Synthetic Turbulence Research Articles

Boundary layers affected by different pressure gradients investigated computationally by a zonal RANS-LES method

The reformulated synthetic turbulence generation (RSTG) method is used to compute by a fully coupled zonal RANS-LES approach turbulent non-zero-pressure gradient boundary layers. The quality of the RSTG method, which is based on the same shape functions and length scale distributions as in zero-pressure gradient flow, is discussed by comparing the zonal RANS-LES findings with pure LES, pure RANS, direct numerical simulation (DNS), and experimental data. For the favorable pressure gradient (FPG) simulation the RANS-to-LES transition occurs in the accelerated flow region and for the adverse pressure gradient (APG) case it is located in the decelerated flow region. The results of the time and spanwise averaged skin-friction distributions, velocity profiles, and Reynolds stress distributions of the zonal RANS-LES simulation show a satisfactory to good agreement with the pure LES, reference DNS, and experimental data. The quality of the findings shows that the rigorous formulation of the synthetic turbulence generation makes the RSTG method applicable without a priori knowledge of the flow properties but those determined by the RANS solution and without using additional control planes to regulate the shear stress budget to a wide range of Reynolds numbers and pressure gradients. The method is a promising approach to formulate embedded RANS-to-LES boundaries in flow regions where the Pohlhausen or acceleration parameter satisfies -1·10-6⩽K⩽2·10-6.

Boundary Conditions and SGS Models for LES of Wall-Bounded Separated Flows: An Application to Engine-Like Geometries

The implementation and the combination of advanced boundary conditions and subgrid scale models for Large Eddy Simulations are presented. The goal is to perform reliable cold flow LES simulations in complex geometries, such as in the cylinders of internal combustion engines. The implementation of an inlet boundary condition for synthetic turbulence generation and of two subgrid scale models, the local Dynamic Smagorinsky and the Wall-Adapting Local Eddy-viscosity SGS model ( WALE) is described. The WALE model is based on the square of the velocity gradient tensor and it accounts for the effects of both the strain and the rotation rate of the smallest resolved turbulent fluctuations and it recovers the proper y3 near-wall scaling for the eddy viscosity without requiring dynamic pressure; hence, it is supposed to be a very reliable model for ICE simulation. Model validation has been performed separately on two steady state flow benches: a backward facing step geometry and a simple IC engine geometry with one axed central valve. A discussion on the completeness of the LES simulation (i.e. LES simulation quality) is given.

Unstructured large eddy simulations of heated supersonic twin jets

Unstructured large eddy simulations are performed with the compressible flow solver ”Charles” developed at Cascade Technologies, for heated supersonic over-expanded jets issued from a twin nozzle. The study focuses on the modeling of the internal flow and its effects on the flow-field in the jet plume and ultimately on the radiated noise. In this work, near-wall adaptive mesh refinement, synthetic inflow turbulence and wall modeling are used inside the nozzle. In addition to the converging-diverging nozzle geometry, the computational domain includes the Y-duct, S-ducts, and angle adapter upstream of the nozzles, to realistically reproduce the elements of the experimental configuration where the internal flow conditions and wall modeling can be expected to affect the external flow-field and sound-field predictions. Comparisons between the numerical predictions and experimental PIV and far-field noise measurements from NASA Glenn Research Center will be presented and discussed.

Toward low‐noise synthetic turbulent inflow conditions for aeroacoustic calculations

SUMMARYThe accuracy of boundary conditions for computational aeroacoustics is a well‐known challenge, due in part to the necessity of truncating the flow domain and replacing the analytical boundary conditions at infinity with numerical boundary conditions. In particular, the inflow boundary condition involving turbulent velocity or scalar fields is likely to introduce spurious waves into the domain, therefore degrading the flow behavior and deteriorating the physical acoustic waves. In this work, a method to generate low‐noise, divergence‐free, synthetic turbulence for inflow boundary conditions is proposed. It relies on the classical view of turbulence as a superposition of random eddies convected with the mean flow. Within the proposed model, the vector potential and the requirement that the individual eddies must satisfy the linearized momentum equations about the mean flow are used. The model is tested using isolated eddies convected through the inflow boundary and an experimental benchmark data for spatially decaying isotropic turbulence. Copyright © 2013 John Wiley & Sons, Ltd.

A New Divergence Free Synthetic Eddy Method for the Reproduction of Inlet Flow Conditions for LES

This paper describes a recent development of the Synthetic Eddy Method (SEM) proposed by Jarrin et al. (Int J Heat Fluid Flow 30(3):435–442, 2009) for generation of synthetic turbulence. The present scheme is designed to produce a divergence-free turbulence field that can reproduce almost all possible states of Reynolds stress anisotropy. This improved representation, when used to provide inlet conditions for an LES, leads to reduced near-inlet pressure fluctuations in the LES and to a reduced development length, both of which lead to lower computer resource requirements. An advantage of this method with respect to forcing approaches (which require an iterative approach) is the suitability for direct usage with embedded LES. Results for a turbulent channel flow are reported here and compared to those from the original SEM, and other direct approaches such as the VORTEX method of Sergent (2002) and the Synthesized Turbulence approach of Davidson and Billson (Int J Heat Fluid Flow 27(6):1028–1042, 2006), showing overall improved performance and a more accurate representation of turbulence structures immediately downstream of the inlet.

Synthetic atmospheric turbulence and wind shear in large eddy simulations of wind turbine wakes

A method of generating a synthetic ambient wind field in neutral atmosphere is described and verified for modelling the effect of wind shear and turbulence on a wind turbine wake using the flow solver EllipSys3D. The method uses distributed volume forces to represent turbulent fluctuations, superimposed on top of a mean deterministic shear layer consistent with that used in the IEC standard for wind turbine load calculations. First, the method is evaluated by running a series of large-eddy simulations in an empty domain, where the imposed turbulence and wind shear is allowed to reach a fully developed stage in the domain. The performance of the method is verified by comparing the turbulence intensity and spectral distribution of the turbulent energy to the spectral distribution of turbulence generated by the IEC suggested Mann model. Second, the synthetic turbulence and wind shear is used as input for simulations with a wind turbine, represented by an actuator line model, to evaluate the development of turbulence in a wind turbine wake. The resulting turbulence intensity and spectral distribution, as well as the meandering of the wake, are compared to field data. Overall, the performance of the synthetic methods is found to be adequate to model atmospheric turbulence, and the wake flow results of the model are in good agreement with field data. An investigation is also carried out to estimate the wake transport velocity, used to model wake meandering in lower-order models. The conclusion is that the appropriate transport velocity of the wake lies somewhere between the centre velocity of the wake deficit and the free stream velocity. Copyright © 2013 John Wiley & Sons, Ltd.

Divergence-free turbulence inflow conditions for large-eddy simulations with incompressible flow solvers

Synthetic turbulence for inflow conditions formulated on a 2-D plane generally produces unphysically large pressure fluctuations in direct numerical and large-eddy simulations. To reduce such artificial fluctuations a divergence-free method is developed with incompressible flow solvers. The procedure of the velocity–pressure solvers is slightly modified on a vertical plane near (rather than at) the inlet by inserting the synthetic turbulence on that plane during the procedure. Simple analytic and numerical error estimations are used to show that the impact of the modified solvers on solution accuracy is small. The final synthetic turbulence satisfies the divergence-free condition. No additional CPU time is required to achieve this condition. The method was tested via simulations of a plane channel flow with Reτ=395. Reynolds stresses, wall skin friction and power spectra of velocity fluctuations are compared with those obtained from using periodic inlet–outlet boundary conditions. In particular, the variances and power spectra of pressure fluctuations are shown to be accurately predicted only when the divergence-free inlet condition is used.

Assessment of Reynolds stresses tensor reconstruction methods for synthetic turbulent inflow conditions. Application to hybrid RANS/LES methods

Hybrid or zonal RANS/LES approaches are recognized as the most promising way to accurately simulate complex unsteady flows under current computational limitations. One still open issue concerns the transition from a RANS to a LES or WMLES resolution in the stream-wise direction, when near wall turbulence is involved. Turbulence content has then to be prescribed at the transition to prevent from turbulence decay leading to possible flow relaminarization. The present paper aims to propose an efficient way to generate this switch, within the flow, based on a synthetic turbulence inflow condition, named Synthetic Eddy Method (SEM). As the knowledge of the whole Reynolds stresses is often missing, the scope of this paper is focused on generating the quantities required at the SEM inlet from a RANS calculation, namely the first and second order statistics of the aerodynamic field. Three different methods based on two different approaches are presented and their capability to accurately generate the needed aerodynamic values is investigated. Then, the ability of the combination SEM+Reconstruction method to manufacture well-behaved turbulence is demonstrated through spatially developing flat plate turbulent boundary layers. In the mean time, important intrinsic features of the Synthetic Eddy method are pointed out. The necessity of introducing, within the SEM, accurate data, with regards to the outer part of the boundary layer, is illustrated. Finally, user’s guidelines are given depending on the Reynolds number based on the momentum thickness, since one method is suitable for low Reynolds number while the second is dedicated to high ones with a transition located around Reθ=3000.

A reformulated synthetic turbulence generation method for a zonal RANS–LES method and its application to zero-pressure gradient boundary layers

A synthetic turbulence generation (STG) method for subsonic and supersonic flows at low and moderate Reynolds numbers to provide inflow distributions of zonal Reynolds-averaged Navier–Stokes (RANS) – large-eddy simulation (LES) methods is presented. The STG method splits the LES inflow region into three planes where a local velocity signal is decomposed from the turbulent flow properties of the upstream RANS solution. Based on the wall-normal position and the local flow Reynolds number, specific length and velocity scales with different vorticity content are imposed at the inlet plane of the boundary layer. The quality of the STG method for incompressible and compressible zero-pressure gradient boundary layers is shown by comparing the zonal RANS–LES data with pure LES, pure RANS, and direct numerical simulation (DNS) solutions. The distributions of the time and spanwise wall-shear stress, Reynolds stress distributions, and two point correlations of the zonal RANS–LES simulations are smooth in the transition region and in good agreement with the pure LES and reference DNS findings. The STG approach reduces the RANS-to-LES transition length to less than four boundary-layer thicknesses.

Towards the LES Simulation of IC Engines with Parallel Topologically Changing Meshes

Rare presented. The goal is to perform reliable cold flow LES simulations in complex geometries, such as in the cylinders of internal combustion engines. The implementation of a boundary condition for synthetic turbulence generation upstream of the valve port and of the compressible formulation of the Wall-Adapting Local Eddy-viscosity sgs model (WALE) is described. The WALE model is based on the square of the velocity gradient tensor and it accounts for the effects of both the strain and the rotation rate of the smallest resolved turbulent fluctuations and it recovers the proper y 3 near-wall scaling for the eddy viscosity without requiring dynamic procedure; hence, it is supposed to be a very reliable model for ICE simulation. Validation of the implemented models has been performed separately on two steady state flow benches, where experimental data were available: a backward facing step geometry and a simple IC engine geometry with one axed central valve. A method to evaluate the turbulence resolution of LES simulation based on a Length Scale Resolution (LSR) parameter has been proposed and a discussion about the completeness of the LES simulation (i.e. LES simulation quality) is presented. Finally, the implementation of a fully parallel algorithm for the management of topologically changing meshes in the OpenFOAM R

Numerically determined geometric collision kernels in spatially evolving isotropic turbulence relevant for droplets in clouds

The collision probability of cloud droplets in a turbulent flow has been investigated using direct numerical simulation. A novel simulation method is used in which synthetic turbulence is generated at the inlet and is transported through the flow domain with a mean carrier flow. For the dispersed phase a Lagrangian point particle model is applied. Collision statistics have been gathered for ten droplet sizes ranging from 5 to 50μm in different statistic volumes in the turbulent flows with dissipation rates between 30 and 250cm2s−3 and Taylor-scale Reynolds numbers between 16.4 and 22.4. It is found that turbulence enhances the collision probability by factors up to 1.66 relative to gravitational settling. The resulting geometric collision kernel is decomposed into its primary contributions: the radial distribution function (RDF) and the mean radial relative velocity. The RDF quantifying the preferential droplet concentration reaches values up to 8.6, while a random distribution corresponds to 1. The mean radial relative velocity is enhanced by factors up to 1.18 relative to gravitational settling. The findings are in good quantitative agreement with results from other studies reported in the literature.

Suppression of particle dispersion by sweeping effects in synthetic turbulence

Synthetic models of Eulerian turbulence like so-called kinematic simulations (KS) are often used as computational shortcuts for studying Lagrangian properties of turbulence. These models have been criticized by Thomson and Devenish (2005), who argued on physical grounds that sweeping decorrelation effects suppress pair dispersion in such models. We derive analytical results for Eulerian turbulence modeled by Gaussian random fields, in particular for the case with zero mean velocity. Our starting point is an exact integrodifferential equation for the particle pair separation distribution obtained from the Gaussian integration-by-parts identity. When memory times of particle locations are short, a Markovian approximation leads to a Richardson-type diffusion model. We obtain a time-dependent pair diffusivity tensor of the form K(ij)(r,t)=S(ij)(r)τ(r,t), where S(ij)(r) is the structure-function tensor and τ(r,t) is an effective correlation time of velocity increments. Crucially, this is found to be the minimum value of three times: the intrinsic turnover time τ(eddy)(r) at separation r, the overall evolution time t, and the sweeping time r/v(0) with v(0) the rms velocity. We study the diffusion model numerically by a Monte Carlo method. With inertial ranges like the largest achieved in most current KS (about 6 decades long), our model is found to reproduce the t(9/2) power law for pair dispersion predicted by Thomson and Devenish and observed in the KS. However, for much longer ranges, our model exhibits three distinct pair-dispersion laws in the inertial range: a Batchelor t(2) regime, followed by a Kraichnan-model-like t(1) diffusive regime, and then a t(6) regime. Finally, outside the inertial range, there is another t(1) regime with particles undergoing independent Taylor diffusion. These scalings are exactly the same as those predicted by Thomson and Devenish for KS with large mean velocities, which we argue hold also for KS with zero mean velocity. Our results support the basic conclusion of Thomson and Devenish (2005) that sweeping effects make Lagrangian properties of KS fundamentally differ from those of hydrodynamic turbulence for very extended inertial ranges.

Embedded Large-Eddy Simulation Using the Partially Averaged Navier–Stokes Model

An embedded large-eddy-simulation modeling approach is explored and verified using the partially averaged Navier–Stokes model as a platform. With the same base model, the turbulence-resolving large-eddy simulation region is embedded by setting the partially averaged Navier–Stokes model coefficient to as distinguished from its neighboring Reynolds-averaged Navier–Stokes region, where is specified. The embedded large-eddy simulation approach is verified in computations of a turbulent channel flow and a turbulent flow over a hump. Emphasis is placed on the impact of turbulent conditions at the Reynolds-averaged Navier–Stokes/large-eddy simulation interface using anisotropic velocity fluctuations generated from synthetic turbulence. The effect of the spanwise size of the computational domain is investigated. It is shown that the embedded large-eddy-simulation method based on the partially averaged Navier–Stokes modeling approach is computationally feasible and able to provide reasonable turbulence-resolving predictions in the embedded large-eddy simulation region. The wall-adapting local eddy-viscosity model is also evaluated for the hump flow and it is found that its performance is worse than that of the the low-Reynolds-number partially averaged Navier–Stokes model when the results are compared with experiments.

Large and Small Eddies Matter: Animating Trees in Wind Using Coarse Fluid Simulation and Synthetic Turbulence

AbstractAnimating trees in wind has long been a problem in computer graphics. Progress on this problem is important for both visual effects in films and forestry biomechanics. More generally, progress on tree motion in wind may inform future work on two‐way coupling between turbulent flows and deformable objects. Synthetic turbulence added to a coarse fluid simulation produces convincing animations of turbulent flows but two‐way coupling between the enriched flow and objects embedded in the flow has not been investigated. Prior work on two‐way coupling between fluid and deformable models lacks a subgrid resolution turbulence model. We produce realistic animations of tree motion by including motion due to both large and small eddies using synthetic subgrid turbulence and porous proxy geometry. Synthetic turbulence at the subgrid scale is modulated using turbulent kinetic energy (TKE). Adding noise after sampling the mean flow and TKE transfers energy from small eddies directly to the tree geometry. The resulting animations include both global sheltering effects and small scale leaf and branch motion. Viewers, on average, found animations, which included both coarse fluid simulation and TKE‐modulated noise to be more accurate than animations generated using coarse fluid simulation or noise alone.

A time and space correlated turbulence synthesis method for Large Eddy Simulations

In the present work the problem of generating synthesized turbulence at inflow boundaries of the simulation domain is addressed in the context of the Large Eddy Simulation (LES) method. To represent adequately certain statistical properties of a turbulent process, we propose a synthesized turbulence method which is based on previous works (Huang et al., 2010; Smirnov et al., 2001) [15,28]. For this purpose, time and space correlations are introduced strictly in the mathematical formulation of the synthetic turbulence inflow data. It is demonstrated that the proposed approach inherits the properties of the methods on which it is based while presents some particular advantages as well.The strategy of imposing conditions on the inlet velocity field through turbulence synthesis is implemented in the parallel multiphysics code called PETSc-FEM (http://www.cimec.org.ar/petscfem) primarily targeted to calculations throughout finite elements on general unstructured 2D and 3D grids. We present several numerical tests in order to validate and evaluate the method describing the dynamic phenomena that take place in “real-life” problems, such as a swirling turbulent flow inside a diffuser and the airflow around a vehicle model inside a wind tunnel at high Reynolds number.

Synthetic Turbulence Constructed by Self-Learning Fractal Interpolation

A self-learning fractal interpolation algorithm to construct synthetic fields with statistical properties close to real turbulence is proposed. Different from our previous work [Phys. Rev. E 84 (2011) 026328, 82 (2010) 036311], the position mapping and stretching factors between the adjacent large and small scales are learned from the initial information. Using this method, a turbulence-like field with K41 spectra and without dissipation is constructed well through a coarse grid velocity signal from one experiment's data. After filtering the interpolated signal appropriately, the probability distribution of velocity, velocity structure functions and the anomalous scaling law of the synthetic field are close to those of the original signal.

Random particle methods applied to broadband fan interaction noise

Predicting broadband fan noise is key to reduce noise emissions from aircraft and wind turbines. Complete CFD simulations of broadband fan noise generation remain too expensive to be used routinely for engineering design. A more efficient approach consists in synthesizing a turbulent velocity field that captures the main features of the exact solution. This synthetic turbulence is then used in a noise source model. This paper concentrates on predicting broadband fan noise interaction (also called leading edge noise) and demonstrates that a random particle mesh method (RPM) is well suited for simulating this source mechanism. The linearized Euler equations are used to describe sound generation and propagation. In this work, the definition of the filter kernel is generalized to include non-Gaussian filters that can directly follow more realistic energy spectra such as the ones developed by Liepmann and von Kármán. The velocity correlation and energy spectrum of the turbulence are found to be well captured by the RPM. The acoustic predictions are successfully validated against Amiet’s analytical solution for a flat plate in a turbulent stream. A standard Langevin equation is used to model temporal decorrelation, but the presence of numerical issues leads to the introduction and validation of a second-order Langevin model.

Large-Eddy Simulation of Flow and Convective Heat Transfer in a Gas Turbine Can Combustor With Synthetic Inlet Turbulence

Large eddy simulations of swirling flow and the associated convective heat transfer in a gas turbine can combustor under cold flow conditions for Reynolds numbers of 50,000 and 80,000 with a characteristic Swirl number of 0.7 are carried out. A precursor Reynolds averaged Navier-Stokes (RANS) simulation is used to provide the inlet boundary conditions to the large-eddy simulation (LES) computational domain, which includes only the can combustor. A stochastic procedure based on the classical view of turbulence as a superposition of the coherent structures is used to simulate the turbulence at the inlet plane of the computational domain using the mean flow velocity and Reynolds stress data from the precursor RANS simulation. To further reduce the overall computational resource requirement and the total computational time, the near wall region is modeled using a zonal two layer model (WMLES). A novel formulation in the generalized co-ordinate system is used for the solution of effective tangential velocity and temperature in the inner layer virtual mesh. The WMLES predictions are compared with the experimental data of Patil et al. (2011, “Experimental and Numerical Investigation of Convective Heat Transfer in Gas Turbine Can Combustor,” ASME J. Turbomach., 133(1), p. 011028) for the local heat transfer distribution on the combustor liner wall obtained using robust infrared thermography technique. The heat transfer coefficient distribution on the liner wall predicted from the WMLES is in good agreement with experimental values. The location and the magnitude of the peak heat transfer are predicted in very close agreement with the experiments.

Momentum and scalar transport in a localised synthetic turbulence in a channel flow with a short contraction

A numerical simulation is undertaken to investigate the transport of momentum and a passive scalar in a localised turbulence in a channel with a contraction. The simulation is carried out using a hybrid method which combines the lattice Boltzmann method (LBM, for the velocity field) and the energy equation (for the temperature field). The localised turbulence is generated through pulsed jets issued in the Poiseuille flow developing in the channel at a Reynolds number of about 1000. The aim of the study is twofold : i) determine effect of the contraction on the localised turbulence, and ii) study how the passive scalar behaves in such contracted localised turbulence.The contraction increase the averaged vorticity in the channel flow, which is accompanied by an increase in the averaged kinetic energy. The contraction also tends to reduce the Reynolds stresses. These results are similar those obtained in turbulent pipe flow with an axisymmetric contraction and in a turbulent boundary layer subjected to a favourable pressure gradient. However, it is found that the heat transport in the normal to the wall direction is more dramatically affected (reduced) than that in the direction of the flow.

Fractal patterns in turbulent flow for laden particles

We use Kinematic Simulation as a particular kind of synthetic turbulence model to study the preferential accumulation of laden particles with inertia and gravity. Particles are released as a unifrom cloud in the periodic simulation box. We allow particles to settle in synthetic flow and after some times particles concentrate in a particular sub-domain. We study the dimensional properties of these attractors as functions of drift parameter and Stokes number. The attractor's topology varies from curve(D = 1) to fractal plane.