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 F29: Turbulent and Planetary Boundary Layers |
Hide Abstracts |
Chair: William Anderson, University of Texas, Dallas Room: Georgia World Congress Center B401 |
Monday, November 19, 2018 8:00AM - 8:13AM |
F29.00001: On the shape of resolvent modes in shear-driven turbulence Scott Dawson, Beverley J McKeon The resolvent formulation of the Navier-Stokes equations gives a means for prediction of turbulent structures and statistics using the singular value decomposition of the resolvent operator based on the appropriate turbulent mean, following the framework developed by McKeon & Sharma (2010). This talk will discuss analytic approximations to the shape of resolvent modes in shear-driven turbulent flows. Such systems typically exhibit large spectral gaps in the singular values of their associated family of resolvent operators, which can make resolvent-based decompositions particularly efficient for reduced-complexity modeling. Here, we use results concerning the pseudospectra of scalar operators (e.g., Reddy et al. 1993, Trefethen 2005) to derive analytic approximations to components of leading pseudospectral (resolvent) modes. This provides a theoretical framework for understanding the origin of observed structures, and gives a method for mode estimation without the need for large numerical computations. We will discuss the implications of these findings for real-time estimation and control, and will further demonstrate the utility of this approach for modeling passive scalar dynamics. |
Monday, November 19, 2018 8:13AM - 8:26AM |
F29.00002: Fully-resolved temperature/velocity measurements in a turbulent thermal boundary layer Marcel Michael Louis, Matthew Fu, Mark Miller, Marcus Hultmark Experimental measurements of scalar transport in turbulent flows are often limited by spatial and temporal resolutions of available probes, especially at high Reynolds numbers. Most investigations of temperature transport have relied on the use of cold-wires to measure the temperature fluctuations. However, the minimal length requirements typically necessary to avoid end-conduction effects conflict with the fine spatial resolution required to fully resolve the smallest scales of turbulence. The goal of this study is to provide fully resolved measurements of the scalar transport in a turbulent channel flow using temperature as a passive scalar. The use of NSTAP and T-NSTAP (temperature nanoscale thermal anemometry probe) for velocity and temperature measurements respectively, presents a unique experimental approach as their small sizes allow for increased temporal and spatial resolution. The main component of the experimental setup is a fully developed channel flow using air as a working fluid with an accompanying developing thermal boundary layer formed from a change in the wall temperature. The setup represents a case where a well-studied velocity field (turbulent channel flow) governs the lesser studied scalar field (thermal boundary layer). |
Monday, November 19, 2018 8:26AM - 8:39AM |
F29.00003: Measurements of unsteady turbulence over wind-driven surface water waves using time-resolved PIV in an atmospheric boundary layer wind-wave tunnel Corey D Markfort, Matthew Stegmeir Wind generated surface waves mediate the transfer of momentum, energy, and scalars across the air-water interface. Models of the atmospheric boundary layer (ABL) and the surface mixing layers of lakes and oceans rely on parameterizations for surface fluxes which are typically based on rough-wall boundary layer theory. Waves are often out of equilibrium with the ABL flow and energy is commonly transferred back to the ABL. There is limited information about the details of the coupled boundary layers which are needed to better understand the role of turbulence in these interactions. We employ time-resolved PIV to measure the detailed structure of air and water boundary layer turbulence under varying wind and wave conditions in the new IIHR Boundary-Layer Wind-Wave Tunnel. The facility combines a 30-m long recirculating water channel with an open-return boundary layer wind tunnel. A thick turbulent boundary layer is developed in the 1 m high air channel, over the water surface, allowing for the study of large-scale boundary layer turbulence interacting with a wind-driven wave field. The first time-resolved turbulence measurements are presented for varying wind speeds over breaking waves and details of the time-evolution of boundary layer fluxes are reported. |
Monday, November 19, 2018 8:39AM - 8:52AM |
F29.00004: Buoyancy effects on large scale motions and amplitude modulation Scott Salesky, William Anderson Recent studies of high-Re wall turbulence have revealed the existence of large scale motions (LSMs) that populate the logarithmic layer and amplitude modulate small-scale turbulent fluctuations near the wall. However, the extent to which LSMs occur in flows with unstable thermal stratification (low Richardson number) and Coriolis forcing (low Rossby number) is not well understood. In this study, we consider the effects of buoyancy on LSMs through large eddy simulations (LES) of the convective atmospheric boundary layer spanning weakly to highly unstable thermal stratification. It is found that, as the CBL becomes increasingly unstable, the inclination angle of structures near the ground increases from 12-15° to nearly 90°. Furthermore, the scale separation between the inner and outer peaks in the premultiplied velocity spectra decreases until only a single peak remains (comparable in magnitude to the boundary layer depth). Under weakly convective conditions, significant amplitude modulation occurs due to both the large-scale streamwise and vertical velocity; however, under highly convective conditions amplitude modulation occurs due to the large-scale vertical velocity alone. Connections between the topology of turbulent structures and flow modulation will also be discussed. |
Monday, November 19, 2018 8:52AM - 9:05AM |
F29.00005: Turbulent boundary layers over a porous surfaces Intesaaf Ashraf, Eduardo Rodriguez Lopez, Manuel Ferreira, Bharathram Ganapathisubramani Turbulent flow over the porous surface is one of the least documented problems, despite having the potential to affect a wide range of applications. This study presents an experimental investigation of turbulent boundary layer over open cell porous foam where the pore size and foam thickness are systematically varied. Hot-wire anemometry, Planar and Stereoscopic particle image velocimetry (PIV) are used for flow characterisation and floating element drag balance for drag measurement. All the measurements are performed over a range of freestream velocities at a fetch of 3.5m in a turbulent boundary layer that has developed over the porous media. The skin-friction coefficient is found to be considerably affected by the porosity as well as the thickness of the porous foam. However, at higher freestream speeds, a Reynolds number invariant skin-friction behaviour appears to exist, similar to rough walls. Hot-wire and PIV data shows that it is very difficult to isolate a clear logarithmic region in the mean velocity profiles. Preliminary assessment of the mean flow and turbulence statistics also indicates that outer layer similarity may be invalid for the surfaces considered here. Further analysis of the energy spectra as well as the spatial structure of the flow will be presented. |
Monday, November 19, 2018 9:05AM - 9:18AM |
F29.00006: Relation between pressure and spanwise velocity fluctuations in turbulent channel flows Ali Mehrez, Yoshinobu Yamamoto, Yoshiyuki Tsuji We study the logarithmic behavior of the mean-square pressure fluctuations ⟨p+2⟩ in fully developed turbulent channel flows. Data of direct numerical simulation of channel flow is used in the present research for Reynolds number flows up to Reτ=4000 . The profile of ⟨p+2⟩ is found to follow a logarithmic behavior at some distances from the wall which is similar to that of the mean-square spanwise velocity fluctuations ⟨w+2⟩ at the same Reynolds number. The results are also emphasized for the higher-order moments of p and w where the logarithmic behavior is confirmed. Spectral analysis in the streamwise and spanwise directions indicate that the ridge of the spectra of p and w are overlapping each other within the regions where ⟨p+2⟩ and ⟨w2⟩ show the log-trends. In addition, within these regions, the shape of the two-dimensional spectra of p and w collapse with each other, and the contours of the spectra at a constant level are bounded by a linear relationship between the streamwise and spanwise length scales This linear relation indicates that p and w may be associated with geometrically shaped self-similar eddies. The shape of these self-similar eddies is investigated, and the scaling of their sizes with the distance from the wall are discussed.
|
Monday, November 19, 2018 9:18AM - 9:31AM |
F29.00007: Direct numerical simulation of a shear driven three-dimensional turbulent boundary layer Hiroyuki Abe Direct numerical simulations have been carried out in a flat-plate turbulent boundary layer with an impulsive spanwise velocity at the wall, thus yielding a skewed three-dimensional turbulent boundary layer (3DTBL). Particular attention is given to the effects of cross flow and Reynolds number. The present inlet Reynolds number is equal to momentum thickness Reynolds number Reθ=300, 600 and 900. The latter Reθ is close to that of Kannepalli and Piomelli (2000) who performed the wall-resolved large eddy simulation at Reθ≈1100. After imposing cross flow, a decrease in magnitude is observed for the turbulent kinetic energy, associated with the behavior that the shear stress lags in change behind the strain rate. In a near equilibrium 3DTBL, the spanwise Reynolds stress ⟨vw⟩ and the Reynolds shear stress ⟨uv⟩ become more and less energetic with increasing cross flow. A large magnitude of cross flow leads to not only short streamwise extent of streaks but also clustering of vortical structures, where the wall pressure fluctuation also shows a large-scale pattern. In the recovery region, a significant Re effect is found for the streamwise skin friction coefficient, but not for the spanwise skin friction one. In this region, the recovery to the zero-pressure-gradient state is also slow. |
Monday, November 19, 2018 9:31AM - 9:44AM |
F29.00008: Roughness Model for Unresolved Sub-grid Structures in Turbulent Flow Over Fractal Urban-like Topographies Xiaowei Zhu, William Anderson Turbulent flows over multiscale, fractal-like topographies are investigated via large-eddy simulation (LES). Iterated function system (IFS) topographies were used to construct urban-like, fractal geometries. The IFS featured the same central square-based prism for generation one, while predefined changes to the mapping function altered the number of descendant generations and thus the fractal dimension. In this way, fractal dimension could be nominated a priori. These idealized synthetical geometries possess the salient features of real urban topographies without the confounding complexities that would otherwise inhibit general scientific deductions. Five fractal dimensions were used to investigate the parameterization of unresolved generations. We quantified the momentum deficit associated with changing attributes, which enabled a posteriori deduction of roughness length parameters needed to model aerodynamic surface stress. We further showed that the aerodynamic stress associated with descendant, sub-generation elements can be parameterized, with only the first few generations resolved on the computational mesh. |
Monday, November 19, 2018 9:44AM - 9:57AM |
F29.00009: Large-eddy simulations of a diurnal cycle driven by mesoscale and observational profile assimilation Dries Allaerts, Caroline Draxl, Matthew Churchfield The coupling of microscale large-eddy simulations (LES) to mesoscale models enables a more realistic representation of turbulent atmospheric boundary layers and may allow for accurate prediction of wind-turbine performance and loading under a wide variety of inflow conditions. One of the challenges associated with meso-micro coupling is the specification of driving conditions for the microscale simulation. One approach is to use spatially and temporally averaged budget terms or tendencies from the mesoscale simulation, but the choice of averaging parameters affects the microscale solution. In this study, we investigate an alternative approach to drive LES based on data-assimilation techniques. The assimilation approach employs Newtonian relaxation to nudge the mean profiles in the microscale simulation towards a desired velocity and temperature profile. We simulate a diurnal cycle observed at the Scaled Wind Farm Technology (SWiFT) facility in Texas, and we compare LES driven by mesoscale profile assimilation, mesoscale tendencies, and observational profile assimilation. |
Monday, November 19, 2018 9:57AM - 10:10AM |
F29.00010: Effect of domain size and mesh resolution on stable and unstable planetary boundary layer turbulence statistics Srinidhi Nagarada Gadde, Richard Stevens Wall-modeled large-eddy simulations (LES) of quasi-equilibrium stable and convective planetary boundary layers (PBLs) are carried out for different domain sizes and mesh resolutions. Wall stresses and fluxes are modeled using the Monin-Obukhov similarity theory. A Lagrangian averaged scale-dependent dynamic model is extended for a scalar field to model the sub-grid scale fluxes in the pseudo-spectral based code. The simulations are carried out on meshes varying from 323 to 10243. For the convective PBL, mean quantities such as velocity and temperature numerically converge with increasing grid resolution. However, higher order moments, especially the skewness of the vertical velocity, are extremely sensitive to mesh resolution. For stable PBL (GABLS1 case), doubling the domain size has no significant effect on the first order statistics, but the statistics are very sensitive to mesh resolution. With increasing mesh resolution, the height of the low-level jet (LLJ) and the boundary layer height reduces. |
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