Bulletin of the American Physical Society
67th Annual Meeting of the APS Division of Fluid Dynamics
Volume 59, Number 20
Sunday–Tuesday, November 23–25, 2014; San Francisco, California
Session G26: Rough Wall Boundary Layers I |
Hide Abstracts |
Chair: Karen Flack, United States Naval Academy Room: 2007 |
Monday, November 24, 2014 8:00AM - 8:13AM |
G26.00001: Turbulent transport of momentum and scalars in urban-like geometries Qi Li, Elie Bou-Zeid, William Anderson, Sue Grimmond A numerical study is carried out using large-eddy simulations to investigate the mechanisms of turbulent transport of momentum and passive scalars over urban-like geometries. The immersed boundary method is used to represent buildings; this induces ``ringing,'' i.e. the Gibbs phenomenon associated with the use of spectral discretization in domains with sharp discontinuities. We present a new approach to reduce this ringing and improve the numerical accuracy of the method. Topological parameters, such as the frontal area index and the plan area index, were varied to examine their impact on turbulence and transport characteristics. The heterogeneity of the surface is shown to increase both the heteorogeneity and anisotropy of the flow, and to significantly modulate the efficiencies of momentum and passive scalar transport. [Preview Abstract] |
Monday, November 24, 2014 8:13AM - 8:26AM |
G26.00002: Optimizing the determination of roughness parameters of urban canopies Auvi Rahman, Pablo Huq We present an optimization procedure to determine the roughness parameters for an urban canopy. The mean velocity profile above an urban canopy is described by the log law via the roughness parameters: zero-plane displacement height $d$, roughness length $z_0$, and friction velocity $u_*$. Traditionally these parameters are obtained from a single mean velocity profile. We have devised a new procedure which is akin to the bootstrap or jackknife resampling method where multiple mean velocity profiles are generated from a single mean velocity profile. Each of the generated profiles are then best fit to the log law and sets of $d$, $z_0$, and $u_*$ are estimated. These sets of values show distinct clusters when plotted against the relative sensitivity of the log law to the zero-plane displacement height $d$. A single representative or optimal value of the roughness parameters are then obtained automatically by utilizing a standard clustering procedure. Application of this method is also presented for field and laboratory data. [Preview Abstract] |
Monday, November 24, 2014 8:26AM - 8:39AM |
G26.00003: Turbulent boundary layer flow over broad-banded roughness Geno Pawlak, Payam Aghsaee, Saeed Mazrouei, Stefano Leonardi, Krishnakumar Rajagopalan, Marcelo Kobayashi The response of the boundary layer to a regular roughness is often parameterized in terms of the length scales defining the roughness. Difficulty arises in the case of broad-banded and highly irregular roughness distributions such as over coral reefs or urban canopies where the length scale that determines the response of the boundary layer is not clear. Here we use a spectral description for roughness to create idealized two-dimensional irregular roughness profiles, using square waves as a basis function. Laboratory experiments along with Direct Numerical Simulations (DNS) are used to examine the hydrodynamic response to the broad-banded roughness and flow characteristics are related to geometric characteristics of the boundary. The simulations and experiments show that the nature of the flow over two-dimensional irregular walls can be determined as a function of the hydrodynamic origin, which, in turn, can be determined as a function of a mean cavity shape. Results are interpreted in terms of the spectral characteristics of the roughness. The contribution of the various spectral components to the total drag is analyzed for each case. The roughness spectrum influences the flow through the shape of the cavities on the wall and can provide some guidance in predicting the nature of the flow. [Preview Abstract] |
Monday, November 24, 2014 8:39AM - 8:52AM |
G26.00004: Effect of surface morphology on drag and roughness sublayer in flows over regular roughness elements Marco Placidi, Bharathram Ganapathisubramani The effects of systematically varied roughness morphology on bulk drag and on the spatial structure of turbulent boundary layers are examined by performing a series of wind tunnel experiments. In this study, rough surfaces consisting of regularly and uniformly distributed LEGO\texttrademark~ bricks are employed. Twelve different patterns are adopted in order to methodically examine the individual effects of frontal solidity ($\lambda_F$, frontal area of the roughness elements per unit wall-parallel area) and plan solidity ($\lambda_P$, plan area of roughness elements per unit wall-parallel area), on both the bulk drag and the turbulence structure. A floating element friction balance based on Krogstad $\&$ Efros (2010) was designed and manufactured to measure the drag generated by the different surfaces. In parallel, high resolution planar and stereoscopic Particle Image Velocimetry (PIV) was applied to investigate the flow features. This talk will focus on the effects of each solidity parameter on the bulk drag and attempt to relate the observed trends to the flow structures in the roughness sublayer. [Preview Abstract] |
Monday, November 24, 2014 8:52AM - 9:05AM |
G26.00005: Near-wall 3D velocity measurements above biomimetic shark skin denticles using Digital In-line Holographic Microscopy Mostafa Toloui, David Brajkovic, Jiarong Hong Digital In-line Holography is employed to image 3D flow structures in the vicinity of a transparent rough surface consisting of closely packed biomimetic shark skin denticles as roughness elements. The 3D printed surface replicates the morphological features of real shark skin, and the denticles have a geometrical scale of 2 mm, i.e.~10 times of the real ones. In order to minimize optical aberrations near the fluid-roughness interface and enable flow measurements around denticles, the optical refractive index of the fluid medium is maintained the same as that of the denticle model in an index-matched flow facility using NaI solution as the working fluid. The experiment is conducted in a 1.2 m long test section with 50 mm $\times$ 50 mm cross section. The sampling volume is located in the downstream region of a shark skin replica of 12'' stretch where the turbulent flow is fully-developed and the transitional effect from smooth to the rough surface becomes negligible. Several instantaneous realizations of the 3D velocity field are obtained and are used to illustrate turbulent coherent structures induced by shark-skin denticles. This information will provide insights on the hydrodynamic function of shark's unique surface ornamentation. [Preview Abstract] |
Monday, November 24, 2014 9:05AM - 9:18AM |
G26.00006: The impact of algal biofilms on skin-friction in a turbulent channel flow Michael Schultz, Karen Flack, Cecily Steppe, Jessica Walker Experiments were carried out in a fully-developed, turbulent channel flow facility over a wide Reynolds number range. The wall shear stress was determined using the bulk flow rate and the streamwise pressure gradient in the downstream section of the channel. A biofilm dominated by three species of diatoms developed on acrylic test surfaces exposed for four days in a brackish tidal environment at the United States Naval Academy. The resulting biofilm had an average thickness of 200 $\mu$m. This biofilm had a significant effect on the flow showing a doubling of the skin-friction compared to the hydraulically-smooth condition at the highest Reynolds number. Scale up of the present results to ship scale indicates that this biofilm would generate an 18\% powering penalty for a mid-sized naval ship at cruising speed. [Preview Abstract] |
Monday, November 24, 2014 9:18AM - 9:31AM |
G26.00007: Numerical simulation of adverse-pressure-gradient boundary layer with or without roughness Pouya Mottaghian, Junlin Yuan, Ugo Piomelli Large-eddy and direct numerical simulations are carried out on flat-plate boundary layer over smooth and rough surfaces, with adverse pressure gradient.The deceleration is achieved by imposing a wall-normal freestream velocity profile, and is strong enough to cause separation at the wall. The Reynolds number based on momentum thickness and freestream velocity at inlet is 600. Numerical sandgrain roughness is applied based on an immersed boundary method, yielding a flow that is transitionally rough. The turbulence intensity increases before separation, and reaches a higher value for the rough case, indicating stronger mixing. Roughness also causes higher momentum deficit near the wall, leading to earlier separation. This is consistent with previous observation made on rough-wall flow separation over a ramp. In both cases, the turbulent kinetic energy peaks inside the shear layer above the detachment region, with higher values in the rough case; it then decreases approaching the reattachment region. Near the wall inside the separation bubble, the near-zero turbulent intensity indicates that the turbulent structures are lifted up in the separation region. Compared with the smooth case, the shear layer is farther from the wall and the reattachment length is longer on the rough wall. [Preview Abstract] |
Monday, November 24, 2014 9:31AM - 9:44AM |
G26.00008: DNS of turbulent channel flow over hemispherical roughness Sicong Wu, Kenneth Christensen, Carlos Pantano Turbulent channel flows over certain rough surfaces have been studied using direct numerical simulation (DNS) in recent years. Most of these previous studies have focused on roughness with closely packed cubic or cylindrical ribs and it is well documented that the near-wall flow is strongly affected by the roughness but the outer region is relatively unaffected, in agreement with previous experiment evidence. In this study, DNS of turbulent channel flow with hexagonally packed hemispheres on a wall is performed. The roughness height k/h is about 10 and the average spacing between hemispheres from center to center is of the order of 6 times of the roughness height. The friction Reynolds number is approximately 400 and the simulation employs the NEK5000 solver, an incompressible Navier-Stokes code based on spectral elements. We will discuss detailed turbulence statistics in the near-wall region and forces on the rough surface, with the aim of improving understanding of flow physics to guide development of reliable LES models. [Preview Abstract] |
Monday, November 24, 2014 9:44AM - 9:57AM |
G26.00009: Effects of roughness on accelerating boundary layer Junlin Yuan, Ugo Piomelli Large-eddy simulation is carried out on a rough-wall boundary layer with favourable pressure gradient (FPG) to study the combined effects of FPG and roughness. The acceleration is strong enough to start relaminarization on a smooth wall; the fully rough regime is achieved in the FPG region. Unlike the flow over a smooth wall, where FPG causes significant Reynolds-stress anisotropy and decoupling between the inner and outer layers, on the rough wall, the interaction between inner and outer layers is amplified near the roughness crest by FPG, due to the increased Reynolds shear stress associated with strong sweeping events, which cause large fluxes of turbulent kinetic energy towards the wall. Spatial variations of time-averaged velocities in the roughness sublayer are observed. They scale with friction velocity and the roughness length scale and, in the FPG region, they increase in magnitude due to the work of the mean flow against the form drag. The spatial disturbances of time-averaged Reynolds stresses play an important role, by sustaining the vertical turbulent motions through a production mechanism, and subsequently lead to the increase in Reynolds shear stress. As a result, relaminarization is not achieved on the rough wall. [Preview Abstract] |
Monday, November 24, 2014 9:57AM - 10:10AM |
G26.00010: Direct numerical simulation of a turbulent pipe with systematically varied three-dimensional roughness Leon Chan, Michael MacDonald, Daniel Chung, Nicholas Hutchins, Andrew Ooi Direct Numerical Simulations (DNS) are conducted at low to medium Reynolds numbers for a turbulent pipe flow with roughness. The roughness, which is comprised of three-dimensional sinusoidal elements, causes a downward shift in the mean velocity profile known as the Hama roughness function $\Delta U^+$. In engineering applications, $\Delta U^+$ (which is related to the coefficient of drag $C_f$) is an important parameter as it is used to quantify the increase in drag and the decrease in efficiency. To have a better understanding of roughness and how it affects the flow, a range of numerical studies were conducted where the roughness height $h^+$, wavelength $\lambda ^+$ and Reynolds number of the flow are varied. For the range of cases simulated, it is found that the roughness average height $k_a^+$ (which is proportional to $h^+$) is strongly correlated to the roughness function $\Delta U^+$ whereas $\lambda^+$ has a weaker influence on the flow. Results from simulations of more complicated surfaces comprised of two superimposed modes of different wavelength are also presented. Analysis of the turbulence statistics convincingly supports Townsend's outer-layer hypothesis for all of the cases simulated. [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