SEARCH · Engineering Papers

Results for “Statistical simulation”

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.

SourceResearch connectionFrom yearThrough year

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

At least 1,243 records · Page 69

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Uncertainty of 21st Century western U.S. snowfall loss derived from regional climate model large ensemble

Abstract The western United States is dependent on winter snowfall over its major mountain ranges, which gradually melts each year, serving as a natural reservoir for water resources. In a future warmer climate, much of this snowfall could be replaced by rain, making it more challenging to capture and store water. In this study, we utilize an ensemble of dynamically downscaled simulations forced by 14 global climate models (GCMs). These GCMs project wildly different futures, in terms of both temperature and precipitation change, producing significant uncertainty in snowfall projections. Here we exploit the robust statistics of the downscaled ensemble, and diagose the sensitivity of end-of-century snowfall loss across the region to both warming and regional wetting/drying in the driving GCM. The windward slopes of the Sierra Nevada and Cascades are particularly sensitive to warming (losing ~ 15% annual snowfall per degree warming), with little influence of precipitation. By contrast, snowfall loss in the inter-mountain west is less sensitive to warming (~ 5% K −1 ), but is significantly offset/exacerbated by precipitation changes (~ 0.5% snow per 1% precipitation). Combining such sensitivities with the warming and regional precipitation signals in the full CMIP6 ensemble, we can fully quantify likely snowfall loss and its uncertainty at any location, for any emissions scenario. We find that the western U.S. as a whole will lose 34 ± 8% of its total volumetric snowfall by end-of-century under the high-emissions SSP3-7.0 scenario, but 25 ± 6% and 17 ± 6% under the lower-emissions SSP2-4.5 and SSP1-2.6 scenarios.

Norris, Jesse (ORCID:0000000289883326)↗

DOE OSTI2022

Internal conversion and intersystem crossing dynamics based on coupled potential energy surfaces with full geometry-dependent spin–orbit and derivative couplings. Nonadiabatic photodissociation dynamics of NH 3 (A) leading to the NH(X 3 Σ – , a 1 Δ) + H 2 channel

Here we simulate the photodissociation of NH 3 originating from its first excited singlet state S 1 into the NH 2 + H (radical) and NH + H 2 (molecular) channels. The states considered are the ground singlet state S 0 , the first excited singlet state S 1 and the lowest-lying triplet state T 1 , which permit for the first time a uniform treatment of the internal conversion and intersystem crossing. The simulations are based on a diabatic potential energy matrix (DPEM) of S 0 , S 1 coupled by a conical intersection seam, as well as a potential energy surface (PES) for T 1 coupled by spin-orbit coupling (SOC) to the two singlet states. The DPEM and PES are fitted to ab initio electronic structure data (ESD) including energies, energy gradients, and derivative couplings. The DPEM also defines an adiabatic to diabatic state (AtD) transformation, which is used to transform the singular adiabatic SOC into a smooth function of the nuclear coordinates in the diabatic representation, allowing the diabatic SOC to be fit to an analytical functional form. ESD and SOC data obtained from these surfaces can serve as input for either quantum or semi-classical characterization of the nonadiabatic dynamics. Using the SHARC suite of programs, nonadiabatic simulations based on over 40 000 semi-classical trajectories assess the convergence of our results. The production of NH + H 2 is not direct, but is only achieved through a quasi-statistical dissociation mechanism after internal conversion to the ground electronic state. This leads to a much lower yield comparing with the main NH 2 + H channel. The NH(X 3 Σ_) radical produced through the intersystem crossing from S 0 to T 1 is rare (~0.2%) compared to NH(a 1 Δ) due to the process being spin forbidden.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2020

Large scale structure reconstruction with short-wavelength modes

Large scale density modes are difficult to measure because they are sensitive to systematic observational errors in galaxy surveys, but we can study them indirectly by observing their impact on small scale perturbations. Cosmological perturbation theory predicts that second-order density inhomogeneities are a convolution of a short- and a long-wavelength mode. This arises physically because small scale structures grow at different rates depending on the large scale environment in which they reside. This induces an off-diagonal term in the two-point statistics in Fourier space that we use as the basis for a quadratic estimator for the large scale field. Here, we demonstrate that this quadratic estimator works well on an N-body simulation of size (2.5 h -1 Gpc) 3 . In particular, the quadratic estimator successfully reconstructs the long-wavelength modes using only small-scale information. This opens up novel opportunities to study structure on the largest observable scales.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2025

Neutrino-nucleus cross section impacts on neutrino oscillation measurements

The challenges in neutrino-nucleus cross section modeling and its impact on neutrino oscillation experiments are widely recognized. However, a comprehensive and theoretically robust estimation of cross section uncertainties has been lacking, and few studies have quantitatively examined their impact on oscillation measurements. In this work, we evaluate the effect of cross section uncertainties on oscillation parameters using setups inspired by NOvA and DUNE. To characterize these uncertainties, we adopt multiple neutrino-nucleus event generators and simulate a realistic experimental procedure that incorporates near-detector data and near-to-far-detector extrapolation. Our results confirm that cross section uncertainties cannot significantly bias oscillation results in current statistics-dominated experiments like NOvA. However, they could lead to substantial bias for future systematics-dominated experiments like DUNE, even when near-detector data are employed to mitigate uncertainties. These findings underscore the need for further studies on the quantitative impacts of cross section modeling, improved strategies to utilize near-detector data and the PRISM concept, and more robust cross section models to optimize the success of future experiments.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2022

Statistical perspective on embrittling potency for intergranular fracture

Embrittling potency is a thermodynamic metric that assesses the influence of solute segregation to a grain boundary (GB) on intergranular fracture. Historically, authors of studies have reported embrittling potency as a single scalar value, assuming a single segregation site of importance at a GB and a particular cleavage plane. However, the topography of intergranular fracture surfaces is not generally known a priori. Accordingly, we, in this study, present a statistical ensemble approach to compute embrittling potency, where many free surface (FS) permutations are systematically considered to model fracture of a GB. The result is a statistical description of the thermodynamics of GB embrittlement. As a specific example, embrittling potency distributions are presented for Cr segregation to sites at two Ni $\langle 111 \rangle$ symmetric tilt GBs using atomistic simulations. We show that the average embrittling potency for a particular GB site, considering an ensemble of FS permutations, is not equal to the embrittling potency computed using the lowest energy pair of FSs. A mean GB embrittlement is proposed, considering both the likelihood of formation of a particular FS and the probability of solute occupancy at each GB site, to compare the relative embrittling behavior of two distinct GBs.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Adaptive Uncertainty Quantification for Stochastic Hyperbolic Conservation Laws

Here, we propose a predictor-corrector adaptive method for the study of hyperbolic partial differential equations (PDEs) under uncertainty. Constructed around the framework of stochastic finite volume (SFV) methods, our approach circumvents sampling schemes or simulation ensembles while also preserving fundamental properties, in particular hyperbolicity of the resulting systems and conservation of the discrete solutions. Furthermore, we augment the existing SFV theory with a priori convergence results for statistical quantities, in particular push-forward densities, which we demonstrate through numerical experiments. By linking refinement indicators to regions of the physical and stochastic spaces, we drive anisotropic refinements of the discretizations, introducing new degrees of freedom where deemed profitable. To illustrate our proposed method, we consider a series of numerical examples for nonlinear hyperbolic PDEs based on Burgers’ and Euler’s equations.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Time-Based CAN IDS Paper Results Code

Modern vehicles are complex cyber-physical systems made of hundreds of electronic control units (ECUs) that communicate over controller area networks (CANs). This inherited complexity has expanded the CAN attack surface which is vulnerable to message injection attacks. These injections change the overall timing characteristics of messages on the bus, and thus, to detect these malicious messages, time-based intrusion detection systems (IDSs) have been proposed. However, time-based IDSs are usually trained and tested on low-fidelity datasets with unrealistic, labeled attacks. This makes difficult the task of evaluating, comparing, and validating IDSs. Here we detail and benchmark four time-based IDSs against the newly published ROAD dataset, the first open CAN IDS dataset with real (non-simulated) stealthy attacks with physically verified effects. We found that methods that perform hypothesis testing by explicitly estimating message timing distributions have lower performance than methods that seek anomalies in a distribution related statistic. In particular, these “distribution-agnostic” based methods outperform “distribution-based” methods by at least 55% in area under the precision-recall curve (AUC-PR). Our results expand the body of knowledge of CAN time-based IDSs by providing details of these methods and reporting their results when tested on datasets with real advanced attacks. Finally, we develop an after-market plug-in detector using lightweight hardware, which can be used to deploy the best performing IDS method on nearly any vehicle.

Moriano, Pablo [Oak Ridge National Lab. (ORNL), Oa↗

DOE OSTI2023

Near-Cloud Atmospheric Ingredients for Deep Convection Initiation

A lack of routine environmental observations located near deepening cumulus congestus clouds limits verification of important theorized and simulated updraft–environment interaction processes occurring during deep convection initiation (CI). We analyze radiosonde profiles collected during several hundred CI events near a mountain range in central Argentina during the CACTI field campaign. Statistical analyses illustrate environmental conditions supporting radar-observed CI outcomes that span a spectrum of convective cell depths, widths, and durations, as well as events lacking precipitating convection. Tested environmental factors include a large variety of sounding-derived measurements of CAPE, CIN, moisture, terrain-relative winds, vertical shear, and lifted parcel properties, with supplemental model reanalysis of background larger-scale vertical motion. CAPE and CIN metrics do not consistently differentiate CI success from failure. Only a few environmental factors contain consistent monotonic relationships among the spectrum of cloud depths achieved during CI: (i) the depth and strength of background ascent, and (ii) the component of low-level flow oriented parallel to the ridgeline. These metrics suggest that the ability of the surrounding flow to lift parcels to their LFC and terrain-modified flow are consistently relevant processes for CI. Low- to midlevel relative humidity strongly discriminated between CI and non-CI events, likely reflecting entrainment-driven dilution processes. However, we could not confidently conclude that relative humidity similarly discriminated robust from marginal CI events. Circumstantial evidence was found linking cell width, an important cloud property governing the probability of CI, to LCL height, boundary layer depth, depth and magnitude of the CIN layer, and ambient wind shear.

Storm environments↗

DOE OSTI2020

Quantifying Uncertainties of Microscopic Nuclear Theories

The goal of this project was to produce modernized nuclear data libraries of light ion fusion reactions and actinide fission product yields. Such libraries are essential input in both stockpile science but also fundamental science (nucleosynthesis, superheavy nuclei). When such libraries are used to model complex processes involving exotic, very short-lived nuclei that have not been measured experimentally, the underlying nuclear data come from theoretical models, the predictive power of which needs to be quantified, together with its uncertainties. By teaming up nuclear theorists and statisticians, we built comprehensive theoretical frameworks based on the most fundamental theories of fusion and fission that we coupled for the first time with a Bayesian framework with machine learning techniques for uncertainty quantification. Our results include: (1) statistical emulators for neutron-alpha scattering, and the fission product mass yields of neutron-induced fission of 239 Pu, which will enable developers to generate random realizations of the corresponding data for uncertainty quantification (UQ) in applications; (2) the first complete and most accurate ab initio prediction of the S-factor for proton capture on 7 Be, which – combined with UQ – may become the recommended value at solar energies in the next review of solar fusion cross sections; (3) statistical emulators for fission mass yields in the neutron-induced fission of 239 Pu; (4) the first quantification of how uncertainties in nuclear forces impact fission properties that are relevant in nucleosynthesis simulations to explain how heavy elements are formed in the Cosmos.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2024

Conceptual Design of Neutron Sensor Qualification Device

The Advanced Sensors and Instrumentation Program at Idaho National Laboratory has been formulating strategies to qualify sensors for use in nuclear environments, particularly in irradiation experiments and advanced reactors. When qualifying neutron sensors for use in high-temperature environments, the wide range of neutron flux levels and representative energy spectra presents significant challenges. This paper discusses the development of the Neutron Sensor Qualification Device (NQD), which is designed to test neutron sensors in high temperature controlled environments with known neutron spectra, addressing the spatial and spectral complexities of neutron fluxes in reactor cores. The proposed NQD will be situated in the exposure room at the Armed Forces Radiobiology Research Institute, thus affording a unique capability to expose sensors to high neutron and gamma fluxes. To achieve thermal control, the device will utilize a radiation-hardened tube furnace, accommodating multiple sensors and neutron activation dosimetry wires. Titanium, iron, and cobalt dosimeter wires are chosen from the American Society for Testing and Materials and International Reactor Dosimetry and Fusion File libraries as references for providing energy-dependent fluence measurements. The design ensures precise sensor positioning to minimize mutual shielding and flux perturbation, which are evaluated via Monte Carlo N particle Transport Code (MCNP) simulations. These simulations have informed the development of guidelines on sensor placement within the NQD. The NQD is essential to the qualification of neutron sensors for advanced reactor technologies. It enables controlled testing of a statistically significant number of sensors, thereby supporting assessments of sensor performance across various neutron flux levels and temperatures. This paper highlights the detailed planning for the NQD prototype, along with its inaugural irradiation (scheduled for fiscal year [FY] 2025). The results from this initial testing will be fundamental in evaluating the device’s performance and establishing measurement uncertainty for in-pile neutron sensor measurements.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2025

Evolution of Magnetohydrodynamic Turbulence in the Expanding Solar Wind: Residual Energy and Intermittency

We conduct 3D magnetohydrodynamic simulations of decaying turbulence in the context of the solar wind. To account for the spherical expansion of the solar wind, we implement the expanding box model. The initial turbulence comprises uncorrelated counterpropagating Alfvén waves and exhibits an isotropic power spectrum. Our findings reveal the consistent generation of negative residual energy whenever nonlinear interactions are present, independent of the normalized cross helicity σc and compressibility. The spherical expansion facilitates this process. The resulting residual energy is primarily distributed in the perpendicular direction, with S 2 (b) − S 2 (u) ∝ l ⊥ or equivalently –E r ∝ k$^{–2}_{⟂}$. Here S 2 (b) and S 2 (u) are second-order structure functions of magnetic field and velocity respectively. In most runs, S 2 (b) develops a scaling relation S 2 (b) ∝ $l$$^{1/2}_{⊥}$ (E b ∝ k$^{–3/2}_{⟂}$). In contrast, S 2 (u) is consistently shallower than S 2 (b), which aligns with in situ observations of the solar wind. We observe that the higher-order statistics of the turbulence, which act as a proxy for intermittency, depend on the initial σc and are strongly affected by the expansion effect. Generally, the intermittency is more pronounced when the expansion effect is present. Finally, we find that in our simulations, although the negative residual energy and intermittency grow simultaneously as the turbulence evolves, the causal relation between them seems to be weak, possibly because they are generated on different scales.

Interplanetary turbulence↗

DOE OSTI2021

Investigation of high-pressure turbulent jets using direct numerical simulation

Direct numerical simulations of free round jets at a Reynolds number (Re D ) of 5000, based on jet diameter (D) and jet-exit bulk velocity (U e ), are performed to study jet turbulence characteristics at supercritical pressures. The jet consists of nitrogen (N 2 ) that is injected into N 2 at the same temperature. To understand turbulent mixing, a passive scalar is transported with the flow at unity Schmidt number. Two sets of inflow conditions that model jets issuing from either a smooth contraction nozzle (laminar inflow) or a long pipe nozzle (turbulent inflow) are considered. By changing one parameter at a time, the simulations examine the jet-flow sensitivity to the thermodynamic condition (characterized in terms of the compressibility factor ($Z$) and the normalized isothermal compressibility), inflow condition and ambient pressure (p ∞ ) spanning perfect- to real-gas conditions. The inflow affects flow statistics in the near field (containing the potential core closure and the transition region) as well as further downstream (containing fully developed flow with self-similar statistics) at both atmospheric and supercritical p ∞ . The sensitivity to inflow is larger in the transition region, where the laminar-inflow jets exhibit dominant coherent structures that produce higher mean strain rates and higher turbulent kinetic energy than in turbulent-inflow jets. Decreasing $Z$ at a fixed supercritical p ∞ enhances pressure and density fluctuations (non-dimensionalized by local mean pressure and density, respectively), but the effect on velocity fluctuations depends also on the local flow dynamics. When $Z$ is reduced, large mean strain rates in the transition region of laminar-inflow jets significantly enhance velocity fluctuations (non-dimensionalized by local mean velocity) and scalar mixing, whereas the effects are minimal in jets from turbulent inflow.

42 ENGINEERING↗

DOE OSTI2026

Stochastic fracture generation and thermo-hydro-mechanical modeling in an equivalent continuum framework for enhanced geothermal systems

Enhanced geothermal systems (EGS) involve fracturing low permeability material to establish well connectivity and then injecting and circulating fluid into the fractured subsurface for geothermal power production. Changes in fracture aperture from contraction of the cooling matrix rock may alter network connectivity and risk thermal short-circuiting. Thermo-hydro-mechanical (THM) models are a useful tool to study these processes. However, as fracture networks are complex, and data may be limited, fracture networks in THM models are often stochastically generated. Given reliance on stochastic fracture networks and THM modeling to represent the subsurface and assess productivity of EGS, increased understanding of the influence of such statistically derived fracture networks on flow and heat transport in THM models is needed. Here, a new fracture process model is developed in the reactive transport code PFLOTRAN to stochastically generate fracture families and simulate changes in fracture aperture over time due to temperature changes of the rock matrix. Sixty-four different fracture networks ranging from well to poorly-connected, are modeled in PFLOTRAN with and without mechanical processes (THM vs TH). Results indicate that for well-connected fracture networks, thermal short-circuiting is less of a concern due to the abundance of available alternative flowpaths. For poorly-connected fracture networks, inclusion of mechanical processes showed steep thermal drawdown coincident with increase in fracture aperture along developing colder flowpaths, demonstrating the risk of thermal short-circuiting. Simulations with additional, larger fractures engineered to establish connectivity in a poorly-fractured subsurface, indicate that while stochastic variation of fracture orientation of the background network had limited influence, such variation in the engineered fractures significantly affected flow and heat transport.

Discrete fracture networks (DFN)↗

DOE OSTI2025

Measurements of Pion and Muon Nuclear Capture at Rest on Argon in the LArIAT Experiment

We report the measurement of the final-state products of negative pion and muon nuclear capture at rest on argon by the LArIAT experiment at the Fermilab Test Beam Facility. We measure a population of isolated MeV-scale energy depositions, or blips, in 296 LArIAT events containing tracks from stopping low-momentum pions and muons. The average numbers of visible blips are measured to be 0.74 ± 0.19 and 1.86 ± 0.17 near muon and pion track endpoints, respectively. The 3.6⁢𝜎 statistically significant difference in blip content between muons and pions provides the first demonstration of a new method of pion-muon discrimination in neutrino liquid argon time projection chamber experiments. LArIAT Monte Carlo simulations predict substantially higher average blip counts for negative muon (1.22 ± 0.08) and pion (2.34 ± 0.09) nuclear captures. We attribute this difference to geant4’s inaccurate simulation of the nuclear capture process.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2022

SNIa Cosmology Analysis Results from Simulated LSST Images: From Difference Imaging to Constraints on Dark Energy

Abstract The Vera Rubin Observatory Legacy Survey of Space and Time (LSST) is expected to process ∼10 6 transient detections per night. For precision measurements of cosmological parameters and rates, it is critical to understand the detection efficiency, magnitude limits, artifact contamination levels, and biases in the selection and photometry. Here we rigorously test the LSST Difference Image Analysis (DIA) pipeline using simulated images from the Rubin Observatory LSST Dark Energy Science Collaboration Data Challenge (DC2) simulation for the Wide-Fast-Deep survey area. DC2 is the first large-scale (300 deg 2 ) image simulation of a transient survey that includes realistic cadence, variable observing conditions, and CCD image artifacts. We analyze ∼15 deg 2 of DC2 over a 5 yr time span in which artificial point sources from Type Ia supernova (SNIa) light curves have been overlaid onto the images. The magnitude limits per filter are u = 23.66 mag, g = 24.69 mag, r = 24.06 mag, i = 23.45 mag, z = 22.54 mag, and y = 21.62 mag. The artifact contamination levels are ∼90% of all detections, corresponding to ∼1000 artifacts deg –2 in g band, and falling to 300 deg –2 in y band. The photometry has biases <1% for magnitudes 19.5 < m < 23. Our DIA performance on simulated images is similar to that of the Dark Energy Survey difference-imaging pipeline on real images. We also characterize DC2 image properties to produce catalog-level simulations needed for distance bias corrections. We find good agreement between DC2 data and simulations for distributions of signal-to-noise ratio, redshift, and fitted light-curve properties. Applying a realistic SNIa cosmology analysis for redshifts z < 1, we recover the input cosmology parameters to within statistical uncertainties.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2026

Extracting the Breakout Distance from the ECOT Trajectories: Gaussian Process Regression Approach

Enhanced Corner Turning (ECOT) experiments provide an important metric of performance of high explosive (HE) formulations. The breakout distance is a single scalar value that characterizes the corner turning efficiency of an HE. Extracting the breakout distance from the raw ECOT results, whether experimental or simulated, is a conceptually straightforward procedure which, however, is non-unique, especially in the presence of noise. More specifically, this procedure involves numerical smoothing and selecting particular values for parameters of this smoothing introduces human bias. In this work, we propose to use the Gaussian process regression to analyze ECOT results. This analysis involves the effective smoothing of the data, thus allowing for accurate extraction of the breakout distance. Most importantly, the parameters of this smoothing can be inferred from the ECOT data itself, rendering the approach effectively parameter-free and thus diminishing the human bias. An additional benefit of the Gaussian process regression, being a statistical inference method, is that not just the value of the breakout distance, but also its confidence interval can be extracted from the data. This report introduces the Gaussian process regression, as applied to ECOT, and demonstrates its usefulness by extracting the breakout distances for a selection of experimental and simulated data.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

DOE OSTI2020

Microphysical Sensitivity of Superparameterized Precipitation Extremes in the Contiguous United States Due to Feedbacks on Large-Scale Circulation

Superparameterized (SP) global climate models have been shown to better simulate various features of precipitation relative to conventional models, including its diurnal cycle as well as its extremes. While various studies have focused on the effect of differing microphysics parameterizations on precipitation within limited-area cloud-resolving models, we examine here the effect on contiguous U.S. (CONUS) extremes in a global SP model. We vary the number of predicted moments for hydrometeor distributions, the character of the rimed ice species, and the representation of raindrop self-collection and breakup. Using a likelihood ratio test and accounting for the effects of multiple hypothesis testing, we find that there are some regional differences, particularly during spring and summer in the Southwest and the Midwest, in both the current climate and a warmer climate with uniformly increased sea surface temperatures. These differences are most statistically significant and widespread when the number of moments is changed. To determine whether these results are due to (fast) local effects of the different microphysics or the (slower) ensuing feedback on the large-scale atmospheric circulation, we run a series of short, 5-day simulations initialized from reanalysis data. We find that the differences largely disappear in these runs and therefore infer that the different parameterizations impact precipitation extremes indirectly via the large-scale circulation. Finally, we compare the present-day results with hourly rain gauge data and find that SP underestimates extremes relative to observations regardless of which microphysics scheme is used given a fixed model configuration and resolution.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

Lagrangian large eddy simulations via physics-informed machine learning

High-Reynolds number homogeneous isotropic turbulence (HIT) is fully described within the Navier–Stokes (NS) equations, which are notoriously difficult to solve numerically. Engineers, interested primarily in describing turbulence at a reduced range of resolved scales, have designed heuristics, known as large eddy simulation (LES). LES is described in terms of the temporally evolving Eulerian velocity field defined over a spatial grid with the mean-spacing correspondent to the resolved scale. This classic Eulerian LES depends on assumptions about effects of subgrid scales on the resolved scales. Here, we take an alternative approach and design LES heuristics stated in terms of Lagrangian particles moving with the flow. Our Lagrangian LES, thus L-LES, is described by equations generalizing the weakly compressible smoothed particle hydrodynamics formulation with extended parametric and functional freedom, which is then resolved via Machine Learning training on Lagrangian data from direct numerical simulations of the NS equations. The L-LES model includes physics-informed parameterization and functional form, by combining physics-based parameters and physics-inspired Neural Networks to describe the evolution of turbulence within the resolved range of scales. The subgrid-scale contributions are modeled separately with physical constraints to account for the effects from unresolved scales. We build the resulting model under the differentiable programming framework to facilitate efficient training. We experiment with loss functions of different types, including physics-informed ones accounting for statistics of Lagrangian particles. We show that our L-LES model is capable of reproducing Eulerian and unique Lagrangian turbulence structures and statistics over a range of turbulent Mach numbers.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗