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 MB: Turbulence Modelling III |
Hide Abstracts |
Chair: Sharath Girimaji, Texas A&M University Room: 101B |
Tuesday, November 24, 2009 8:00AM - 8:13AM |
MB.00001: Progress in Algebraic Reynolds Stress Model development for compressible flows Carlos Gomez, Sharath Girimaji The standard algebraic Reynolds stress modeling (ARSM) approach is used to develop closures for compressible turbulence. A new rapid-pressure strain correlation model that is sensitive to gradient (Ma$_{g})$ and turbulent (Ma$_{t})$ Mach numbers is employed in the closure development. At low Ma$_{g}$, the pressure-strain correlation model assumes its standard incompressible form and the ARSM of Girimaji (1996) is recovered. Based on the analysis of the Navier-Stokes equation in the rapid distortion limit, it is suggested that at intermediate Ma$_{g}$ the effects of pressure, on an average, is to counter that of inertia. This leads to a balance between production and pressure-strain correlation leading to no temporal change in Reynolds stresses in this regime. In the high Ma$_{g}$ regime, the effect of pressure is negligible compared to inertia thus yielding a fully explicit algebraic Reynolds stress model. The complete proposed model is tested in a 2D supersonic mixing layer. Similarity profiles of mean streamwise velocity and turbulence quantities are compared with the experimental data of Goebel and Dutton (1991). Numerical mixing layer spreading rates are found to be in relatively good agreement with experimental data. [Preview Abstract] |
Tuesday, November 24, 2009 8:13AM - 8:26AM |
MB.00002: Progress in the development of explicit Algebraic passive Scalar Flux Model (ASFM) for compressible flows Mona Karimi, Sharath Girimaji Closure modeling of turbulent scalar flux represents an important field of active research in turbulent combustion. Invoking the weak-equilibrium limit of turbulence, the evolution equation for the scalar flux is simplified. Utilizing Representation theory, the resulting algebraic equation is solved for normalized scalar flux. The novelty of the work lies in the extension of this procedure to compressible flows, employing the pressure-scalar gradient correlation that is valid for a wide range of Gradient Mach numbers: (i) the incompressible model of the pressure-scalar gradient is retained at low Mach numbers, (ii) the pressure effects are take to be negligible compared to inertial term at very large Mach numbers, and (iii) the action of pressure is assumed to nullify the inertial effects at intermediate Mach numbers. It is demonstrated that the eddy diffusivity depends, not just on the mean scalar gradient, but also on the anisotropy of the velocity field. The dependence of eddy diffusivity on strain-rate, rotation-rate, and time scale ratio at different Gradient Mach numbers is examined. Additionally, simulations employing the model will be presented. [Preview Abstract] |
Tuesday, November 24, 2009 8:26AM - 8:39AM |
MB.00003: A Method for Interface-Turbulence Forcing in Hybrid LES/RANS Simulations Suad Jakirlic, Bjoern Kniesner A computational strategy coupling near-wall, eddy-viscosity-based RANS models with LES in a two-layer Hybrid LES/RANS (HLR) scheme is proposed in the present work. Key questions concerning the coupling of both methods, the inherently steady RANS method and highly-unsteady LES method, are closely connected to the treatment at the interface separating both sub-regions: (1) the exchange of the variables across the LES/RANS interface was adjusted by implicit imposition of the condition of equality of the modeled turbulent viscosities; (2) second issue is the utilisation of a self-adjusting interface position in the course of the simulation; (3) the third issue, the present work is focussing on, addresses the usage of a special forcing technique, which compensates the loss of information due to strong damping in the RANS region by creation of artificial and correlated fluctuations using a method originating from a digital-filter-based generation of inflow data for spatially developing DNS and LES due to Klein et al. (2003). Herewith, the recovery of the fluctuations on the LES side of the interface is accelerated. The model validation is conducted by computing numerous wall-bounded configurations of different geometrical complexity featuring different mean flow and turbulence phenomena. [Preview Abstract] |
Tuesday, November 24, 2009 8:39AM - 8:52AM |
MB.00004: Towards a Dynamic DES model Pramod Subbareddy, Graham Candler Hybrid RANS/LES methods are being increasingly used for turbulent flow simulations in complex geometries. Spalart's detached eddy simulation (DES) model is one of the more popular ones. We are interested in examining the behavior of the Spalart-Allmaras (S-A) Detached Eddy Simulation (DES) model in its ``LES mode.'' The role of the near-wall functions present in the equations is analyzed and an explicit analogy between the S-A and a one-equation LES model based on the sub-grid kinetic energy is presented. A dynamic version of the S-A DES model is proposed based on this connection. Validation studies and results from DES and LES applications will be presented and the effect of the proposed modification will be discussed. [Preview Abstract] |
Tuesday, November 24, 2009 8:52AM - 9:05AM |
MB.00005: Investigation of the one-point statistics of the scalar dissipation determined from scalar fields in large-eddy simulation Robert Knaus, Carlos Pantano, Joseph Oefelein In large-eddy simulation (LES), molecular scalar mixing is completely modeled since it generally occurs at scales below the cutoff. In this case, only the resolved scalar field is simulated. In many situations the statistics of the resolved scalar fields are accurate, e.g., if the supply of unmixed fluids is provided through boundary conditions. In combustion, there are usually subgrid processes, e.g., flames, controlling the rate of chemical conversion whose coupling with the turbulence can not be resolved completely in LES. If the subgrid process is strongly sensitive to the rate of mixing, as in nonpremixed combustion, one question of interest is to what extent can one recover the true statistics of mixing from those available in the LES fields so that the accuracy of the coupling with combustion can be improved. We investigate theoretically the relationship between the resolved and true scalar dissipation and propose a framework to recover the missing one-point statistical information. [Preview Abstract] |
Tuesday, November 24, 2009 9:05AM - 9:18AM |
MB.00006: Simulation of anisotropic pollution dispersion in urban environments Rafael Izarra-Garcia The prediction accuracy of air flow in urban environments is strongly dependent on the selection of the turbulence model for Reynolds stresses. In case of pollution dispersion simulations, a model for the turbulent scalar fluxes is also needed and it is usually performed with the simple isotropic gradient diffusion assumption. In the present work, the influence of advanced anisotropic models for the turbulent scalar fluxes is studied and compared with two wind tunnel experiments (2D and 3D). Five anisotropic algebraic flux models, two second moment models and the simple gradient diffusion model (at various Sct) were implemented in FLUENT 6.3 and compared using statistical performance measures. The results showed general good predictions for most scalar flux models with unexpected poor improvements of the anisotropic models over the isotropic and a lack of model coefficient calibration for pollution dispersion applications. Finally, simulation improvements, sensitivity of model coefficients and several parameters and assumptions needed for this application are discussed. [Preview Abstract] |
Tuesday, November 24, 2009 9:18AM - 9:31AM |
MB.00007: DNS and RANS Modeling of Dispersion in the Wake of an Obstacle David Philips, Riccardo Rossi, Gianluca Iaccarino We present a numerical study of the dispersion of a passive scalar in turbulent separated flows to establish the predictive capabilities of algebraic flux models against the standard eddy-diffusivity representation. The scalar dispersion from a point source over a wavy wall is initially investigated to carefully evaluate scalar flux models through comparisons with DNS data. The roof-top release of a passive plume from a wall-mounted cube in a turbulent boundary layer is then presented to demonstrate that algebraic models can also be applied successfully to atmospheric dispersion at street-scale. Despite the questionable validity of local-equilibrium conditions, the numerical experiments show that algebraic models provide a significant improvement for scalar dispersion simulations of complex flows with respect to the standard eddy-diffusivity model. [Preview Abstract] |
Tuesday, November 24, 2009 9:31AM - 9:44AM |
MB.00008: Lag-model for dissipation of scalar variance in LES Sergei Chumakov Modeling the dissipation term in transport equation for the subgrid-scale (SGS) scalar variance is of high importance in LES of reacting flows. We propose a new approach to model the dissipation which is based on averaging of the source term for the SGS variance along LES Lagrangian trajectories. The averaging is performed backwards in time using a particular weight function with the peak at some non-zero characteristic time $T$, the ``lag'' time. The method is based on the notion of the non-zero cascade time scale, i.e., the production and dissipation terms are not assumed to be correlated spatially but rather temporally along LES Lagrangian trajectories. The approach results in the addition of one extra transport equation to the system, bringing the total number of auxiliary equations to two. {\it A priori} tests show reasonably good prediction of the SGS variance dissipation rate. Results of an {\it a posteriori} test for a reacting flow will be shown. [Preview Abstract] |
Tuesday, November 24, 2009 9:44AM - 9:57AM |
MB.00009: A posteriori analysis of numerical errors in transport equation models for subfilter scalar variance Colleen M. Kaul, Venkat Raman Conserved-scalar based large eddy simulations (LES) of non-premixed combustion require accurate models for subfilter scalar variance. Because most subfilter variance models depend strongly on the smallest resolved scales, they are susceptible to large numerical errors when calculated using finite difference methods. To evaluate models based on solution of a variance transport equation, errors must be assessed in a dynamic framework. The current work presents the results of a versatile a posteriori analysis method using modified wavenumbers to emulate finite difference errors in a pseudospectral code. Filtered scalar and variance fields corresponding to schemes of varying accuracy are evolved along with DNS velocity and scalar fields. This approach permits discretionary inclusion of various error sources, allowing detailed characterization of numerical and modeling issues in transport equation based subfilter variance prediction. [Preview Abstract] |
Tuesday, November 24, 2009 9:57AM - 10:10AM |
MB.00010: Pressure Effects in Compressible Flows at the Rapid Distortion Limit Rebecca Bertsch, Sharath S. Girimaji We investigate the role of pressure in rapidly sheared homogeneous compressible turbulence using the Rapid Distortion Theory (RDT). Our objective is to develop a simple dynamical model that reproduces the 3-stage turbulent kinetic energy evolution observed in compressible RDT calculations. The three stages of turbulent kinetic energy evolution are: (1) initial growth at a rate higher than equivalent incompressible flow; (2) an intermediate period of stabilization where the turbulent kinetic energy grows very slowly or not at all; and (3) final period of growth at a rate similar to incompressible flows. The dynamical model is a function of gradient and turbulent Mach numbers and attempts to shed light on the pressure-strain correlation process in compressible flows. [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