Large Eddy Simulation Code Research Articles

Large-eddy-simulation of 3-dimensional Rayleigh-Taylor instability in incompressible fluids

The 3-dimensional incompressible Rayleigh-Taylor instability is numerically studied through the large-eddy-simulation (LES) approach based on the passive scalar transport model. Both the instantaneous velocity and the passive scalar fields excited by sinusoidal perturbation and random perturbation are simulated. A full treatment of the whole evolution process of the instability is addressed. To verify the reliability of the LES code, the averaged turbulent energy as well as the flux of passive scalar are calculated at both the resolved scale and the subgrid scale. Our results show good agreement with the experimental and other numerical work. The LES method has proved to be an effective approach to the Rayleigh-Taylor instability.

Un code de simulation des grandes échelles La plate-forme CEA/EDF PRICELES

Click to increase image sizeClick to decrease image sizeA 3D Large Eddy Simulation code is presented in this paper. With this development, CEA and EDF goal is to be able to study the type of complex flows, that are found when industrial processes are considered. Models, numerical models and parallel speed-up of the code are considered.

Towards Technical Application of Large Eddy Simulation

AbstractThe paper is concerned with the computation of high Reynolds number circular cylinder flow (Re = 3900/140,000) based on the large eddy simulation (LES) technique. Because this flow involves a variety of complex flow features encountered in technical applications, successful simulations for this test case, especially at high Reynolds numbers, can be considered as the first step to real world applications of LES. Based on an efficient finite‐volume LES code, a detailed study on different aspects influencing the quality of LES results was carried out. In the present paper, some of the results are presented and compared with experimental measurements available.

Numerical Simulation of Sound Generated by Vortex Pairing in a Mixing Layer

A numerical code solving the filtered Navier-Stokes equations is developed using special techniques of computational aeroacoustics. This approach allows a direct determination of the compressible field on a computational domain including the acoustic far field. A two-dimensional mixing layer between two flows at M 1 = 0.12 and M 2 = 0.48 is simulated. The Reynolds number built up from the initial vorticity thickness and the velocity difference across the mixing layer is Re ω = 1.28 x 10 4 . An appropriate forcing of the mixing layer is defined to have only one pairing in the computational domain at a fixed location. The sound generation pattern for a single pairing displays a double spiral structure, corresponding to the rotating quadrupole associated to two corotative vortices. Successive pairings produce an acoustic radiation at the frequency of this mechanism. The directly computed far-field sound is then compared to the prediction of Lighthill's acoustic analogy (Lighthill, M. J., On Sound Generated Aerodynamically-I. General Theory, Proceedings of the Royal Society of London, Vol. 211, Series A 1107, 1952, pp. 564-587) based on the aerodynamic fluctuations provided by the large-eddy simulation code. Two integral formulations of the analogy, based on spatial derivatives and time derivatives, respectively, are tested. Results are in good qualitative agreement with the results of direct simulation. The accuracy is, however, greater with the formulation using time derivatives instead of spatial derivatives.

Statistics of filtered velocity in grid and wake turbulence

Data on spatially filtered turbulence are commonly needed for a priori sub-grid model studies and for a posteriori testing of large eddy simulation (LES) codes. In this paper, hot-wire anemometry is used to record very long records of such data, required for good convergence of high-order statistics. An array consisting of four X-wire probes placed transversal to the flow direction is built. Unlike previous single-probe hot-wire measurements, which only allowed stream-wise filtering using Taylor’s hypothesis, the array permits cross-stream filtering as well. Measurements which are spatially filtered at a length-scale Δ pertaining to the inertial-range of turbulence are performed in grid and wake turbulence. The data can be used directly to compare with results from LES. From the data, fundamental differences between filtered and unfiltered velocity fields are examined through probability density functions and the scaling behavior of high-order structure functions. A comparative study of probability density functions of filtered and unfiltered velocity increments shows that the tails of the distributions are affected by the filtering even at scales much larger than the filter scale. Significant differences are also observed in regard to the scaling of structure functions. It is shown that extended self-similarity, a recent technique for measuring inertial range scaling exponents, yields questionable results when applied to structure functions of filtered velocity.

Effects of turbulence and heterogeneous emissions on photochemically active species in the convective boundary layer

Photochemistry is studied in a convective atmospheric boundary layer. The essential reactions that account for the ozone formation and depletion are included in the chemical mechanism which, as a consequence, contains a wide range of timescales. The turbulent reacting flow is modeled with a large‐eddy simulation (LES) code. The deviations from chemical equilibrium that are caused by turbulent motions are investigated in terms of the intensity of segregation. For the studied cases it is found that the volume‐averaged concentrations calculated with the LES code agree well with the concentrations calculated with a box model. The reaction rate between RH (a generic hydrocarbon emitted at the surface) and OH is most strongly affected (3% slower than in the box model). However, if RH is emitted nonuniformly at the surface, or if the RH‐OH reaction rate is increased, the volume‐averaged RH destruction by OH may be slowed down by as much as 30% compared to a box model. Sensitivity studies showed that the intensity of segregation between RH and OH not only depends on the strength and spatial distribution of the RH emissions but also on the way NO is emitted in the model atmosphere. The results obtained indicate that the assumption that localized emissions of reactive hydrocarbons, for example, isoprene or terpenes, are instantaneously mixed may lead to an underestimation of their atmospheric lifetime.

Spectral characteristics of edge magnetic turbulence in COMPASS-D

Edge fluctuation data from both COMPASS-D and calculations with the large-eddy simulation code CUTIE have been analysed with a number of techniques, revealing coherent structures which exhibit high-frequency, standing-wave oscillations; some of those observed during edge localized modes (ELMs) have an `inverse-chirp' character and these are related to a disturbance of the plasma boundary in the lower-inboard quadrant of the poloidal plane. A `precursor' mode, seen in Ohmic discharges at about 220 kHz just before large ELMs, appears to be outward ballooning in character. Although the CUTIE calculations do not yet include ELM simulations, the results seem to correspond qualitatively with those seen between ELMs or during ELM-free periods on COMPASS-D.

LES of Chemical and Acoustic Forcing of a Premixed Dump Combustor

This paper describes the first steps in the development of a large eddy simulation (LES) code able to compute combustion instabilities in gas turbines. This code was used to compute the forcing of an experimentally investigated premixed dump combustor. It is shown that the main effect of acoustic waves entering the combustion chamber is to create large vortices and unsteady heat release when these vortices burn. Another effect of waves entering the combustor is to modulate the fuel and air flow rates produced by the feeding lines. In this case the equivalence ratio of the mixture entering the combustor may also vary. This was investigated in a “chemical effect” simulation where the inlet equivalence ratio fluctuates but the total flow rate remains constant. For perturbations from stoichiometric burning, this mechanism was shown to induce less destabilizing effects than the purely aerodynamical mechanism due to vortex formation and combustion. It is shown that the LES methodology developed is able to reproduce the experimentally observed phase shift between acoustic excitation and total reaction rate in the chamber.

A flame surface density approach to large-eddy simulation of premixed turbulent combustion

The flame surface density approach to the modeling of premixed turbulent combustion is well established in the context of Reynolds-averaged simulations. For the future, it is necessary to consider large-eddy simulation (LES), which is likely to offer major advantages in terms of physical accuracy, particularly for unsteady combustion problems. LES relies on spatial filtering for the removal of unresolved phenomena whose characteristic length scales are smaller than the computational grid scale. Thus, there is a need for soundly based physical modeling at the subgrid scales. The aim of this paper is to explore the usefulness of the flame surface density concept as a basis for LES modeling of premixed turbulent combustion. A transport equation for the filtered flame surface density is presented, and models are proposed for unclosed terms. Comparison with Reynolds-averaged modeling is shown to reveal some interesting similarities and differences. These were exploited together with known physics and statistical results from experiment and from direct numerical stimulation in order to gain insight and refine the modeling. The model has been implemented in a combustion LES code together with standard models for scalar and momentum transport. Computational results were obtained for a simple three-dimensional flame propagation test problem, and the relative importance of contributing terms in the modeled equation for flame surface density was assessed. Straining and curvature are shown to have a major influence at both the resolved and subgrid levels.

Calcul direct du rayonnement acoustique d'une couche de mélange par macrosimulation

A Large Eddy Simulation code (ALESIA for Appropriate Large Eddy Simulation for Aeroacoustics) was developed to compute the noise radiated by vortex pairing in a 2-D mixing layer. The code uses numerical techniques specific to Computational AeroA-coustics (CAA) in order to obtain the compressible field with high precision. We show that it is possible to determine directly the acoustic radiation, and that vortex pairing noise corresponds to a rotating quadrupôle source.

Surface length scales and shear stress: Implications for land‐atmosphere interaction over complex terrain

A large eddy simulation (LES) code of the atmospheric boundary layer (ABL) has been developed and applied to study the effect of spatially variable surface properties on the areally averaged surface shear stress at the land‐atmosphere interface. The LES code simulates the space and time evolution of the large‐scale turbulent eddies and their transport effects in the ABL. We report here on simulations of flow over spatially variable roughness fields. The dynamics are simulated, and the resulting space‐time fields are averaged to explore the effects of the surface variability length scales on the average surface shear stress, as used in large‐scale models to estimate scalar fluxes, such as evaporation. We observe asymmetrical response of the smooth‐to‐rough and rough‐to‐smooth transitions, such that the effects of the transitions accumulate rather than cancel. It is shown that the presence of abrupt changes in surface roughness and the atmosphere's response to these patches create a marked dependence of the statistical structure of surface shear stress on the length scale of the surface patches. An increase in regionally averaged surface stress for decreasing horizontal patch length scale is found.

Noise computation of a mixing layer using LES simulations

A study of noise computation of a two-dimensional subsonic mixing layer is performed. The first concern is to develop a large eddy simulation (LES) code solving the filtered Navier–Stokes equations to obtain the nonlinear generation of sound. The compact DRP scheme of Tam and Webb is used in space, and the time integration is carried out by a four-step Runge–Kutta algorithm. The boundary conditions of Dong and Tam are implemented. Then, a two-dimensional mixing layer is simulated. The spatial growth, turbulence rates, and spectra are in good agreement with other available studies. The goal is now to use the LES data to calculate the acoustic field generated by the mixing layer. Various methodologies have been investigated. First, the direct exploitation of the LES pressure field reveals tremendous difficulties due to the different orders of magnitude between aerodynamic and acoustic fluctuations. The other approaches consist of using source terms calculated from the LES velocity fluctuations. Two formulations are tested: Lighthill’s analogy based on a volume integral over the region encompassing sound sources, and linearized Euler’s equations where a source term is added. Numerical results and future developments will be discussed.

Structure of the Entrainment Zone Capping the Convective Atmospheric Boundary Layer

The authors use large-eddy simulation (LES) to investigate entrainment and structure of the inversion layer of a clear convectively driven planetary boundary layer (PBL) over a range of bulk Richardson numbers, Ri. The LES code uses a nested grid technique to achieve fine resolution in all three directions in the inversion layer. Extensive flow visualization is used to examine the structure of the inversion layer and to illustrate the temporal and spatial interaction of a thermal plume and the overlying inversion. It is found that coherent structures in the convective PBL, that is, thermal plumes, are primary instigators of entrainment in the Ri range 13.6 # Ri # 43.8. At Ri 5 13.6, strong horizontal and downward velocities are generated near the inversion layer because of the plume‐interface interaction. This leads to folding of the interface and hence entrainment of warm inversion air at the plume’s edge. At Ri 5 34.5, the inversion’s strong stability prevents folding of the interface but strong horizontal and downward motions near the plume’s edge pull down pockets of warm air below the nominal inversion height. These pockets of warm air are then scoured off by turbulent motions and entrained into the PBL. The structure of the inversion interface from LES is in good visual agreement with lidar measurements in the PBL obtained during the Lidars in Flat Terrain field experiment. A quadrant analysis of the buoyancy flux shows that net entrainment flux (or average minimum buoyancy flux wu min) is identified with quadrant IV w2u1 , 0 motions, that is, warm air moving downward. Plumes generate both large negative quadrant II w1u2 , 0 and positive quadrant III w2u2 . 0 buoyancy fluxes that tend to cancel. The maximum vertical gradient in potential temperature at every (x, y) grid point is used to define a local PBL height, zi(x, y). A statistical analysis of zi shows that skewness of zi depends on the inversion strength. Spectra of zi exhibit a sensitivity to grid resolution. The normalized entrainment rate we/w * , where we and w * are entrainment and convective velocities, varies as ARi21 with A 0.2 in the range 13.6 # Ri # 43.8 and is in good agreement with convection tank measurements. For a clear convective PBL, the authors found that the finite thickness of the inversion layer needs to be considered in an entrainment rate parameterization derived from a jump condition.

Large-eddy simulation of stratified channel flow

An investigation of large-eddy simulation (LES) for turbulent channel flow with buoyancy effects was performed by solving the resolved incompressible Navier-Stokes equations under the Boussinesq approximation. The Smagorinsky eddy-viscosity model and Yoshizawa eddy-viscosity model were used to describe the unresolved subgrid scale (SGS) fluctuations respectively. After some numerical testing, the latter was further simplified so that it can be used in the dynamic model closure. A LES code was developed for parallel computations by using the parallel technique, and was run on the Dawn-1000 parallel computer. To demonstrate the viability and accuracy of the code, our results are compared with and found in good agreement with available LES results.

Development of a parallel code to simulate skewed flow over a bluff body

This paper outlines a new complex geometry large-eddy simulation code using finite differences and a multigrid Poisson solver written for a parallel computer. The flow domain may be constructed from an arbitrary arrangement of rectangular blocks thus permitting flow in regions with complicated shapes, and may be mapped to any number of processors up to the number of blocks. The code is used to simulate turbulent flow at a Reynolds number of 3000 past a cube placed at ground level in rough terrain with its sides set 45° to mean flow direction. Preliminary results are presented which reproduce the conical vortices on the top of the cube.

LES study of flow around a CO2-droplet plume in the ocean

Abstract Transport of liquid CO2 to depths of 500–2000 m, and subsequent solution of CO2 in the sea water, is one of several CO2 mitigation options which have the potential to become both technically and economically feasible, and environmentally acceptable. A three dimensional Large Eddy Simulation (LES) code with conservation equations for salt and total carbond concentration is used to simulate spreading from a CO2 dropet plume, in an ocean linearly stratified through a salinity gradient of 0.025 psu km−1. Results from an integrated model for release of CO2 droplets rising in the water column are used to prescribe the carbon source for the LES model. For a plume with release of 25 kg-CO2 s−1 the CO2 enriched water spreads out at a neutral depth in a layer with center at a depth about 300 meter below the release port ranging from an upper bound at the center of the plume down to 1600 meter. The pH value in this layer becomes low and will affect the marine life.

Local Grid Refinement in Large-Eddy Simulation

A method for performing nested-grid calculations with a Large-Eddy Simulation (LES) code is described. The grid consists of a coarse mesh and a fine mesh which overlaps the coarse in some region. A standard finite-volume method is used on both meshes. By means of grid communication, the velocity and pressure at both meshes are matched. To check that large eddies which are already resolved on the coarse grid are not affected by the nesting procedure, a simple two-dimensional mixing layer is simulated. Several simulations of this flow have been carried out with a different number of grid points on the nested grid. It is found that, without much extra computational effort, the grid-nesting improves the turbulent statistics with respect to the results found on the coarse mesh. This improvement occurs first of all in the region where grid refinement is applied, but better results are also found on the coarse mesh outside the grid-refinement region. Furthermore, it is shown that the large-scale structures in this flow are not influenced by the boundary between the coarse and fine grid.

A grid nesting method for large-eddy simulation of planetary boundary-layer flows

A method for performing nested grid calculations with a large-eddy simulation code is described. A common numerical method is used for all meshes, and the grid architecture consists of a single outer or coarse grid, and nested or fine grids, which overlap in some common region. Inter-grid communication matches the velocity, pressure and potential temperature fields in the overlap region. Resolved and sub-grid scale (SGS) turbulent fluxes and kinetic energy on the fine grid are averaged to the coarse grid using a conservation rule equivalent to Germano's identity used to develop dynamic SGS models. Simulations of a slightly convective, strong shear planetary boundary layer were carried out with varying surface-layer resolutions. Grid refinements in the (x, y, z) directions of up to (5, 5, 2) times were employed. Two-way interaction solutions on the coarse and fine meshes are successfully matched in the overlap region on an instantaneous basis, and the turbulent motions on the fine grid blend smoothly into the coarse grid across the grid interface. With surface-layer grid nesting, significant increases in resolved eddy fluxes and variances are found. The energy-scale content of the vertical velocity, and hence vertical turbulent fluxes, appear to be most influenced by increased grid resolution. Vertical velocity spectra show that the dominant scale shifts towards higher wavenumbers (smaller scales) and the magnitude of the peak energy is increased by more than a factor of 3 with finer resolution. Outside of the nested region the average heat and momentum fluxes and spectra are slightly influenced by the fine resolution in the surface layer. From these results we conclude that fine resolution is required to resolve the details of the turbulent motions in the surface layer. At the same time, however, increased resolution in the surface layer does not appreciably alter the ensemble statistics of the resolved and SGS motions outside of the nested region.

Simulation of a Stratocumulus-Topped Planetary Boundary Layer: Intercomparison among Different Numerical Codes

This paper reports an intercomparison study of a stratocumulus-topped planetary boundary layer (PBL) generated from ten 3D large eddy simulation (LES) codes and four 2D cloud-resolving models (CRMs). These models vary in the numerics, the parameterizations of the subgrid-scale (SGS) turbulence and condensation processes, and the calculation of longwave radiative cooling. Cloud-top radiative cooling is often the major source of buoyant production of turbulent kinetic energy in the stratocumulus-topped PBL. An idealized nocturnal stratocumulus case was selected for this study. It featured a statistically horizontally homogeneous and nearly solid cloud deck with no drizzle, no solar radiation, little wind shear, and little surface heating. Results of the two-hour simulations showed that the overall cloud structure, including cloud-top height, cloud fraction, and the vertical distributions of many turbulence statistics, compared well among all LESs despite the code variations. However, the entrainment rate was found to differ significantly among the simulations. Among the model uncertainties due to numerics, SGS turbulence, SGS condensation, and radiation, none could be identified to explain such differences. Therefore, a follow-up study will focus on simulating the entrainment process. The liquid water mixing ratio profiles also varied significantly among the simulations; these profiles are sensitive to the algorithm used for computing the saturation mixing ratio. Despite the obvious differences in eddy structure in two- and three-dimensional simulations, the cloud structure predicted by the 2D CRMs was similar to that obtained by the 3D LESs, even though the momentum fluxes, the vertical and horizontal velocity variances, and the turbulence kinetic energy profiles predicted by the 2D CRMs all differ significantly from those of the LESs.

Large-eddy simulation of a neutrally stratified boundary layer: A comparison of four computer codes

A neutrally stratified Ekman layer limited in depth by a stress-free rigid lid has been simulated using four different large-eddy simulation computer codes. This is the second large-eddy simulation intercomparison study in a series of simulations of different flows aimed at a systematic analysis of large-eddy simulation sensitivity to subgrid scale model assumptions and numerical methods. The resolution was chosen to be coarse so that the results might be sensitive to the details of the various subgrid scale model formulations. Stochastic backscatter of subgrid kinetic energy and scalar variance was included in one subgrid scale model. Profiles of first and second moments, budgets of resolved-scale second moments, and spectra have been analyzed. It is found that significant differences in the lowest third of the boundary layer result from moderate changes in subgrid scale eddy diffusivities or from the inclusion of backscatter. To isolate effects of the various numerical methods, a set of tests with a fixed subgrid scale K-profile were made. When run with a common subgrid scale model the results from the four large-eddy simulation codes converge. This test indicates that the large-eddy simulation results are more sensitive to the subgrid scale model formulations than to the various numerics.