15:00
Wall-bounded flows 5
Chair: Arne Johansson
15:00
15 mins
|
Rayleigh-Bénard convection: local scaling exponent in confined thermal convection
Robert Kaiser, Ronald du Puits
Abstract: Local heat flux measurements at the heating plate of a cubic Rayleigh-Bénard cell show that the scaling exponent a in the Nu~Ra^a scaling law depends on the position at the plate surface. Using infrared thermography, the two-dimensional local wall heat flux is studied with a high spatial resolution. The results have been obtained in a small aspect ratio cell with Gamma_x=1, Gamma_y=0.26 and clearly show an effect of the sidewalls on the local and therefore also on the global scaling of the heat transport.
|
15:15
15 mins
|
DNS OF TURBULENT CHANNEL FLOWWITH A FLEXIBLE SQUARE CYLINDER
Koichi Tsujimoto
Abstract: Fluid-Structure Interaction (FSI) problem is concerned with in various research fields such as mechanical, aerospace, civil and medical engineering. Their accurate prediction and control are desired. So far, in order to improve the performance of various applications, many kind of research, on the heat transfer enhancement due to vortex generator in heat exchangers, on the drag reduction
through the setting of bluff body in pipe-line systems, and on the reduction of flow induced vibration, are conducted. In particular, since the wake of wall-mounted cylinder is a common flow regime in above-mentioned research, the detail of the flow has been aggressively investigated so far[1]. The present study, we pay attention to the flow control using flexible structures in the above mentioned flows.
To investigate the potentiality of the control in advance, both high accurate and stable computational scheme is needed so that theactual phenomena including turbulence is well predicted. Therefore, in order to analyze the fluid-structure interaction, we propose aweak-coupling method[2] in which for flexible structures, the rigorous equations of motion are discretized with finite volume method
(FVM[3]); for a flow computation, the finite difference method (FDM) is used and the flexible structures is reproduced via immersed boundary method[4]. In this present paper, we demonstrate on the result of flow structure around of rigid and elastic cylinder in turbulent channel flow.
|
15:30
15 mins
|
Comparative assessment of grid-spacing-based filter width formulations for Very Large Eddy Simulation
Anastasia Kondratyuk, Michael Schäfer, Suad Jakirlic
Abstract: When applying hybrid LES/RANS turbulence models the relevant filter width plays a crucial role concerning switching between different operating modes. Presently, the influence of the filter width and the choice of the corresponding criterion within the so-called VLES (Very Large Eddy Simulation) computational framework is investigated. Results obtained by using three eddy-viscosity-based background RANS models in conjunction with different filter-width formulations and two representative wall-bounded flow configurations are presented.
|
15:45
15 mins
|
The streamwise turbulence intensity in the intermediate layer of high Reynolds turbulent pipe flow
John Christos Vassilicos, Jean-Philippe LAVAL, Jean-Marc FOUCAUT, Michel STANISLAS
Abstract: A modification of the Townsend-Perry attached eddy model is derived in order to reproduce a more realistic variation of the integral length scale. A new wavenumber range is introduced to the model at wavenumbers smaller than the Townsend-Perry k^(-1) spectrum. This necessary addition can also account for the high Reynolds number outer peak of the turbulent kinetic energy in the intermediate layer. An analytic expression is obtained for this outer peak in agreement with extremely high Reynolds number data by Hultmark et al (2012, 2013). The finding of Dallas et al (2009) that it is the eddy turnover time and not the mean flow gradient which scales with distance to the wall and skin friction velocity in the intermediate layer implies, when combined with Townsend's (1976) production-dissipation balance, that the mean flow gradient has an outer peak at the same location as the turbulent kinetic energy.
|
16:00
15 mins
|
Quasi-steady quasi-homogeneous (QSQH) description of the relationship between large-scale and small-scale motions in near-wall turbulence
Chi Zhang, Sergei Chernyshenko
Abstract: The validity of the recently proposed hypothesis that the influence of large-scale motions on the near-wall turbulence is quasisteady is investigated by quantitative comparisons with the data obtained by direct numerical simulations of a turbulent channel flow.
Large-scale motions are filtered by a Fourier cut-off filter multi-objectively optimised to increase the correlation between large-scale
motions near and away from the wall while decreasing the correlation between small-scale motions. The quasi-steady hypothesis is
found to be approximate. It is also found that adding non-linear terms into calculations will improve the accuracy of the prediction.
|
16:15
15 mins
|
Phase diagram of turbulent Taylor-Couette flow
Detlef Lohse, Rodolfo Ostilla-Monico, Erwin van der Poel, Roberto Verzicco, Siegfried Grossmann
Abstract: We will present the results of our recent numerical work on the nature of the phase diagram of turbulent
Taylor-Couette (TC) flow, both with co- and counterrotating cylinder. The work can be seen
as the extension of the famous experimental Andereck et al.
phase diagram for Taylor-Couette flow just above the onset of instabilities, towards the ultimate turbulence
regime, and now obtained numerically. In particular, we will understand when and why optimal transport of
angular velocity from the inner to the outer cylinder is achieved and how this is connected to the angular velocity
profile and the structures in the flow.
|
16:30
15 mins
|
Reconstruction of turbulent pipe-flow profiles from laser Doppler velocimetry data
Denis F. Hinz
Abstract: Laser Doppler velocimetry (LDV) is a powerful experimental tool that finds increasing application in industrial research and product development. For example, LDV may be used to support the development of water, heat, and cooling meters as well as the design and validation of the associated testing and calibration facilities. For such applications, a basic quantity of interest is the volume flow rate Q that can be determined through numerical integration of the velocity profile over the measurement grid. However, under realistic experimental conditions, optical disturbances and reflections may cause corrupted and unreliable data at various measurement points. Here, we study a criterium to identify and reconstruct such data points. The reconstruction criterium is based on the standard error associated with each measurement point. Using some of our LDV data, we show that the error in Q with respect to a reference flow rate may be reduced significantly through the reconstruction. Consequently, the reconstruction criterium allows to establish accurate and reliable flow rates that have potential to be used as a primary validation reference.
|
16:45
15 mins
|
Perturbations evolution and streaks formation in turbulent channel flow.
Nikolay Nikitin
Abstract: Evolution of initially small perturbations superimposed on turbulent channel flow is investigated via DNS. On the linear stage perturbations grow exponentially in average as $\sim \exp(\lambda t)$ with $\lambda^+\approx0.021$ in accordance with previous findings \cite{NV1,NV2}. Maximum perturbation amplitude is located at distance $y^+=12$ from the wall, thus, the perturbation growth may be attributed to streak instability. However, in contrast to existing models of streak instability the growing patterns appear and disappear occasionally as localized spots in $x-z$ plane. Streaks in perturbation field appear on the nonlinear stage of evolution. Perturbation amplitude saturates finally on the level of $\sqrt{2}$ of that of turbulence fluctuations in underlying turbulent flow indicating that disturbed flow is uncorrelated with original undisturbed flow field.
|
|