turbulent flows has been successfully demonstrated in the previous research (Bose, Moin \& You, Phys. Fluids, 2010; Singh, You \& Bose, Phys. Fluids, 2012). However, the use of broad-width filters accompanies significant increase in computational cost in terms of memory space and communication load for a distributed memory (MPI-based) parallel computation. To overcome the difficulty, a recursive filtering algorithm which can effectively replace a broad-width commutative filter with a series of narrow-width filters. The efficacy of the commutative recursive filtering method is evaluated in explicit-filter LES of turbulent channel flow, with particular attention to the performance of commutative recursive filters in terms of computational cost and memory requirement for a parallel computation. [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