Bulletin of the American Physical Society
70th Annual Meeting of the APS Division of Fluid Dynamics
Volume 62, Number 14
Sunday–Tuesday, November 19–21, 2017; Denver, Colorado
Session G31: Computational Fluid Dynamics: RANS and Hybrid RANS-LESCFD
|
Hide Abstracts |
Chair: Mark Kimber, Texas A&M University Room: 108 |
Monday, November 20, 2017 10:35AM - 10:48AM |
G31.00001: Hybrid RANS-LES using high order numerical methods Marc Henry de Frahan, Shashank Yellapantula, Ganesh Vijayakumar, Robert Knaus, Michael Sprague Understanding the impact of wind turbine wake dynamics on downstream turbines is particularly important for the design of efficient wind farms. Due to their tractable computational cost, hybrid RANS/LES models are an attractive framework for simulating separation flows such as the wake dynamics behind a wind turbine. High-order numerical methods can be computationally efficient and provide increased accuracy in simulating complex flows. In the context of LES, high-order numerical methods have shown some success in predictions of turbulent flows. However, the specifics of hybrid RANS-LES models, including the transition region between both modeling frameworks, pose unique challenges for high-order numerical methods. In this work, we study the effect of increasing the order of accuracy of the numerical scheme in simulations of canonical turbulent flows using RANS, LES, and hybrid RANS-LES models. We describe the interactions between filtering, model transition, and order of accuracy and their effect on turbulence quantities such as kinetic energy spectra, boundary layer evolution, and dissipation rate. [Preview Abstract] |
Monday, November 20, 2017 10:48AM - 11:01AM |
G31.00002: ABSTRACT WITHDRAWN |
Monday, November 20, 2017 11:01AM - 11:14AM |
G31.00003: An investigation into the reduction of log-layer mismatch in wall-modeled LES with a hybrid RANS/LES approach Riccardo Balin, Philippe R. Spalart, Kenneth E. Jansen Hybrid RANS/LES modeling approaches used in the context of wall-modeled LES (WMLES) of channel flows and boundary layers often suffer from a mismatch in the RANS and LES log-layer intercepts of the mean velocity profile. In the vicinity of the interface between the RANS and LES regions, the mean velocity gradient is too steep causing a departure from the log-law, an over-prediction of the velocity in the outer layer and an under-prediction of the skin-friction. This steep gradient is attributed to inadequate modeled Reynolds stresses in the upper portion of the RANS layer and at the interface. Channel flow computations were carried out with the IDDES approach of Shur et al. in WMLES mode based on the Spalart-Allmaras RANS model. This talk investigates the robustness of this approach for unstructured grids and explores changes required for grids where insufficient elevation of the Reynolds stresses is observed. M. L. Shur, P. R. Spalart, M. Kh. Strelets, and A. K. Stravin, A hybrid RANS-LES approach with delayed-DES and wall-modelled LES capabilities, International Journal of Heat and Fluid Flow 29, 1638 (2008). [Preview Abstract] |
Monday, November 20, 2017 11:14AM - 11:27AM |
G31.00004: Application of a New Hybrid RANS/LES Modeling Paradigm to Compressible Flow Todd Oliver, Clark Pederson, Sigfried Haering, Robert Moser It is well-known that traditional hybrid RANS/LES modeling approaches suffer from a number of deficiencies. These deficiencies often stem from overly simplistic blending strategies based on scalar measures of turbulence length scale and grid resolution and from use of isotropic subgrid models in LES regions. A recently developed hybrid modeling approach has shown promise in overcoming these deficiencies in incompressible flows [Haering, 2015]. In the approach, RANS/LES blending is accomplished using a hybridization parameter that is governed by an additional model transport equation and is driven to achieve equilibrium between the resolved and unresolved turbulence for the given grid. Further, the model uses an tensor eddy viscosity that is formulated to represent the effects of anisotropic grid resolution on subgrid quantities. In this work, this modeling approach is extended to compressible flows and implemented in the compressible flow solver SU2 (http://su2.stanford.edu/). We discuss both modeling and implementation challenges and show preliminary results for compressible flow test cases with smooth wall separation. [Preview Abstract] |
Monday, November 20, 2017 11:27AM - 11:40AM |
G31.00005: Comparison of LES and PANS calculations with available DNS data for the flow past a square cylinder Branislav Basara, Zoran Pavlovic The recent Direct Numerical Simulation (DNS) data of Trias et al. (2015) for the flow around a square cylinder at Re$=$22000 was used to assess the accuracy of Large Eddy Simulation (LES) and Partially-Averaged Navier-Stokes (PANS) calculations. The triple decomposition of the velocity into the mean, coherent and stochastic components allows useful and detailed analysis of calculation results. This is of a special interest for PANS calculations, which usually use efficient but inadequate treatments for the main model resolution parameter. This issue may get pronounced in a strong vortex shedding flow as presented in this work. Namely, the unresolved to total kinetic energy ratio known as the resolution parameter, is obtained from the grid spacing and the integral length scale of turbulence, which is usually calculated by summing up unresolved turbulent kinetic energy obtained from its own equation and the resolved kinetic energy calculated as a difference between the instantaneous and averaged velocity. Consequently, such summation procedure includes coherent and stochastic parts possible leading to higher integral length scales and too low values of the resolution parameter. An assessment of this deficiency by using DNS data as well as by comparing to LES calculations results is essential for further PANS modelling improvements. This work will assess the recent proposals for the calculation of the resolution parameter including one proposed by Basara and Girimaji (2013), which was related to the solution of the modelled resolved kinetic energy equation and thus avoiding the long and expensive flow averaging. [Preview Abstract] |
Monday, November 20, 2017 11:40AM - 11:53AM |
G31.00006: Unsteady Computational Tests of a Non-Equilibrium Adam Jirasek, Peter Hamlington, Andrew Lofthouse A non-equilibrium turbulence model is assessed on simulations of three practically-relevant unsteady test cases; oscillating channel flow, transonic flow around an oscillating airfoil, and transonic flow around the Benchmark Super-Critical Wing. The first case is related to piston-driven flows while the remaining cases are relevant to unsteady aerodynamics at high angles of attack and transonic speeds. Non-equilibrium turbulence effects arise in each of these cases in the form of a lag between the mean strain rate and Reynolds stresses, resulting in reduced kinetic energy production compared to classical equilibrium turbulence models that are based on the gradient transport (or Boussinesq) hypothesis. As a result of the improved representation of unsteady flow effects, the non-equilibrium model provides substantially better agreement with available experimental data than do classical equilibrium turbulence models. This suggests that the non-equilibrium model may be ideally suited for simulations of modern high-speed, high angle of attack aerodynamics problems. [Preview Abstract] |
Monday, November 20, 2017 11:53AM - 12:06PM |
G31.00007: Prediction of an internal boundary layer on a flat plate after a step change in roughness using a near-wall RANS model Minghan Chu, Fanxiao Meng, Donald J. Bergstrom An in-house computational fluid dynamics code was used to simulate turbulent flow over a flat plate with a step change in roughness, exhibiting a smooth-rough-smooth configuration. An internal boundary layer (IBL) is formed at the transition from the smooth to rough (SR) and then the rough to smooth (RS) surfaces. For an IBL the flow far above the surface has experienced a wall shear stress that is different from the local value. Within a Reynolds-Averaged-Navier-Stokes (RANS) formulation, the two-layer k-$ \epsilon$ model of Durbin et al. (2001) was implemented to analyze the response of the flow to the change in surface condition. The numerical results are compared to experimental data, including some in-house measurements and the seminal work of Antonia and Luxton (1971,72). This problem captures some aspects of roughness in industrial and environmental applications, such as corrosion and the earth’s surface heterogeneity, where the roughness is often encountered as discrete distributions. It illustrates the challenge of incorporating roughness models in RANS that are capable of responding to complex surface roughness profiles. [Preview Abstract] |
Monday, November 20, 2017 12:06PM - 12:19PM |
G31.00008: On a turbulent wall model to predict hemolysis numerically in medical devices. Seunghun Lee, Minwook Chang, Seongwon Kang, Nahmkeon Hur, Wonjung Kim Analyzing degradation of red blood cells is very important for medical devices with blood flows. The blood shear stress has been recognized as the most dominant factor for hemolysis in medical devices. Compared to laminar flows, turbulent flows have higher shear stress values in the regions near the wall. In case of predicting hemolysis numerically, this phenomenon can require a very fine mesh and large computational resources. In order to resolve this issue, the purpose of this study is to develop a turbulent wall model to predict the hemolysis more efficiently. In order to decrease the numerical error of hemolysis prediction in a coarse grid resolution, we divided the computational domain into two regions and applied different approaches to each region. In the near-wall region with a steep velocity gradient, an analytic approach using modeled velocity profile is applied to reduce a numerical error to allow a coarse grid resolution. We adopt the Van Driest law as a model for the mean velocity profile. In a region far from the wall, a regular numerical discretization is applied. The proposed turbulent wall model is evaluated for a few turbulent flows inside a cannula and centrifugal pumps. The results present that the proposed turbulent wall model for hemolysis improves the computational efficiency significantly for engineering applications. [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