Search

The objective of the present work is to analyse the behaviour of a turbulent decelerating boundary layer under the effect of both passive and active jets vortex generators (VGs). The stereo PIV database of Godard and Stanislas [1,2] obtained in an adverse pressure gradient boundary layer is used for this study. After presenting the effect on the mean velocity field and the turbulent kinetic energy, the line of analysis is extended with two points spatial correlations and vortex detection in instantaneous velocity fields. It is shown that the actuators concentrate the boundary layer turbulence in the region of upward motion of the flow, and segregate the near-wall streamwise vortices of the boundary layer based on their vorticity sign. [ABSTRACT FROM AUTHOR]

The objective of the present work is to evaluate the steady and unsteady Reynolds-averaged Navier-Stokes (RANS) approach turbulence model for separated flow applications. A υ2- f model based on the standard k - ω model is employed and referred to as the k - ω - υ2- f model. This model is validated by flow around two- and three-dimensional bluff bodies. The selected configurations are a single square cylinder, tandem square cylinders and cube. The flow predictions are discussed and compared with available experimental and numerical data. For these cases, none of the previously published numerical predictions obtained by steady-state RANS produced a good match with experimental data. Because these flow cases are inherently unsteady at high Reynolds numbers, unsteady RANS computations should be employed. It is found that the results obtained with this mode are in agreement with the available experimental and numerical data. [ABSTRACT FROM AUTHOR]

Regularization-based subgrid-scale (SGS) turbulence models for large eddy simulations (LES) are quantitatively assessed for decaying homogeneous turbulence (DHT) and transition to turbulence for the Taylor-Green vortex (TGV) through comparisons to laboratory measurements and direct numerical simulations (DNS). LES predictions using the Leray-α, LANS-α, and Clark-α regularization-based SGS models are compared to the classic nondynamic Smagorinsky model. Regarding the regularization models, this work represents their first application to relatively high-Re decaying turbulence with comparison to the active-grid-generated decaying turbulence measurements of Kang et al. (J. Fluid Mech., 2003) at Reλ ≈ 720 and the Re = 3000 DNS of transition to turbulence in the TGV of Drikakis et al. (J. Turbulence, 2007). For DHT the nondynamic Smagorinsky model is in excellent agreement with measurements for the turbulent kinetic energy and energy spectra but higher-order moments show slight discrepancies. For TGV too, the energy decay rates of Smagorinsky agree reasonably well with DNS. Regarding the regularization models stable results are not obtained as compared to the Smagorinsky model at the same grid resolution for various values of α; and at higher grid resolutions, they are in worse agreement. However, with additional dissipation such as in mixed α-Smagorinsky models, results are acceptable and show only slight deviations from Smagorinsky. [ABSTRACT FROM AUTHOR]

The present paper focuses on a dynamic version of the variational multiscale model and investigates its performance in large eddy simulations (LES) of turbulent channel flow at Reynolds numbers (based on friction velocity) of 180, 395, and 590. The dynamic procedure was implemented for the variational multiscale model on the basis of the classical Smagorinsky model and the wall-adapting local eddy-viscosity (WALE) model. One base regular grid and three different distorted grids are employed in the evaluation. The dynamic multiscale models are compared with the direct numerical simulation (DNS) data and the conventional dynamic models. Next, the error components involved in LES are separated and the grid irregularity effect on the shortcoming of the subgrid scale models and the error of numerical approximation is assessed. A main outcome is that the dynamic multiscale models in comparison with the dynamic Smagorinsky and the dynamic WALE are less affected by grid distortion, leading to better behavior on the irregular grids. [ABSTRACT FROM AUTHOR]

Characteristics of the process of entrainment in plane mixing layers, and the changes with compressibility and heat release, were studied using temporal DNS with simultaneous fluid packet tracking. Convective Mach numbers of the simulations are 0.15, 0.7 and 1.1. The Reynolds number is quite high (between 11 000 and 37 000 based on layer width and velocity difference), and is above the mixing transition. The study agrees with recent findings in round jets: first, engulfed fluid volume and its growth rate are both very small compared with the volume of the turbulent region and its growth rate, respectively. Secondly, most often, the process occurs close to the turbulent-nonturbulent boundaries. A new finding is that both compressibility and heat release retard the entrainment process so that it takes an O(1) time for vorticity or scalar levels to grow even after growth has been initiated. This delay is manifested as the fall in mixing layer growth rates as compressibility and heat release levels increase. [ABSTRACT FROM AUTHOR]

This paper aims at determining the most influential inlet turbulence parameters on the downstream transitional mixing region. To this end, a stochastic method is developed to generate a divergence-free random velocity field with a prescribed energy distribution in physical and wave-number space. In addition, predetermined integral length scales can be established. Ten direct numerical simulations of a temporally evolving transitional turbulent mixing layer are examined in detail. Simulation results show a large disparity in mean and instantaneous turbulent quantities, mainly effected by the energy spectrum, the integral length scale and the divergence freeness of the initial field. [ABSTRACT FROM AUTHOR]

Direct numerical simulations (DNS) are performed to investigate mixing in free round coaxial jets. A great attention has been put on the influence of upstream conditions upon the global flow structure and the mixing process. The mixing behavior is studied through the spatial and temporal development of the mixture fraction of the annular and the inner fluids, and examined by means of flow visualization and statistics. It is shown that the turbulent mixing process and the mixture fraction field in coaxial jets depend on the upstream conditions, even though a quasi self-similar state is reached. The mixing alterations are explained by the understanding of the flow dynamics modifications implied by the different upstream conditions. These alterations are mainly due to the intense generation of streamwise vortices, favored by high inlet velocity gradients and velocity ratios, as well as low ratios between the inner and the outer jet diameters. This is associated with a high quality of mixing, as far as global mixedness is concerned. It is also shown that the annular fluid reaches the inner fluid and mixes swiftly into it. Conversely, the latter remains confined. Additionally, spots of pure unmixed species are observed at the end of the computational domain, and shown to be due to the annular jet. [ABSTRACT FROM AUTHOR]

Conventional large-eddy simulation (LES) and monotone integrated LES (MILES) are tested in emulating the dynamics of transition to turbulence in the Taylor-Green vortex (TGV). A variety of subgrid scale (SGS) models and high-resolution numerical methods are implemented in the framework of both incompressible and compressible fluid flow equations. Comparisons of the evolution of characteristic TGV integral measures are made with previously reported and new direct numerical simulation (DNS) data. The computations demonstrate that the convective numerical diffusion effects in the MILES methods can consistently capture the physics of flow transition and turbulence decay without resorting to an explicit SGS model, while providing accurate prediction of established theoretical findings for the kinetic energy dissipation, energy spectra, enstrophy and kinetic energy decay. All approaches tested provided fairly robust computational frameworks. [ABSTRACT FROM AUTHOR]

A direct numerical simulation has been carried out for turbulent flow around a periodically deforming film. The film deformation is sinusoidal in the streamwise direction, whose nodes are fixed in space. The maximum amplitude of the deformation is limited within the linear sublayer. An unsteady generalized curvilinear coordinate and the collocated grid system are applied to the finite differencing of the NS equation. The computational results show that the turbulence intensities, the Reynolds shear stress and the wall-shear stress increase, in the case where the period of wall deformation is longer than the ratio of the wavelength of deformation to the fluid velocity in the buffer region. In contrast to this, these statistical quantities do not increase, in the case where the period of wall deformation is slightly shorter than the ratio. This difference is caused by the enhancement of mixing in the wall-normal direction and the modifications of the Q4 event due to the deformation. [ABSTRACT FROM AUTHOR]

We demonstrate that the spacetime statistics of the birth of turbulent spots in boundary layers can be reconstructed qualitatively from the average behaviour of macroscopic measures in the transition zone. The conclusion in [1] that there exists a connection between the patterns in laminar instability and the birth of turbulent spots is strengthened. We examine why the relationship between instability and transition to turbulence is manifest in some cases and appears to be totally absent in others. Novel cellular automaton-type simulations of the transition zone are conducted, and the pattern of spot birth is obtained from secondary instability analysis. The validity of the hypothesis of concentrated breakdown, according to which most turbulent spots originate at a particular streamwise location, is assessed. The predictions made lend themselves to straightforward experimental verification. [ABSTRACT FROM AUTHOR]

The first direct numerical simulation that is sufficiently large to study the self-similar behaviour of a turbulent wall jet is performed. The investigation is an extension of the simulation performed by Levin et al. (2005, A study of the Blasius wall jet. Journal of Fluid Mechanics, 539, 313–347). The same numerical method is used, but a significantly larger computational domain enables us to follow the development of the flow throughout the transition into its early turbulent evolution. Two-dimensional waves and streamwise elongated streaks, matched to measured disturbances, are introduced in the flow to trigger a natural transition mechanism. The Reynolds number is 3090 based on the inlet velocity and the nozzle height. The simulation provides detailed visualisations of the flow structures and statistics of mean flow and turbulent stresses. A weak subharmonic behaviour in the transition region is revealed by animations of the flow. The averaged data are presented in both inner and outer scalings in order to identify self-similar behaviour. Despite the low Reynolds number and the short computational domain, the turbulent flow exhibits a reasonable self-similar behaviour, which is most pronounced with inner scaling in the near-wall region. [ABSTRACT FROM AUTHOR]

We discuss the effect of large-scale anisotropy of the shear type on the small-scale structure of turbulence. Our analysis is based on numerical solutions of the Lagrangian tetrad model of Chertkov M., Pumir A. and Shraiman B.I. (1999, Physics of Fluids , 11, 2394) adapted to the model case with large-scale anisotropy. The model, formulated in terms of a set of stochastic differential equations for the coarse-grained velocity gradient and tensor of inertia of a typical shape, naturally connects Lagrangian and Eulerian parameterizations of turbulence. We use diagnostics of Chertkov al. (1999) which allows us to analyse and interpret different correlation functions at the resolved scale in terms of the flow geometry. Our main conclusion, concerning the issue of anisotropy, is that even though overall the local isotropy is restored with the scale decrease, the particular pace of the isotropy restoration depends very much on the object analysed. We found that the vorticity-dominated objects, such as enstrophy, tend to restore the isotropy much faster than their strain-dominated counterparts, e.g. energy flux and strain variance. [ABSTRACT FROM AUTHOR]

We describe how outer flow turbulence phenomena depend on the interaction with the wall. We quantify the influence of highly rough surfaces on turbulence statistics, eigenmodes and spectra of eigenvalues in the outer part of the wall shear layer of an isothermal turbulent flow. Surface geometries with different amplitude-to-wavelength ratios can be regarded as surfaces of defined wall roughness. The wavelength, ?, and the wave amplitude, 2 a , of a water channel are varied by changing the bottom wall in the test section. We consider three sinusoidal wall profiles defined by the amplitude-to-wavelength ratio a = 2 a /? = 0.2 (? = 30 mm), a = 0.2 (? = 15 mm), and a = 0.1 (? = 30 mm). Digital particle image velocimetry is performed to examine the spatial variation of the streamwise, spanwise and wall-normal velocity components. Statistical quantities that are determined are the Reynolds stresses, turbulent kinetic energy, and turbulent energy production. Measurements are performed at a Reynolds number of 11200, defined by the half channel height and the bulk velocity. A characteristic length scale in the most dominant eigenmodes can be found that agrees for all wall profiles and is independent of the amplitude-to-wavelength ratio. The turbulence quantities compared at defined streamwise coordinates in wall-normal direction are independent of the wavy wall profile in the outer region when scaled with the friction velocity, u * w . [ABSTRACT FROM AUTHOR]

We report results from a high-resolution numerical study of fluid particles transported by a fully developed turbulent flow at Rλ = 280. Single-particle trajectories were followed for a time range spanning more than three decades, from less than a tenth of the Kolmogorov timescale up to one large-eddy turnover time. We present results concerning acceleration statistics and the statistics of trapping by vortex filaments conditioned to the local values of vorticity and enstrophy. We found that a suitable observable, a ⊥ = a × v, connected to the centripetal acceleration, is strongly sensitive to the trapping of particles into vortex filaments. In particular, we found that its sign may persist up to 10–15 τη, while the typical oscillation of the longitudinal acceleration happens on a time lag of a few τη. [ABSTRACT FROM AUTHOR]

The effects of subgrid-scale (SGS) velocity and scalar and SGS models on the resolvable-scale velocity-scalar joint probability density function (JPDF) are studied. It is shown that the SGS stress and SGS scalar flux evolve the JPDF though the conditional mean SGS stress and flux, their production rates, and the SGS scalar variance production rate. The mean SGS stress and SGS scalar flux production rate are predicted using Lumley's assumption (Lumley, J. 1967 Physics of Fluids , 16, 3859–3867), which is consistent with Kolmogorov's hypothesis. Analyses of these statistics using data obtained in a slightly heated turbulent jet show that the mean SGS stress, SGS scalar flux, and their production rates have filter-scale dependences consistent with our predictions, suggesting that the SGS turbulence has diminishing influences on the lower-order resolvable-scale statistics for inertial-range filter scales. The Smagorinsky model predictions also have similar filter-scale dependences, suggesting that the models have diminishing effects on lower-order LES statistics. The measured conditional SGS stress and SGS flux as well as the conditional production rates decrease much slower than the predicted filter dependences, indicating that the SGS turbulence has non-trivial effects on the high-order resolvable-scale statistics even for inertial-range filter scales. The Smagorinsky models predictions of these conditional statistics decrease slower with the filter scale than our predictions but do not agree with measurements; therefore the models are likely to influence the high-order LES statistics but in ways different from that of the SGS turbulence. [ABSTRACT FROM AUTHOR]

In this paper we study a well-known three-dimensional turbulence model, the filtered Clark model, or Clark–α model. This is a large eddy simulation (LES) tensor-diffusivity model of turbulent flows with an explicit spatial filter of width α. We show the global well-posedness of this model with constant Navier–Stokes viscosity. Moreover, we establish the existence of a finite dimensional global attractor for this dissipative evolution system, and we provide an analytical estimate for its fractal and Hausdorff dimensions. Our estimate is proportional to ( L / l d ) 3 , where L is the integral spatial scale and l d is the viscous dissipation length scale. This explicit bound is consistent with the physical estimate for the number of degrees of freedom based on heuristic arguments. Using semi-rigorous physical arguments we show that the inertial range of the energy spectrum for the Clark–å model has the usual k -5/3 Kolmogorov power law for wave numbers k å ≪ 1 and k -3 decay power law for k å ≫ 1. This is an evidence that the Clark–α model parameterizes efficiently the large wave numbers within the inertial range, k å ≫ 1, so that they contain much less translational kinetic energy than their counterparts in the Navier–Stokes equations. [ABSTRACT FROM AUTHOR]

We report recent results from a high-resolution numerical study of fluid particles transported by a fully developed turbulent flow. Single-particle trajectories were followed for a time range spanning more than three decades, from less than a tenth of the Kolmogorov timescale up to one large-eddy turnover time. We present some results concerning acceleration statistics and the statistics of trapping by vortex filaments. [ABSTRACT FROM AUTHOR]

A comparison between two different ways of extracting coherent vortices in three-dimensional (3D) homogeneous isotropic turbulence is performed, using either orthogonal or biorthogonal wavelets. The method is based on a wavelet decomposition of the vorticity field and a subsequent thresholding of the wavelet coefficients. The coherent vorticity is reconstructed from a few strong wavelet coefficients, while the incoherent vorticity is reconstructed from the remaining weak coefficients. The choice of the threshold, which has no adjustable parameters, is motivated for the orthogonal case from the denoising theory. Using only 3 % of the coefficients we show that both decompositions, that is orthogonal and biorthogonal, extract the coherent vortices. They contain most of the energy (around 99 % in both cases) and retain 74 % and 68 % of the enstrophy in the orthogonal and biorthogonal cases, respectively. The incoherent background flow for the orthogonal decomposition, which corresponds to 97 % of the wavelet coefficients, is structureless, decorrelated, and has a Gaussian velocity probability distribution function (PDF). In contrast, for the biorthogonal decomposition, the background flow exhibits quasi-two-dimensional (2D) structures and yields an exponential velocity PDF. Moreover, the biorthogonal decomposition loses 3.7% of both enstrophy and helicity, while they are conserved by the orthogonal decomposition. [ABSTRACT FROM AUTHOR]

The surge of turbulence research in the first half of the twentieth century is reviewed with the focus on the institutional environment that provided an umbrella for the activities in this field. Before the Second World War, the International Congresses for Applied Mechanics served as the main stage for presenting research results on turbulence. After the War, the International Union of Theoretical and Applied Mechanics (IUTAM), organizations such as the Division of Fluid Dynamics of the American Physical Society and new journals (theJournal of Fluid MechanicsandThe Physics of Fluids) added to the institutional framework for turbulence activities and publications. The development of these activities is illustrated with examples from the correspondence of some of the involved actors (such as George K. Batchelor and Francois N. Frenkiel, the editors of the new journals; both were also involved with the organization of the Marseille events in 1961). [ABSTRACT FROM AUTHOR]

A stochastic form of linear stability analysis is used to construct initial conditions for variable density moment closures for flows due to the unstable interface between two different density fluids. The material interface can be driven by a mechanism, such as gravity, a pressure gradient (Rayleigh-Taylor), or by a shock (Richtmyer-Meskoff). Initial moments for second moment closure strategies that carry the variables k, 〈uiuj〉, Rij, ε, and ai for the hydrodynamical field and 〈ρ2〉, 〈ρv〉, and ερ, ερv for the material field are derived. The method produces a one-to-one correspondence between the initial conditions of a statistical model and the initial energy production - crucially important to the early time problem that sets up the late time problem and is not addressed in models for these flows. The analysis is applied to the Rayleigh-Taylor transition and turbulence problem. Several scalings emerge from the analysis that parameterize the initial condition dependence of RT turbulence on statistics of the interfacial morphology. Computations are conducted to assess the importance of nondimensional parameters on the transition process. An interfacial Reynolds number, involving two statistically defined length scales, collapses the simulation data exceptionally well and is thus a parameter that can be used to effect the transition to turbulence. About 97% of the variability of the turbulence and bulk Reynolds numbers is explained by the initial Taylor-Reynolds number, Reλ0. The initial Taylor-Reynolds number, Reλ0, includes two length scales of the interface, δ is the root mean square (rms) thickness and κis rms zero-crossing wavenumber. Minimizing δ2/κ appearing in Reλ0 stabilizes the interface and reduces the speed of transition to a turbulent flow. Maximimizing δ2/κ destabilizes the interface and increases the rate of transition to a developed turbulent flow and may be of use in reducing the computational costs of transition in the direct numerical simulations of fully developed Rayleigh-Taylor turbulence. [ABSTRACT FROM AUTHOR]

A new technique for extracting information about cyclic and periodic features in turbulent-flow fields is proposed. This is based on a combination of proper orthogonal decomposition, applied to DNS and LES flow data, and the use of 'recurrence plots', the latter being a tool often adopted for analysing dynamical systems outside the field of fluid mechanics. The principal premise underpinning the technique is that the properties of a flow, viewed as a dynamical system, can be characterised by the temporal coefficients of the related proper orthogonal decomposition analysis of the flow. While such an analysis, on its own, yields useful information on the mode shapes, ordered according to their energy content, and on the temporal variation of these modes, it is not suited to the identification of 'cyclicity', here understood to be recurrent, as opposed to periodic, events that characterise drastic changes in the flow when this is viewed as a dynamical system. Following illustration with reference to a classical dynamical system, the Lorenz set, the method is applied to four turbulent flows: a three-dimensional transitional mixing layer, a flow separating from a ducted two-dimensional hump, that same flow controlled by means of a synthetic slot jet, and a flow separating from a three-dimensional ducted hill. [ABSTRACT FROM AUTHOR]