Bulletin of the American Physical Society
68th Annual Meeting of the APS Division of Fluid Dynamics
Volume 60, Number 21
Sunday–Tuesday, November 22–24, 2015; Boston, Massachusetts
Session M21: Turbulence: Modeling II |
Hide Abstracts |
Chair: Kunlun Bai, Yale University Room: 209 |
Tuesday, November 24, 2015 8:00AM - 8:13AM |
M21.00001: A low-dimensional model for large-scale coherent structures Kunlun Bai, Dandan Ji, Eric Brown We demonstrate a methodology to predict the dynamics of the large-scale coherent structures in turbulence using a simple low dimensional stochastic model proposed by Brown and Ahlers (Phys. Fluids, 2008). The model terms are derived from the Navier-Stokes equations, including a potential term depending on the geometry of the system. The model has previously described several dynamical modes of the large-scale circulation (LSC) in turbulent Rayleigh-B\'{e}nard convection. Here we test a model prediction for the existence of a new mode where the LSC stochastically changes direction to align with different diagonals of a cubic container. The model successfully predicts the switching rate of the LSC at different tilting conditions. The success of the prediction of the switching mode demonstrates that a low-dimensional turbulent model can quantitatively predict the existence and properties of different dynamical states that result from boundary geometry. [Preview Abstract] |
Tuesday, November 24, 2015 8:13AM - 8:26AM |
M21.00002: Revisit on Proper Orthogonal Decomposition Method Mahdi Hosseinali, Joseph Hall Understanding the underlying mechanisms of seemingly random movements in turbulent flows is the most challenging ongoing area of fluid dynamics. Structures with characteristic length scale comparable to the geometry of the flow, so called coherent structures, are assumed to be responsible for the major characteristic behaviors of the flow. These structures then break down to smaller structures and so on until they get damped on viscose level. Identification of coherent structures thus is of paramount importance in fluid dynamics. Among numerous methods POD seems to be the most successful approach to breaks the sophisticated turbulent field into a series of unbiased modes. Since its introduction to fluid dynamic community by Lumley the only major improvement was method of snapshots by Sirovich which is used today on PIV measurements. This talk is aimed to look at different forms of POD kernels which are mostly based on a physical point of view rather than pure mathematics. [Preview Abstract] |
Tuesday, November 24, 2015 8:26AM - 8:39AM |
M21.00003: Interaction of two-dimensional turbulence with a sheared channel flow: a numerical study Leon Kamp, Vitor Marques Rosas Fernandes, GertJan van Heijst, Herman Clercx Interaction of large-scale flows with turbulence is of fundamental and widespread importance in geophysical fluid dynamics and also, more recently for the dynamics of fusion plasma. More specifically the interplay between two-dimensional turbulence and so-called zonal flows has gained considerable interest because of its relevance for transport and associated barriers. We present numerical results on the interaction of driven two-dimensional turbulence with typical sheared channel flows (Couette and Poiseuille). It turns out that a linear shear rate that is being sustained by moving channel walls (Couette flow) is far more effective in suppressing turbulence and associated transport than a Poiseuille flow. We explore the mechanisms behind this in relation to the width of the channel and the strength of the shear of the background flow. Also the prominent role played by the no-slip boundaries and the Reynolds stress is discussed. [Preview Abstract] |
Tuesday, November 24, 2015 8:39AM - 8:52AM |
M21.00004: The effect of a solid boundary on homogeneous isotropic turbulence: an experimental investigation Blair Johnson, Edwin Cowen An experimental study is performed to investigate the turbulent boundary layer at a smooth solid boundary in the absence of mean shear. Driven by a spatio-temporally varying randomly actuated synthetic jet array suspended above an enclosed water tank, high Reynolds number horizontally homogeneous isotropic turbulence is generated with negligible mean flow. Acoustic Doppler velocimetry and particle image velocimetry measurement techniques are used to characterize the near-boundary flow with statistical metrics such as turbulence intensities, turbulent kinetic energy, temporal and spatial spectra, and integral length scales. We compare various methods of computing dissipation rates and evaluate the assumptions of isotropy that are typically invoked. Furthermore, we consider Eulerian frequency spectra to improve dissipation estimates from single-point velocity measurements. Our investigations examine the effect of altering jet firing parameters on the integral length scale and resulting turbulent structures. We conclude with thoughts on the use of~the dissipation rate to parameterize the bed stress in the absence of mean shear where traditional friction velocity methods struggle to fully capture the local stresses and energy present in turbulence. [Preview Abstract] |
Tuesday, November 24, 2015 8:52AM - 9:05AM |
M21.00005: Counter gradient diffusion in a plane wall jet O. Ramesh, V. Dhamotharan Turbulent wall-jets are important in a variety of applications such as the Coanda effect for boundary layer separation control, film-cooling applications in a jet engine etc., One of the important features of a wall jet is the existence of a region of counter gradient diffusion of momentum. The counter-gradient region is a sort of pathological situation for RANS based turbulence models as they would not be able to predict this region. In this study we show from our experiments in a wall jet the counter-gradient region of diffusion can be understood from simple structural models for a wall jet eddy. Towards this flow visualization and hotwire measurements have been performed. It is seen from the smoke flow visualizations that the outward portion of the flow is backward leaning i.e. in the upstream direction. This is consistent with the orientation of eddy structure obtained from two-point correlation measurements. A building block eddy of a wall jet is proposed that has aspects of a boundary layer eddy in the inner wall region and a jet eddy in the outer region. It is argued by a simple vortex dynamics model that the counter-gradient region occurs due to the influence of the jet eddies in the near-wall region. [Preview Abstract] |
Tuesday, November 24, 2015 9:05AM - 9:18AM |
M21.00006: Hierarchical Structure of Fast Stretching Vortices in Turbulent Flows Masato Hirota, Yu Nishio, Seiichiro Izawa, Yu Fukunishi Geometric relations between fast stretching vortices of a certain scale and vortices twice larger are investigated to understand the energy cascade process in a turbulent flow from a view point of vortex interactions. Multi-scaled vortices are extracted from a homogeneous isotropic turbulence using a band-pass filter based on the Fourier decomposition. An extracted vortex is reconstructed as a set of short cylindrical vortex segments. The stretching speed of a vortex segment, caused by the velocity field, which vortices twice the scale generate, is measured. Then, the vortex segments are classified into the three categories by their stretching speeds: The first is the "super fast" stretching vortices, whose stretching speed is in top 1{\%} of total segments. The second is "moderately fast" stretching vortices, whose stretching speed is in top 10{\%} of total segments. The third is "slowly or not" stretching vortices, whose stretching speed is lower than the top 10{\%}. The geometric relations between the vortex segments and its surrounding vortices which are larger are analyzed in terms of the distances and the relative angles. The result shows that vortex segments tend to be aligned parallel or anti-parallel to the larger vortices for the "slowly or not" stretching vortices. For the "super fast" stretching vortices, it is found that they tend to be orthogonal to the vortices of double size. Meanwhile, no particular tendency is found for the "moderately fast" stretching vortices. [Preview Abstract] |
Tuesday, November 24, 2015 9:18AM - 9:31AM |
M21.00007: Evolution of the velocity gradient tensor in the near field of a square cylinder Massimiliano Breda, Oliver Buxton The condition of the velocity gradient tensor (VGT) is analysed in the near field of the flow past a square cylinder. The data was acquired by tomographic particle image velocimetry at a moderate Reynolds number (Re = 16,000). The analysis focused on the evolution of the joint pdf (jpdf) between the second and third invariants of the characteristic equation for the VGT. These invariants are known to fully characterise the state of the VGT and have been previously used to observe the transition to a fully developed turbulent state in the interface region between turbulent and non-turbulent flows. We analyse the flow very close to the cylinder, where developed turbulence is not necessarily expected. The findings show that in the mean recirculation region, where the intermittency of the shear layer is low no tear drop shaped jpdf, indicative of fully developed turbulence, is found. Once the intermittency increases, tear drop shaped jpdfs are found where the velocity fluctuations are not Gaussian distributed, suggesting the small scales reach a fully developed turbulent state ahead of the large ones. This is further investigated by analysing the geometry of the local straining, the vorticity-strain rate alignment and enstrophy production. [Preview Abstract] |
Tuesday, November 24, 2015 9:31AM - 9:44AM |
M21.00008: On the Distribution of Velocity Gradients, Viscosity and Reynolds Stresses in Varied Bed Elevation Turbulent Flow Hanieh Tabkhi, Arash Nayebzadeh There is wide variation of depth across a cross section along the axis of the estuaries and coastal regions. In addition to bed turbulence and secondary circulations, structure of turbulence flow at these regions is affected by variation in the depth of the bed. Lateral variations of depth cause strong transverse free shear layers due to steep velocity gradient. Many experimental and laboratory studies have mentioned these shear layers in previous studies. At present study, three dimensional modeling in a Cartesian coordinate system has been performed for varied bed elevation flow. Normal and shear Reynolds stresses have been calculated applying k-epsilon and k-omega turbulence models. Model is validated by previous related studies and mutual effects of velocity gradients on turbulence viscosity and Reynolds stresses have been investigated. Results show that velocity gradients monotonically increase by increasing magnitude of turbulence viscosity and Reynolds stresses. [Preview Abstract] |
Tuesday, November 24, 2015 9:44AM - 9:57AM |
M21.00009: On the viscosity stratification in temporal mixing layer Luminita Danaila, Noureddine Taguelmimt, Abdellah Hadjadj We assess the effects of viscosity variations in low-speed temporally-evolving turbulent mixing layer. The two streams are density-matched, but the slow fluid is Rv times more viscous than the rapid stream. Direct Numerical Simulations (DNS) are performed for several viscosity ratios, Rv varying between 1 and 9. The space-time evolution of Variable-Viscosity Flow (VVF) is compared with that of the Constant-Viscosity Flow (CVF). The velocity fluctuations occur earlier and are more enhanced for VVF. In particular, the kinetic energy peaks earlier and is up to three times larger for VVF than for CVF at the earliest stages of the flow. Over the first stages of the flow, the temporal growth rate of the fluctuations kinetic energy is exponential, in full agreement with linear stability theory. The transport equation for the fluctuations kinetic energy is favourably compared with simulations data. The enhanced kinetic energy for VVF is mainly due to an increased production at the interface between the two fluids, in tight correlation with enlarged values of mean velocity gradient at the inflection point of the mean velocity profile. The transport equations of the one-and two-point kinetic energy show that self-preservation cannot be complete in variable-viscosity flows. [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