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 D29: Turbulent Boundary Layers II |
Hide Abstracts |
Chair: Dennice Gayme, Johns Hopkins University Room: Georgia World Congress Center B401 |
Sunday, November 18, 2018 2:30PM - 2:43PM |
D29.00001: Turbulence: The view from the wall Miguel Encinar, Adrian Lozano-Duran, Javier Jimenez For many practical applications, from naive observation to active control strategies, it is desirable to know the internal state of wall-bounded turbulence. However, practical measurements are limited to the wall, where everything vanishes except for the pressure and the wall-parallel vorticities. We focus on the reconstruction of the second invariant of the velocity gradient tensor, $\Pi$, away from the wall from these three measurements, using its linear relation with the inertial component of the pressure, $p$; $\Nabla^2p= 2\Pi$. Statistically optimal approximations are computed for the velocity fields, $\mathbf{u}$, recovering similar results to those of $\Pi$. Their relative error close to the wall ($y^+\approx 10$) is less than $10\%$ of the turbulent kinetic energy. While both the reconstructions of $\Pi$ and $\mathbf{u}$ capture most of the flow features at $y^+ < 30$, they degrade rapidly, holding reasonable errors only for scales comparable or larger than the distance to the wall, or ``attached''. |
Sunday, November 18, 2018 2:43PM - 2:56PM |
D29.00002: Multiscale analysis of non-equilibrium boundary layers at moderate Reynolds numbers Adrian Lozano-Duran, Marco Giometto, George Ilhwan Park, Parviz Moin Turbulent boundary layers with mean-flow three dimensionality occur frequently in complex flow configurations. In these cases, the mean flow direction changes continuously across the boundary layer thickness due to the cross-stream pressure gradient induced by the geometry of the immersed body. This results in counter intuitive reduction of turbulent stresses, with important implications for turbulence modeling. We investigate temporally developing three |
Sunday, November 18, 2018 2:56PM - 3:09PM |
D29.00003: Modelling inner--outer interactions of average wall-shear-stress in rough-wall turbulent channel flows using amplitude modulation Sicong Wu, Carlos Pantano DNS of turbulent channel flows over hemispherical roughness were performed with friction Reynolds numbers $Re_\tau=200$, $400$ and $600$. The inner scaled roughness height $k^+=20$ was maintained for all Reynolds numbers while the spacing between hemispheres was varied from $d/k=2$--$4$. Wall-shear-stresses on both walls were extracted by integrating the stress tensor over the rough surfaces and averaged within arrays of cells of various sizes. Analysis of amplitude modulation was applied to the cell-averaged wall-shear-stress to investigate the inner--outer interactions between the outer large-scale structures and the wall-shear-stress. Pre-multiplied streamwise spectra of the universal wall-shear-stress signals (based on AM model) was investigated for all roughness and a scaling relation between the spectra and cell sizes was found, which collapses all spectra at large wavelengths between various roughness parameters. However, roughness shedding and spectral aliasing modify the scaling at small wavelengths. Spherical harmonics was used to extract the detailed power-spectrum of the wall-shear-stress to help quantify the observed behavior. |
Sunday, November 18, 2018 3:09PM - 3:22PM |
D29.00004: Characterizing Energy Transfer in Restricted Nonlinear Wall-Bounded Turbulence Benjamin Minnick, Dennice F Gayme The computational expense of resolving all scales of turbulence in wall-bounded flows is impractical for engineering applications, prompting the development of reduced-order models. Numerical and experimental evidence showing the existence of structures elongated in the streamwise direction of turbulent shear flows motivates the exploration of a streamwise coherent modeling framework. The restricted nonlinear (RNL) modeling paradigm takes such an approach by omitting the streamwise varying part of the perturbation-perturbation nonlinearity in the Navier-Stokes equations, thereby reducing the number of active streamwise Fourier modes in order to reduce computational cost. This model has shown promise in reproducing first- and second-order statistics with as few as one nonzero streamwise mode. Despite the promise of this approach, certain aspects of the dynamics have yet to be fully understood. Here we focus on the production, dissipation, and transport of energy on these restricted modes. Spectra and energy budgets from the RNL system supported by its natural set and selected set of streamwise modes are compared to DNS data in the low-to-moderate Reynolds number range. |
Sunday, November 18, 2018 3:22PM - 3:35PM |
D29.00005: Evidence of an inverse cascade in a plane wall jet through large-scale forcing. Shibani Bhatt, Sravan kumar Artham, Ebenezer Gnanamanickam The present work focuses on understanding the scale interactions within a turbulent boundary layer through large-scale, large-perturbation forcing. A plane wall jet (PWJ), the model flow field, was subject to acoustic forcing. The streamwise wavelengths of the forcing $\lambda_{ex}$, were larger than the local integral length scale $\delta(x)$, where $x$ is the streamwise direction. When the forcing wavelength $\lambda_{ex}$ was larger than the wavelengths of the naturally occurring outer large-scales of the unperturbed jet $\lambda_{n}$, the forcing energy was transferred to scales smaller than $\lambda_{ex}$ in a forward cascade. As the PWJ develops downstream, the local integral length scale $\delta(x)$ gets larger due to the spreading of the PWJ. Hence, as the PWJ develops at downstream locations where $\lambda_{ex}<\lambda_n$, the forcing energy was transferred to scales larger than $\lambda_{ex}$ indicating an inverse cascade. This transfer was accompanied by a transfer of momentum away from the near-wall region to the outer region of the plane wall jet and reduction of the friction velocity for all $\lambda_{ex}$ considered. |
Sunday, November 18, 2018 3:35PM - 3:48PM |
D29.00006: Laminar boundary layer-free-stream turbulence interface, turbulent spot-free-stream turbulence interface, and boundary layer turbulence-free-stream turbulence interface James Martin Wallace, Xiaohua Wu, Jean-Pierre Hickey The BTFTI concept refers to the interface between boundary layer turbulence and free-stream turbulence (Wu et al, PNAS, 114, 2017) . Here, we further introduce the TSFTI and the LBFTI to denote the interfaces separating, respectively, boundary layer transitional-turbulent spots and the laminar boundary layer from free-stream turbulence. Previous work has focused almost exclusively on the traditional concept of the turbulent/non-turbulent interface (TNTI), often ignoring the possibility that the boundary layer’s free-stream can be occupied by finite amplitude vortical turbulent flow. In this work, we study the BTFTI, LBFTI and TSFTI using DNS of a spatially-developing, smooth, zero-pressure-gradient, flat-plate boundary layer beneath a decaying free-stream of homogeneous isotropic turbulence at momentum-thickness Reynolds numbers from 80 to 3000. These interfaces are quantified using the iso-surface of temperature at a threshold value of 1 percent of the wall temperature. Geometric characteristics of the interfaces were measured. Conditional sampling of velocity, scalar, swirling strength and vorticity across the interfaces was performed. Connections between the interfaces with boundary layer vortices have been established through high-quality flow visualizations. |
Sunday, November 18, 2018 3:48PM - 4:01PM |
D29.00007: Turbulent/non-turbulent interface detection for wall-bounded flows Nico Reuther, Christian J. Kaehler The intermittency between laminar and turbulent flow is a well known feature in the outer region of turbulent boundary layers. Detecting the complex interface between irrotational fluid and inhomogeneous turbulence and understanding the physical processes occurring at the interface have achieved a lot of scientific attention. However, no universal method to detect the turbulent/non-turbulent interface (TNTI) has been established. This holds in particular for high Reynolds number experimental data where the spatial resolution is limited due to the requirement of resolving the entire intermittent region. Therefore, this work analyzes the sensitivity of TNTI detection approaches. The techniques considered are based on the turbulent kinetic energy, the homogeneity of the potential flow region and the particle image density. The effect of bias errors is studied on both instantaneous snapshots and also with respect to statistically averaged quantities such as the intermittency factor and the length of the TNTI. In addition, the suitable working range for the techniques is outlined and several possible applications that can contribute to a better understanding of turbulent boundary layers are indicated. |
Sunday, November 18, 2018 4:01PM - 4:14PM |
D29.00008: Influence of wall-attached structures on the interface of the quiescent core region in turbulent pipe flow Jongmin Yang, Jinyul Hwang, Hyung Jin Sung The quiescent core region in pipe flow is explored using the direct numerical simulation data of turbulent pipe flow at Reτ = 926. The quiescent core region is one of the uniform momentum zones and contains the highest convection velocity in turbulent pipe flow. The interface of the quiescent core region has a similar physical behavior to the turbulent/non-turbulent interface and underlying internal layer. The turbulent structures in pipe flow are detected by using the percolation theory. The wall-attached structure is closely associated with geometrical characteristics of the interface and entrainment, which is a transport phenomenon at the interface. The wall-attached structure has a huge influence on dynamics in the vicinity of the interface. The wall-attached u-structure acquired in the present work is a structure that is subject to the attached-eddy hypothesis, and the influence of the wall-attached structures near the interface varies according to height change of the wall-attached structures. Finally, dynamics of the interface is reconstructed using conditional statistics according to the growth of the wall-attached structure. |
Sunday, November 18, 2018 4:14PM - 4:27PM |
D29.00009: Wall-attached structures of streamwise velocity fluctuations in turbulent pipe flow Jinyul Hwang, Hyung Jin Sung The wall-attached structures of streamwise velocity fluctuations (u) are investigated using direct numerical simulation data of turbulent pipe flow at Reτ ≈ 930 and 3000. The attached clusters of u are extracted in instantaneous flow fields. At Reτ ≈ 3000, the attached structures exist over a broad range leading to the presence of the linear relationship between the streamwise length and height (ly) for the taller structures (ly+ > 400). The width of the objects follows the linear law over 100 < ly+ < R+. The population density of the attached structures is inversely proportional to ly over 0.3 < ly/R < 0.6, reminiscent of the distribution of hierarchy scales. The streamwise velocity is reconstructed through the superposition of these structures. The log-law indicator function shows a plateau over 164 (≈ 3Reτ1/2) < y+ < 0.18R+ for the positive u, whereas no flat region for the negative u. The magnitude of the wall-shear stress for the positive u is approximately ten times larger than that for the negative u, representing that the logarithmic region observed in the positive u is attributed to high local Reτ. |
Sunday, November 18, 2018 4:27PM - 4:40PM |
D29.00010: Wall-attached structures of velocity fluctuations in a turbulent boundary layer subjected to adverse pressure gradient Min Yoon, Jinyul Hwang, Jongmin Yang, Hyung Jin Sung We demonstrate the logarithmic region of turbulent intensity by identifying wall-attached structures of velocity fluctuations through the direct numerical simulation of an adverse pressure gradient turbulent boundary layer (APG TBL) (β = 1.43 and Reτ = 775). For comparison, the attached structures of zero pressure gradient (ZPG) TBL (Reτ = 825) are included. The wall-attached structures are self-similar with respect to their heights (ly), and in particular the population density of the streamwise component (u) scales inversely with ly, reminiscent of the hierarchy of attached eddies. The turbulence intensities contained within the wall-parallel components (u and w) exhibit the logarithmic behavior. The tall attached structures of u are composed of multiple uniform momentum zones with long streamwise extents. The magnitude of the near-wall peak observed in the streamwise turbulent intensity increases with increasing ly, reflecting the nested hierarchies of the attached u structures. The spanwise length (lz) of the wall-attached structures in APG TBL increases more than that of ZPG TBL, while the streamwise length (lx) decreases. |
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. |
© 2025 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