Large Eddy Simulation Model Research Articles

Multiscale simulation of respiratory airflow using physiologically consistent geometry and boundary conditions in OpenFOAM.

This study examines the aerodynamics of a wing-in-ground (WIG) airfoil subjected to heaving (plunging) & pitching motion, specifically examining the distinctions in flow between transitional and turbulent states, which are essential for assessing its aerodynamic efficiency. The numerical investigation employs a validated Large Eddy Simulation (LES) model to evaluate WIG aerodynamics for a symmetrical airfoil. By analyzing variables such as pitching angle, phase shift angle, and pitching-to-heaving (PTH) ratios, the study explores their effects on thrust, lift, and drag. Findings indicate that adjusting critical heaving & pitching parameters significantly improves aerodynamic performance in turbulent flow, with thrust increasing 4 to 10-fold compared to transitional flow. Greater thrust and lift were observed in proximity to the ground. Modifying the phase-shift angle, pitching angle, and PTH led to thrust improvements of 64.9%, 2.9 times, and 4 times, respectively. The enhanced aerodynamic properties near the ground are attributed to improved flow consistency and decreased separation during heaving & pitching in turbulent flow.

Abstract. Understanding and modeling the turbulent transport of surface layer fluxes are essential for numerical weather forecasting models. The presence of heterogeneous surface obstacles (buildings) that have dimensions comparable to the model vertical resolution requires further complexity and design in the planetary boundary layer (PBL) scheme. In this study, we develop a numerical method to couple a recently validated PBL scheme, TKE-ACM2, with multi-layer Building Effect Parameterization (BEP) in the Weather Research and Forecasting (WRF) model. Subsequently, the performance of TKE-ACM2+BEP is examined under idealized convective atmospheric conditions with a simplified building layout. Furthermore, its reproducibility is benchmarked with a state-of-the-art large-eddy simulation model, PALM, which explicitly resolves the building aerodynamics. The results indicate that TKE-ACM2+BEP outperforms another operational PBL scheme (Boulac) coupled with BEP by reducing bias in both the potential temperature (θ) and wind speed (u). Following this, real case simulations are conducted for a highly urbanized domain, namely the Pearl River Delta (PRD) region in China. High-resolution wind speed LiDAR observations suggest that TKE-ACM2+BEP reduces overestimation in the lower part of the boundary layer compared with the Bulk method, which lacks an urban scheme, at a LiDAR site located in a densely built environment. In addition, the surface temperature and relative humidity given by TKE-ACM2+BEP at surface stations in urbanized areas are more accurate than those given by TKE-ACM2 without BEP. However, it is revealed that BEP does not always improve the accuracy of the surface wind speed, as it can introduce excessive aerodynamic drag.

ABSTRACT High fidelity Large Eddy Simulations (LES) are performed on a non-premixed hydrogen-air annular rotating detonation combustor. The simulations utilize a 22-step hydrogen-air reaction mechanism and an existing flow solver that is optimized for high-speed reacting flows. The selected combustor, designed by the US Air Force Research Laboratory (AFRL), has been extensively studied experimentally across various configurations and operating conditions. In this study, a specific geometrical configuration is examined, with variations in the mass flow rate while maintaining a global equivalence ratio of ϕ = 1. Two mass flow rates are investigated. After initiation and an unsteady transition phase, the detonation waves settle into a stable rotational mode. The concerned cases predict either one or two co-rotating waves, with higher flow rates leading to an increased number of waves. The LES model successfully captures key characteristics of rotating detonation waves, and its predictive capability is quantitatively assessed using available experimental data. The computed time-averaged chamber pressures closely align with experimental values and patterns; the simulated detonation wave frequencies fall into similar ranges as those clocked in the AFRL hardware firings. The time-resolved LES data reveals insights into mixing and combustion phenomena within a rotating detonation combustion, ranging from pre- to post-wave. The dynamics of the propagating wave(s) and its interactions with the surrounding flow field are analyzed, the most significant of which include sharp temperature and pressure rises, volume expansion, fuel suppression and vorticity traces, at and right behind the detonation front.

The study investigates the impact of motions of floating offshore wind turbine platforms on wake evolution and overall wind farm performance, employing large-eddy simulation (LES) and dynamic wake modeling method. First, the differences between wakes of floating and bottom-fixed wind turbines under forced motion are examined. Subsequently, a systematic comparative analysis is performed for four representative floating platform configurations—Spar, Semi-submersible, Tension-Leg Platform (TLP), and Monopile (Mnpl)—to assess wake dynamics and downstream turbine responses within tandem-arranged arrays. Results indicate that platform pitch motion, by inducing periodic variations in the rotor’s relative inflow angle, significantly enhances wake unsteadiness, accelerates kinetic energy recovery, and promotes vortex breakdown. Tandem-arrange turbines simulations further reveal that platform-dependent motion characteristics substantially influence wake center displacement, velocity deficit, downstream turbine thrust, and overall power fluctuations at the wind farm scale. Among the examined configurations, the Spar platform exhibits the most pronounced wake disturbance and the largest downstream load and power oscillations, with rotor torque and thrust increasing by 10.2% and 10.6%, respectively, compared to other designs. This study elucidates the coupled mechanisms among 6-DOFs (Six Degrees Of Freedom) motions, wake evolution, and power performance, providing critical insights for optimizing floating wind farm platform design and developing advanced cooperative control strategies.

Herein, a coupled three‐dimensional large eddy simulation model and volume of fluid model is established to systematically investigate the effect of the argon injection through single‐channel and multi‐channel stopper rods, casting speed, and argon flow rate on the molten steel flow, spatial distribution of bubbles, and jet characteristics of a bifurcated submerged entry nozzle (SEN). The comparison with the water model shows that the current model can accurately predict the bubble distribution in the SEN. The multi‐channel argon blowing makes the argon distribution more uniform and generates bubbles with smaller diameters and larger quantities. The average diameter of bubbles is 16.72 mm in the single‐channel blowing, while the average diameter of bubbles is 12.03 mm in the multi‐channel blowing. The dispersion degree of argon bubbles increases with the increase of casting speed. The jet speed and backflow speed increase with the increase of the casting speed, while the jet vertical angle and the proportion of the backflow zone decrease gradually. With the increase of the argon flow rate, the fluctuation of the backflow speed at the outport will also increase. The injection of argon has a significant impact on the jet characteristics at the outport.

Abstract This study leverages the Field Experiment on Sub‐mesoscale Spatio‐Temporal Variability in Lindenberg (FESSTVaL), including comprehensive observations of the surface, atmospheric boundary layer (ABL), and clouds, to compare the performance of a numerical weather prediction (NWP) model at sub‐km resolutions with traditional large‐eddy simulations (LES). This comparison is both relevant and timely: as the typical grid spacings of both techniques are converging, so should their results if the LES is assumed to be a good virtual laboratory for the ABL and if NWP is resolving the turbulence well. The representative of NWP is the Icosahedral Nonhydrostatic (ICON) model run at horizontal resolutions ranging from 2.5 km to 78 m. The LES model MicroHH is run at resolutions from 75 m to 38 m. ICON is set up in a limited area with realistic boundary conditions and heterogeneous land surface, whereas the setup of MicroHH is doubly periodic above a homogeneous surface and flat terrain. We focus our comparison at a horizontal grid resolution of about 78 m, where the two models overlap. ICON can represent the ABL processes with high fidelity. It approaches the performance of the MicroHH‐LES in representing surface turbulent and radiation fluxes due to better‐resolved ABL dynamics, clouds, and land‐surface properties at sub‐km resolutions. The modeled turbulence shows good agreement with observations, highlighting the equal importance of resolved and subgrid turbulent mixing at sub‐km resolutions. Our modeling setup also reproduces deep convective cold pools, although with only a qualitatively realistic onset and development. In addition to the FESSTVaL observations used here, extensive datasets are available for follow‐up studies focusing on specific processes. The complete set of ICON simulations covers the intensive observation period between June 5 and July 5, 2021, and can be used in process studies as well as providing a benchmark for model development.

The growth of small perturbations in isotropic turbulence is studied using massive ensembles of direct numerical simulations. These ensembles capture the evolution of the ensemble-averaged flow field and the ensemble variance in the fully nonlinear regime of perturbation growth. Evolution equations for these two fields are constructed by applying the ensemble average operator to the Navier–Stokes equations and used to study uncertainty growth in scale and physical space. It is shown that uncertainty growth is described by a flux of energy from the ensemble-averaged flow to the ensemble variance. This flux is formally equivalent to the subgrid scale (SGS) energy fluxes of the turbulence cascade, and can be interpreted as an inverse uncertainty cascade from small to large scales. In the absence of information sources (measurements), the uncertainty cascade is unsteady and leads to the progressive filtering of the small scales in the ensemble-averaged flow, a process that represents the loss of predictability due to chaos. Similar to the kinetic energy cascade, the uncertainty cascade displays an inertial range with a constant average uncertainty flux, which is bounded from below by the average kinetic energy dissipation. Locally in space, uncertainty fluxes differ from the SGS energy fluxes at the same scale, but both have similar statistics and are significantly correlated with each other in space. This suggests that uncertainty propagation is partly connected to the energy cascade and that they share similar mechanisms. These findings open avenues to model uncertainty propagation in turbulence following an approach similar to the SGS models in large-eddy simulations. This is relevant not only to efficiently assess the reliability and accuracy of turbulence forecasts, but also to design uncertainty-robust reconstruction techniques for data assimilation or SGS modelling.

Abstract With increased urbanization, fires in the wildland urban interface (WUI) have become a severe problem worldwide. The unique features of WUI may influence the atmospheric flows in the vicinity of fire. This study utilizes the parallelized large eddy simulation model (PALM) system for fire‐atmosphere simulations of Bottle Lake Forest, Christchurch, New Zealand. Over 3,000 residential buildings are situated around the 7 forest, with many homes only 50 m away from the forest edge. We conducted high‐fidelity fire‐atmosphere simulations with the finest grid spacing of 4 m. Wildland forest (WF) and flat terrain simulations were conducted to provide a reference for comparison with WUI simulations. Fire‐weather conditions for the 2022/2023 New Zealand fire season were selected based on the Fire Weather Index (FWI). Data from previous fire field campaigns were obtained to represent a low‐intensity fire heat forcing. The results reveal a pulsing behavior in downwind heat transport when the forest canopy is included. Furthermore, the presence of the WUI is associated with extended downwind fire heat transport compared to WF and flat terrain scenarios. This study is the first to simulate atmospheric flows near fires in a WUI setting with such high fidelity. The findings highlight the critical role of WUI features in shaping fire‐atmosphere dynamics, though further research is required to disentangle the contributions of individual WUI components to these effects.

This study presents the implementation and validation of C++ library modules within the OpenFOAM Technology for high-fidelity simulations of multi-regime combustion in aero-engines. The dynamic thickened flame combustion model for large eddy simulations is integrated with ordinary differential equation solvers to enable fast direct integration of finite-rate chemistry on graphics processing units. The reacting solvers are extended to incorporate a mixture-averaged formulation for viscosity and thermal conductivity. Additionally, diffusion models are generalized to multicomponent mixtures to account for individual species contributions, along with corrections that support non-unity Lewis numbers in the diffusion terms of the convection-diffusion equations for chemical species. The methodology is validated against experimental data from the hydrogen/air flame produced by the hydrogen low NOx burner. The numerical results demonstrate the robustness, accuracy, and computational efficiency of the proposed approach, paving the way for future investigations into the complex physics of aero-engine installations powered by sustainable aviation fuels.

Abstract The sea-breeze front (SBF) intruding into coastal cities can lead to sudden changes in local weather and air quality. Its structures and disturbances over the urban surface with dense tall buildings have been a challenge for numerical modeling. In this study, we present the fine-scale structures of an actual SBF simulated by mesoscale-to–large-eddy simulation (LES) models at 3-m resolution in a 15-km domain of Sendai City, Japan. The head of SBF is shown to develop remarkable three-dimensional structures in the downtown area, where it interacts strongly with the microscale turbulent flows induced by buildings. Localized strong updrafts form at the lees of high-rise buildings within a few minutes just after the passing of the SBF, as ejections emerge in urban warm air mass surrounded by the cool air mass of the sea breeze. Downdrafts in the sea breeze are seen at the windward sides of high-rise buildings and help to increase wind speed in adjacent streamwise streets. After the passage of SBF, near-surface wind speeds exhibit high-frequency fluctuations with a period of 2–3 min, in good agreement with in situ observations in the downtown area. These results demonstrate the good capabilities of the building-resolving LES of both mesoscale weather conditions and microscale turbulent flows over a large city for the prediction of urban weather and environment in the streets. Significance Statement The sea-breeze front (SBF) has significant impacts on the local weather in coastal cities around the world. It is challenging to simulate the fine-scale structures of SBF and their interactions with turbulent flows over the cities. In this study, we present an improved building-resolving numerical simulation of an actual SBF that develops fine-scale structures in the downtown area with high-rise buildings. The SBF head is shown to be strongly disturbed by buildings and interacts with urban turbulent flows, which drive local strong updrafts and vertical transports of heat. The passing SBF also leads to high-frequency fluctuations in the simulated surface wind speed, in good agreement with observations. The results are helpful for improving our understanding and prediction of the local winds at street scales during the passing of the front.

Abstract Subkilometer-grid numerical simulations are highly desired for many weather-sensitive applications. A U-Net-based least squares generative adversarial network (U-LSGAN) model that integrates generative adversarial networks with least-squared loss in a U-Net framework is developed to downscale 1 km × 1 km mesoscale model outputs to 200 m × 200 m large-eddy simulation (LES) grids for Lushan and surrounding regions in eastern China, an area characterized by steep mountains and complex land cover. Three key meteorological variables—2-m temperature (T2), 2-m relative humidity (RH2), and 10-m wind speed (WSPD10)—are downscaled, and the impacts of auxiliary inputs including terrain height, land uses, accumulated precipitation, and surface pressure are evaluated. The model uses mean absolute error (MAE), smooth L1 (smooth MAE), and reduced-weight adversarial losses to optimize the downscaling process between coarse-resolution inputs and fine-scale targets, enabling accurate reproduction of subkilometer weather patterns. Results based on 2522 test samples show that U-LSGAN can effectively capture the magnitude and fine-scale structures of all three variables with power spectra closely matching the LES model outputs. In comparison to bilinear interpolation, U-LSGAN achieves dramatic improvements, with MAEs of RH2, T2, and WSPD10 reduced by 36.05%, 43.16%, and 52.02%, respectively. Significance Statement Many industrial applications, such as air traffic control, wind turbine siting, and wind power forecasting, as well as hazardous gas or aerosol accident response, and cloud-seeding weather modification operations require weather analyses and forecasts at very fine scale (grids at intervals of tens to hundreds of meters). Although such refined weather modeling can be done with large-eddy simulation (LES) models, it is typically very computing costly and cannot provide timely service to meet urgent decision needs. Based on 2522 test samples, it is demonstrated that the deep learning model developed in this study can accurately and efficiently downscale 1-km grid surface mesoscale weather fields to 200-m grid LES outputs within seconds and it works well for both flat and complex terrain regions.

The present study examines turbulent flow around a three-dimensional tripile submerged foundation experimentally, using laser Doppler velocimetry, and numerically, using Reynolds-averaged Navier–Stokes (RANS) and large eddy simulation (LES) techniques. The study is conducted at a Reynolds number of 104 with a tripile spacing ratio of 3. Flow measurements are compared to assess the predictive capabilities of the selected turbulence models. All RANS models succeeded in predicting primary mean flow phenomena, including flow detachment, vortex recirculation, and downstream reattachment. The LES model performed adequately well both near-wake and far-wake regions. Within the near wake region, the standard k−ϵ model exhibited the largest deviation from experimental data, although it performed appropriately well in the far-wake region. The k−ω shear stress transport model overpredicted the wake recovery. The observed discrepancies are likely due to limitations in modeling flows with large pressure gradients. Also, detailed structural analysis was conducted using the instantaneous flow data obtained from the LES simulations. Key flow features such as the horseshoe vortex, arch vortex, and a dipole structure composed of counter-rotating vortices are identified, exhibiting qualitative agreement with previous high-Reynolds number studies on an isolated cylinder. Instantaneous flow visualization revealed an antler-shaped vortex structure in the downstream wake, which resulted from the interaction of streamwise and spanwise vortices. Time-averaged surface streamlines were used to identify saddle points, attachment nodes, and swirl patterns on the tripiles. Notably, visualization at the free end of the downstream cylinder showed inward-shifted foci and a crescent-shaped recirculation region.

Abstract. High-resolution data on reactive nitrogen deposition are needed to inform cost-effective policies. Large eddy simulation models coupled to a dry deposition module present a valuable tool for obtaining these high-resolution data. In this paper we describe the implementation of a dry deposition module, which is an extension of DEPAC v3.11 with co-deposition (DEPAC v3.11_ext; hereafter simply referred to as DEPAC), in a large eddy simulation code (DALES v4.4) and its first application in a real-world case study. With this coupled model, we are able to represent the turbulent surface–atmosphere exchange of passive and reactive tracers at the hectometer resolution. A land surface module was implemented to solve the surface energy budget and provide detailed information for the calculation of deposition fluxes per land use (LU) class. Both the land surface model and the dry deposition module are extensively described, as are the inputs that are needed to run them. To show the advantages of this new modeling approach, we present a case study for the city of Eindhoven in the Netherlands, focusing on the emission, dispersion, and deposition of NOx and NH3. We find that DALES is able to reproduce the main features of the boundary layer development and the diurnal cycle of local meteorology well, with the exception of the evening transition. DALES calculates the dispersion and deposition of NOx and NH3 in great spatial detail, clearly showing the influence of local LU patterns on small-scale transport, removal efficiencies, and mixing characteristics.

Abstract This study investigates the roles of clear air entrainment and shallow cloud ventilation, alongside rainforest ‐assimilation, in the turbulent exchange of within the lower tropical troposphere under clear‐to‐cloudy conditions. Constrained by comprehensive observations from the CloudRoots‐Amazon22 campaign, spanning leaf stomatal to upper atmosphere, we design and evaluate a representative shallow convective numerical experiment with the turbulence‐resolving Dutch Atmospheric Large Eddy Simulation model, incorporating a bulk rainforest representation. We assess contributions from the rainforest, clouds, and environment through the vertically integrated, domain‐averaged budget by comparing simulations with and without the dynamic effects of clouds. Our findings reveal three distinct diurnal regimes named: entrainment‐diluting, cloud‐ventilation‐and‐entrainment, and ‐assimilation. Shallow convective clouds (23%), clear air entrainment (21%), and rainforest ‐assimilation (56%) collectively influence the diurnal evolution and vertical exchange of within the clear‐to‐cloudy boundary layer, with their relative importance varying per diurnal regime. In the absence of clouds, ventilation ceases and exchange is driven solely by entrainment and ‐assimilation, resulting in a 20%–25% reduction in mixing effectiveness. In the vertical, shallow clouds ventilate to heights reaching twice the boundary layer depth, significantly affecting the vertical distribution until late afternoon. Analysis of the correlation between and O shows that shallow convective clouds organize the turbulent exchange at shallow cloud‐scales, shaping a vertical pattern of negative to positive ‐O correlation from the roughness sublayer into the cloud layer. These findings highlight key processes crucial for accurately representing the lower tropical tropospheric budget across clear‐to‐cloudy conditions.

Understanding wave transformation and its interaction with coastal structures is critical for shoreline protection and design. While physical modeling has traditionally supported such studies, its high cost has led to increased reliance on numerical modeling. This study uses FLOW-3D to simulate wave propagation in a scaled 2D coastal wave channel and compares the performance of three turbulence models: Laminar, Renormalized Group (RNG) k-ε, and Large Eddy Simulation (LES). Simulations are based on a 3-meter wave height, 12-second period, and 7-meter water depth, with wave elevation recorded at six probes along the domain. Results show that the LES model achieved the most accurate prediction, with a significant wave height of 3.01 meters at the structure location—an error of only 0.33%—outperforming RNG and laminar models. These findings highlight the superior turbulence resolution of LES in capturing energy dissipation and wave evolution. The study provides practical guidance for coastal engineers in selecting turbulence models based on accuracy and computational trade-offs. Future research should include model validation with experimental or field data and extend to irregular wave conditions to enhance real-world applicability.

ABSTRACT We developed a framework for rapid monitoring of radioactive plumes in the vicinity of nuclear facilities based on a quick and practical high-resolution atmospheric dispersion simulation method that combines a large-eddy simulation (LES) model pre-simulation database (pre-sim DB) of wind conditions and onsite meteorological observation results, as proposed by the previous study. However, this framework was not quantitatively demonstrated using measurement data. In this study, we evaluated the performance of the wind condition reproduction and plume dispersion analysis methods. Air dose rates observed at monitoring posts around the stack were compared with the values reproduced by the method using the pre-sim DB, and the reproducibility of both air dose rate and flow field was discussed. The pre-sim DB-based method successfully captured the temporal variation of air dose rates at the monitoring posts, though it tended to overestimate the peak values. Particularly when the vertical wind shear was pronounced, the method using the pre-sim DB could cause significant errors. This is likely because the method relies on wind conditions from a single observation point, which inherently limits its ability to represent vertical wind shear within the pre-sim DB. Despite these limitations, particularly in reproducing complex wind fields, the method utilizing the pre-sim DB offers a valuable and practical tool for rapid dose rate simulation due to its lower computational cost compared to unsteady simulations using an LES model.

ABSTRACT Previously, second-order moment (SOM) combustion model has been proposed and used for large-eddy simulation (LES) of turbulent flames. Although it gives good results, its accuracy remains to be further improved. Therefore, modified algebraic second-order moment (MASOM) and direct-moment closure (DMC) models were developed. The simulation results were compared with experiments and direct numerical simulation (DNS) results and also with those obtained by existing models, indicating their advantage over the original SOM and some other combustion models. This paper gives a description of MASOM and DMC combustion models, their application and validation in LES of different flames. The instantaneous results show flame structures.

Abstract This study compares two advanced numerical approaches in the context of investigating flame stabilization in hydrogen jets in crossflow (JICF). The first approach utilizes a highly resolved large-eddy simulation (LES) coupled with a strain-based flamelet manifold combustion model. The second approach involves a high-fidelity direct numerical simulation (DNS) with detailed chemical kinetics for the same JICF case. To capture the complex flame dynamics, the LES solver uses a static, highly resolved mesh, whereas the DNS solver employs adaptive mesh refinement (AMR) based on flame thickness. Both simulations are validated against experimental data, with a specific focus on flow regimes characterized by the Reynolds number, particularly mixing effects on flame stabilization. The flame dynamics are further analyzed through hydroxyl radical mass fraction contours and spatiotemporal scatter plots of key quantities. The study reveals that turbulent mixing regions are captured well by the LES model. However, on the lee side of the flame, differential diffusion significantly affects flame stabilization, which occurs further downstream in the DNS simulations. This comparison presents a holistic perspective on the capabilities of these methods in terms of spatial and chemical basis, highlighting their potential to study the dynamics of practical combustion systems under varying conditions.

Abstract This is part II of a companion paper which introduced the implementation of a Wall-Modeled Large-Eddy Simulation (WMLES) model combined with the Immersed Boundary Method (IBM) to analyze transonic flow in a gas turbine nozzle guide vane. In particular, Part I focused on a fully transonic configuration, validating the model against experimental data, identifying the most cost-effective model in terms of accuracy and computational effort, and demonstrating the benefits of scale-resolved approaches by characterizing the primary scales of wake motion. Part II expands on this by investigating the effects of flow compressibility and comparing high subsonic and fully transonic cases within the same environment. In particular, after initial verification of the numerical model robustness, the instantaneous, near-wall, and averaged flow dynamics are investigated as a function of the cascade expansion ratio. Steady Reynolds-Averaged Navier-Stokes solutions are also compared with current WMLES, showing the latter with a superior ability to capture transitional behaviors of the boundary layers, turbulent kinetic energy production/convection and dissipation, and predicting a more consistent behavior of wall thermal properties. Such initial stages of the analysis pave the way for characterizing the vane's momentum and thermal losses. Consequently, a novel thermal loss coefficient is proposed, accounting for the localized cooling effects. Finally, Lagrangian statistics within the scale-resolved framework are presented, underscoring the role of compressibility in the wake turbulent behavior and primary frequencies of the system.