SEARCH · Engineering Papers

Results for “model discovery”

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 955 records · Page 53

All sources NASA NTRS DOE OSTI

DOE OSTI2022

On-the-fly autonomous control of neutron diffraction via physics-informed Bayesian active learning

We demonstrate the first live, autonomous control over neutron diffraction experiments by developing and deploying ANDiE: the autonomous neutron diffraction explorer. Neutron scattering is a unique and versatile characterization technique for probing the magnetic structure and behavior of materials. However, instruments at neutron scattering facilities in the world is limited, and instruments at such facilities are perennially oversubscribed. We demonstrate a significant reduction in experimental time required for neutron diffraction experiments by implementation of autonomous navigation of measurement parameter space through machine learning. Prior scientific knowledge and Bayesian active learning are used to dynamically steer the sequence of measurements. We show that ANDiE can experimentally determine the magnetic ordering transition of both MnO and Fe 1.09 Te all while providing a fivefold enhancement in measurement efficiency. Furthermore, in a hypothesis testing post-processing step, ANDiE can determine transition behavior from a set of possible physical models. ANDiE's active learning approach is broadly applicable to a variety of neutron-based experiments and can open the door for neutron scattering as a tool of accelerated materials discovery.

36 MATERIALS SCIENCE↗

DOE OSTI2023

Benchmarking the PCMCI Causal Discovery Algorithm for Spatiotemporal Systems

Causal discovery algorithms construct hypothesized causal graphs that depict causal dependencies among variables in observational data. While powerful, the accuracy of these algorithms is highly sensitive to the underlying dynamics of the system in ways that have not been fully characterized in the literature. In this report, we benchmark the PCMCI causal discovery algorithm in its application to gridded spatiotemporal systems. Effectively computing grid-level causal graphs on large grids will enable analysis of the causal impacts of transient and mobile spatial phenomena in large systems, such as the Earth’s climate. We evaluate the performance of PCMCI with a set of structural causal models, using simulated spatial vector autoregressive processes in one- and two-dimensions. We develop computational and analytical tools for characterizing these processes and their associated causal graphs. Our findings suggest that direct application of PCMCI is not suitable for the analysis of dynamical spatiotemporal gridded systems, such as climatological data, without significant preprocessing and downscaling of the data. PCMCI requires unrealistic sample sizes to achieve acceptable performance on even modestly sized problems and suffers from a notable curse of dimensionality. This work suggests that, even under generous structural assumptions, significant additional algorithmic improvements are needed before causal discovery algorithms can be reliably applied to grid-level outputs of earth system models.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRS2002

Laboratory Astrophysics Needs of the Herschel Space Observatory

The science teams of the Herschel Space Observatory have identified a number of areas where laboratory study is required for proper interpretation of Herschel observational data. The most critical is the collection and compilation of laboratory data on spectral line frequencies, transition probabilities and energy levels for the known astrophysical atomic and molecular species in 670 to 57 micron wavelength range of Herschel. The second most critical need is the compilation of collisional excitation cross sections for the species known to dominate the energy balance in the ISM and the temperature dependent chemical reaction rates. On the theoretical front, chemical and radiative transfer models need to be prepared in advance to assess calibration and identify instrument anomalies. In the next few years there will be a need to incorporate spectroscopists and theoretical chemists into teams of astronomers so that the spectroscopic surveys planned can he properly calibrated and rapidly interpreted once the data becomes available. The science teams have also noted that the enormous prospects for molecular discovery will be greatly handicapped by the nearly complete lack of spectroscopic data for anything not already well known in the ISM. As a minimum, molecular species predicted to exist by chemical models should be subjected to detailed laboratory study to ensure conclusive detections. This has the greatest impact on any astrobiology program that might be proposed for Herschel. Without a significant amount of laboratory work in the very near future Herschel will not be prepared for many planned observations, much less addressing the open questions in molecular astrophysics.

Pearson, J. C.↗

NASA NTRS2019

A 3 Gyr White Dwarf with Warm Dust Discovered via the Backyard Worlds: Planet 9 Citizen Science Project

Infrared excesses due to dusty disks have been observed orbiting white dwarfs with effective temperatures between 7200 and 25,000 K, suggesting that the rate of tidal disruption of minor bodies massive enough to create a coherent disk declines sharply beyond 1 Gyr after white dwarf formation. We report the discovery that the candidate white dwarf LSPM J0207+3331, via the Backyard Worlds: Planet 9 citizen science project and Keck Observatory follow-up spectroscopy, is hydrogen dominated with a luminous compact disk (L IR/L star = 14%) and an effective temperature nearly 1000 K cooler than any known white dwarf with an infrared excess. The discovery of this object places the latest time for large-scale tidal disruption events to occur at ~3 Gyr past the formation of the host white dwarf, making new demands of dynamical models for planetesimal perturbation and disruption around post-main-sequence planetary systems. Curiously, the mid-infrared photometry of the disk cannot be fully explained by a geometrically thin, optically thick dust disk as seen for other dusty white dwarfs, but requires a second ring of dust near the white dwarf's Roche radius. In the process of confirming this discovery, we found that careful measurements of WISE source positions can reveal when infrared excesses for white dwarfs are co-moving with their hosts, helping distinguish them from confusion noise.

Debes, John H.↗

DOE OSTI2026

Adaptable Standards for Discovery, Access, and Usability of Oak Ridge National Laboratory’s Data Portals and Catalogs

Oak Ridge National Laboratory (ORNL) is leveraging its established capabilities and subject matter expertise in data curation, governance, management, national security, and risk assessment and mitigation to support the US Department of Energy (DOE) Grid Modernization Initiative. Using standards modeled by the National Institute of Standards and Technology (NIST), the Data Curation Network (DCN), the Oak Ridge Leadership Computing Facility (OLCF), and other leading organizations in the fields of energy research, high-performance computing, and national and homeland security, ORNL seeks to provide a federated approach to research data discovery, use, and interoperability.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

DOE OSTI2021

Inverse design of nanoporous crystalline reticular materials with deep generative models

Reticular frameworks are crystalline porous materials that form via the self-assembly of molecular building blocks in different topologies, with many having desirable properties for gas storage, separation, catalysis, biomedical applications and so on. The notable variety of building blocks makes reticular chemistry both promising and challenging for prospective materials design. Here we propose an automated nanoporous materials discovery platform powered by a supramolecular variational autoencoder for the generative design of reticular materials. We demonstrate the automated design process with a class of metal–organic framework (MOF) structures and the goal of separating carbon dioxide from natural gas or flue gas. Our model shows high fidelity in capturing MOF structural features. Here, we show that the autoencoder has a promising optimization capability when jointly trained with multiple top adsorbent candidates identified for superior gas separation. MOFs discovered here are strongly competitive against some of the best-performing MOFs/zeolites ever reported.

36 MATERIALS SCIENCE↗

DOE OSTI2023

Defect graph neural networks for materials discovery in high-temperature clean-energy applications

We present a graph neural network approach that fully automates the prediction of defect formation enthalpies for any crystallographic site from the ideal crystal structure, without the need to create defected atomic structure models as input. Here we used density functional theory reference data for vacancy defects in oxides, to train a defect graph neural network (dGNN) model that replaces the density functional theory supercell relaxations otherwise required for each symmetrically unique crystal site. Interfaced with thermodynamic calculations of reduction entropies and associated free energies, the dGNN model is applied to the screening of oxides in the Materials Project database, connecting the zero-kelvin defect enthalpies to high-temperature process conditions relevant for solar thermochemical hydrogen production and other energy applications. The dGNN approach is applicable to arbitrary structures with an accuracy limited principally by the amount and diversity of the training data, and it is generalizable to other defect types and advanced graph convolution architectures. In conclusion, it will help to tackle future materials discovery problems in clean energy and beyond.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Towards Automated Design of Corrosion Resistant Alloy Coatings with an Autonomous Scanning Droplet Cell

We present an autonomous scanning droplet cell platform designed for on-demand alloy electrodeposition and real-time electrochemical characterization. Automation and machine learning are currently driving rapid innovation in high-throughput and autonomous materials design and discovery. We present two alloy design vignettes: one focusing on a multi-objective corrosion resistant alloy optimization and a study highlighting the complexity of the multimodal characterization needed to provide insight into the underlying structural and chemical factors that drive observed material behavior. This motivates a close coupling between autonomous research platforms and scientific machine learning methodology that blends mechanistic physical models and black box machine learning models. Lastly, we reflect on our early efforts in on-demand alloy deposition, highlighting some of the challenges. This emerging research area presents new opportunities to accelerate materials synthesis, evaluation, and hence discovery and design.

36 MATERIALS SCIENCE↗

DOE OSTI2020

TOI-1235 b: A Keystone Super-Earth for Testing Radius Valley Emergence Models around Early M Dwarfs

Small planets on close-in orbits tend to exhibit envelope mass fractions of either effectively zero or up to a few percent depending on their size and orbital period. Models of thermally driven atmospheric mass loss and of terrestrial planet formation in a gas-poor environment make distinct predictions regarding the location of this rocky/nonrocky transition in period–radius space. Here we present the confirmation of TOI-1235 b (P = 3.44 days, r{sub p}=1.738{sub −0.076}{sup +0.087} R{sub ⊕}), a planet whose size and period are intermediate between the competing model predictions, thus making the system an important test case for emergence models of the rocky/nonrocky transition around early M dwarfs (R {sub s} = 0.630 ± 0.015 R{sub ⊙}, M {sub s} = 0.640 ± 0.016 M{sub ⊙}). We confirm the TESS planet discovery using reconnaissance spectroscopy, ground-based photometry, high-resolution imaging, and a set of 38 precise radial velocities (RVs) from HARPS-N and HIRES. We measure a planet mass of 6.91{sub −0.85}{sup +0.75} M{sub ⊕}, which implies an iron core mass fraction of 20{sub −12}{sup +15}% in the absence of a gaseous envelope. The bulk composition of TOI-1235 b is therefore consistent with being Earth-like, and we constrain an H/He envelope mass fraction to be <0.5% at 90% confidence. Our results are consistent with model predictions from thermally driven atmospheric mass loss but not with gas-poor formation, suggesting that the former class of processes remains efficient at sculpting close-in planets around early M dwarfs. Our RV analysis also reveals a strong periodicity close to the first harmonic of the photometrically determined stellar rotation period that we treat as stellar activity, despite other lines of evidence favoring a planetary origin (P=21.8{sub −0.8}{sup +0.9} days, m{sub p}sin i=13.0{sub −5.3}{sup +3.8} M{sub ⊕}) that cannot be firmly ruled out by our data.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS2012

FUV Emission from AGB Stars: Modeling Accretion Activity Associated with a Binary Companion

It is widely believed that the late stages of evolution for Asymptotic Giant Branch (AGB) stars are influenced by the presence of binary companions. Unfortunately, there is a lack of direct observational evidence of binarity. However, more recently, strong indirect evidence comes from the discovery of UV emission in a subsample of these objects (fuvAGB stars). AGB stars are comparatively cool objects (< or =3000 K), thus their fluxes falls off drastically for wavelengths 3000 Angstroms and shorter. Therefore, ultraviolet observations offer an important, new technique for detecting the binary companions and/or associated accretion activity. We develop new models of UV emission from fuvAGB stars constrained by GALEX photometry and spectroscopy of these objects. We compare the GALEX UV grism spectra of the AGB M7 star EY Hya to predictions using the spectral synthesis code Cloudy, specifically investigating the ultraviolet wavelength range (1344-2831 Angstroms). We investigate models composed of contributions from a photoionized "hot spot" due to accretion activity around the companion, and "chromospheric" emission from collisionally ionized plasma, to fit the UV observations.

ultraviolet emission↗

DOE OSTI2020

Imaging grain microstructure in a model ceramic energy material with optically generated coherent acoustic phonons

Characterization of microstructure, chemistry and function of advanced energy materials remains a challenge for instrumentation science with tremendous opportunity for new discoveries. This active area of research is starting to make considerable strides with examples ranging from methodologies that employ bright X-rays, electron microscopy and optical spectroscopy. However, further development of instruments, capable of multimodal measurements, is necessary to fully reveal complex microstructure evolution in realistic environments and dynamical systems that are far from equilibrium. In this regard, laser-based instruments have a unique advantage as multiple measurement methodologies are easily combined into a single instrument using standard optics. A new pump-probe method that uses optically generated acoustic phonons is expanding standard optical characterization by providing depth resolved information. Here we report on a noteworthy extension of this pump-probe method to image grain microstructure in ceria. Rich information regarding the orientation of individual crystallites is obtained by monitoring acoustic phonon velocities and noting how the polarization of the probe beam influences the detected signal amplitude. When paired with other optical microscopies, this methodology will provide new perspectives for optical characterization of ceramic materials.

36 MATERIALS SCIENCE↗

NASA NTRS2012

Resolving the Origin of the Diffuse Soft X-ray Background

In January 1993, the Diffuse X-ray Spectrometer (DXS) measured the first high-resolution spectrum of the diffuse soft X-ray background between 44-80A. A line-dominated spectrum characteristic of a 10(exp 6)K collisionally ionized plasma' was expected but while the observed spectrum was clearly line-dominated, no model would fit. Then in 2003 the Cosmic Hot Interstellar Plasma Spectrometer (CHIPS) launched and observed the diffuse extreme-ultraviolet (EUV) spectrum between 90- 265A. Although many emission lines were again expected; only Fe IX at 171.1A was detected. The discovery of X-rays from comets led to the realization that heavy ions (Z=6-28) in the solar wind will emit soft X-rays as the ions interact via charge exchange with neutral atoms in the heliosphere and geocorona. Using a new model for solar wind charge exchange (SWCX) emission, we show that the diffuse soft X-ray background can be understood as a combination of emission from charge exchange onto the slow and fast solar wind together with a more distant and diffuse hot (10(exp 6)K) plasma.

Smith, Randall K.↗

DOE OSTI2025

Surrogate Neural Architecture Codesign Package (SNAC-Pack)

Neural Architecture Search is a powerful approach for automating model design, but existing methods struggle to accurately optimize for real hardware performance, often relying on proxy metrics such as bit operations. We present Surrogate Neural Architecture Codesign Package (SNAC-Pack), an integrated framework that automates the discovery and optimization of neural networks focusing on FPGA deployment. SNAC-Pack combines Neural Architecture Codesign's multi-stage search capabilities with the Resource Utilization and Latency Estimator, enabling multi-objective optimization across accuracy, FPGA resource utilization, and latency without requiring time-intensive synthesis for each candidate model. We demonstrate SNAC-Pack on a high energy physics jet classification task, achieving 63.84% accuracy with resource estimation. When synthesized on a Xilinx Virtex UltraScale+ VU13P FPGA, the SNAC-Pack model matches baseline accuracy while maintaining comparable resource utilization to models optimized using traditional BOPs metrics. This work demonstrates the potential of hardware-aware neural architecture search for resource-constrained deployments and provides an open-source framework for automating the design of efficient FPGA-accelerated models.

Weitz, Jason [UC, San Diego] (ORCID:00090004631535↗

NASA NTRSDate not supplied

Ice Growth Model for Polar Water Capture Systems

In-situ resource utilization (ISRU) is essential to NASA’s goal of sustainable and cost-effective missions to the Moon and beyond. Water has become a particular source of interest since its discovery in the permanently shadowed regions (PSRs) of the Moon. This water exists as ice bound within the regolith, so one mode of capture is heating to sublimation temperatures and capturing at very low pressure. It then can be stored in ice form and transported out of the PSR and to a sunlit ridge, where processes such as electrolysis can be performed[2]. An understanding of frost growth dynamics under rarefied conditions is necessary for successful water capture and storage due to transient thermal properties of frost. A Diffusion-Limited Aggregation (DLA) approach was taken to analytically model bulk thermal conductivity changes of the growing frost layer under varying wall temperatures and frost layer porosities. The model utilizes well-known correlations for effective thermal conductivity and records the values as the frost layer thickness increases. This thermal conductivity can be applied to a known temperature change and be used to evaluate heat flux through the combined frost layer and tank wall for either a flat plate or cylindrical geometry. An experiment is described that will evaluate this model using one-dimensional frost growth on a horizontal, flat plate under low-pressure. The ice growth will be measured using a millimeter-scale photogrammetry camera system and a load cell measured transient mass growth.

In situ Resource Utilization↗

NASA NTRSDate not supplied

Ice Growth Model for Polar Water Capture Systems

ice growth↗

NASA NTRSDate not supplied

ES2Vec: Earth Science Metadata Suggestions and Analogical Reasoning

As the volume of text-based Earth science research grows, it is increasingly possible to discover latent relationships in the literature. However, traditional methodologies are restricted by limited computational capabilities and intractable problem spaces. Advancements in natural language processing (NLP) have allowed us to use an extensive Earth science corpus to create a domain-specific word vector model, Es2Vec, which we have used to surface latent relationships between Earth science concepts and generate improved keyword tags. Earth science metadata keyword assignment is a challenging problem. Dataset curators select appropriate keywords from the Global Change Master Directory (GCMD) set of keywords. The keywords an are integral part of the search and discovery of these datasets. Hence, the selection of keywords is crucial to increasing the discoverability of datasets. Utilizing machine learning techniques, we provide users with automated keyword suggestions to complement manual selection. We trained a machine learning model that leverages the semantic embedding ability of Word2Vec models to process abstracts and suggest relevant keywords. A user interface tool we built to assist data curators in the assignment of such keywords is also described.

word vectors↗

DOE OSTI2025

Bridging nuclear physics across energy scales: from neutrinoless double-beta decay to high-energy heavy-ion collisions

This paper exemplifies how connecting methods at disparate energy scales can illuminate fundamental questions. By demonstrating that nuclear wave function properties governing rare decay processes also influence collective behavior at extreme temperatures and densities, the authors have opened a new pathway for constraining physics beyond the Standard Model. With multiple ton-scale 0νββ experiments under construction, any method reducing NME uncertainties will directly impact our ability to interpret discoveries or constrain neutrino properties. The general principle—that collective phenomena in high-energy collisions can illuminate subtle features of many-body correlations in the colliding nuclei—may find applications across nuclear and particle physics. Furthermore, this intersection of nuclear structure theory, heavy-ion physics, and fundamental symmetry tests represents fertile ground for future discoveries in modern physics.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2021

Identification of earth-abundant materials for selective dehydrogenation of light alkanes to olefins

Catalytic alkane dehydrogenation is of considerable importance in the synthesis of olefins industrially. However, discovery of highly active, selective, and stable heterogeneous catalysts to replace noble metal Pt remains challenging. By combining descriptor-based microkinetic modeling, high-throughput computations, machine-learning concepts, and experiments, we efficiently evaluated 1,998 bimetallic alloys and successfully identified Ni 3 Mo as one of the most promising catalysts in selective ethane dehydrogenation. This work will open new possibilities of using earth-abundant materials as catalysts for essential heterogeneous catalytic reactions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗