Three-dimensional Large Eddy Simulation Research Articles

Numerical investigation of internal solitary waves of elevation type propagating on a uniform slope

Numerous laboratory-scale physical experiments and numerical simulations have been carried out to explore the shoaling dynamics of internal solitary waves (ISWs) on slope topographies. Detailed features during wave breaking have been investigated under relatively low Reynolds numbers, but for real ocean-scale or lake-scale scenarios with a much higher Reynolds number, laboratory-scale modeling is inadequate to capture the three-dimensional turbulent characteristics in the wave shoaling process. As a result, a three-dimensional large-eddy simulation (LES) is performed in the present study to investigate the shoaling process of the elevation-typed ISWs traveling on uniform slopes in a two-layer fluid system. Scale effects due to the Reynolds numbers (varied from 103 to 105) and three-dimensional characteristics during wave shoaling are also explored and discussed. Detailed ISW-slope interaction dynamics, including the typical shoaling features, the characteristics of internal boluses, and both the velocity field and the energy transformation, are systematically obtained and analyzed. It is found that, while reaching the maximum vertical displacement (i.e., maximum run-up height), the frontal part of the heavier lower-layer fluid can evolve into the internal bolus if the internal Iribarren number, Ir, defined as the ratio of the topographic slope and the square root of the incident wave steepness, is less than 0.65. The maximum wave-induced velocities and energy loss are also well related to Ir. Empirical regressed equations for seven important physical parameters during the shoaling process are also proposed. The extreme velocities, wave energy loss, and three-dimensionality of the flow field are all identified to be very sensitive to Reynolds numbers, indicating that traditional two-dimensional laboratory-scale modeling tools may be insufficient to accurately capture the shoaling mechanisms of the ISWs of elevation type.

Broadening of Modeled Cloud Droplet Spectra Using Bin Microphysics in an Eulerian Spatial Domain

Abstract This study investigates droplet size distribution (DSD) characteristics from condensational growth and transport in Eulerian dynamical models with bin microphysics. A hierarchy of modeling frameworks is utilized, including parcel, one-dimensional (1D), and three-dimensional large-eddy simulation (LES). The bin DSDs from the 1D model, which includes only vertical advection and condensational growth, are nearly as broad as those from the LES and in line with observed DSD widths for stratocumulus clouds. These DSDs are much broader than those from Lagrangian microphysical calculations within a parcel framework that serve as a numerical benchmark for the 1D tests. In contrast, the bin-modeled DSDs are similar to the Lagrangian microphysical benchmark for a rising parcel in which Eulerian transport is not considered. These results indicate that numerical diffusion associated with vertical advection is a key contributor to broadening DSDs in the 1D model and LES. This DSD broadening from vertical numerical diffusion is unphysical, in contrast to the physical mixing processes that previous studies have indicated broaden DSDs in real clouds. It is proposed that artificial DSD broadening from vertical numerical diffusion compensates for underrepresented horizontal variability and mixing of different droplet populations in typical LES configurations with bin microphysics, or the neglect of other mechanisms that broaden DSDs such as growth of giant cloud condensation nuclei. These results call into question the ability of Eulerian dynamical models with bin microphysics to investigate the physical mechanisms for DSD broadening, even though they may reasonably simulate overall DSD characteristics.

Analysis of mild ignition in a shock tube using a highly resolved 3D-LES and high-order shock-capturing schemes

A highly resolved three-dimensional large-eddy simulation (LES) is presented for a shock tube containing a stoichiometric hydrogen–oxygen ( $$\hbox {H}_2$$ / $$\hbox {O}_2$$ ) mixture, and the results are compared against experimental results. A parametric study is conducted to test the effects of grid resolution, numerical scheme, and initial conditions before the 3D simulations are presented in detail. An approximate Riemann solver and a high-order interpolation scheme are used to solve the conservation equations of the viscous, compressible fluid and to account for turbulence behind the reflected shock. Chemical source terms are calculated by a finite-rate model. Simultaneous results of pseudo-Schlieren, temperature, pressure, and species are presented. The ignition delay time is predicted in agreement with the experiments by the three-dimensional simulations. The mechanism of mild ignition is analysed by Lagrangian tracer particles, tracking temperature histories of material particles. We observed strongly increased temperatures in the core region away from the end wall, explaining the very early occurrence of mild ignition in this case.

Large Eddy Simulations for Indirect Combustion Noise Assessment in a Nozzle Guide Vane Passage

The combustion noise in aero-engines is known to originate from two different sources. First, the unsteady heat release in the combustion chamber generates the direct combustion noise. Second, hot and cold spots of air generated by the combustion process are convected and accelerated by the turbine stages and give rise to the so-called indirect combustion noise. The present work targets, by using a numerical approach, the generation mechanism of indirect combustion noise for a simplified geometry of a turbine stator passage. Periodic temperature fluctuations are imposed at the inlet, permitting to simulate hot and cold packets of air coming from the unsteady combustion. Three-dimensional Large Eddy Simulation (LES) calculations are conducted for transonic operating conditions to evaluate the blade acoustic response to the forced temperature perturbations at the inlet plane. Transonic conditions are characterized by trailing edge expansion waves and shocks. It is notably shown that their movement can be excited if disturbances with a particular frequency are injected in the domain.

Initial dilution equations for wastewater discharge: Example of non-buoyant jet in wave-following-current environment

In this study, three semi-empirical equations are developed to quantify the wave effect on the initial dilution of wastewater discharge, which usually forms a turbulent jet. To develop the equations, data from a non-buoyant jet in a wave-following-current environment, which is realised using a three-dimensional large eddy simulation model, are used with 27 cases in total. The effects of the wave-to-current velocity ratio, jet-to-current velocity ratio, and Strouhal number are considered in the equations. The initial dilution equations, which focused on the jet visual area, cross-sectional minimum dilution, and vertical location of the cross-sectional minimum dilution, have vital relationships with the characteristic length scale. The reliability of the present equations is discussed by comparing them with existing equations for a jet in a crossflow-only environment. The applicability range of the present equations is illustrated by comparing them with data for a jet in a crossflow-only environment and data for other cases of a jet in a wave-following-current environment. The three equations provide references for the conceptual design of wastewater discharge. This study reveals that a surface wave with low energy has a positive effect on the initial dilution of wastewater discharge.

Passive scalar diffusion in three-dimensional turbulent rectangular free jets with numerical evaluation of turbulent Prandtl/Schmidt number

The passive scalar spreading of fluids with laminar Prandtl or Schmidt number, Pr, Sc, equal to 1 in turbulent rectangular submerged free jets is analyzed by means of numerical simulation and theoretical analysis in the Reynolds number range 5000–40,000. The numerical investigation is carried out by means of a three-dimensional (3D) Large Eddy Simulation (LES) approach with the dynamic Smagorinsky model. A new mathematical model allows to obtain a simplified description of the passive scalar spreading in the largest area of the flow field, the Fully Developed Region (FDR). The present three-dimensional (3D) investigation shows that the passive scalar spreading follows a self-similarity law in the Fully Developed Region (FDR), as well as in the mean Undisturbed Region of Flow (URF) and in the Potential Core Region (PCR), similarly to what found in the Near Field Region (NFR) of rectangular submerged free jets, investigated with a two-dimensional (2D) approach. The turbulent Prandtl or Schmidt number is evaluated numerically and is found to be inversely proportional to the mean velocity gradient in the PCR. The present 3D numerical results show that the turbulent Prandtl or Schmidt number is zero in most part of the mean URF, and PCR, while it assumes different values outside. In the FDR the turbulent Prandtl or Schmidt number is constant and approximately equal to 0.7, in agreement with the literature, showing that turbulence affects momentum and passive scalar in a different way.

Dynamic Mode Decomposition of a Wing-Body Junction Flow

Junction flows are subject to an intense adverse pressure gradient and three-dimensional separation when encountering a wall-mounted obstacle. A dynamically rich horseshoe vortex system is formed in this region. In this study the junction flow at the interaction of a wing and a flat plate is investigated. The numerical modelling is carried out using the three-dimensional large eddy simulation (LES) approach at the Reynolds number Re = 1.15×105 based on the wing’s maximum thickness T and the free stream velocity Uref. The comparison with the experimental results shows that the numerical simulations fairly accurately reproduce the phenomenon under study. The dynamic mode decomposition (DMD) of the resolved flow field is employed to obtain the coherent dynamics of the flow. To clearly demonstrate the oscillation characteristics and the horseshoe vortex structures of junction flow the velocity field in the plane of symmetry is decomposed with eduction of two dominant DMDmodes. These two DMDmodes are reconstituted and developed, together with the mean flow mode to explain the latent dynamics. Mode 1 reveals the merging of the horseshoe vortices and mode 2 is responsible for the process of fission and stretching.

Autoignition flame dynamics in sequential combustors

This numerical study investigates the linear and non-linear flame dynamics in the second stage of a sequential combustor, with methane fuel injection into vitiated hot gas. It focuses on the heat release rate response of the sequential flame to entropy waves. The response is shown to be very sensitive to small changes in operating condition and excitation amplitude. One-dimensional (1-D) flame simulations were performed to identify transitions between three combustion regimes: autoignition, flame propagation and flame propagation assisted by autoignition. Three-dimensional (3-D) large eddy simulations (LES) were performed for two configurations: one that includes the fuel injector and the mixing section, and one with a perfectly premixed inlet. An analytically reduced chemistry (ARC) mechanism, which allows to account for autoignition chemistry, was used in combination with the dynamic thickened flame (DTF) model. For certain conditions, local autoignition events occur upstream of the flame in the advected “hot” streamwise strata that result from the inlet modulation. These auto-ignited kernels get convected downstream and impinge on the stabilized flame front leading to a sudden increase of heat release rate followed by an abrupt decrease due to flame front merging. This is identified as the driving mechanism for the onset of non-linear flame response characterised by high transfer function gains. In particular, this work shows that the gain of the flame describing function increases beyond a certain threshold of the excitation amplitude.

Dependence of square cylinder wake on Reynolds number

The wake of a square cylinder is investigated for Reynolds number Re < 107. Two-dimensional (2D) laminar simulation and three-dimensional (3D) large-eddy simulation are conducted at Re ≤ 1.0 × 103, while experiments of hotwire, particle image velocimetry, and force measurements are carried out at a higher Re range of 1.0 × 103 < Re < 4.5 × 104. Furthermore, data covering a wide Re range, from 100 to 107, in the literature are comprehensively collected for discussion and comparison purposes. The dependence on Re of the recirculation bubble size or vortex formation length, wake width, shear-layer transition, time-mean drag force, and Strouhal number is discussed in detail, revealing five flow regimes, each having distinct variations of the above parameters. With increasing Re, while the streamwise recirculation size enlarges at Re < 50 (steady flow regime), the vortex formation length reduces at 50 < Re < 1.6 × 102 (laminar flow regime), remains unchanged at 1.6 × 102 < Re < 2.2 × 102 (2D-to-3D transition flow regime), and decreases at 2.2 × 102 < Re < 1 × 103 (shear layer transition I regime), approaching asymptotically a constant at Re > 1.0 × 103 (shear layer transition II regime). Meanwhile, the wake width decreases with Re in the laminar flow regime, grows in 2D-to-3D transition and shear layer transition I regimes, and levels off in the shear layer transition II regime. The narrowest wake width is identified in the 2D-to-3D transition flow regime, corresponding to a minimum time-mean drag force and a largest Strouhal number. With increasing Re, the shear-layer transition length rapidly declines in the shear layer transition I regime where the transition occurs downstream of the trailing corner of the cylinder. On the other hand, it slowly tapers off in the shear layer transition II regime where the transition takes place upstream of the trailing corner. An extensive comparison is made between the dependence on Re of a circular cylinder wake and a square cylinder wake, with their distinct natures highlighted.

Three-dimensional large-eddy simulations of the early phase of contrail-to-cirrus transition: effects of atmospheric turbulence and radiative transfer

Three-dimensional large-eddy simulations of the early phase of contrail-to-cirrus transition: effects of atmospheric turbulence and radiative transfer

An Eulerian stochastic field cavitation model coupled to a pressure based solver

Probability density functions (PDF) and the relevant methods have been widely used to describe non-linear phenomena in the realm of turbulence modelling and CFD. In order to solve PDF transport equations, the main trend of previous studies rely on Monte Carlo method with Lagrangian particle tracking. However, as with any Lagrangian based approach, the scalability of the parallelized simulations of such method is less than satisfactory. An Eulerian stochastic field model has been presented recently by Dumond et al. [1] to simulate cavitating flows. Their model uses a fully compressible density based solver. Here we present an adapted version using an iso-thermal cavitation model adopting the homogeneous mixture assumption in a pressure based flow solver which is more relevant to engine simulations. A PDF method is used to represent a distribution of vapour volume fractions, based on which the Eulerian stochastic field (ESF) method is applied to perform a three-dimensional large eddy simulation (LES) of the cavitation phenomena inside an academic fuel injector configuration. The numerical model is based on a volume of fluids approach and coupled with a pressure based solver for the flow field, and is implemented in the framework of the open source C++ toolbox OpenFOAM. The result of the ESF simulation is compared against that from a typical single volume fraction solver for validation. Vortex structures and its correspondence to cavitation are shown, and the behaviour of the PDF at different probe locations at different times are acquired to demonstrate the potential of the ESF model in capturing both transient and stochastically steady cavitation.

The hydrodynamic forces on a circular cylinder in proximity to a wall with intermittent contact in steady current

We present a three-dimensional Large Eddy Simulations (LES) study on the hydrodynamics of a circular cylinder close to a wall. Intermittent gaps between the cylinder and wall, forming spanning and non-spanning sections along the span are considered. This geometry setting represents an idealized model of pipelines laid on an uneven seabed. As compared to that of the corresponding spanwise-uniform cylinders, the drag coefficient on spanning sections is generally smaller, while the drag on non-spanning sections is slightly larger. Large amplitudes of the sectional forces are observed, partially due to the variations of the gap flow under the spanned sections. Flow deflection from the non-spanning section to the spanning section induces changes in local force and pressure distributions. Despite these changes, the mean force on the cylinder with intermittent contact as a whole is shown to be reasonably estimated by pro rata of forces from spanwise-uniform cylinders over the parameter space considered in this study.

Large eddy simulation (LES) calculations of natural convection in cylindrical enclosures with rack and seeds for crystal growth applications

This paper presents three-dimensional (3D) large eddy simulation (LES) calculations of natural convection in laterally-heated cylindrical enclosures using ANSYS FLUENT. The problem considered here is at a Rayleigh number, (Ra) of 8.8×106, based on a characteristic length scale defined by the ratio of the reactor's volume to the lateral area. The numerical simulations are carried out for three different geometrical configurations: an empty reactor, a reactor with a rack and 32 seeds, and thirdly, a reactor with a rack and 80 seeds mounted inside. The objective of the current study is to understand the effects of the rack and seeds on the flow patterns and thermal distribution inside the reactor. In order to investigate these effects, a comprehensive analysis of velocity and temperature on the selected planes and points inside the domain were performed. Results showed that with an increase in seeds within the reactor, the flow slowed down more, the turbulence intensity reduced, and the extent of temperature inversion increased. These have important implications for the choice of ammonothermal method in terms of acidic or basic, and also for species transport and reactions, which eventually affect the deposition rates of GaN crystals in autoclaves.

A large eddy simulation of the breakup and atomization of a liquid jet into a cross turbulent flow at various spray conditions

A three-dimensional large eddy simulation (LES) is conducted to investigate the breakup and atomization of a liquid jet into a cross turbulent flow for several variants of a liquid-gas momentum flux ratio by varying the liquid injection velocity and cross flow temperature. The spray-field dynamics are treated using a combined Eulerian-Lagrangian approach in which the gas phase is discretized using a density-based, finite-volume approach. A Kelvin-Helmholtz and Rayleigh-Taylor (KH-RT) hybrid wave breakup model is implemented to simulate the liquid column and droplet breakup process. While the KH model is applied to the liquid column breakup (primary breakup), the RT model is implemented to the breakup of the small droplets (secondary breakup). The detail flow structures of the counter-rotating vortex pair (CVP) and vortex interaction behind the injector are observed. The spray penetration depth in the crossflow compares well with the experimental data and similar to empirical equations. The Sauter mean diameter (SMD) distribution is analyzed along the flow downstream representing somewhat different, and an analytical correlation model is proposed to pre-evaluate the SMD in the flowfield.

The effect of mountain wind on the falling snow deposition

The air-flow over morphologic prominence is of vital importance for the temporal and spatial variations of snow distribution, which may further affect the hydrological cycle, climate system, ecological evolution as well as other natural processes. Most previous studies have neglected the effect of non-uniform mountain wind on the trajectory of falling snow particle and thus the uneven snow distribution over complex terrain is little known. Here in this paper, we carry out a numerical study on the deposition process of mixed grain size snow particles over mountain terrains. A three-dimensional large eddy simulation (LES) code is used to predict the wind field and the fluid-solid coupling effect is considered, the Lagrangian particle tracking method is employed to track the movement of each falling snow particle. Our results suggest the boundary layer flow over steep slopes or complex terrains exhibits obvious uneven characteristics. The mountain wind has obvious effect on the motion of snow particles and finally results in a non-uniform distribution of snow. This research is of significance to understand the evolution of high resolution snow distribution in cold highland areas.

3D large eddy simulation (LES) calculations and experiments of natural convection in a laterally-heated cylindrical enclosure for crystal growth

The ammonothermal crystal growth technique is one of the most effective methods used for growing Gallium Nitride (GaN) crystals in terms of the final product quality and manufacturing efficiency. Popular applications of GaN crystals include light emitting diodes (LEDs) and high-frequency electronic devices. A laterally-heated cylindrical enclosure with a top to bottom temperature gradient is considered in this study to investigate the fluid mechanics and heat transfer in an ammonothermal crystal growth reactor. Three-dimensional (3D) large eddy simulations (LES) results of natural convection in a laterally heated cylindrical reactor, using the commercial computational fluid dynamics (CFD) software, ANSYS FLUENT, are presented in this paper for a Rayleigh number (Ra) of 8.8×106. The Ra is defined based on a length scale that is equal to the ratio of volume to the lateral area of the cylinder (R/2). In addition, an experiment of a geometrically- and dynamically-similar geometry is developed, and particle image velocimetry (PIV)-based flow visualizations are carried out for the purpose of validating the numerical model. Comparisons between experiments and numerical simulations showed that flow patterns were qualitatively similar, and Fourier transforms of velocity magnitudes at selected points in the domain matched reasonably well. An added interesting observation in the simulation was the existence of temperature inversion, which has potential implications on the choice of mineralizer (acidic/basic).

Numerical simulation of the initial destabilization of an air-blasted liquid layer

Numerical simulations of a planar air/water air-blast atomization are performed using an in-house multiphase Navier–Stokes solver which uses a semi-Lagrangian geometric volume of fluid method to track the position of the interface. This solver conserves mass exactly and mitigates momentum and kinetic energy conservation errors. Excellent agreement with recent experiments is obtained when comparing physical quantities, such as the liquid cone length, maximum wave frequency and spatial growth rate of the primary instability. The inclination of the gas inflow, which mimics the slope of the separator plate, is shown to enhance the primary atomization. A three-dimensional large-eddy simulation, run using physically correct air/water parameters, is used to provide the statistics of the flow. The gas layer is laminar close to the entrance and becomes turbulent at positions further downstream. The liquid wave crests expand in thin sheets, which break into secondary droplets, as observed in experiments.

Comparative studies of numerical methods for evaluating aerodynamic characteristics of two-dimensional airfoil at low Reynolds numbers

ABSTRACTThis study investigates the predictability of the aerodynamic performance of some numerical methods at low Reynolds numbers and their dependency on the geometric shape of airfoil. We conducted three-dimensional large-eddy simulations (3-D LES), two-dimensional laminar simulations (2-D Lam), and Reynolds-averaged Navier–Stokes simulations with Baldwin–Lomax (2-D RANS(BL)) and Spalart–Allmaras (2-D RANS(SA)) turbulence models. Although there is little discrepancy between the 3-D LES, 2-D Lam, and 2-D RANS(SA) results in terms of the lift and drag characteristics, significant differences are observed in the predictability of the separation and reattachment points. The predicted lift, separation, and reattachment points of the 2-D Lam are qualitatively similar to those of the 3-D LES, except for high angles of attack at which a massive separation occurs. The 2-D RANS(SA) shows good predictability of the lift and separation points, but it does not estimate reattachment points accurately. The 2-D RANS(BL) fails to predict the precise separation points, which results in a poor lift predictability. These characteristics appear regardless of the airfoil geometry shapes. The results suggest that a 2-D Lam without any turbulence models can be used to estimate qualitative airfoil aerodynamic characteristics at the low Reynolds numbers.

운전점에 따른 3차원 소형축류홴의 와도 특성에 대한 대규모 와 모사

The unsteady-state, incompressible and three-dimensional large-eddy simulation(LES) was carried out to evaluate the vorticity distribution of a small-size axial fan(SSAF). The X-component vorticity profiles developed around blade tips turn from axial to radial, and diminish the density of distribution according to the increase of static pressure. Otherwise, the Z-component vorticity profiles evenly develop at the region larger than the half radial distance of blade at the operating points of A and B, partly at the trailing-edge region of blade and radially over bellmouth according to the increase of static pressure.

Axisymmetric Tornado Simulations at High Reynolds Number

Abstract This study is the first in a series that investigates the effects of turbulence in the boundary layer of a tornado vortex. In this part, axisymmetric simulations with constant viscosity are used to explore the relationships between vortex structure, intensity, and unsteadiness as functions of diffusion (measured by a Reynolds number Rer) and rotation (measured by a swirl ratio Sr). A deep upper-level damping zone is used to prevent upper-level disturbances from affecting the low-level vortex. The damping zone is most effective when it overlaps with the specified convective forcing, causing a reduction to the effective convective velocity scale We. With this damping in place, the tornado-vortex boundary layer shows no sign of unsteadiness for a wide range of parameters, suggesting that turbulence in the tornado boundary layer is inherently a three-dimensional phenomenon. For high Rer, the most intense vortices have maximum mean tangential winds well in excess of We, and maximum mean vertical velocity exceeds 3 times We. In parameter space, the most intense vortices fall along a line that follows , in agreement with previous analytical predictions by Fiedler and Rotunno. These results are used to inform the design of three-dimensional large-eddy simulations in subsequent papers.