Search NASA⌕ Search

SEARCH · Search NASA

Results for “diffusion surrogate”

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 109 records · Page 6
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Deep Learning-based Surrogate Model for Efficient Reservoir Simulation in Large-scale Geological Carbon Storage: Application in IBDP Dataset

This project introduces an advanced deep learning (DL)-based surrogate modeling approach to enhance the efficiency and accuracy of large-scale geological carbon storage (GCS) simulations. Using the Illinois Basin Decatur Project (IBDP) dataset as training data, the study employs a residual U-Net architecture to predict critical state variables such as pressure and CO₂ saturation, as well as CO₂ plume migration. By incorporating key geological parameters (e.g., porosity, permeability, and rock facies) and physics-informed inputs like the diffusive time of flight and time step, the DL model effectively reduces computational complexity while maintaining robust physical constraints. Compared to traditional simulators like Eclipse, the DL model achieves remarkable accuracy, with a root mean square error (RMSE) of 1.57 psi for pressure and 0.007 for saturation, and dramatically reduces computational time from hours to just 69.9 seconds for 50-step simulations. These results demonstrate the potential of innovative DL methodologies to improve the predictivity and operational efficiency of GCS simulations, providing a reliable foundation for decision-making in CCS operations. Supported by the SMART initiative, this project underscores the success of leveraging computational innovations to advance CCS technologies.

advanced deep learning↗
DOE OSTI2025

GPR_calculator: An on-the-fly surrogate model to accelerate massive nudged elastic band calculations

We present GPR_calculator, a package based on Python and C++ programming languages to build an on-the-fly surrogate model using Gaussian Process Regression (GPR) to approximate computationally expensive electronic structure calculations. The key idea is to dynamically train a GPR model during the simulation that can accurately predict energies and forces with uncertainty quantification. When the uncertainty is high, the costly electronic structure calculation is performed to obtain the ground truth data, which is then used to update the GPR model. To illustrate the effectiveness of GPR_calculator, we demonstrate its application in Nudged Elastic Band (NEB) simulations of surface diffusion and reactions, achieving 3-10 times acceleration compared to pure ab initio calculations. The source code is available at https://github.com/MaterSim/GPR_calculator.

Gaussian process regression↗
NASA NTRS1992

Multiple periodicities in the solar magnetic field - Possible origin in a multiple-mode solar dynamo

The solar magnetic field is generated in an oscillatory mode with a 22 yr full period and gives rise to the 11 yr sunspot cycle. However, analyses of contemporary solar records, as well as other surrogate indicators of solar activity, suggest the presence also of longer term periodicities in the solar magnetic cycle. This paper suggests that the solar dynamo can operate in a multiply periodic state, with several periodicites being generated simultaneously at different depths in the convection zone. A simple two-layer model of the solar convection zone is used to illustrate the physical mechanism of spatially localized, multiple-periodicity-mode dynamo regeneration. The two layers are characterized by differences in their respective turbulent magnetic diffusivities. Although the magnetic modes interact with one another, each mode is produced large in one layer or the other, and has an oscillation period approximately equal to the time characteristic of magnetic diffusion across the layer. The observed complicated periodicity pattern in the solar magnetic field could be a combination of two (or more) dynamo modes generated in this manner. The calculations are carried out using a differential rotation model consistent with recent helioseismological measurements, illustrating the challenge to dynamo theory raised by those observational results.

Boyer, D. W.↗
DOE OSTI2025

Entrapment Behavior of Solid Surrogate Fission Products at Engineered UN Nano‐Hetero‐Interfaces Within Metallic Nuclear Fuels

Nanometric hetero-interfaces provide a wealth of scientific and engineering opportunities due to their complex and often misunderstood properties that can differ from their respective bulk constituents. In this work, the ability for engineered nanostructures within a bulk U─Mo alloy to arrest simulant fission products is investigated experimentally and computationally. Nanostructured 90 wt% U/ 10 wt% Mo (U-10Mo) with 7.1 at% Nd is consolidated using spark-plasma- sintering (SPS) techniques and is heat-treated at 500 °C under vacuum for 24, 100, 500, and 1000 h. Analysis on the sintered and heat-treated U-10Mo reveals rapid kinetics in Nd diffusion to nanocluster sites, with evidence of Nd diffusion occurring during sintering and during the following heat-treatment. The segregation behavior of Nd at two different U─Mo/UN interfaces is computationally verified using density functional theory (DFT) to reinforce experimental data.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

Active learning of ternary alloy structures and energies

Abstract Machine learning models with uncertainty quantification have recently emerged as attractive tools to accelerate the navigation of catalyst design spaces in a data-efficient manner. Here, we combine active learning with a dropout graph convolutional network (dGCN) as a surrogate model to explore the complex materials space of high-entropy alloys (HEAs). We train the dGCN on the formation energies of disordered binary alloy structures in the Pd-Pt-Sn ternary alloy system and improve predictions on ternary structures by performing reduced optimization of the formation free energy, the target property that determines HEA stability, over ensembles of ternary structures constructed based on two coordinate systems: (a) a physics-informed ternary composition space, and (b) data-driven coordinates discovered by the Diffusion Maps manifold learning scheme. Both reduced optimization techniques improve predictions of the formation free energy in the ternary alloy space with a significantly reduced number of DFT calculations compared to a high-fidelity model. The physics-based scheme converges to the target property in a manner akin to a depth-first strategy, whereas the data-driven scheme appears more akin to a breadth-first approach. Both sampling schemes, coupled with our acquisition function, successfully exploit a database of DFT-calculated binary alloy structures and energies, augmented with a relatively small number of ternary alloy calculations, to identify stable ternary HEA compositions and structures. This generalized framework can be extended to incorporate more complex bulk and surface structural motifs, and the results demonstrate that significant dimensionality reduction is possible in thermodynamic sampling problems when suitable active learning schemes are employed.

Chemistry↗
DOE OSTI2024

Evaluation of lithium as a tritium storage medium for betavoltaics

Lithium foils were demonstrated to absorb surrogate protium for tritium-powered betavoltaics. 20μm thick lithium foils were hole-punched from a ribbon of electrodeposited lithium on copper foil. The lithium foils were loaded with hydrogen in a custom Sievert apparatus where the pressure drop showed full hydriding at a hydrogen pressure of 2 bar and at all loading temperatures above the lithium melting point at 190, 200, 225, 250, and 300. Lithium hydride formation was confirmed with Raman spectroscopy after hydrogen loading. The kinetics of experimental hydride formation was compared to the diffusion-limited Mintz–Bloch model. While the Mintz–Bloch model showed good fit with the experimental loadings, the model overpredicted the loading kinetics starting at 250 °C and at higher temperatures. In conclusion, The overprediction was either caused by lithium hydride outgassing due to some reduction with some residual lithium hydroxide created from brief air exposure when sealing the lithium in the reactor or a transition from diffusion-limited hydride growth to surface or metal–hydride interface-limited hydride growth.

25 ENERGY STORAGE↗
NASA NTRS2007

Opposed Jet Burner Extinction Limits: Simple Mixed Hydrocarbon Scramjet Fuels vs Air

Opposed Jet Burner tools have been used extensively by the authors to measure Flame Strength (FS) of laminar non-premixed H2 air and simple hydrocarbon (HC) air counterflow diffusion flames at 1-atm. FS represents a strain-induced extinction limit based on air jet velocity. This paper follows AIAA-2006-5223, and provides new HC air FSs for global testing of chemical kinetics, and for characterizing idealized flameholding potentials during early scramjet-like combustion. Previous FS data included six HCs, pure and N2-diluted; and three HC-diluted H2 fuels, where FS decayed very nonlinearly as HC was added to H2, due to H-atom scavenging. This study presents FSs on mixtures of (candidate surrogate) HCs, some with very high FS ethylene. Included are four binary gaseous systems at 300 K, and a hot ternary system at approx. 600 K. The binaries are methane + ethylene, ethane + ethylene, methane + ethane, and methane + propylene. The first three also form two ternary systems. The hot ternary includes both 10.8 and 21.3 mole % vaporized n-heptane and full ranges of methane + ethylene. Normalized FS data provide accurate means of (1) validating, globally, chemical kinetics for extinction of non-premixed flames, and (2) estimating (scaling by HC) the loss of incipient flameholding in scramjet combustors. The n-heptane is part of a proposed baseline simulant (10 mole % with 30% methane + 60% ethylene) that mimics the ignition of endothermically cracked JP-7 like kerosene fuel, as suggested by Colket and Spadaccini in 2001 in their shock tube Scramjet Fuels Autoignition Study. Presently, we use FS to gauge idealized flameholding, and define HC surrogates. First, FS was characterized for hot nheptane + methane + ethylene; then a hot 36 mole % methane + 64% ethylene surrogate was defined that mimics FS of the baseline simulant system. A similar hot ethane + ethylene surrogate can also be defined, but it has lower vapor pressure at 300 K, and thus exhibits reduced gaseous capacity. The new FS results refine our earlier idealized reactivity scale that shows wide ranging (50 x) diameter-normalized FSs for various HCs. These range from JP-10 and methane to H2 air, which produces an exceptionally strong flame that agrees within approx. 1% of recent 2-D numerically simulations. Finally, we continue advocating the FS approach as more direct and fundamental, for assessing idealized scramjet flameholding potentials, than measurements of unstrained laminar burning velocity or blowout in a Perfectly Stirred Reactor.

Pellett, Gerald L.↗
NASA NTRS2001

Trioxane-Air Counterflow Diffusion Flames in Normal and Microgravity

Trioxane, a weakly bound polymer of formaldehyde (C3H6O3, m.p. 61 C, b.p. 115 C), is a uniquely suited compound for studying material flammability. Like many of the more commonly used materials for such tests (e.g., delrin, polyethylene, acrylic sheet, wood, and paper), it displays relevant phenomena (internal heat conduction, melting, vaporization, thermal decomposition, and gas phase reaction of the decomposition products). Unlike the other materials, however, it is non-sooting and has simple and well-known chemical kinetic pathways for its combustion. Hence it should prove to be much more useful for numerical modeling of surface combustion than the complex fuels typically used. We have performed the first exploratory tests of trioxane combustion in the counterflow configuration to determine its potential as a surrogate solid fuel which allows detailed modeling. The experiments were performed in the spring and summer of 1998 at the National Institute of Standards and Technology in Gaithersburg, MD, and at NASA-GRC in Cleveland. Using counterflow flames at 1-g, we measured the fuel consumption rate and the extinction conditions with added N2 in the air; at mg conditions, we observed the ignition characteristics and flame shape from video images. We have performed numerical calculations of the flame structure, but these are not described here due to space limitations. This paper summarizes some burning characteristics of trioxane relevant to its use for studying flame spread and fire suppression.

Linteris, Gregory T.↗
DOE OSTI2023

Toward DMC Accuracy Across Chemical Space with Scalable Δ-QML

In the past decade, quantum diffusion Monte Carlo (DMC) has been demonstrated to successfully predict the energetics and properties of a wide range of molecules and solids by numerically solving the electronic many-body Schrödinger equation. With O(N 3 ) scaling with the number of electrons N, DMC has the potential to be a reference method for larger systems that are not accessible to more traditional methods such as CCSD(T). Assessing the accuracy of DMC for smaller molecules becomes the stepping stone in making the method a reference for larger systems. We show that when coupled with quantum machine learning (QML)-based surrogate methods, the computational burden can be alleviated such that quantum Monte Carlo (QMC) shows clear potential to undergird the formation of high-quality descriptions across chemical space. We discuss three crucial approximations necessary to accomplish this: the fixed-node approximation, universal and accurate references for chemical bond dissociation energies, and scalable minimal amons-set-based QML (AQML) models. Numerical evidence presented includes converged DMC results for over 1000 small organic molecules with up to five heavy atoms used as amons and 50 medium-sized organic molecules with nine heavy atoms to validate the AQML predictions. Finally, numerical evidence collected for Δ-AQML models suggests that already modestly sized QMC training data sets of amons suffice to predict total energies with near chemical accuracy throughout chemical space.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Atomistic simulations and machine learning of solute grain boundary segregation in Mg alloys at finite temperatures

Understanding solute segregation thermodynamics is the first step in investigating grain boundary (GB) properties, such as strong yttrium (Y) effects on grain growth and texture evolution in micro-scale polycrystalline magnesium (Mg) alloys. To estimate the average GB segregation behavior in low-solute-concentration Mg alloys (e.g., 2 at.% Y), a state-of-the-art spectral approach is applied based on a per-site segregation energy spectrum for Y solute atoms at zero K obtained from molecular statistics (MS) simulations of ~10 4 GB sites in Mg symmetric tilt GBs (STGBs). Although selected MS simulation results are consistent with verification by density functional theory (DFT) calculations, estimates of average segregation tendency based on the zero-K energy spectrum deviate from experimental observations. To resolve this problem, thermodynamic integration (TI) methods based on molecular dynamics (MD) simulations are used to determine the per-site segregation free energies of Y at representative GB sites, which show contributions beyond harmonic approximations can be important for certain GB sites at high temperatures. A surrogate model of per-site segregation free energy is constructed from a small set of TI data points using stacking cross-validation regressors and physics-informed descriptors. This model is applied to predict the Y segregation free energy spectra for thousands of GB sites in Mg STGBs with uncertainty quantification. Finally, the average segregation tendency predicted by the spectral approach based on the free energy spectra agrees well (within the uncertainty range) with experimental observations for micro-scale polycrystalline Mg alloys at typical thermomechanical processing temperatures (500 ~ 800 K), where thermodynamic equilibrium states are likely to be achieved due to fast diffusion.

Atomistic simulations↗
DOE OSTI2025

MMMnet: A Neural Network Surrogate for Real-Time Transport Prediction Based on the Updated Multi-Mode Model

The Multi-Mode Model (MMM) is a physics-based anomalous transport model integrated into TRANSP for predicting electron and ion thermal transport, electron and impurity particle transport, and toroidal and poloidal momentum transport. While MMM provides valuable predictive capabilities, its computational cost, although manageable for standard simulations, is too high for real-time control applications. MMMnet, a neural network-based surrogate model, is developed to address this challenge by significantly reducing computation time while maintaining high accuracy. Trained on TRANSP simulations of DIII-D discharges, MMMnet incorporates an updated version of MMM (9.0.10) with enhanced physics, including isotopic effects, plasma shaping via effective magnetic shear, unified correlation lengths for ion-scale modes, and a new physics-based model for the electromagnetic electron temperature gradient mode. A key advancement is MMMnet’s ability to predict all six transport coefficients, providing a comprehensive representation of plasma transport dynamics. MMMnet achieves a two-order-of-magnitude speed improvement while maintaining strong correlation with MMM diffusivities, making it well-suited for real-time tokamak control and scenario optimization.

DIII-D↗
DOE OSTI2021

A NIR fluorescent smart probe for imaging tumor hypoxia

Abstract Background Tumor hypoxia is a characteristic of paramount importance due to low oxygenation levels in tissue negatively correlating with resistance to traditional therapies. The ability to noninvasively identify such could provide for personalized treatment(s) and enhance survival rates. Accordingly, we recently developed an NIR fluorescent hypoxia‐sensitive smart probe ( NO 2 ‐Rosol ) for identifying hypoxia via selectively imaging nitroreductase (NTR) activity, which could correlate to oxygen deprivation levels in cells, thereby serving as a proxy. We demonstrated proof of concept by subjecting a glioblastoma (GBM) cell line to extreme stress by evaluating such under radiobiological hypoxic ( p O 2 ≤ ~0.5%) conditions, which is a far cry from representative levels for hypoxia for brain glioma ( p O 2 = ~1.7%) which fluctuate little from physiological hypoxic ( p O 2 = 1.0‐3.0%) conditions. Aim We aimed to evaluate the robustness, suitability, and feasibility of NO 2 ‐Rosol for imaging hypoxia in vitro and in vivo via assessing NTR activity in diverse GBM models under relevant oxygenation levels ( p O 2 = 2.0%) within physiological hypoxic conditions that mimic oxygenation levels in GBM tumor tissue in the brain. Methods We evaluated multiple GBM cell lines to determine their relative sensitivity to oxygenation levels via measuring carbonic anhydrase IX (CAIX) levels, which is a surrogate marker for indirectly identifying hypoxia by reporting on oxygen deprivation levels and upregulated NTR activity. We evaluated for hypoxia via measuring NTR activity when employing NO 2 ‐Rosol in in vitro and tumor hypoxia imaging studies in vivo. Results The GBM39 cell line demonstrated the highest CAIX expression under hypoxic conditions representing that of GBM in the brain. NO 2 ‐Rosol displayed an 8‐fold fluorescence enhancement when evaluated in GBM39 cells ( p O 2 = 2.0%), thereby establishing its robustness and suitability for imaging hypoxia under relevant physiological conditions. We demonstrated the feasibility of NO 2 ‐Rosol to afford tumor hypoxia imaging in vivo via it demonstrating a tumor‐to‐background of 5 upon (i) diffusion throughout, (ii) bioreductive activation by NTR activity in, and (iii) retention within, GBM39 tumor tissue. Conclusion We established the robustness, suitability, and feasibility of NO 2 ‐Rosol for imaging hypoxia under relevant oxygenation levels in vitro and in vivo via assessing NTR activity in GBM39 models.

Hettie, Kenneth S.↗
DOE OSTI2021

Deep-learning-enabled Bayesian inference of fuel magnetization in magnetized liner inertial fusion

Fuel magnetization in magneto-inertial fusion (MIF) experiments improves charged burn product confinement, reducing requirements on fuel areal density and pressure to achieve self-heating. By elongating the path length of 1.01 MeV tritons produced in a pure deuterium fusion plasma, magnetization enhances the probability for deuterium–tritium reactions producing 11.8-17.1 MeV neutrons. Nuclear diagnostics thus enable a sensitive probe of magnetization. Characterization of magnetization, including uncertainty quantification, is crucial for understanding the physics governing target performance in MIF platforms, such as magnetized liner inertial fusion (MagLIF) experiments conducted at Sandia National Laboratories, Z-facility. We demonstrate a deep-learned surrogate of a physics-based model of nuclear measurements. A single model evaluation is reduced from O ( 10 – 100 ) (10–100) CPU hours on a high-performance computing cluster down to O ( 10 – 100 ) (10) ms on a laptop. This enables a Bayesian inference of magnetization, rigorously accounting for uncertainties from surrogate modeling and noisy nuclear measurements. The approach is validated by testing on synthetic data and comparing with a previous study. We analyze a series of MagLIF experiments systematically varying preheat, resulting in the first ever systematic experimental study of magnetic confinement properties of the fuel plasma as a function of fundamental inputs on any neutron-producing MIF platform. We demonstrate that magnetization decreases from B R ∼ 0.2 ~0.5 to B R ∼ 0.2 ~0.2 MG cm as laser preheat energy deposited increases from E preheat ∼ 460 ~460 J to E preheat ∼ 460 ~1.4 kJ. This trend is consistent with 2D LASNEX simulations showing Nernst advection of the magnetic field out of the hot fuel and diffusion into the target liner.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

REBOUND: Reverse Engineering Bidirectional Outflow Under Non-Equilibrium Diffusion

Rare-earth elements (REEs) are essential for electronics, renewable energy, and defense technologies. However, the current supply of REEs relies on mining concentrated in a few countries and energy-intensive separations. DOE’s Basic Energy Sciences (BES) program has launched a grand challenge which aims to ensure a sustainable supply of critical REEs by developing innovative and environmentally friendly separation methods. As an alternative to costly and harmful traditional methods, the Non-Equilibrium Transport Driven Separations (NETS) initiative has created a microfluidic Y-channel co-flow method that applies external fields to exploit magneto- and electrohydrodynamic effects for separating dilute REE ions from complex feedstocks. Computational fluid dynamics (CFD) studies have identified a few operating conditions with promising ion selectivity and separation efficiency. However, challenges remain regarding Y-channel versatility across feedstocks and accurate incorporation of physical phenomena into CFD models. In this work, we develop a multi-fidelity modelling approach which integrates experimental results with CFD simulation to build a surrogate model for the dependence of separation efficiency to variation of design parameters. The surrogate model enables a reinforcement learning (RL) method to adaptively launch CFD and experimental runs, improving model fidelity around optimal Y-channel parameters.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Deep Learning Emulation of Subgrid-Scale Processes in Turbulent Shear Flows

Deep neural networks (DNNs) are developed from a data set obtained from the dynamic Smagorinsky model to emulate the subgrid-scale (SGS) viscosity (ν sgs ) and diffusivity (κ sgs ) for turbulent stratified shear flows encountered in the oceans and the atmosphere. These DNNs predict ν sgs and κ sgs from velocities, strain rates, and density gradients such that the evolution of the kinetic energy budget and density variance budget terms is similar to the corresponding values obtained from the original dynamic Smagorinsky model. These DNNs also compute ν sgs and κ sgs ~2–4 times quicker than the dynamic Smagorinsky model resulting in a ~2–2.5 times acceleration of the entire simulation. This study demonstrates the feasibility of deep learning in emulating the subgrid-scale (SGS) phenomenon in geophysical flows accurately in a cost-effective manner. In a broader perspective, deep learning-based surrogate models can present a promising alternative to the traditional parameterizations of the subgrid-scale processes in climate models.

58 GEOSCIENCES↗
DOE OSTI2023

On the influence of over-parameterization in manifold based surrogates and deep neural operators

Constructing accurate and generalizable approximators (surrogate models) for complex physico-chemical processes exhibiting highly non-smooth dynamics is challenging. The main question is what type of surrogate models we should construct and should these models be under-parameterized or over-parameterized. In this work, we propose new developments and perform comparisons for two promising approaches: manifold-based polynomial chaos expansion (m-PCE) and the deep neural operator (DeepONet), and we examine the effect of over-parameterization on generalization. While m-PCE enables the construction of a mapping by first identifying low-dimensional embeddings of the input functions, parameters, and quantities of interest (QoIs), a neural operator learns the nonlinear mapping via the use of deep neural networks. Here, we demonstrate the performance of these methods in terms of generalization accuracy by solving the 2D time-dependent Brusselator reaction-diffusion system with uncertainty sources, modeling an autocatalytic chemical reaction between two species. We first propose an extension of the m-PCE by constructing a mapping between latent spaces formed by two separate embeddings of the input functions and the output QoIs. To further enhance the accuracy of the DeepONet, we introduce weight self-adaptivity in the loss function. We demonstrate that the performance of m-PCE and DeepONet is comparable for cases of relatively smooth input-output mappings. However, when highly non-smooth dynamics is considered, DeepONet shows higher approximation accuracy. We also find that for m-PCE, modest over-parameterization leads to better generalization, both within and outside of distribution, whereas aggressive over-parameterization leads to over-fitting. In contrast, an even highly over-parameterized DeepONet leads to better generalization for both smooth and non-smooth dynamics. Furthermore, we compare the performance of the above models with another recently proposed operator learning model, the Fourier Neural Operator, and show that its over-parameterization also leads to better generalization. Taken together, our studies show that m-PCE can provide very good accuracy at very low training cost, whereas a highly over-parameterized DeepONet can provide better accuracy and robustness to noise but at higher training cost. In both methods, the inference cost is negligible.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Learning turbulent flows with generative models for super resolution and sparse flow reconstruction

Neural operators are promising surrogates for dynamical systems but when trained with standard L 2 losses they tend to oversmooth fine-scale turbulent structures. Here, we show that combining operator learning with generative modeling overcomes this limitation. We consider three practical turbulent-flow challenges where conventional neural operators fail: spatio-temporal super-resolution, forecasting, and sparse flow reconstruction. For Schlieren jet super-resolution, an adversarially trained neural operator (adv-NO) reduces the energy-spectrum error by 15 × while preserving sharp gradients at neural operator-like inference cost. For 3D homogeneous isotropic turbulence, adv-NO trained on only 160 timesteps from a single trajectory forecasts accurately for five eddy-turnover times and offers 114 × wall-clock speed-up at inference than the baseline diffusion-based forecasters, enabling near-real-time rollouts. For reconstructing cylinder wake flows from highly sparse Particle Tracking Velocimetry-like inputs, a conditional generative model infers full 3D velocity and pressure fields with correct phase alignment and statistics. These advances enable accurate reconstruction and forecasting at low compute cost, bringing near-real-time analysis and control within reach in experimental and computational fluid mechanics.

Fluid dynamics↗