Bulletin of the American Physical Society
71st Annual Meeting of the APS Division of Fluid Dynamics
Volume 63, Number 13
Sunday–Tuesday, November 18–20, 2018; Atlanta, Georgia
Session E39: Turbulence Modeling I |
Hide Abstracts |
Chair: Luca Biferale, University of Rome Tor Vergata Room: Georgia World Congress Center Ballroom 3/4 |
Sunday, November 18, 2018 5:10PM - 5:23PM |
E39.00001: Time-averaged active model-splitting hybrid RANS/LES Sigfried Haering, Todd A. Oliver, Robert D Moser We discuss progress in the development and validation of a new approach for hybrid RANS/LES simulations. Referred to as the time-averaged active model-splitting approach (TAMS) , it has been specifically constructed to overcome challenges associated with existing hybrid approaches related to LES/RANS blending techniques and inconsistencies between the resolved and modeled turbulence. The new approach is based on a hybridization strategy in which the RANS and LES components act through separate models formulated using the mean and fluctuating velocity, respectively, as approximated by time averaging over the local turbulent timescales. Further, resolved fluctuating turbulent energy is actively transferred from the modeled to the resolved scales in regions identified to be capable of resolving more turbulence. Finally, the model makes use of an anisotropic LES model that is intended to represent the effects of grid anisotropy with potential to directly account for grid heterogeneity as well. The approach is demonstrated on fully-developed, incompressible channel flow with encouraging results. |
Sunday, November 18, 2018 5:23PM - 5:36PM |
E39.00002: Subgrid-Scale Model Development Using Approximate Bayesian Computation Olga A Doronina, Colin AZ Towery, Peter E Hamlington The predictive power of large eddy simulations (LES) depends on the accuracy of closure models used to represent subgrid-scale (SGS) fluxes. Traditionally, model parameters have been determined through either direct inversion of model equations given some reference data or using optimization techniques. However, the former approach becomes complicated for models with many different parameters or when the model consists of partial differential equations, and the latter approach precludes the quantification of parameter uncertainty. In this talk, we use Approximate Bayesian Computation (ABC) and Markov Chain Monte Carlo (MCMC) methods to estimate parameter values, as well as their uncertainties, in SGS models. The MCMC-ABC approach avoids the need to directly compute a likelihood function during the parameter estimation, enabling a substantial speed-up as compared to full Bayesian analyses. The approach also naturally provides uncertainties in parameter estimates, avoiding the artificial certainty implied by optimization methods for parameter estimation. The MCMC-ABC approach is outlined, and both a priori and a posteriori test results for homogeneous isotropic turbulence are provided to demonstrate the accuracy and computational cost of the approach. |
Sunday, November 18, 2018 5:36PM - 5:49PM |
E39.00003: A problem with near-wall Reynolds stress models William Kenneth George, Michel Stanislas, Jean-Philippe Laval, Jean-Marc Foucaut, Christophe Cuvier A common mistake in modern turbulence practice is to confuse the quantity $D_{ik} = \nu \langle \partial u_i/\partial x_j~ \partial u_k/\partial x_j \rangle$ with the true dissipation tensor $\varepsilon_{ik} = 2 \nu \langle s_{ij}~ s_{kj} \rangle$, where $s_{ij}$ is the fluctuating strain-rate. The traces $D_{kk}$ and $\varepsilon_{kk}$ are in fact equal ONLY when the turbulent flow is homogeneous. Many flows at high Reynolds number are approximately locally homogeneous; but even when the traces are nearly equal, the component dissipations are not$^{1}$, and they are never the same inside $y^+=30$ of a wall-bounded flow.$^{2}$ It is argued that a major problem with RS-models in wall-bounded flows is the use of $D_{ik}$ in the modeled equations. \noindent 2) Foucaut et al. 2018 SPIV meas. of full dissip. tensor in a turb. b.l. (submitted) |
Sunday, November 18, 2018 5:49PM - 6:02PM |
E39.00004: Macroscopic Forcing Method and its application in assessment of RANS models Danah Park, Yasaman Shirian, Ali Mani In this talk we introduce a statistical method, which we call the Macroscopic Forcing Method (MFM), that can be combined with solutions from DNS data to reveal quantitative information about the differential operators in RANS models. We introduce MFM by applying it to example canonical problems, and demonstrate its capability in revealing errors associated with assumed model forms, e.g. standard eddy diffusivity, regardless of the tuning of their coefficient distribution. We will also demonstrate examples of MFM-inspired corrections to existing RANS models leading to significant improvement in their predictive capability. |
Sunday, November 18, 2018 6:02PM - 6:15PM |
E39.00005: Optimal sub-grid model for accurate determination of intermittent inertial range properties of turbulent flows Michele Buzzicotti, Patricio Clark Di Leoni, Fabio Bonaccorso, Kartik P Iyer, Luca Biferale We present a study of the statistical properties of homogeneous and isotropic turbulence evolved by fully resolved direct numerical simulations (DNS) and by large eddy simulations (LES) equipped with different sub-grid-scales (SGS) models. We compare the effects produced on the high-order, multi-scales correlation functions by the well known Smagorinsky closure with the performances of a new model based on the introduction of a Lagrange multiplier which constraints exactly the turbulent energy spectrum scaling law to the Kolmogorov k^(-5/3) slope at high wavenumbers. Data from hyperviscous DNS are also analyzed. To identify the optimal model which maximizes the extension of the inertial range dynamics, we compare the statistics measured from LES and hyperviscous simulations with the statistics of DNS performed on a more refined grid. We also consider fine-tuning SGS models by "nudging" the smaller LES resolved scales from the dynamics obtained at the same scales in the inertial range of a higher resolution DNS. |
Sunday, November 18, 2018 6:15PM - 6:28PM |
E39.00006: Reconstructed turbulent fields using 4D variational data assimilation: reproduction of instantaneous structures. Naseer Abdullah We look into a turbulent system where incomplete velocity measurement data over a time period are available. Assuming that the dynamical evolution of the system is governed by the Navier-Stokes (NS) equation, we use the 4D variational data assimilation, which is a valuable tool for recovering missing information about a system from what is directly measurable to reconstruct the initial state of the system, such that the evolution of the system matches the measurement data. We formulate the problem as an optimization problem, where the initial field is taken as the control variable. The goal is to find the optimal control variable to minimize the differences between the measurement and the velocity field evolved from the reconstructed initial field, subject to the constraint imposed by the NS equation. The reconstructed fields are compared with direct numerical simulation (DNS) data. We look into the difference between the instantaneous variations of the reconstructed and the DNS field. The examples of the averaged geometries show that the differences decrease with time within the optimization horizon. Also, the instantaneous high vorticity structures for the reconstructed fields are reproduced with good agreement with the DNS field. |
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