Search NASA⌕ Search

SEARCH · Search NASA

Results for “Direct numerical simulation (DNS)”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Developing And Scaling an OpenFOAM Model to Study Turbulent Flow in a HFIR Coolant Channel

Improving the understanding of how computational fluid dynamics (CFD) direct numerical simulations (DNS) of flows in the High Flux Isotope Reactor (HFIR) perform when run in parallel using the high performance computing (HPC) platform Summit at the Oak Ridge Leadership Computing Facility (OLCF) is of particular importance to boost the computational tools used to support HFIR conversion to low enriched fuel (LEU). Evaluation of scaling performance was driven by the increasing importance of graphics processing unit (GPU) usage in HPC, which is becoming the standard for modern supercomputers such as Summit. The desired results are to obtain a strong positive correlation between the computational resources dedicated to a problem and the relative speed-up of the simulation in comparison to a benchmark. This capability will allow substantially improvement in HFIR flow analytical capabilities, specifically when predicting turbulence properties at high Reynolds numbers. The study leverages previous simulation results performed with code PHASTA (finite element) on HPC platforms Cori (NERSC) and Theta (ALCF) [1] with computing options provided in the computing platform OpenFOAM (finite volume) at OLCF. Transitioning from PHASTA to OpenFOAM will (1) eliminate dependence on third-party software for mesh generation and manipulation, (2) reduce resource needs by employing modern architectures, and (3) build expertise for future modeling of HFIR-specific problems like heat transfer in involute geometry, entrance effects, flow structure in channel corners, and so on—all important issues when defining the available thermal margins in the transition to LEU. CPUs and GPUs differ significantly in their architecture and utilization, as discussed in the literature [2]. The most important differences are in the approach to computations and their memory. A single GPU contains a large quantity of cores, enabling it to perform with a much higher throughput than a CPU, but execution requires a different approach. GPU codes execute instructions using the Single-Instruction Multiple-Thread (SIMT) approach in which a single instruction is used for groups of threads called warps. A warp typically consists of 32 threads which must execute the same set of instructions, although on separate threads. Alternately, a CPU has far fewer cores that are much more flexible in their operation, excelling at quickly performing more complex serial computations. This is why GPUs have greater throughput when properly utilized. The second important difference is seen when comparing their memory spaces. Limited memory allocations and CPU–GPU communications cause a significant bottleneck in GPU-accelerated programs. Further study was required to properly take advantage of GPU resources. A comprehensive analysis of code performance and the model-specific features of turbulence constitutes the core of this work. In this study, a DNS simulation of HFIR channel turbulence was performed with the finite volume CFD code OpenFOAM v2112 and CUDA v11.0 on Red Hat Enterprise Linux v8.2. The OpenFOAM installation had AMGx integrated to enable GPU acceleration and utilizes the PETSc4FOAM library. The computational resources and the problem size were scaled on CPU and CPU + GPU architectures to gain a better understanding of the performance of a DNS problem on modern computing hardware. The study aimed to analyze the scaling of the code exclusively on CPUs and then to examine the scaling of the codes with GPU acceleration enabled. Scaling studies included CPU and GPU acceleration on a mesh of varying resolution to analyze the impact of problem size relative to computational resources. In the course of preparing the GPU configuration on Summit, mainly using the AMGX solvers, difficulties were encountered stemming from constant changes resulting from extensive ongoing development activities and the changing environment. This resulted in the inability to complete the GPU portion of the work. The code was compiled and tested, but production runs to assess acceleration were not performed because the used discretional compute time allocation expired as year-end approached. The Summit HPC platform is scheduled for decommissioning in 2024, making it unattractive for future use with Nvidia-based GPUs. Therefore, the work will be moved onto NERSC machines in FY24. An application was prepared and submitted, and sufficient node-hours were awarded to continue the research in the next calendar year. This report summarizes work performed thus far, which mostly focused on CPU OpenFOAM computing.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2024

Computational Modeling of Proppants Transport in Rock Fractures

Presentation describing the direct numerical simulations (DNS) of fluid and particulate transport through rough-walled rock fractures showing transport properties of proppants as a function of fracture properties, injection velocity, and proppant size. Results demonstrate the particulate transport is highly influenced by the ratio of proppant diameter to mean fracture aperture, and quantified in terms of areal space that proppant coverage occurs.

Rousta, Farid↗
DOE OSTI2024

Symbolic diagnostics to interpret and analyze neural network models

Embedded machine-learned models (EMLMs) have the promise to improve the predictive accuracy of engineering simulators in environments of national interest. EMLMs often comprise complex input-output maps (e.g., neural networks), which make them unamenable to rigorous analysis and generally difficult to interpret. In the face of decades of theory, this lack of interpretability is a significant barrier to building confidence in these models. This work outlines an approach to interpret EMLMs using sparse polynomial regression for comparison with theoretical understanding. To do so, we build on the concept of Locally Interpretable Model-agnostic Explanations (LIME) using physics-informed clustering, prototype selection, and library construction. While general, we demonstrate our method on tensor-basis neural networks used in Reynolds-Averaged Navier-Stokes simulations of hypersonic fluid flows. Results are presented for a simulated toy model and for direct numerical simulations (DNS) of turbulent flows over a flat plate.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Hierarchically Informed Engineering Models for Predictive Modeling of Turbulent Premixed Flame Propagation in Pre- chamber Turbulent Jet Ignition

The goal of the project is to improve the predictive accuracy and efficiency of turbulent combustion sub-models for pre-chamber turbulent jet ignition (TJI). This goal is achieved through the development of a hierarchically informed engineering model for turbulent combustion in TJI. The model development starts with the highest level of model description of turbulent combustion with direct numerical simulation (DNS) from which fundamental characteristics and scaling properties of turbulent premixed flame propagation under TJI relevant conditions are obtained.

42 ENGINEERING↗
DOE OSTI2025

Challenge Problem 1: Preliminary Results of the Direct Numerical Simulation of Transient Flows

This report presents the first direct numerical simulations (DNS) of transient mixed convection in an idealized downcomer-like channel (Challenge Problem 1, Phase II). Using the GPU-accelerated NekRS solver, we modeled a sudden decay in driving pressure, mimicking loss-of-flow events, and tracked the resulting evolution of Reynolds number, boundary-layer structure, turbulence statistics, and heat-transfer metrics. Key findings include the systematic thickening and eventual asymmetry of velocity and thermal boundary layers under buoyant deceleration; minimal “memory” lag in Reynolds shear stress and TKE profiles when sampled at matching Re, yet clear shifts of peak locations toward the cooled wall; overshoots in transient eddy-viscosity and eddy-diffusivity (and corresponding sub-unity turbulent Prandtl numbers) on the cooled side; and a pronounced transient Nusselt-number enhancement driven by wall-temperature inertia and residual eddy mixing. These effects combined to offer a temporary cooling margin above steady-state predictions during reactor LOF transients. Future work will extend this work to a more complex “Case II” geometry (90° turn + lower plenum) and generate multi-Re/Pr datasets for data-driven turbulence closures.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2020

Impact of Vibrational Nonequilibrium on the Supersonic Mode in Hypersonic Boundary Layers

The supersonic mode in hypersonic boundary layers has recently been shown to achieve a larger peak growth rate than the traditional second mode in particular circumstances relevant to free flight. In this environment, high levels of aerodynamic heating are encountered such that incorporation of chemical and vibrational nonequilibrium effects can become critical. The impact of thermal nonequilibrium (TNE) on the supersonic mode in particular has not been thoroughly and systematically investigated. This work uses thermochemical nonequilibrium (TCNE) direct numerical simulation (DNS) and TCNE linear stability theory (LST) to obtain a more complete investigation of the supersonic mode using both nonequilibrium and frozen gas models. The simulation is Mach 5 flow over a 1-mm-nose-radius axisymmetric cone 1 m in length, representative of experimental conditions in ground test facilities. Both LST and DNS results indicate that TNE effects are destabilizing compared with frozen flow and can lead to an N-factor difference of approximately +1 at the end of the cone. LST results also indicate that TNE effects are stabilizing to the second and supersonic modes in comparison with thermal equilibrium. The supersonic mode appeared to be slightly more affected by TNE than the second mode. In conclusion, collectively, the findings in this work suggest that even in ground test facilities where TCNE effects are relatively small compared with free flight, incorporating nonequilibrium models in both the mean flow and stability solvers can have a practical effect on hypersonic boundary-layer transition prediction analyses.

42 ENGINEERING↗
DOE OSTI2024

Direct Numerical Simulation Database of High-Speed Flow over Parameterized Curved Walls

This study presents a direct numerical simulation (DNS) database of high-speed turbulent boundary layers (TBLs) subject to pressure gradients due to parametrically varied backward-facing and forward-facing wall curvatures, with an inflow Mach number of 4.9 and a friction Reynolds number of [Formula: see text] immediately before the onset of wall curvature. The Mach and Reynolds numbers are significantly higher than those reported in the literature for the DNS of pressure-gradient TBLs. The flow conditions and baseline wall geometries are representative of experiments in the high-speed blowdown wind tunnel at the National Aerothermochemistry Laboratory at Texas A&M University. The wall steepness of the baseline geometry for both the backward-facing and forward-facing walls was systematically varied to cause attached, incipiently separated, and fully separated flows. Precomputed flow statistics, including turbulent kinetic energy budgets, are available on the website of the Turbulence Modeling Resource of the NASA Langley Research Center, allowing other investigators to query any property of interest.

Engineering↗
DOE OSTI2021

A priori examination of reduced chemistry models derived from canonical stirred reactors using three-dimensional direct numerical simulation datasets

Data-driven approaches to construct reduced chemical kinetic models, that rely heavily on thermo-chemical datasets with full chemical kinetics, have been gaining popularity. Datasets from direct numerical simulations (DNS) under three-dimensional (3-D) realistic turbulent flow conditions are desirable but limited to carefully designed parametric conditions due to the computational cost. Constructing datasets from a large ensemble of zero-dimensional stirred reactors like perfectly stirred reactor (PSR) and partially stirred reactor (PaSR) is a computationally efficient solution to consider the turbulence-chemistry interactions and cover a broad range of parametric conditions. In this paper, we derive reduced chemistry models from solutions of a large number of PSR and PaSR reactors using autoencoder (AE) neural networks and principal component analysis (PCA), and conduct a priori examination of the reduced models in three temporally evolving 3-D DNS jet flames featuring local extinction and re-ignition. The results show that the reduced models derived from PaSR datasets, i.e., AE-PaSR and PCA-PaSR, generally show significant improvement over the ones derived from PSR datasets. Among all the reduced models, AE-PaSR shows the best agreement with DNS results on the reconstruction accuracy and the representation of temporally evolving local extinction and re-ignition events.

Zhang, Pei↗
DOE OSTI2023

As a Matter of Dynamical Range – Scale Dependent Energy Dynamics in MHD Turbulence

Abstract Magnetized turbulence is ubiquitous in many astrophysical and terrestrial plasmas but no universal theory exists. Even the detailed energy dynamics in magnetohydrodynamic (MHD) turbulence are still not well understood. We present a suite of subsonic, super-Alfvénic, high plasma beta MHD turbulence simulations that only vary in their dynamical range, i.e., in their separation between the large-scale forcing and dissipation scales, and their dissipation mechanism (implicit large eddy simulation, ILES, and direct numerical simulation (DNS)). Using an energy transfer analysis framework we calculate the effective numerical viscosities and resistivities, and demonstrate that all ILES calculations of MHD turbulence are resolved and correspond to an equivalent visco-resistive MHD turbulence calculation. Increasing the number of grid points used in an ILES corresponds to lowering the dissipation coefficients, i.e., larger (kinetic and magnetic) Reynolds numbers for a constant forcing scale. Independently, we use this same framework to demonstrate that—contrary to hydrodynamic turbulence—the cross-scale energy fluxes are not constant in MHD turbulence. This applies both to different mediators (such as cascade processes or magnetic tension) for a given dynamical range as well as to a dependence on the dynamical range itself, which determines the physical properties of the flow. We do not observe any indication of convergence even at the highest resolution (largest Reynolds numbers) simulation at 2048 3 cells, calling into question whether an asymptotic regime in MHD turbulence exists, and, if so, what it looks like.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Cloud droplet diffusional growth in homogeneous isotropic turbulence: bin microphysics versus Lagrangian super-droplet simulations

The increase in the spectral width of an initially monodisperse population of cloud droplets in homogeneous isotropic turbulence is investigated by applying a finite-difference fluid flow model combined with either Eulerian bin microphysics or a Lagrangian particle-based scheme. The turbulence is forced applying a variant of the so-called linear forcing method that maintains the mean turbulent kinetic energy (TKE) and the TKE partitioning between velocity components. The latter is important for maintaining the quasi-steady forcing of the supersaturation fluctuations that drive the increase in the spectral width. We apply a large computational domain (64 3 m 3 ), one of the domains considered in Thomas et al. (2020). The simulations apply 1 m grid length and are in the spirit of the implicit large eddy simulation (ILES), that is, with small-scale dissipation provided by the model numerics. This is in contrast to the scaled-up direct numerical simulation (DNS) applied in Thomas et al. (2020). Two TKE intensities and three different droplet concentrations are considered. Analytic solutions derived in Sardina et al. (2015), valid for the case when the turbulence integral timescale is much larger than the droplet phase relaxation timescale, are used to guide the comparison between the two microphysics simulation techniques. The Lagrangian approach reproduces the scalings relatively well. Representing the spectral width increase in time is more challenging for the bin microphysics because appropriately high resolution in the bin space is needed. The bin width of 0.5 µm is only sufficient for the lowest droplet concentration (26 cm -3 ). For the highest droplet concentration (650 cm -3 ), an order of magnitude smaller bin size is barely sufficient. The scalings are not expected to be valid for the lowest droplet concentration and the high-TKE case, and the two microphysics schemes represent similar departures. Finally, because the fluid flow is the same for all simulations featuring either low or high TKE, one can compare point-by-point simulation results. Such a comparison shows very close temperature and water vapor point-by-point values across the computational domain and larger differences between simulated mean droplet radii and spectral width. The latter are explained by fundamental differences in the two simulation methodologies, numerical diffusion in the Eulerian bin approach and a relatively small number of Lagrangian particles that are used in the particle-based microphysics.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

JHTDB-wind: a web-accessible large-eddy simulation database of a wind farm with virtual sensor querying

This paper introduces JHTDB-wind (https://turbulence.idies.jhu.edu/datasets/windfarms, last access: 11 November 2025), a publicly accessible database containing large-eddy simulation (LES) data from wind farms. Building on the framework of the Johns Hopkins Turbulence Database (JHTDB), which hosts direct numerical simulation (DNS) and some LES datasets of canonical turbulent flows, JHTDB-wind stores the 4D space–time history of the flow and provides users the ability to access and query the data via a web-based virtual sensor interface. The initial dataset comprises LES results from a large wind farm with 10×6 turbines, modeled using a filtered actuator line method, under conventionally neutral atmospheric conditions. These data comprise 1 h (hour) of flow field data (velocity, pressure, potential temperature deviation, subgrid-scale (SGS) eddy viscosity, and turbine forces, approximately 15 TB (terabytes) and wind turbine data – including both turbine-level operational quantities and blade-level aerodynamic quantities (approximately 1.3 TB) – stored in Zarr and Parquet formats, respectively. Data retrieval is facilitated by the giverny Python package, allowing remote users to query the database in Python or MATLAB (C and Fortran support are available for flow field data). This paper details the simulation setup and demonstrates data access through examples that analyze wind farm flow structures and turbine performance. The framework is extensible to future datasets, including the JHTDB-wind diurnal cycle simulation analyzed in Xiao et al. (2025).

17 WIND ENERGY↗
NASA NTRS1988

Time-accurate simulations of a shear layer forced at a single frequency

Calculations are presented for the forced shear layer studied experimentally by Oster and Wygnanski, and Weisbrot. Two different computational approaches are examined: Direct Numerical Simulation (DNS) and Large Eddy Simulation (LES). The DNS approach solves the full three dimensional Navier-Stokes equations for a temporally evolving mixing layer, while the LES approach solves the two dimensional Navier-Stokes equations with a subgrid scale turbulence model. While the comparison between these calculations and experimental data was hampered by a lack of information on the inflow boundary conditions, the calculations are shown to qualitatively agree with several aspects of the experiment. The sensitivity of these calculations to factors such as mesh refinement and Reynolds number is illustrated.

Claus, R. W.↗
NASA NTRS1987

Angular distribution of turbulence in wave space

An alternative to the one-point closure model for turbulence, the large eddy simulation (LES), together with its more exact relative, direct numerical simulation (DNS) are discussed. These methods are beginning to serve as partial substitutes for turbulence experiments. The eddy damped quasi-normal Markovian (EDQNM) theory is reviewed. Angular distribution of the converted data was examined in relationship to EDQNM.

Coleman, G.↗
NASA NTRS1987

Premixed turbulent flame calculation

The importance of turbulent premixed flames in a variety of applications has led to a substantial amount of effort towards improving the understanding of these flames. Although these efforts have increased the understanding, many questions still remain. The use of direct numerical simulation (DNS) in solving these questions is examined.

El-Tahry, S.↗
NASA NTRS1988

Time-accurate simulations of a shear layer forced at a single frequency

This report presents calculations of the forced shear layer studied experimentally by Oster and Wygnanski (1982) and Weisbrot (1984). Two different computational approaches are examined: Direct Numerical Simulation (DNS) and Large Eddy Simulation (LES).The DNS appraoch solves the full three-dimensional, Navier-Stokes equations for a temporally evolving mixing layer, while the LES approach (as used in this report) solves the two-dimensional, Navier-Stokes equations with a subgrid scale turbulence model. While the comparison between these calculations and experimental data was hampered by a lack of information on the inflow boundary conditions, the calculations are shown to qualitatively agree with several aspects of the experiment. The sensitivity of these calculations to factors such as mesh refinement and Reynolds number is illustrated.

Claus, R. W.↗
NASA NTRS1989

The effects of particle loading on turbulence structure and modelling

The objective of the present research was to extend the Direct Numerical Simulation (DNS) approach to particle-laden turbulent flows using a simple model of particle/flow interaction. The program addressed the simplest type of flow, homogeneous, isotropic turbulence, and examined interactions between the particles and gas phase turbulence. The specific range of problems examined include those in which the particle is much smaller than the smallest length scales of the turbulence yet heavy enough to slip relative to the flow. The particle mass loading is large enough to have a significant impact on the turbulence, while the volume loading was small enough such that particle-particle interactions could be neglected. Therefore, these simulations are relevant to practical problems involving small, dense particles conveyed by turbulent gas flows at moderate loadings. A sample of the results illustrating modifications of the particle concentration field caused by the turbulence structure is presented and attenuation of turbulence by the particle cloud is also illustrated.

Squires, Kyle D.↗
NASA NTRS1990

A numerical study of the turbulent Ekman layer

The three-dimensional time-dependent turbulent flow in a neutrally stratified Ekman layer over a smooth flat surface was numerically simulated by directly solving the Navier-Stokes equations. Issues addressed using the direct numerical simulation (DNS) fields include the presence or absence of large-scale coherent structures ('longitudinal' or 'roll' vortices) in neutrally stratified Ekman-layer turbulence, the effects of the horizontal component of the angular velocity vector (i.e., latitude), and implications for models of the PBL. Experimental and DNS profiles are compared.

Coleman, G. N.↗
NASA NTRS1990

Large-eddy simulation of transitional channel flow

A large-eddy simulation (LES) of transition in plane channel flow was carried out. The LES results were compared with those of a fine direct numerical simulation (DNS), and with those of a coarse DNS that uses the same mesh as the LES, but does not use a residual stress model. While at the early stages of transition, LES and coarse DNS give the same results: the presence of the residual stress model was found to be necessary to predict accurately mean velocity and Reynolds stress profiles during the late stages of transition (after the second spike stage). The evolution of single Fourier modes is also predicted more accurately by the LES than by the DNS. As small scales are generated, the dissipative character of the residual stress starts to reproduce correctly the energy cascade. As transition progresses, the flow approaches its fully developed turbulent state, the subgrid scales tend towards equilibrium, and the model becomes more accurate.

Piomelli, Ugo↗