• Volume/Page
  • Keyword
  • DOI
  • Citation
  • Advanced
   
 
 
 

Flickr Twitter iResearch App Facebook

Search Issue | RSS Feeds RSS
Previous Issue Next Issue

Aug 2007

Volume 19, Issue 8, Articles (08xxxx)

Issue Cover Spotlight Figure

Phys. Fluids 19, 083601 (2007); http://dx.doi.org/10.1063/1.2754346 (14 pages)

G. J. Sheard, T. Leweke, M. C. Thompson, and K. Hourigan
back to top
RSS Feeds
back to top Turbulent Flows

Consistent modeling of interphase turbulent kinetic energy transfer in particle-laden turbulent flows

Ying Xu and Shankar Subramaniam

Phys. Fluids 19, 085101 (2007); http://dx.doi.org/10.1063/1.2756579 (9 pages) | Cited 2 times

Online Publication Date: 8 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
The interphase transfer of turbulent kinetic energy (TKE) is an important term that affects the evolution of TKE in fluid and particle phases in particle-laden turbulent flow. This work shows that the interphase TKE transfer terms must obey a mathematical constraint, which in the limiting case of statistically homogeneous flow with zero mean velocity in both phases, requires these terms be equal and opposite. In the single-point statistical approach called the two-fluid theory, the interphase TKE transfer terms are unclosed and need to be modeled. Multiphase turbulence models that satisfy this constraint of conservative interphase TKE transfer admit a term-by-term comparison with true direct numerical simulations (DNS) that enforce the exact velocity boundary condition on each particle’s surface. Analysis of three models reveals that not all models satisfy the requirement of conservative interphase TKE transfer. DNS that invoke the point-particle assumption also do not obey this principle of conservative interphase TKE transfer, and this precludes the comparison of model predictions of TKE budgets in each phase with point-particle DNS. This study motivates the development of multiphase turbulence models based on the insights revealed by this analysis, leading to a meaningful comparison of TKE budgets with true DNS.
Show PACS
47.27.ek Direct numerical simulations
47.27.eb Statistical theories and models
47.55.-t Multiphase and stratified flows

A term-by-term direct numerical simulation validation study of the multi-environment conditional probability-density-function model for turbulent reacting flows

S. T. Smith and R. O. Fox

Phys. Fluids 19, 085102 (2007); http://dx.doi.org/10.1063/1.2757699 (19 pages) | Cited 2 times

Online Publication Date: 17 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
The multi-environment conditional probability-density-function (MECPDF) approach for modeling extinction and re-ignition in turbulent nonpremixed reacting flows is analyzed. A unique derivation of the model is given, which makes use of numerical Gaussian quadrature in addition to physical assumptions. The new derivation offers insight into the physical meaning of model terms and offers a more rigorous method for model validation. The assumptions required to close the dissipation terms are validated term by term using data from direct numerical simulations of an inert and a reacting scalar in decaying isotropic turbulence. Results show convergence of the numerical quadrature with an increasing number of quadrature points. Also, good agreement is shown for the physical model assumptions required to close the mixed dissipation and the progress-variable dissipation terms. The MECPDF method is also demonstrated to offer the flexibility to incorporate either micromixing or otherwise more sophisticated models for the mixing between regions of the flow that exhibit differing degrees of extinction.
Show PACS
47.27.ek Direct numerical simulations
47.27.Gs Isotropic turbulence; homogeneous turbulence
47.70.Fw Chemically reactive flows

Spontaneous generation of vortex crystals from forced two-dimensional homogeneous turbulence

Javier Jiménez and Alan Guegan

Phys. Fluids 19, 085103 (2007); http://dx.doi.org/10.1063/1.2757713 (6 pages) | Cited 1 time

Online Publication Date: 17 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
The long-term limit of statistically stationary two-dimensional turbulence is shown to depend on the form of the large-scale forcing, in agreement with previous results. That effect is studied systematically by continuously varying the forcing from deterministic to Brownian in direct numerical simulations in doubly periodic boxes. As expected, this switches on or off the enstrophy cascade and the presence of strong coherent structures, but the transition is not monotonic. Under intermediate forcing conditions, the flow evolves to a stationary vortex crystal with triangular lattice, which appears to be stable and to last indefinitely. Deterministic forcings frustrate crystallization through the formation of fast-moving dipoles, and very random ones melt the crystal. The dispersion properties of the different regimes are studied, and it is shown that efficient particle dispersion depends on the presence of multiscale turbulence. The relation with other two-dimensional systems is discussed.
Show PACS
47.32.-y Vortex dynamics; rotating fluids
47.27.-i Turbulent flows

Mixing properties of a stably stratified parallel shear layer

Paolo Monti, Giorgio Querzoli, Antonio Cenedese, and Simonetta Piccinini

Phys. Fluids 19, 085104 (2007); http://dx.doi.org/10.1063/1.2756580 (9 pages) | Cited 1 time

Online Publication Date: 21 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
The results of a laboratory investigation on turbulent mixing of mass and momentum associated with the collapse of Kelvin-Helmholtz waves which develop in a stably stratified parallel shear layer with O(103) Schmidt number are presented. The waves were generated in a stratified, tilting tank using a denser fluid underlying a lighter fluid with a velocity shear between the layers. Detailed velocity and density measurements were conducted simultaneously using particle tracking velocimetry and laser-induced fluorescence techniques. Estimation of the efficiency of mixing, the ratio between the diffusive flux of density responsible for mixing, to the turbulent kinetic energy dissipation rate, suggest it is independent of the initial values of the Reynolds and the bulk Richardson numbers. The measurements of the eddy diffusivities of mass and momentum showed a reasonable agreement with other laboratory-based predictions of turbulent mixing and observational data in atmospheric stably stratified flows.
Show PACS
47.55.Hd Stratified flows
47.27.wj Turbulent mixing layers
47.80.Jk Flow visualization and imaging

The elementary energy transfer between the two-point velocity mean and difference

M. Germano

Phys. Fluids 19, 085105 (2007); http://dx.doi.org/10.1063/1.2760283 (5 pages) | Cited 2 times

Online Publication Date: 21 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
In this paper, the elementary energy transfer between the two-point average of a velocity field and the two-point difference has been examined. The equations related to the two-point quantities are derived from the Navier-Stokes equations applied to incompressible flows. Some interesting aspects of this simple transfer of energy are discussed both from the point of view of the filtered equations associated with the two-point velocity average and from the point of view of the properties of the structure functions associated with the two-point velocity difference. A new inertial relation valid for homogeneous flows is finally presented.
Show PACS
47.10.ad Navier-Stokes equations
47.27.Ak Fundamentals

Energy analysis of turbulent channel flow using biorthogonal wavelets

Vivek J. Joshi and Dietmar Rempfer

Phys. Fluids 19, 085106 (2007); http://dx.doi.org/10.1063/1.2760277 (12 pages) | Cited 2 times

Online Publication Date: 23 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
Turbulent flows exhibit many different length and time scales. Hence, it is important to study energy transfer between the scales, in order to gain deeper insight into turbulence. Wavelets offer potential for the analysis of the energy transfer in a turbulent flow. This is mainly due to their locality and scalability property. The present work involves a wavelet decomposition of the terms in the turbulent kinetic energy transport equation of a fully developed channel flow and a study of the behavior of the important terms. A detailed analysis of the energy transfer term is performed. An attempt is made to identify some well-known structures in the flow at different scales of decomposition. The dynamics of these coherent structures is examined based on their contribution to energy transfer in the flow at discrete scales. This is performed by correlating the velocity and energy transfer in the flow. To analyze energy transfer due to spatial structures, we utilize the concept of “triad interaction” in wavelet space. The study involves computing all possible combinations of wavelet modes and finding the most energetic ones. We investigate these interactions to capture the dynamics of the local structures and their contribution to the energy transfer.
Show PACS
47.60.-i Flow phenomena in quasi-one-dimensional systems
02.30.Uu Integral transforms
47.27.nd Channel flow
47.27.Ak Fundamentals

Local heat fluxes in turbulent Rayleigh-Bénard convection

Olga Shishkina and Claus Wagner

Phys. Fluids 19, 085107 (2007); http://dx.doi.org/10.1063/1.2756583 (13 pages) | Cited 17 times

Online Publication Date: 24 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
The vertical turbulent heat flux Ω in Rayleigh-Bénard convection, its spatial distribution, and some mean characteristics are investigated by means of direct numerical simulations for the Rayleigh numbers Ra = 106 and 107 and well resolved large-eddy simulations for Ra = 108. All simulations were performed for Prandtl number Pr = 0.7 and aspect ratio of a cylindrical container Γ = 5. Analyzing the spatial distribution of Ω, it is shown that the fluid volume with negative Ω values increases with Ra and reaches one-third of the total volume for Ra = 108. The spread in the local heat flux values expands with increasing distance from the top or the bottom plates. For example, for Ra = 107, about 31% and 19% of the center horizontal cross section reflects, respectively, negative and large positive ( ≥ 2Nu) values of Ω, while at the plates the local heat flux values vary basically between 0 and 2Nu. Further, it is shown that with growing Rayleigh numbers, the zones of higher values of the time-averaged local heat flux move toward the corners, where horizontal and vertical walls intersect. Analytical relations between the components of Ω and the thermal dissipation rates, proven in the paper, show that the square root of the thermal dissipation rate describes well the spatial distribution of the local heat transport close to the top or the bottom plates.
Show PACS
47.55.pb Thermal convection
47.27.te Turbulent convective heat transfer
47.54.-r Pattern selection; pattern formation
47.11.Df Finite volume methods
47.27.ek Direct numerical simulations
47.27.ep Large-eddy simulations

Outer-layer similarity in the presence of a practical rough-wall topography

Y. Wu and K. T. Christensen

Phys. Fluids 19, 085108 (2007); http://dx.doi.org/10.1063/1.2741256 (15 pages) | Cited 22 times

Online Publication Date: 24 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
High-resolution particle image velocimetry measurements are made in the streamwise–wall-normal plane of a zero-pressure-gradient turbulent boundary layer over smooth and rough walls at Reθ ≈ 13000. The roughness considered herein is replicated from a surface scan of a turbine blade damaged by deposition of foreign materials and its topography is highly irregular and contains a broad range of topographical scales. Two physical scalings of the same roughness topography are considered, yielding two different rough surfaces: RF1 with k = 4.2 mm and RF2 with k = 2.1 mm, where k is the average peak-to-valley roughness height. At Reθ ≈ 13000, these roughness conditions yield k+k/y* = 207, δ/k = 28, ks+ = 115, and δ/ks = 48 for RF1 and k+ = 91, δ/k = 50, ks+ = 29, and δ/ks = 162 for RF2 (where δ is the boundary-layer thickness, ks is the equivalent sand-grain height, and y* is the viscous length scale). The mean velocity deficits along with the Reynolds normal and shear stress profiles for both roughness conditions collapse on the smooth-wall baseline in the outer layer when appropriately scaled by the friction velocity, uτ. Probability density functions and quadrant analysis of the instantaneous events contributing to the mean Reynolds shear stress show similar outer-layer consistency between the smooth and rough cases when scaled appropriately with uτ. In addition, one-dimensional, two-point streamwise, and wall-normal velocity autocorrelation coefficients are also found to collapse in the outer region, indicating a similarity in the spatial structure of the outer-layer turbulence. The observed collapse of the smooth- and rough-wall turbulence statistics in the outer layer supports Townsend’s wall similarity hypothesis for flow over the unique surface topography considered herein.
Show PACS
47.27.nb Boundary layer turbulence
47.80.Jk Flow visualization and imaging
47.85.-g Applied fluid mechanics

Large-eddy simulations of the turbulent Hartmann flow close to the transitional regime

I. E. Sarris, S. C. Kassinos, and D. Carati

Phys. Fluids 19, 085109 (2007); http://dx.doi.org/10.1063/1.2757710 (9 pages) | Cited 5 times

Online Publication Date: 28 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
A series of large-eddy simulations (LES) of turbulent and transitional channel flows of a conductive fluid under the effect of a uniform magnetic field applied in the wall-normal direction, usually referred to as Hartmann flows, are performed using the dynamic Smagorinsky model. The flow is characterized by the hydrodynamic Reynolds number Re and the Hartmann number Ha that is related to the strength of the external magnetic field. Previous measurements and works on stability analysis have shown that a critical modified Reynolds number, based on the Hartmann layer thickness R = Re/Ha, exists at approximately R = 380. Here, the LES are used to investigate the laminarization of flow when decreasing R. Also, similarities of the turbulent flows are explored for different values of Re and Ha but the same values of R or of the interaction parameter N = Ha2/Re. The LES confirm that R is the relevant parameter that describes the transition and a critical Reynolds number for relaminarization is observed at R ≈ 500. However, for turbulent flows at moderate values of the Ha and Re as considered in this study, the similarity with respect to N appears to be better than for R, especially for the first order statistics. Finally, the importance of the dynamic Smagorinsky model as the Hartmann number increases is discussed.
Show PACS
47.27.E- Turbulence simulation and modeling
47.60.-i Flow phenomena in quasi-one-dimensional systems

On the use of the Kolmogorov-Landau approach in deriving various correlation functions in two-dimensional incompressible turbulence

Sagar Chakraborty

Phys. Fluids 19, 085110 (2007); http://dx.doi.org/10.1063/1.2760282 (5 pages) | Cited 2 times

Online Publication Date: 28 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
We look at various correlation functions, which include those that involve both the velocity and the vorticity fields, in two-dimensional (2D) isotropic homogeneous decaying turbulence. We adopt the more intuitive approach due to Kolmogorov (and subsequently Landau in his text on fluid dynamics) and show that how the 2D turbulence results, obtainable using other methods, may be established in a simpler way. Also, some experimentally verifiable correlation functions in the dissipation range have been derived for the same system. The paper also showcases the inability of the Kolmogorov-Landau approach to get the “one-eighth law” in the enstrophy cascade region. As discussed in the paper, this may raise the spectre of logarithmic corrections once again in 2D turbulence.
Show PACS
47.27.Gs Isotropic turbulence; homogeneous turbulence
47.32.C- Vortex dynamics

A comparison of spectral sharp and smooth filters in the analysis of nonlinear interactions and energy transfer in turbulence

J. Andrzej Domaradzki and Daniele Carati

Phys. Fluids 19, 085111 (2007); http://dx.doi.org/10.1063/1.2760281 (13 pages) | Cited 8 times

Online Publication Date: 29 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
Energy transfer in turbulence is a result of nonlinear interactions among different scales of motion. For the purposes of quantitative analyses, various definitions of scales of motion can be used. This nonuniqueness leads to the possibility, raised in the literature on the subject, that properties of the energy transfer deduced from such analyses can be qualitatively affected by the scale definitions employed. We address this question by computing detailed energy exchanges between different scales of motion in direct numerical simulations databases of isotropic turbulence and employing different scale definitions. The scales of motion are defined by decomposing velocity fields using three specific filters: sharp spectral, Gaussian, and tangent hyperbolic. The traditional analysis of the energy transfer in terms of sharp spectral filters is generalized to smooth filters with broad support in the spectral space. The computed detailed energy transfer functions show only a minor quantitative dependence on the filter type. The qualitative conclusions obtained using sharp spectral filters are the same as for smooth filters, unless the latter has a very broad spectral support.
Show PACS
47.27.ek Direct numerical simulations
47.27.Gs Isotropic turbulence; homogeneous turbulence

An analysis of the energy transfer and the locality of nonlinear interactions in turbulence

J. Andrzej Domaradzki and Daniele Carati

Phys. Fluids 19, 085112 (2007); http://dx.doi.org/10.1063/1.2772248 (13 pages) | Cited 12 times

Online Publication Date: 29 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
Using the results of direct numerical simulations of isotropic turbulence, we compute detailed energy exchanges between different scales of motion and investigate how they contribute to the global quantities such as the energy transfer, the spectral energy flux, and the subgrid-scale dissipation. The scales of motion are defined by decomposing velocity fields using sharp spectral and smooth, tangent hyperbolic filters. The analysis of detailed interactions reveals that individual nonlocal contributions are large but significant cancellations lead to the global quantities asymptotically dominated by the local interactions. Implications of these results for turbulence modeling are discussed.
Show PACS
47.27.-i Turbulent flows
47.10.ad Navier-Stokes equations

Large eddy simulation and measurements of turbulent enclosed rotor-stator flows

Éric Séverac, Sébastien Poncet, Éric Serre, and Marie-Pierre Chauve

Phys. Fluids 19, 085113 (2007); http://dx.doi.org/10.1063/1.2759530 (17 pages) | Cited 8 times

Online Publication Date: 30 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
Turbulent flows are studied in an actual enclosed rotor-stator configuration with a rotating hub and a stationary shroud. Besides its fundamental importance—the disk boundary layer is one of the simplest platforms for investigating the underlying structure of three-dimensional boundary layers—this cavity models more complex configurations relevant to rotating machinery. Large eddy simulation is performed using a spectral vanishing viscosity technique that is shown leading to stable discretizations without sacrificing the formal accuracy of the spectral approximation. Numerical results and velocity measurements have been favorably compared for a large range of rotational Reynolds numbers (105 ⩽ Re = Ωb2/ν ⩽ 106) in an annular cavity of curvature parameter Rm = (b+a)/(ba) = 1.8 and of aspect ratio G = (ba)/h = 5, where a and b are, respectively, the inner and outer radii of the rotating disk and h is the interdisk spacing. In the detailed picture of the flow structure that emerges, the turbulence is confined mainly in the boundary layers including in the Stewartson layer along the external cylinder. For Reynolds numbers Re ≥ 105, the stator boundary layer is turbulent over most of the cavity. On the other hand, the rotor layer becomes progressively turbulent from the outer radial locations, although the rotating hub is shown to destabilize the inner part of the boundary layers. The isosurface maps of the Q-criterion reveal that the three-dimensional spiral arms observed in the unstable laminar regime evolve to more axisymmetric structures when turbulence occurs. At Re = 106, the flow is fully turbulent and the anisotropy invariant map highlights turbulence structuring, which can be either a “cigar-shaped” structuring aligned on the tangential direction or a “pancake-shaped” structuring depending on the axial location. The reduction of the structural parameter a1 (the ratio of the magnitude of the shear stress vector to twice the turbulence kinetic energy) under the typical limit 0.15, as well as the misalignment between the shear stress vector and the mean velocity gradient vector, highlight the three-dimensional nature of both rotor and stator boundary layers with a degree of three-dimensionality much higher than in the idealized system studied by Lygren and Andersson [ J. Fluid Mech. 426, 297 (2001) ; ZAMP 55, 268 (2004) ; and Int. J. Heat Fluid Flow 27, 551 (2006) ].
Show PACS
47.85.-g Applied fluid mechanics
47.80.-v Instrumentation and measurement methods in fluid dynamics
47.10.-g General theory in fluid dynamics
47.27.nb Boundary layer turbulence
47.60.-i Flow phenomena in quasi-one-dimensional systems
47.32.Ef Rotating and swirling flows

Comparison of direct numerical simulations and particle-image velocimetry data of turbulent channel flow rotating about the streamwise axis

I. Recktenwald, T. Weller, W. Schröder, and M. Oberlack

Phys. Fluids 19, 085114 (2007); http://dx.doi.org/10.1063/1.2760278 (11 pages) | Cited 2 times

Online Publication Date: 31 August 2007

Full Text: Read Online (HTML) | Download PDF

Show Abstract
The influence of rotation about the streamwise axis on a turbulent channel flow is discussed by analyzing and comparing numerical and experimental data. The numerical study uses direct numerical simulations (DNS), while the experiments were based on particle-image velocimetry (PIV). The flow properties in the numerical and experimental analyses were maintained as similar as possible. For comparability with data from the literature the nonrotating channel flow at a Reynolds number Reτ = 180 serves as a reference problem. Depending on the rotation rate of the channel the development of a secondary flow perpendicular to the main flow has been found in both investigations. The rotation primarily influences those components of the Reynolds shear stresses, which contain the spanwise velocity component. The size of the correlation areas and thus the length scales of the flow increase in all three coordinate directions, leading to longer structures. The increased momentum exchange can be deducted from the behavior of the mean main velocity profile, which shows a decreased centerline velocity and an increased velocity near the wall at growing rotation rates. The qualitative development of the mean flow and the statistical flow properties is similar for the simulation and the experiment. The growth of the coherent turbulent structures with the rotation rate is confirmed by both investigations.
Show PACS
47.27.nd Channel flow
47.27.ek Direct numerical simulations
47.32.Ef Rotating and swirling flows
47.80.-v Instrumentation and measurement methods in fluid dynamics
Close
Google Calendar
ADVERTISEMENT

close