On the genesis of quasi-steady vortices in a rotating turbulent flow

1987 ◽  
Vol 185 ◽  
pp. 121-136 ◽  
Author(s):  
Mathieu Mory ◽  
Philippe Caperan
Keyword(s):  
Kinetic Energy ◽  
Turbulent Flow ◽  
Turbulent Kinetic Energy ◽  
Thermal Convection ◽  
Rotation Axis ◽  
Rotation Period ◽  
Marginal Stability ◽  
Rotating Fluid ◽  
Mean Flow ◽  
Perturbation Equations

Turbulent flows subjected to rotation display vortices parallel to the rotation axis and exhibiting a long timescale compared to the turbulent turnover time and the rotation period. A similar flow pattern is observed arising from the thermal instability in a rotating fluid. We demonstrate the analogy between turbulence and thermal convection in a rotating fluid. A basic quasi-geostrophic turbulent flow is considered which is forced at the bottom of the layer by a stochastic component of velocity parallel to the rotation axis. The turbulent basic state has no mean flow and the gradient along the rotation axis of the turbulent kinetic energy −∂z〈ω2〉 is analogous to the mean temperature profile in thermal convection. The linear perturbation equations of this basic turbulent state are given, where the thermal diffusion equation is replaced by the turbulent kinetic energy equation. Using a simple closure of this equation the model demonstrates the occurrence of an instability when the Reynolds number exceeds a critical value. Marginal stability curves are deduced by numerical integration of the perturbation equations. The results show order-of-magnitude agreement with laboratory experiments.

Wall Pressure Fluctuations in the Reattachment Region of a Backward Facing Step

2007 ◽  
Author(s):  
Yu-Tai Lee ◽  
Theodore M. Farabee ◽  
William K. Blake
Keyword(s):  
Kinetic Energy ◽  
Turbulent Flow ◽  
Turbulent Kinetic Energy ◽  
Source Term ◽  
Pressure Fluctuations ◽  
Wall Pressure ◽  
Mean Flow ◽  
Backward Facing Step ◽  
Fluctuating Pressure ◽  
Wall Pressure Fluctuations

Steady mean flow fields and turbulent flow characteristics obtained from solving the Reynolds Averaged Navier Stokes (RANS) equations with a k-ε isotropic turbulence model are used to predict the frequency spectrum of wall-pressure fluctuations for flow past a backward facing step. The linear source term (LST) of the governing fluctuating-pressure equation is used in deriving the final double integration formula for the fluctuating wall pressure. The integrand of the solution formula includes the mean-flow velocity gradient, modeled turbulence normal fluctuation, Green’s function and the spectral model for the interplane correlation. An anisotropic distribution of the turbulent kinetic energy is implemented using a function named anisotropic factor. This function represents a ratio of the turbulent normal Reynolds stress to the turbulent kinetic energy and is developed based on an equilibrium turbulent flow or flows with zero streamwise pressure gradient. The spectral correlation model for predicting the wall-pressure fluctuations is obtained through modeling of the streamwise and spanwise wavenumber spectra. The nonlinear source term (NST) in the original fluctuating-pressure equation is considered following the conclusion of Kim’s direct numerical simulation (DNS) study of channel flow. Predictions of frequency spectra for the reattachment flow past a backward facing step (BFS) are investigated to verify the validity of the current modeling. Detailed turbulence features and wall-pressure spectra for the flow in the reattachment region of the BFS are predicted and discussed. DNS and experimental data for BFSs are used to develop and validate these calculations. The prediction results based on different modeling characteristics and flow physics agree with the observed turbulence field.

Measurements and Predictions of a Highly Turbulent Flowfield in a Turbine Vane Passage

2000 ◽  
Vol 122 (4) ◽  
pp. 666-676 ◽  
Author(s):  
R. W. Radomsky ◽  
K. A. Thole
Keyword(s):  
Kinetic Energy ◽  
Turbulent Flow ◽  
Gas Turbine ◽  
Turbulent Kinetic Energy ◽  
Reynolds Stress ◽  
Complex Geometry ◽  
Turbulence Models ◽  
Reynolds Stress Model ◽  
Stress Model ◽  
Mean Flow

As highly turbulent flow passes through downstream airfoil passages in a gas turbine engine, it is subjected to a complex geometry designed to accelerate and turn the flow. This acceleration and streamline curvature subject the turbulent flow to high mean flow strains. This paper presents both experimental measurements and computational predictions for highly turbulent flow as it progresses through a passage of a gas turbine stator vane. Three-component velocity fields at the vane midspan were measured for inlet turbulence levels of 0.6%, 10%, and 19.5%. The turbulent kinetic energy increased through the passage by 130% for the 10% inlet turbulence and, because the dissipation rate was higher for the 19.5% inlet turbulence, the turbulent kinetic energy increased by only 31%. With a mean flow acceleration of five through the passage, the exiting local turbulence levels were 3% and 6% for the respective 10% and 19.5% inlet turbulence levels. Computational RANS predictions were compared with the measurements using four different turbulence models including the k-ε, Renormalization Group (RNG) k-ε, realizable k-ε, and Reynolds stress model. The results indicate that the predictions using the Reynolds stress model most closely agreed with the measurements as compared with the other turbulence models with better agreement for the 10% case than the 19.5% case. [S0098-2202(00)00804-X]

Analysis of Turbulent Thermal Convection Between Horizontal Plates

1983 ◽  
Vol 105 (4) ◽  
pp. 789-794 ◽  
Author(s):  
M. Kaviany ◽  
R. Seban
Keyword(s):  
Temperature Distribution ◽  
Kinetic Energy ◽  
Turbulent Kinetic Energy ◽  
Thermal Convection ◽  
Equation Model ◽  
Experimental Results ◽  
Mixing Length ◽  
Mean Temperature ◽  
The Mean ◽  
The One

The one-equation model of turbulence is applied to the turbulent thermal convection between horizontal plates maintained at constant temperatures. A pseudo-three-layer model is used consisting of a conduction sublayer adjacent to the plates, a turbulent region within which the mixing length increases linearly, and a turbulent core within which the mixing length is a constant. It is assumed that the Nusselt number varies with the Rayleigh number to the one-third power. As a result, the steady-state distributions of the turbulent kinetic energy and the mean temperature are obtrained and presented in closed forms. These results include the effects of Prandtl number. The predictions are compared with the available experimental results for different Prandtl and Rayleigh numbers. Also included are the predictions of Kraichnan, which are based on a less exact analysis. The results of the one-equation model are in fair agreement with the experimental results for the distribution of the turbulent kinetic energy and the mean temperature distribution. The predictions of Kraichnan are in better agreement with the experimental results for the mean temperature distribution.

The Application of Eddy-Viscosity Stress Limiters for Modeling Cross-Flow Separation

2010 ◽  
Vol 132 (9) ◽  
Author(s):  
P. A. Gregory ◽  
P. N. Joubert ◽  
M. S. Chong ◽  
A. Ooi
Keyword(s):  
Kinetic Energy ◽  
Skin Friction ◽  
Turbulent Kinetic Energy ◽  
Flow Separation ◽  
Eddy Viscosity ◽  
Cross Flow ◽  
The Body ◽  
Mean Flow ◽  
Viscosity Models ◽  
Experimental Apparatus

The ability of eddy-viscosity models to simulate the turbulent wake produced by cross-flow separation over a curved body of revolution is assessed. The results obtained using the standard k−ω model show excessive levels of turbulent kinetic energy k in the vicinity of the stagnation point at the nose of the body. Additionally, high levels of k are observed throughout the wake. Enforcing laminar flow upstream of the nose (which replicates the experimental apparatus more accurately) gives more accurate estimates of k throughout the flowfield. A stress limiter in the form of Durbin’s T-limit modification for eddy-viscosity models is implemented for the k−ω model, and its effect on the computed surface pressures, skin friction, and surface flow features is assessed. Additionally, the effect of the T-limit modification on both the mean flow and the turbulent flow quantities within the wake is also examined. The use of the T-limit modification gives significant improvements in predicted levels of turbulent kinetic energy and Reynolds stresses within the wake. However, predicted values of skin friction in regions of attached flow become up to 50% greater than the experimental values when the T-limit is used. This is due to higher values of near-wall turbulence being created with the T-limit.

Effect of Viscoelasticity on the Turbulence Kinetic Energy Budget of the Non-Newtonian Fluid Turbulent Flow Accompanied by Drag Reduction

2007 ◽  
Author(s):  
Y. Kagawa ◽  
B. Yu ◽  
Y. Kawaguchi
Keyword(s):  
Kinetic Energy ◽  
Turbulent Flow ◽  
Drag Reduction ◽  
Turbulent Kinetic Energy ◽  
Newtonian Fluid ◽  
Viscoelastic Fluid ◽  
Energy Budget ◽  
Rheological Parameters ◽  
Fluid Turbulence ◽  
Kinetic Energy Budget

For the purpose of elucidating the mechanism of drag reduction by additives and finding a way to judge optimum drag-reducing additives through a simple rheological test, we performed DNS analysis of viscoelastic fluid turbulent flow in a two-dimensional channel. In this calculation, we employed the Giesekus constitutive equation to model the interaction between water-soluble polymer, or the elastic micellar network structure, and solvent. We calculated the fluid flow by varying the rheological parameters of the model. We examined the turbulent kinetic energy budget and studied the “viscoelastic contribution” term in the budget equation for turbulent intensity, which is not apparent in normal Newtonian fluid turbulence. Viscoelastic contribution has a characteristic effect on viscoelastic fluid turbulence. We concluded that the viscoelastic contribution plays a major role in turbulent frictional drag reduction. Dissipation and viscoelastic contribution serve as a key factor of turbulent kinetic energy loss in most areas of the channel. From the visualization of local and instantaneous eddy behavior, we discussed the relationship between viscoelastic contribution, elastic energy and turbulent production. We found that viscoelastic contribution serves as a direct local source of turbulent production, and that energy is stored in the elasticity.

On the structure of shear stress and turbulent kinetic energy flux across the roughness layer of a gravel-bed channel flow

2009 ◽  
Vol 638 ◽  
pp. 423-452 ◽  
Author(s):  
EMMANUEL MIGNOT ◽  
D. HURTHER ◽  
E. BARTHELEMY
Keyword(s):  
Shear Stress ◽  
Kinetic Energy ◽  
Turbulent Kinetic Energy ◽  
Channel Flow ◽  
Mean Flow ◽  
Local Flow ◽  
Gravel Bed ◽  
Wall Flows ◽  
The Mean ◽  
Conditional Statistics

This study examines the structure of shear stress and turbulent kinetic energy (TKE) flux across the roughness layer of a uniform, fully rough gravel-bed channel flow (ks+ ≫ 100, δ/k = 20) using high-resolution acoustic Doppler velocity profiler measurements. The studied gravel-bed roughness layer exhibits a complex random multi-scale roughness structure in strong contrast with conceptualized k- or d-type roughness in standard rough-wall flows. Within the roughness layer, strong spatial variability of all time-averaged flow quantities are observed affecting up to 40% of the boundary layer height. This variability is attributed to the presence of bed zones with emanating bed protuberances (or gravel clusters) acting as local flow obstacles and bed zones of more homogenous roughness of densely packed gravel elements. Considering the strong spatial mean flow variability across the roughness layer, a spatio-temporal averaging procedure, called double averaging (DA), has been applied to the analysed flow quantities. Three aspects have been addressed: (a) the DA shear stress and DA TKE flux in specific bed zones associated with three classes of velocity profiles as previously proposed in Mignot, Barthélemy & Hurther (J. Fluid Mech., vol. 618, 2009, p. 279), (b) the global and per class DA conditional statistics of shear stress and associated TKE flux and (c) the contribution of large-scale coherent shear stress structures (LC3S) to the TKE flux across the roughness layer. The mean Reynolds and dispersive shear structure show good agreement between the protuberance bed zones associated with the S-shape/accelerated classes and recent results obtained in standard k-type rough-wall flows (Djenidi et al., Exp. Fluids, vol. 44, 2008, p. 37; Pokrajac, McEwan & Nikora, Exp. Fluids, vol. 45, 2008, p. 73). These gravel-bed protuberances act as local flow obstacles inducing a strong turbulent activity in their wake regions. The conditional statistics show that the Reynolds stress contribution is fairly well distributed between sweep and ejection events, with threshold values ranging from H = 0 to H = 8. However, the TKE flux across the roughness layer primarily results from the residual shear stress between ejection and sweep of very high magnitude (H = 10–20) and of small turbulent scale. Although LC3S are seen to penetrated the interfacial roughness layer, their TKE flux contribution is found to be negligible compared to the very energetic small-scale sweep events. These sweeps are dominantly produced in the bed zones of local gravel protuberances where the velocity profiles are inflexional of S-shape type and the mean flow properties are of mixing-layer flow type as previously shown in Mignot et al. (2009).

On the interactions between large-scale structure and fine-grained turbulence in a free shear flow I. The development of temporal interactions in the mean

1976 ◽  
Vol 352 (1669) ◽  
pp. 213-247 ◽  
Keyword(s):  
Energy Transfer ◽  
Kinetic Energy ◽  
Shear Layer ◽  
Turbulent Kinetic Energy ◽  
Viscous Dissipation ◽  
Mean Flow ◽  
Fine Grained ◽  
Large Eddy ◽  
The Mean ◽  
Three Components

In this problem a mean turbulent shear layer originally exists, homogeneous in the streamwise direction, formed perhaps by previous instabilities, but in equilibrium with the fine-grained turbulence. At a given time, a large eddy of a fixed horizontal wavenumber is initiated. We study the subsequent time development of the non-equilibrium interactions between the three components of flow as they adjust towards ultimate simultaneous equilibrium, using the integrated energy-balance conservation equations to derive the amplitude equations. This necessarily involves the usual averaging procedure and a conditional or phase-averaging procedure by which the large structure motion is educed from the total fluctuations. In general, the mean flow growth is due to the energy transfer to both fluctuating components, the large eddy gains energy from the mean motion and exchanges energy with the fine-grained turbulence, while the fine-grained turbulence gains energy from the mean flow and exchanges with the large eddy and converts its energy to heat through viscous dissipation of the smallest scales. The closure problem is obtained via the shape assumptions which enter into the interaction integrals. The situation in which the fine-grained turbulent kinetic energy production and viscous dissipation are in local balance is considered, the displacement from equilibrium being due only to the energy transfer from the large eddy. The large eddy shape is taken to be two-dimensional, instability-wavelike, with its vorticity axis perpendicular to the direction of the mean outer stream. Prior to averaging, detailed but approximate calculations of the wave-induced turbulent Reynolds stresses are obtained; the product of these stresses with the appropriate large-eddy rates of strain give the energy transfer mechanism between the two disparate scales of fluctuations. Coupled, nonlinear amplitude or energy density equations for the three components of motion are obtained, the coefficients of which are the interaction integrals guided by the shape assumptions. It is found that for the special case of parallel flow, the energy of the large eddy first undergoes a hydrodynamic-instability type of amplification but eventually decays due to the energy transfer to the fine-grained turbulence, while the turbulent kinetic energy is displaced from an original level of equilibrium to a new one because of the ability of the large eddy to negotiate an indirect energy transfer from the mean flow. For the growing shear layer, approximate considerations show that if the mechanism of energy transfer from the large to the small scale is eventually weakened by the shear layer growth compared to the large-eddy production mechanism so that the amplification and decay process repeats, ‘bursts’ of the remnant of the same large eddy will occur repeatedly until an ultimate equilibrium is reached among the three interacting components of motion. However, for the large eddy whose wavenumber corresponds to that of the initially most amplified case, the ‘bursting’ phenomenon is much less pronounced and equilibrium is very nearly reached at the end of the very first ‘burst’.

The evolution of a uniformly sheared thermally stratified turbulent flow

1997 ◽  
Vol 334 ◽  
pp. 61-86 ◽  
Author(s):  
PAUL PICCIRILLO ◽  
CHARLES W. VAN ATTA
Keyword(s):  
Kinetic Energy ◽  
Turbulent Kinetic Energy ◽  
Richardson Number ◽  
Initial Conditions ◽  
Velocity Shear ◽  
Spectral Energy ◽  
Small Scale ◽  
Mean Flow ◽  
Mean Velocity ◽  
Small Scales

Experiments were carried out in a new type of stratified flow facility to study the evolution of turbulence in a mean flow possessing both uniform stable stratification and uniform mean shear.The new facility is a thermally stratified wind tunnel consisting of ten independent supply layers, each with its own blower and heaters, and is capable of producing arbitrary temperature and velocity profiles in the test section. In the experiments, four different sized turbulence-generating grids were used to study the effect of different initial conditions. All three components of the velocity were measured, along with the temperature. Root-mean-square quantities and correlations were measured, along with their corresponding power and cross-spectra.As the gradient Richardson number Ri = N2/(dU/dz)2 was increased, the downstream spatial evolution of the turbulent kinetic energy changed from increasing, to stationary, to decreasing. The stationary value of the Richardson number, Ricr, was found to be an increasing function of the dimensionless shear parameter Sq2/∈ (where S = dU/dz is the mean velocity shear, q2 is the turbulent kinetic energy, and ∈ is the viscous dissipation).The turbulence was found to be highly anisotropic, both at the small scales and at the large scales, and anisotropy was found to increase with increasing Ri. The evolution of the velocity power spectra for Ri [les ] Ricr, in which the energy of the large scales increases while the energy in the small scales decreases, suggests that the small-scale anisotropy is caused, or at least amplified, by buoyancy forces which reduce the amount of spectral energy transfer from large to small scales. For the largest values of Ri, countergradient buoyancy flux occurred for the small scales of the turbulence, an effect noted earlier in the numerical results of Holt et al. (1992), Gerz et al. (1989), and Gerz & Schumann (1991).

Sign in / Sign up

Export Citation Format

Share Document