Bulletin of the American Physical Society
72nd Annual Meeting of the APS Division of Fluid Dynamics
Volume 64, Number 13
Saturday–Tuesday, November 23–26, 2019; Seattle, Washington
Session B11: Turbulent Boundary Layers: Wall Modeling |
Hide Abstracts |
Chair: Xiang Yang, Pennsylvania State University Room: 3B |
Saturday, November 23, 2019 4:40PM - 4:53PM |
B11.00001: Comparing a data-based approach and a physics-based approach for wall modeled LES of flow in a spanwise rotating channel Xiang Yang, Xinyi Huang The logarithmic law of the wall is a poor approximation of the mean flow in a spanwise rotating channel, and therefore a wall model that leads to the logarithmic mean flow at equilibrium conditions is inadequate for wall modeled LES (WMLES) of spanwise rotating channels. In this presentation, we will develop a wall modeling capability for flow in a spanwise rotating channel. In particular, we will compare a data-based modeling approach and a physics-based approach: the physics based model relies on an eddy viscosity that yields the correct linear mean flow at high rotation numbers, and the data-based model is a feed-forward neural network (FNN) that is trained using the available mean flow data. FNNs are often criticized for not generalizing well. Here, informed about our (albeit limited) knowledge of the flow, the FNN is found to provide good mean flow predictions within and outside the training dataset. In conclusion, data-based models are useful alternatives to physics-based models. [Preview Abstract] |
Saturday, November 23, 2019 4:53PM - 5:06PM |
B11.00002: Examination of wall-model fidelity with spatially diffuse non-equilibrium forcing Michael Adler, David Gonzalez, Logan Riley, Datta Gaitonde Wall-flux modelling is necessary to facilitate large-eddy simulation of high Reynolds number, wall-bounded turbulence. We examine the effect of wall-model fidelity and complexity on the accuracy of wall-modeled simulations of non-equilibrium turbulent boundary layers. A database of moderate Reynolds number wall-resolved simulations of non-equilibrium turbulent boundary layers is constructed, including scenarios of adverse/favorable pressure gradient and heated/cooled walls. The non-equilibrium forcing is spatially diffuse, providing significantly large regions where equilibrium scaling laws cannot be reasonably assumed in the inner layer, and necessitating non-equilibrium models for inner-layer accuracy. A database of wall-modeled simulations is then constructed to assess the model ability to reproduce non-equilibrium effects in both the inner and outer flow. Several model classes are employed including ODE, algebraic, and dynamic slip variants. The favorable performance of several new non-equilibrium models of algebraic complexity (efficient and easy to implement) is assessed relative to the equilibrium models and benchmark wall-resolved simulations. [Preview Abstract] |
Saturday, November 23, 2019 5:06PM - 5:19PM |
B11.00003: Wall model for large eddy simulations accounting for particle effect. Ping Wang Computational investigations of high Reynolds number turbulent multiphase flow is rather time-consuming even in large eddy simulation (LES) framework. The integral wall model (iWM) for large eddy simulation of particle-free wall-bounded turbulent flows proposed by Yang et al (2015) is further improved to include the body forces that particles exert on the fluid. The slightly modified iWMLES method is tested in the context of a finite-difference LES code for turbulent two phase flow. For model testing, the data from wall-resolved LES of particle-laden flow at Reynolds number of Re$_{\mathrm{\tau }} \quad =$550 \textasciitilde 4200 are treated as “standard data”. The results show that the mean velocity profiles, particle concentration, particle mean velocity and mass flux profiles compare well with “standard data” once particle force is properly included in WMLES of particle-laden flow. Meanwhile, it allows significant reduction of the required CPU time over simulations of turbulent two phase flow in which no-slip conditions are applied. [Preview Abstract] |
Saturday, November 23, 2019 5:19PM - 5:32PM |
B11.00004: Equilibrium/Non-Equilibrium Wall-Modeled LES of Airfoil Stall Phenomena at High-Reynolds Number Yoshiharu Tamaki, Soshi Kawai In this talk, we discuss the predictability of the airfoil stall phenomena using the wall-modeled LES with and without the equilibrium assumption. Through the wall-modeled LES of the Aerospatiale A-airfoil at a realistic high Reynolds number (Reynolds number based on the airfoil chord and free-stream velocity $Re_c=2.1\times 10^6$), we analyze the behavior of the equilibrium and non-equilibrium wall models by comparing with the wall-resolved LES database. One significant difference between the two wall models is observed near the leading edge, where the flow is accelerated by the favorable pressure gradient. The equilibrium wall model cannot capture the skin friction peak near the leading edge, which results in the underestimation of the streamwise boundary-layer development. It is also indicated that the difference in the boundary-layer development consequently affects the flow separation near the trailing edge and the resultant prediction of the lift coefficient. [Preview Abstract] |
Saturday, November 23, 2019 5:32PM - 5:45PM |
B11.00005: Wall-modeled LES of 3D turbulent boundary layer in a square duct with a 30$^\circ$ bend Xiaohan Hu, Minjeong Cho, George Park Mean three dimensionality and high Reynolds number are inherent features of turbulent boundary layers (TBL) in nature and realistic geometries of engineering applications. Wall models for large-eddy simulation (LES) have not been assessed in such nonequilibrium flows, where some of the key modeling assumptions are deemed invalid (e.g., unidirectional flow, alignment of mean shear and turbulent stresses). To this end, we are currently investigating predictive capability of two wall models in a spatially developing three-dimensional TBL. The experiment of Schwarz \& Bradshow (J. Fluid Mech. (1994), vol. 212, pp. 183-209) is considered, where a skewed mean velocity profile is generated as TBL along the floor of a square duct deflects through a 30$^\circ$ bend. The Reynolds number is moderately high ($Re_\theta = 4000 \sim 9000$). Wall-modeled LES (WMLES) with an equilibrium ODE wall model and a non-equilibrium PDE wall model are conducted. Accuracy of the WMLES calculations will be discussed in terms of prediction quality of key mean-flow statistics at two grid resolutions, including boundary-layer integral parameters, skin-friction and wall-pressure coefficients, and $y$-dependent flow turning angle. Cost of wall modeling for the two wall models will be discussed. [Preview Abstract] |
Saturday, November 23, 2019 5:45PM - 5:58PM |
B11.00006: Assessment of Wall-Modeled LES for Turbulent Boundary Layers with Heated/Cooled Wall Ryo Hirai, Soshi Kawai In this talk, we first investigate the effects of wall heat flux induced by heated and cooled walls in zero-pressure-gradient flat-plate turbulent boundary layers by using wall-resolved large-eddy simulation (LES). Based on the wall-resolved LES database, we then assess the capability of the equilibrium wall model (Kawai and Larsson, PoF, 2012) for predicting the thermal turbulent boundary layers. By considering the behaviors of density and viscosity variations in the inner-layer of the thermal boundary-layer, one source of the major errors induced by the existing model is analyzed and a method that allows for the errors to be removed is proposed. Additionally, we also seek the possibility of further improvements in the wall model by modifying the mixing-length model based on the scaling laws derived from the analysis of the wall-resolved LES database. [Preview Abstract] |
Saturday, November 23, 2019 5:58PM - 6:11PM |
B11.00007: A near-wall eddy viscosity for compressible turbulent flows based on velocity transformation with application to wall models Prahladh S. Iyer, Pedro S. Volpiani, Johan Larsson, Sergio Pirozzoli, Mujeeb R. Malik A near-wall eddy viscosity is derived for zero-pressure gradient compressible turbulent flows based on a velocity transformation kernel. This is only valid in the inner layer of the flow with application to wall models. A priori analysis is performed using high-speed Direct Numerical Simulation (DNS) databases of turbulent boundary layers and channels over a range of Mach numbers and thermal conditions using the new formulation to assess its accuracy, and compare it with existing eddy viscosity models. Preliminary results are encouraging, and indicate that the errors in the wall model prediction is directly correlated with the error in the velocity transformation. More specifically, the eddy viscosity obtained with the proposed methodology and using the recent compressible velocity transformation of Volpiani et al. (2019) yields accurate results for supersonic and hypersonic turbulent boundary layers. [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. |
© 2020 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
1 Research Road, Ridge, NY 11961-2701
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700