Accuracy Of Large Eddy Simulation Research Articles

Effect of inlet conditions on the accuracy of large eddy simulations of a turbulent rectangular wake

Abstract A large eddy simulation (LES) study was performed on a turbulent incompressible wake flow in a rectangular channel. The simulation results were evaluated using particle image velocimetry (PIV) data from a previous experimental study of the same flow (Liu et al., 2013). Comparisons were made of one-point statistics as well as spatial correlations. An extensive pre-simulation study was carried out in which the effect of inlet conditions, grid resolution, time step and subgrid model was investigated and the parameters were optimized. Using the digital filter method of Klein et al. (2003), turbulent inflow velocities were generated based on velocity mean and variance obtained from the experimental data and correlation length scales. It was found that the simulation results in large parts of the domain were strongly dependent on the inlet length scales specified. With a suitable set of length scales, the inlet method was successful at providing inlet conditions that generated accurate simulation results. The very good agreement seen between experiment and simulation demonstrates LES as a method that, with carefully selected inlet conditions, not only can predict the pointwise turbulence statistics of a liquid wake flow, but also capture key features of its large-scale turbulent structures.

Synthesis of CFD and Monte-Carlo simulations for improved design and operation of Dense Medium Cyclones

Accurate resolution of the complex flow features within a Dense Medium Cyclone used in coal beneficiation contributes not only to the design of cyclones of greater separation efficiency but also the optimization of operational process parameters. A three step approach to numerical simulation of flow within such a cyclone, starting with the approximate and simpler Reynolds stress model, transiting to the more complex and accurate Large Eddy Simulation after a certain level of maturity, and then activating a multi-phase model after flow stabilization in the suction core, is developed. A systematic approach based on Monte-Carlo simulations is then synthesized, relating the distribution of particle misplacements stochastically to the distribution of simulated flow and particle conditions like turbulence, size, density, etc. The relationship of misplacements via these particle conditions to the antecedent feed coal conditions is established. Every significant step in the evolution of this approach is validated against experiment or production data and also verified for all necessary conditions of simulation accuracy like grid effects, air-core features, outflow split, misplacement distribution across the size-density matrix and separation efficiency. Thus a coupled Computational Fluid Dynamics and Monte-Carlo simulations approach is synthesized for application to improved design and operation of a Dense Media Cyclone.

Evaluation of scale resolving turbulence generation methods for Large Eddy Simulation of turbulent flows

Large Eddy Simulation (LES) has become an attractive simulation method even for technical processes and it usually provides space and time resolved fluctuations of a significant portion of the spectrum. However, in contrast to a RANS simulation an accurate LES requires the definition of suitable initial and boundary conditions, which includes turbulent structures with physically sound spatial and temporal correlations. Such turbulent structures are usually generated artificially at the boundary. Three different algorithms for generating turbulent fluctuations are evaluated in the present work. The investigated methods are Filtered noise [1], Diffused noise [2] and an Inverse Fourier approach [3,4]. These techniques were developed for generating inflow data for LES and have already been used in published research [5–15], e.g. for investigating turbulent combustion processes. In the present work the turbulent statistics i.e. energy spectra and velocity correlations as well as derived quantities such as turbulent kinetic energy and subgrid scale viscosity are investigated in more detail in a comparative fashion for the generated turbulent velocity fields. As a simple test case, the decay of turbulence in a cubical box, is considered here to provide information on the initially generated turbulence as well as its temporal evolution. The results are analyzed in detail and are compared to experimental data. Turbulence fluctuations generated by Filtered noise and Diffused noise lead to similar results. The resulting energy spectra and velocity correlations agree generally well with experimental data despite some discrepancies at very early times after initialization. The Inverse Fourier approach yielded good agreement at all times, but at increased computational cost. In addition, the implementation of Filtered noise and Diffused noise might be easier for most cases of practical interest. In particular, the Diffused noise approach can be used for the generation of inhomogeneous turbulence on arbitrary grids.

Grid resolution assessment in large eddy simulation of dispersion around an isolated cubic building

In the present paper, the influence of grid resolution on prediction accuracy of large eddy simulation (LES) of dispersion around an isolated cubic building was investigated. Several grid resolution assessment techniques in LES, namely, two-point correlations, ratio of SGS viscosity to molecular viscosity and ratio of SGS shear stress to resolved shear stress were considered. It was found that two-point correlations could be regarded as a useful method for evaluation of grid resolution. However, it was highly time consuming and required extra post-processing calculation. Also, it was shown that ratios of SGS viscosity to molecular viscosity and SGS shear stress to resolved shear stress decreased by improving grid resolution and they did not change significantly by further increasing the grid resolution. Thus, both these ratios were concluded to be an applicable technique for the evaluation of grid resolution in LES.

Large-Eddy Simulation of Turbulent Flow with Heat Transfer in a Rod Bundle

To assess the accuracy of large-eddy simulation (LES) predictions for flow without and with heat transfer in a rod bundle, analyses were performed with a constant-coefficient Smagorinsky LES model, and numerical predictions were compared with experimental measurements in a heated triangular rod array. First, flow simulations without heat transfer were performed with one and two channels at the central region of the bundle, and simulation predictions were compared with the experimental data. For the normalized mean axial velocity and the axial component of the turbulent intensity, the predictions of the one-channel model are nearly identical with those of the two-channel model. For the other turbulence parameters, the predictions of the one-channel model are either identical or are mostly in good agreement with those of the two-channel model. LES predictions for the mean axial velocity agree well with experimental measurements. Predictions of the axial component of the turbulent intensity agree well with experimental measurements for most of the points of measurement. Predictions of the other parameters of turbulence agree well to reasonably well with measurements. Because LES simulations are computationally very demanding, the LES simulation of heat transfer was performed only with the one-channel model. LES predicts the temperature of the rod surface within the range of the experimental error. The profile (log law) of the dimensionless fluid temperature T+ predicted by LES has the same slope as that derived from the measurements, but it has a significantly higher constant. The turbulent intensity of temperature is predicted well to reasonably well. The turbulent heat flux in the axial direction and the radial direction is predicted well at points away from the wall, but there is significant discrepancy between predictions and measurements close to the wall. The predicted turbulent heat flux in the azimuthal direction agrees very well to quite well with measurements.

A relative assessment of sub grid scale models for large eddy simulation of co-axial combustor

Large eddy simulation (LES) of a co-axial combustor is reported. Accuracy of LES depends on the ability of the sub-grid scale (SGS) models to predict the turbulent viscosity. The sensitivity of LES results for different SGS models is established for a coaxial annular combustor. The radial, axial and tangential velocity distribution predicted by four sub-grid scale models is compared with the experimental results of Sommerfeld and Qiu. It is observed that the flow physics is captured more accurately by WALE model as compared to other SGS models. The predictions of WALE model are closer to the experimental results for all stations in the flow domain.

LES of turbulent wind and gas dispersion in a city

We have conducted LES (large eddy simulation) for turbulent wind flows and dispersion of hazardous gas emitted from a point source on the ground in a large city, Tokyo. Also, for numerical validation, wind tunnel tests using a Dynamic Particle Image Velocimetry (DPIV) and tracer gas measurement system are performed to estimate turbulence statistics of wind and gas concentration in between the compacted tall buildings. We elucidate the predictive accuracy of LES based on the wind tunnel test data. Finally, we discuss the meaning and the potential of LES prediction for dispersion problems in a complex urban area, taking into consideration the limitation of the predicted results by Reynolds Averaged Navier–Stokes (RANS) modeling.

Influence of the turbulence representation at the inlet on the downstream flow pattern in LES of backward-facing step

The impact of the choice of inlet boundary condition treatment on the fluid flow is studied in this work. The correct representation of the turbulence at the inlet to the domain is essential for the accuracy of Large Eddy Simulation. The inappro- priate specification of the inlet velocity has significant effect on the downstream flow pattern. The case of backward-facing step was used as a test case. Three different approaches of the inlet boundary conditions were studied: uniform velocity profile, mean velocity profile of the fully developed channel flow and velocity obtained from mapping velocity from plane positioned behind the inlet back to the inlet. The results of the simulations were compared with experimental results. It has shown, that using uniform velocity profile on inlet and even prescribing turbulent mean velocity profile without proper representation of turbulence fluctuations leads to unrealistic results.

Large-eddy simulation of thermal striping in unsteady non-isothermal triple jet

Unsteady temperature fluctuations of non-isothermal turbulent jets are encountered in many engineering applications including liquid metal cooled fast reactors (LMFR), and can cause thermal stresses on solid boundaries. An accurate prediction of the temperature fluctuations is important to assess potential thermal fatigue damage to components, and traditionally this has been done by RANS turbulence modelling calculations with limited success. In this study, a large eddy simulation (LES) technique was applied to predict the temperature fluctuations of thermal striping observed in a triple jet. The triple jet model was used as a mock-up of the outlet of fuel subassemblies in a nuclear fast reactor. The results show that LES predicted the highly oscillatory nature of unsteady thermal mixing of the triple jet. The LES results were in good agreement with the available experimental data in terms of mean, RMS, skewness and kurtosis. The large amplitude of the temperature fluctuations associated with the thermal striping was captured correctly, demonstrating that LES can be used to analyse unsteady characteristics of thermal striping. Instantaneous and time mean thermal fields were further analysed to assess the capability and accuracy of LES in the thermal striping study. The Spalart–Allmaras and realisable k − ε turbulence models were also considered along with LES. It is found that these turbulence models produced a very small amplitude of fluctuations, and failed to predict the correct magnitude of unsteady thermal fluctuations, highlighting the limitations of the RANS approach in unsteady heat transfer simulations.

Large Eddy Simulation of Turbulence in a Rod Cluster

To assess the accuracy of large eddy simulation (LES) predictions for a flow in a rod bundle, analyses were performed with different parameters of a constant-coefficient Smagorinsky LES model for a flow in a square-pitch rod bundle, and model predictions are compared with experimental data. The parameters considered are the grid structure, the value of the Smagorinsky constant, the damping of the eddy viscosity, and the size of the channel geometry. Because LES simulations are computationally very demanding, for adequately accurate predictions the grid structure needs to be well optimized in terms of cell size, aspect ratio, and cell orthogonality. The use of hanging nodes can significantly reduce the number of cells without a significant penalty on the accuracy of predictions. For this flow, the change in the value of the Smagorinsky constant from 0.14 to zero did not have a drastic effect on predictions. Although, overall, Lilly damping gave slightly better predictions than van Driest damping, both damping functions gave similar predictions. The LES predictions for the mean axial velocity, for the fluctuating velocity component in the main flow direction, and for the Reynolds stresses are in very good agreement with the experimental measurements. There is also good agreement between predictions and measurements for the wall shear stress, but there is a significant discrepancy between predictions and measurements for the fluctuating velocity components in the lateral directions (u and v).

Large-eddy simulation of atmospheric boundary layer flow through wind turbines and wind farms

Accurate prediction of atmospheric boundary layer (ABL) flow and its interactions with wind turbines and wind farms is critical for optimizing the design (turbine siting) of wind energy projects. Large-eddy simulation (LES) can potentially provide the kind of high-resolution spatial and temporal information needed to maximize wind energy production and minimize fatigue loads in wind farms. However, the accuracy of LESs of ABL flow with wind turbines hinges on our ability to parameterize subgrid-scale (SGS) turbulent fluxes as well as turbine-induced forces. This paper focuses on recent research efforts to develop and validate an LES framework for wind energy applications. SGS fluxes are parameterized using tuning-free Lagrangian scale-dependent dynamic models. These models optimize the local value of the model coefficients based on the dynamics of the resolved scales. The turbine-induced forces (e.g., thrust, lift and drag) are parameterized using two types of models: actuator-disk models that distribute the force loading over the rotor disk, and actuator-line models that distribute the forces along lines that follow the position of the blades. Simulation results are compared to wind-tunnel measurements collected with hot-wire anemometry in the wake of a miniature three-blade wind turbine placed in a boundary layer flow. In general, the characteristics of the turbine wakes simulated with the proposed LES framework are in good agreement with the measurements in the far-wake region. Near the turbine, up to about five rotor diameters downwind, the best performance is obtained with turbine models that induce wake-flow rotation and account for the non-uniformity of the turbine-induced forces. Finally, the LES framework is used to simulate atmospheric boundary-layer flow through an operational wind farm.

Numerical simulation of dispersion around an isolated cubic building: Model evaluation of RANS and LES

Several studies have been carried out on CFD prediction based on a RANS (Reynolds Averaged Navier–Stokes equations) model for dispersion around buildings, but it was reported that a RANS computation often provides extremely high concentration, which are not observed in usual measurements. These results suggest that transient simulations such as the large-eddy simulation (LES) might be required to achieve more accurate results. Nevertheless, very few studies have evaluated the basic performance of LES in modeling the dispersion field for a simple configuration in comparison with the RANS model. Therefore, relative performance of these simulation methods for dispersion problem around buildings should be clarified in order to make it possible to choose a suitable numerical method for its purpose. The purpose of this study is to confirm the accuracy of LES in modeling plume dispersion near and around a simple building model and to clarify the mechanism for the discrepancy in relation to the RANS computation. Simple LES modeling gives better results than RNG modeling of the distribution of concentration, although the difference for mean velocity is not so large. The horizontal diffusion of concentration is well reproduced by LES. This tendency is closely related to the reproduction of unsteady periodic fluctuation around cubical forms in LES.

Large eddy simulation using high-resolution and high-order methods

Restrictions on computing power make direct numerical simulation too expensive for complex flows; thus, the development of accurate large eddy simulation (LES) methods, which are industrially applicable and efficient, is required. This paper reviews recent findings about the leading order dissipation rate associated with high-resolution methods and improvements to the standard schemes for use in highly turbulent flows. Results from implicit LES are presented for a broad range of flows and numerical schemes, ranging from the second-order monotone upstream-centered schemes for conservation laws to very high-order (up to ninth-order) weighted essentially non-oscillatory schemes.

Large-eddy simulation and wind tunnel study of flow over an up-hill slope in a complex terrain

This study examines the accuracy of large-eddy simulation (LES) to simulate the flow around a large irregular sloping complex terrain. Typically, real built up environments are surrounded by complex terrain geometries with many features. The complex terrain surrounding The Hong Kong University of Science and Technology campus was modelled and the flow over an uphill slope was simulated. The simulated results, including mean velocity profiles and turbulence intensities, were compared with the flow characteristics measured in a wind tunnel model test. Given the size of the domain and the corresponding constraints on the resolution of the simulation, the mean velocity components within the boundary layer flow, especially in the stream-wise direction were found to be reasonably well replicated by the LES. The turbulence intensity values were found to differ from the wind tunnel results in the building recirculation zones, mostly due to the constraints placed on spatial and temporal resolutions. Based on the validated mean velocity profile results, the flow-structure interactions around these buildings and the surrounding terrain were examined.

Numerical Study of Thermal Plume Characteristics and Entrainment in an Enclosure with a Point Heat Source

The structure of a buoyant plume above a point heat source in a ventilated enclosure has been investigated using large-eddy simulation (LES). The aim of the work is to assess the performance and the accuracy of LES for modelling buoyancy-driven displacement ventilation of an enclosure and investigate the role of coherent structures in the plume entrainment mechanism which is important in these flow types because, for example, entrainment determines the ventilation flow rate. The Smagorinsky subgrid-scale model is used for the unresolved small-scale turbulence. The Rayleigh number Ra is chosen in the range where spatial transition from laminar to turbulent flow takes place (Ra=1.5×109). The stratification height and temperature of the stratified layer deduced from the mean field of the LES data is in good agreement with the theory of Linden, Lane-Serff and Smeed (1990). The plume entrainment coefficient is in good agreement with experimental values determined by Morton, Taylor and Turner (1956), Rouse, Yih and Humphreys (1952), and Baines and Turner (1969). Instantaneously, the plume develops through expansion and contraction phases, where the expansion phase is associated with the existence of coherent large-scale structures leading to an outward stretching of the plume and the contraction occurs as a result of partial breakdown and/or loss of coherence of these structures. As a result, the instantaneous entrained mass (and thus the entrainment coefficient) at different heights below the mean interface height were found to fluctuate about a mean value. Visualization of the computed flow showed that the stretching mechanism of the large-scale structures, which governs the expansion-contraction behaviour of the plume, occurs in such a way that the coherent structure dominating the flow below the interface height takes a spiral shape.

Large eddy simulation of stratified mixing in a three-dimensional lock-exchange system

Accurate and computationally-efficient modeling of stratified mixing processes are of paramount importance in both coastal and large-scale ocean circulation. In this study, our main objective is to investigate the feasibility and accuracy of large eddy simulation ( LES) as a possible tool to study small-scale oceanic processes. To this end, LES is evaluated in a 3D lock-exchange problem, which contains shear-driven mixing, internal waves, interactions with boundaries and convective motions, while having a simple domain, initial and boundary conditions, and forcing. Two general classes of LES models are tested, namely eddy viscosity (EV) models based on constant–coefficient and dynamic Smagorinsky models, and an approximate deconvolution (AD) model. By noting that the dynamic Smagorinsky and AD models have different strengths in that the former is good in providing appropriate dissipation while the latter in preserving the detail of coherent structures on coarse resolution meshes, a hybrid approach combining EV and AD models is also evaluated. A direct numerical simulation ( DNS) is performed as the benchmark solution, and all LES models are tested on three coarse meshes. The main measure of mixing is taken as the temporal evolution of background potential energy. It is found that constant-coefficient Smagorinsky models can only provide a marginal improvement over under-resolved simulations, while both dynamic Smagorinsky and AD models lead to significant improvements in mixing accuracy. The primary accomplishment of this study is that it is shown that the hybrid approach attains the best agreement with the mixing curve from DNS, while being computationally approximately a thousand times faster.

A Similarity Subgrid Model for Premixed Turbulent Combustion

The accuracy of large-eddy simulation (LES) of a turbulent premixed Bunsen flame is investigated in this paper. To distinguish between discretization and modeling errors, multiple LES, using different grid sizes h but the same filterwidth Δ, are compared with the direct numerical simulation (DNS). In addition, LES using various values of Δ but the same ratio Δ/h are compared. The chemistry in the LES and DNS is parametrized with the standard steady premixed flamelet for stochiometric methane-air combustion. The subgrid terms are closed with an eddy-viscosity or eddy-diffusivity approach, with an exception of the dominant subgrid term, which is the subgrid part of the chemical source term. The latter subgrid contribution is modeled by a similarity model based upon 2Δ, which is found to be superior to such a model based upon Δ. Using the 2Δ similarity model for the subgrid chemistry the LES produces good results, certainly in view of the fact that the LES is completely wrong if the subgrid chemistry model is omitted. The grid refinements of the LES show that the results for Δ = h do depend on the numerical scheme, much more than for h = Δ/2 and h = Δ/4. Nevertheless, modeling errors and discretization error may partially cancel each other; occasionally the Δ = h results were more accurate than the h ≤ Δ results. Finally, for this flame LES results obtained with the present similarity model are shown to be slightly better than those obtained with standard β-pdf closure for the subgrid chemistry.

LES of the flow and building wall pressures in the center of Tokyo

For strong wind problems in the city, we have to consider the effect of turbulence characteristics for the flow field and wind pressure on a building surface. Consequently, we use large eddy simulation (LES) rather than Reynolds averaged Navier–Stokes (RANS) in order to reproduce non-isotropy and non-stationarity of the actual wind flow field. LES can reproduce these characteristics, but in contrast, it is difficult to manage complicated boundary conditions of building surfaces in an urban area. Therefore, we use geographic information system (GIS) data and nested grid technique of several levels to reproduce building shapes in detail. With these techniques, we can use a fine calculation grid for a target area and carry out LES for the strong wind field in the center of metropolitan Tokyo. Finally, we try to estimate the accuracy of LES by comparing local flow field and the local pressure distribution of LES results with those of field measurement data.

Large eddy simulation of turbulent flows around bluff bodies in overlaid grid systems

For accurate large eddy simulation (LES) of complex turbulent flows around bluff bodies with various shapes, we have investigated the effects of the upwind scheme on wake structures and the problems in overset grid systems, which require interpolation of physical values between sub-domains. For the former LES of turbulent flows around a circular cylinder at Reynolds number ( Re)=10,000 was carried out with or without using a third-order upwind scheme. As distorted wake structures can be observed when using an upwind scheme, it is desirable to perform LES without using an upwind scheme. For the latter, we propose an overset grid method under LES without using an upwind scheme, because inaccurate conservation of physical values in the region connecting sub-domains causes numerical instability and inaccurate solution. The proposed method improves the standard deviation of pressure distributions around a circular cylinder at Re=10,000.

Design and Validation of a Scale-Adaptive Filtering Technique for LRN Turbulence Modeling of Unsteady Flow

An adaptive low-pass filtering procedure for the modeled turbulent length and time scales is derived and applied to Wilcox’ original low reynolds number k-ω turbulence model. It is shown that the method is suitable for complex industrial unsteady flows in cases where full large eddy simulations (LESs) are unfeasible. During the simulation, the modeled length and time scales are compared to what can potentially be resolved by the computational grid and time step. If the modeled scales are larger than the resolvable scales, the resolvable scales will replace the modeled scales in the formulation of the eddy viscosity. The filtered k-ω model is implemented in an in-house computational fluid dynamics (CFD) code, and numerical simulations have been made of strongly swirling flow through a sudden expansion. The new model surpasses the original model in predicting unsteady effects and producing accurate time-averaged results. It is shown to be superior to the wall-adpating local eddy-viscosity (WALE) model on the computational grids considered here, since the turbulence may not be sufficiently resolved for an accurate LES. Because of the adaptive formulation, the filtered k-ω model has the potential to be successfully used in any engineering case where an LES is unfeasible and a Reynolds (ensemble) averaged Navier–Stokes simulation is insufficient.