SEARCH · Search NASA

Results for “deep generative models”

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 577 records · Page 32

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Physics-informed, empirically constrained machine learning for designing Fe-9Cr alloys

<span style="font-family: Calibri, sans-serif; font-size: 12pt;">Materials data analytics can be used to significantly shorten development time of specialized alloys needed for next generation energy applications. Incorporation of the domain knowledge into deep-learning graph structure via fuzzy pre-training and causal process imitation presents a viable approach to developing accurate data-driven models and reliable alloy design tools, with limited datasets. It was demonstrated that the domain knowledge-based empirical constraints not only inhibit overfitting but also allow training more accurate and reliable ML models with improved transparency of the output interpretation. In this study, alloy tensile properties were interpreted with three competing virtual-microstructure models.</span>

Romanov, Vyacheslav↗

DOE OSTI2025

Universal Electronic‐Structure Relationship Governing Intrinsic Magnetic Properties in Permanent Magnets

An electronic-structure-centered perspective is presented on permanent-magnet (PM) design, highlighting two key levers, that is, saturation magnetization (M s ), governed by 3d-band filling and exchange physics, and magnetocrystalline anisotropy energy (MAE), arising from spin-orbit coupling (SOC) on anisotropic orbital populations. Reviewing current practices, including DFT-based MAE/J ij extraction, atomistic-spin and micromagnetic modeling, and high-throughput machine learning (ML) pipelines, three bottlenecks limiting predictive discovery is identified that is i) electronic-structure accuracy for small MAE (sensitive to functional choice, Hubbard U, and many-body effects), ii) finite-temperature and kinetic realism (phonon/magnon renormalization, ordering kinetics), and iii) descriptor and multiscale decoupling (lack of SOC-weighted and orbital-resolved fingerprints). Deep dives into the electronic-structure of Nd─Fe─B and Fe─N show how these fingerprints govern magnetic performance, motivating DFT- and quantum-mechanics-based descriptors for discovery. Unbiased, structure-driven exploration, coupled with high-throughput simulations, ML, generative AI, and reasoning models, accelerates candidate identification and propagates insights across scales. Addressing supply-chain risks, on future needs of designing “critical-element-free” magnets with tailored microstructure and high energy products is emphasized. By integrating electronic fingerprints, AI reasoning, and multiscale modeling, a practical roadmap is provided for rare-earth-lean or rare-earth free, high-performance, sustainable PMs.

Singh, Prashant [Ames Laboratory, and Iowa State U↗

DOE OSTI2021

A physics-informed and hierarchically regularized data-driven model for predicting fluid flow through porous media

This paper presents a new deep learning data-driven model for predicting structure dependent pore-fluid velocity fields in rock. The model is based on a Convolutional Auto-Encoder (CAE) artificial neural network capable of learning from image data generated by direct numerical simulations of fluid flow through pore-structures, such as by Lattice Boltzmann or molecular dynamics methods. The main novelty of the model in comparison to previous CAE-based data-driven approaches consists of three parts. The first is a methodology for decomposing the full-domain of the porous media into sub-regions, or “sub-domains”, in order to reduce the overall size of the CAE, batch process the sub-domains in parallel, and enable the CAE to learn local and generalizable nonlinear mappings of pore-fluid velocities. The second consists of embedding the finite difference solutions of the incompressible Navier-Stokes and continuity equations into convolutional layers prior to the CAE in order to provide the CAE with knowledge of fluid dynamics physics (PhyFlow). The third main novelty is that the training of the CAE is regularized with a hierarchical loss function that encourages the learning of fluid flow patterns (in a way similar to ranked modes in principal component analysis), ranking from most to least important. This is shown to increase the stability in learning, reduce over-fitting, and promote interpretability of the CAE neural network layers (HierCAE). The comprehensive new data-driven model, which we call the PhyFlow-HierCAE model, is shown to exhibit improved accuracy and generalizability of flow field predictions over conventional CAE models, attributable to the embedded physical knowledge and the hierarchical regularization, as well as realize orders of magnitude speed-ups in computation times as a surrogate for the direct numerical simulations. Examples of training and forward predictions on unseen pore-structures are provided and evaluated for data from Lattice Boltzmann and molecular dynamics simulations of pore-fluid flow. The model is shown to be a fast and accurate emulator (or “surrogate”) for predicting effective permeability of unseen pore-structures based on learning from relatively small direct numerical simulation datasets.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

NASA NTRS1988

Navier-Stokes calculations of transonic flows past cavities

A computational investigation of subsonic and transonic flows past three dimensional deep and transitional cavities is presented. Computational simulations of these self induced oscillatory flows were generated through time accurate solutions of the Reynolds averaged full Navier-Stokes equations, using the explicit MacCormack scheme. The Reynolds stresses were included through the Baldwin-Lomax algebraic turbulence model with certain modifications. Two cases were computed to demonstrate the capability of the numerical scheme in modeling the complex three dimensional flow features inside a cavity. The results from an experimental investigation were used not only to benchmark the computations, but also to widen the database used for the discussions and conclusions. The computational results include instantaneous and time averaged flow properties everywhere in the computational zone. Time series analyses were performed for the instantaneous pressure values on the cavity floor. The features of deep and transitional cavity flows, and the effects of the sidewall on the cavity flow flowfield are illustrated through computational graphics.

Baysal, Oktay↗

DOE OSTI2019

Learning Deep Neural Network Representations for Koopman Operators of Nonlinear Dynamical Systems

The Koopman operator has recently garnered much attention for its value in dynamical systems analysis and data-driven model discovery. However, its application has been hindered by the computational complexity of extended dynamic mode decomposition; this requires a combinatorially large basis set to adequately describe many nonlinear systems of interest, e.g. cyber-physical infrastructure systems, biological networks, social systems, and fluid dynamics. Often the dictionaries generated for these problems are manually curated, requiring domain-specific knowledge and painstaking tuning. In this paper we introduce a deep learning framework for learning Koopman operators of nonlinear dynamical systems. We show that this novel method automatically selects efficient deep dictionaries, outperforming state-of-the-art methods. We benchmark this method on partially observed nonlinear systems, including the glycolytic oscillator and show it is able to predict quantitatively 100 steps into the future, using only a single timepoint, and qualitative oscillatory behavior 400 steps into the future.

koopman operators, machine learning, deep learning↗

DOE OSTI2019

Valuation of ecosystem services in alternative bioenergy landscape scenarios

Abstract Agricultural land in the Midwest is a source of food and fuel, as well as biodiversity. It is also a cause of excess nutrients that make their way to the Mississippi River and the Gulf of Mexico. To address unsustainable changes to biogeochemical cycles and ecosystem functions, a multidisciplinary approach involving social science, natural science, and engineering is often effective. Given the potential of second‐generation biofuels, and capitalizing on the deep‐rooted perennial bioenergy crops capable of thriving in poor soils, we demonstrated an integrated socio‐environmental analysis of the impacts of growing switchgrass within row‐crop landscapes in Illinois. In this study, we model land use scenarios that incorporate switchgrass as a biofuel crop in a Midwest corn‐belt watershed using the Soil Water Assessment Tool coupled with an economic analysis for the Vermilion Basin in Illinois. We estimated the values of ecosystem services under an alternative bioenergy landscape, including commodity and bioenergy crops, changes in biogeochemistry, and recreational services. The estimated annual values of nitrate and sediment reduction attributed to bioenergy crops range from $38 million to $97 million and $16,000 to $197,000, respectively. The annual value of carbon dioxide emission reduction ranges from $1.8 million to $6.1 million based on the initial crop rotation pattern. Estimated average annual values for wildlife viewing, water‐based recreation, and pheasant hunting are $1.24 million, $0.17 million, and $0.3 million, respectively. To our knowledge, this study represents the first effort to comprehensively quantify ecosystem services using a process‐based model, and estimate their value in an alternative bioenergy landscape. The information we generate could aid in understanding the potential for biomass production from marginal land and the total economic value of the landscape at various spatial scales. The framework is useful in fostering alternative bioenergy landscapes with synergies in a food, energy, and conservation nexus.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Deep learning forecasts the spatiotemporal evolution of fluid-induced microearthquakes

Microearthquakes generated by subsurface fluid injection record the evolving stress state and permeability of reservoirs. Forecasting their spatiotemporal evolution is therefore critical for applications such as enhanced geothermal systems, carbon dioxide sequestration and other geoengineering applications. Here we propose a transformer neural network model that ingests hydraulic stimulation history and prior microearthquake observations to forecast four key quantities: cumulative microearthquake count, cumulative logarithmic seismic moment, and the 50th- and 95th-percentile extents of the microearthquake cloud. Applied to the EGS Collab Experiment 1 dataset, the model achieves R2 > 0.98 for the 1-s forecast horizon and R2 > 0.88 for the 15-s forecast horizon across all targets, and supplies uncertainty estimates through a learned standard deviation term. These accurate, uncertainty-quantified forecasts enable real-time inference of fracture propagation and permeability evolution, demonstrating the strong potential of deep-learning approaches to improve seismic-risk assessment and guide mitigation strategies in future fluid-injection operations.

Chung, Jaehong↗

NASA NTRS2013

Performance Evaluation of Large Aperture 'Polished Panel' Optical Receivers Based on Experimental Data

Recent interest in hybrid RF/Optical communications has led to the development and installation of a "polished-panel" optical receiver evaluation assembly on the 34-meter research antenna at Deep-Space Station 13 (DSS-13) at NASA's Goldstone Communications Complex. The test setup consists of a custom aluminum panel polished to optical smoothness, and a large-sensor CCD camera designed to image the point-spread function (PSF) generated by the polished aluminum panel. Extensive data has been obtained via realtime tracking and imaging of planets and stars at DSS-13. Both "on-source" and "off-source" data were recorded at various elevations, enabling the development of realistic simulations and analytic models to help determine the performance of future deep-space communications systems operating with on-off keying (OOK) or pulse-position-modulated (PPM) signaling formats with photon-counting detection, and compared with the ultimate quantum bound on detection performance for these modulations. Experimentally determined PSFs were scaled to provide realistic signal-distributions across a photon-counting detector array when a pulse is received, and uncoded as well as block-coded performance analyzed and evaluated for a well-known class of block codes.

on-off keying (OOK)↗

DOE OSTI2018

100 Years of Research on Mesoscale Convective Systems

Abstract When cumulonimbus clouds aggregate, developing into a single entity with precipitation covering a horizontal scale of hundreds of kilometers, they are called mesoscale convective systems (MCSs). They account for much of Earth’s precipitation, generate severe weather events and flooding, produce prodigious cirriform anvil clouds, and affect the evolution of the larger-scale circulation. Understanding the inner workings of MCSs has resulted from developments in observational technology and modeling. Time–space conversion of ordinary surface and upper-air observations provided early insight into MCSs, but deeper understanding has followed field campaigns using increasingly sophisticated radars, better aircraft instrumentation, and an ever-widening range of satellite instruments, especially satellite-borne radars. High-resolution modeling and theoretical insights have shown that aggregated cumulonimbus clouds induce a mesoscale circulation consisting of air overturning on a scale larger than the scale of individual convective up- and downdrafts. These layers can be kilometers deep and decoupled from the boundary layer in elevated MCSs. Cooling in the lower troposphere and heating aloft characterize the stratiform regions of MCSs. As a result, long-lived MCSs with large stratiform regions have a top-heavy heating profile that generates potential vorticity in midlevels, thus influencing the larger-scale circulation within which the MCSs occur. Global satellite data show MCSs varying in structure, depending on the prevailing large-scale circulation and topography. These patterns are likely to change with global warming. In addition, environmental pollution affects MCS structure and dynamics subtly. Feedbacks of MCSs therefore need to be included or parameterized in climate models.

Houze, Jr., Robert A.↗

DOE OSTI2020

Unified Medical Language System resources improve sieve-based generation and Bidirectional Encoder Representations from Transformers (BERT)–based ranking for concept normalization

Concept normalization, the task of linking phrases in text to concepts in an ontology, is useful for many downstream tasks including relation extraction, information retrieval, etc. We present a generate-and-rank concept normalization system based on our participation in the 2019 National NLP Clinical Challenges Shared Task Track 3 Concept Normalization. The shared task provided 13 609 concept mentions drawn from 100 discharge summaries. We first design a sieve-based system that uses Lucene indices over the training data, Unified Medical Language System (UMLS) preferred terms, and UMLS synonyms to generate a list of possible concepts for each mention. We then design a listwise classifier based on the BERT (Bidirectional Encoder Representations from Transformers) neural network to rank the candidate concepts, integrating UMLS semantic types through a regularizer. Our generate-and-rank system was third of 33 in the competition, outperforming the candidate generator alone (81.66% vs 79.44%) and the previous state of the art (76.35%). During postevaluation, the model’s accuracy was increased to 83.56% via improvements to how training data are generated from UMLS and incorporation of our UMLS semantic type regularizer. Analysis of the model shows that prioritizing UMLS preferred terms yields better performance, that the UMLS semantic type regularizer results in qualitatively better concept predictions, and that the model performs well even on concepts not seen during training. Our generate-and-rank framework for UMLS concept normalization integrates key UMLS features like preferred terms and semantic types with a neural network–based ranking model to accurately link phrases in text to UMLS concepts.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

DOE OSTI2025

Chemical classification program synthesis using generative artificial intelligence

Accurately classifying chemical structures is essential for cheminformatics and bioinformatics, including tasks such as identifying bioactive compounds of interest, screening molecules for toxicity to humans, finding non-organic compounds with desirable material properties, or organizing large chemical libraries for drug discovery or environmental monitoring. However, manual classification is labor-intensive and difficult to scale to large chemical databases. Existing automated approaches either rely on manually constructed classification rules, or are deep learning methods that lack explainability. This work presents an approach that uses generative artificial intelligence to automatically write chemical classifier programs for classes in the Chemical Entities of Biological Interest (ChEBI) database. These programs can be used for efficient deterministic run-time classification of SMILES structures, with natural language explanations. The programs themselves constitute an explainable computable ontological model of chemical class nomenclature, which we call the ChEBI Chemical Class Program Ontology (C3PO). We validated our approach against the ChEBI database, and compared our results against deep learning models and a naive SMARTS pattern based classifier. C3PO outperforms the naive classifier, but does not reach the performance of state of the art deep learning methods. However, C3PO has a number of strengths that complement deep learning methods, including explainability and reduced data dependence. C3PO can be used alongside deep learning classifiers to provide an explanation of the classification, where both methods agree. The programs can be used as part of the ontology development process, and iteratively refined by expert human curators.

Artificial Intelligence↗

NASA NTRS1989

Navier-Stokes calculations of transonic flows past cavities

A computational investigation of subsonic and transonic flows past 3-D deep transitional cavities is presented. Computational simulations of these self-induced oscillatory flows were generated through time-accurate solutions of the Reynolds averaged full Navier-Stokes equations, using the explicit MacCormack scheme. The Reynolds stresses were included through the Baldwin-Lomax algebraic turbulence model with certain modifications. Two cases were computed to demonstrate the capability of the numerical scheme in modeling the complex 3-D flow features inside a cavity. The results from an experimental investigation were used not only to benchmark the computations, but also to widen the database used for the discussions and conclusions. The computational results include instantaneous and time averaged flow properties everywhere in the computational zone. Time series analyses were performed for the instantaneous pressure values on the cavity floor. The features of deep and transitional cavity flows, and the effect of the sidewall on the cavity flow flowfield are illustrated through computational graphics.

Baysal, Oktay↗

DOE OSTI2018

On the Relationship Between Fault Permeability Increases, Induced Stress Perturbation, and the Growth of Aseismic Slip During Fluid Injection

Fluid injections into the deep subsurface can, at times, generate earthquakes, but often, they only produce aseismic deformations. Here we analyze the influence of fault hydromechanical properties on the growth of injection-induced aseismic slip. Using hydromechanical modeling, we show how permeability enhancement in addition to the background stress and frictional weakening has an important effect on the pressure diffusion and slip growth during injection. We find that the more pronounced the fault permeability enhancement, the stronger is the growth of the aseismic slip zone. The effect of enhanced permeability is more pronounced when the fault is initially close to failure. Our results show that aseismic slip grows beyond the pressurized zone when the fault permeability increases, while slip remains behind the pressurized zone when permeability does not vary from its initial preslip value. Furthermore, fault permeability increases should be considered as complementary mechanism to current models of fluid-induced aseismic slip.

58 GEOSCIENCES↗

DOE OSTI2019

Generation of synthetic dielectric dispersion logs in organic-rich shale formations using neural-network models

Dielectric dispersion (DD) logs acquired in subsurface geologic formations generally are composed of conductivity ([Formula: see text]) and relative permittivity ([Formula: see text]) measurements at four discrete frequencies in the range of 10 MHz to 1 GHz. Acquisition of DD logs in subsurface formations is operationally challenging, and it requires a hard-to-deploy infrastructure. We developed three supervised neural-network-based predictive methods to process conventional, easy-to-acquire subsurface logs for generating the eight DD logs acquired at four frequencies. These predictive methods will improve reservoir characterization in the absence of a DD logging tool. The predictive methods are tested in three wells intersecting organic-rich shale formations of the Permian Basin and the Bakken Shale. The first method predicts the eight dispersion logs simultaneously using a single artificial neural network (ANN) model, whereas the second method simultaneously predicts the four conductivity dispersion logs using one ANN model, followed by simultaneous prediction of four permittivity dispersion logs using a second ANN model. The third method sequentially predicts the eight dispersion logs, one at a time using eight sequential ANN models, based on a predetermined ranking of the prediction accuracy for each of the eight DD logs when simultaneously generated. Considering that the conventional and DD logs are recorded more than 10,000 ft deep in the subsurface using logging tools that are run at different times in rugose boreholes for sensing the near-wellbore geologic formation, the data used in this predictive work is prone to noise and biases that tend to adversely affect the prediction performances of the proposed methods. In terms of normalized root-mean square error (Nrms error), the prediction performances of the second predictive method are 8.5% worse and 6.2% better for the conductivity and permittivity dispersion logs, respectively, as compared with those of the first predictive method. The third method has best prediction performance for permittivity dispersion logs, which is 0.089 in terms of the Nrms error.

Geochemistry & Geophysics↗

DOE OSTI2018

Impact of aerosols on ice crystal size

The interactions between aerosols and ice clouds represent one of the largest uncertainties in global radiative forcing from pre-industrial time to the present. In particular, the impact of aerosols on ice crystal effective radius (R ei ), which is a key parameter determining ice clouds' net radiative effect, is highly uncertain due to limited and conflicting observational evidence. Here we investigate the effects of aerosols on R ei under different meteorological conditions using 9-yearsatellite observations. We find that the responses of R ei to aerosol loadings are modulated by water vapor amount in conjunction with several other meteorological parameters. While there is a significant negative correlation between Rei and aerosol loading in moist conditions, consistent with the “Twomey effect” for liquid clouds, a strong positive correlation between the two occurs in dry conditions. Simulations based on a cloud parcel model suggest that water vapor modulates the relative importance of different ice nucleation modes, leading to the opposite aerosol impacts between moist and dry conditions. When ice clouds are decomposed into those generated from deep convection and formed in situ, the water vapor modulation remains in effect for both ice cloud types, although the sensitivities of R ei to aerosols differ noticeably between them due to distinct formation mechanisms. The water vapor modulation can largely explain the difference in the responses of R ei to aerosol loadings in various seasons. A proper representation of the water vapor modulation is essential for an accurate estimate of aerosol–cloud radiative forcing produced by ice clouds.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRSDate not supplied

GeoNEX-ML: A Machine Learning System for Earth Observations

Improved capabilities of earth monitoring satellites are enabling a wide range of studies on the environmental effects of climate change, often leveraging the recent advancements in machine learning. At the same time, the new capabilities, including higher spatial resolution and temporal frequency, are expanding the amount of data generated at exponential rates. At the NASA Earth eXchange (NEX), we build deep learning methods to learn from cross sensor satellite-based Earth observations for generating new datasets with efficient processing techniques. Using current generation geostationary satellites on NEX, we present an interchangeable set of machine models to perform spectral adjustment, physical model emulation, LEO-GEO emulation, and optical flow. These tools are used to generate consistent virtual observations across sensors, perform atmospheric correction and cloud detection, and estimate land surface temperature and atmospheric winds. This approach aims to improve the robustness of remotely sensed data processing by learning from diverse sets of observations while enabling near real-time and on-demand capabilities.

Geostationary↗

NASA NTRSDate not supplied

Atmospheric winds with deep optical flow

Atmospheric winds↗

DOE OSTI2018

Assessment of Multiple Monitoring Methods for Detection of Brine and CO2 Leakage

We assess the effectiveness of four surface-based geophysical methods and in situ monitoring of pressure and chemical changes for detecting brine and CO2 leakage from legacy wells into underground sources of drinking water (USDW) overlying a CO2 storage reservoir. This is the first such effort to evaluate and compare six monitoring methods for detection of CO2 leakage. This assessment, conducted in support of the USDOE National Risk Assessment Partnership (NRAP), utilizes 3000 synthetic data sets, generated in an uncertainty quantification (UQ) model framework. This framework combines the results of (1) a CO2 storage reservoir model of a hypothetical storage operation of 5 MT/yr for 50 years into the 1700-m deep Vedder Fm. at a site near Kimberlina, California with (2) wellbore leakage models of a legacy well and (3) multi-phase flow and reactive transport models of brine and CO2 plume migration in aquifer layers overlying the storage reservoir. Leakage drives subsurface changes, quantified by pressure plumes, CO2-saturation plumes, and total dissolved solids (TDS) plumes, the latter being related to changes in pH and species concentrations. Depending on the depth of leakage, aquifer permeability, and regional groundwater gradient, pressure plume volumes can become larger than the CO2 and TDS plumes; however, reaching detectable thresholds may take too long to be useful. For a single monitoring well located 500 m downgradient of a leaky legacy well, a depth of 570 m is found to be optimal for sampling pressure and chemical changes. We assess four surface-based geophysical monitoring methods: electrical resistivity tomography (ERT), magnetotellurics (MT), gravity, and seismic methods. Sensors cover a 4-km × 2-km area. Both ERT and MT methods are sensitive to changes in CO2 saturation and TDS concentration, whereas gravity and seismic methods detect changes in CO2 saturation. The six monitoring methods complement each other in space and time, and we find that a minimum of 20 kT of CO2 leakage is required for the leakage plume to be detected. Of the four surface-based geophysical methods, gravity monitoring appears to be the most sensitive to CO2 leaks, while the usefulness of in situ pressure and chemical monitoring will depend on the level of unfilterable background noise. Geophysical monitoring methods may detect a leak anywhere in a large area and in a large depth range, but the effectiveness of pressure and chemical monitoring depends on the number of monitoring wells, where those wells are located, and the depth of sampling. Surface-based geophysical monitoring methods are cost-effective alternatives to in situ pressure and chemical monitoring methods.

NRAP, CCS, monitoring, geophysical methods, gravit↗