Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “attributed graphs”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

ACES-GNN: can graph neural network learn to explain activity cliffs?

Graph Neural Networks (GNNs) have revolutionized molecular property prediction by leveraging graph-based representations, yet their opaque decision-making processes hinder broader adoption in drug discovery. This study introduces the Activity-Cliff-Explanation-Supervised GNN (ACES-GNN) framework, designed to simultaneously improve predictive accuracy and interpretability by integrating explanation supervision for activity cliffs (ACs) into GNN training. ACs, defined by structurally similar molecules with significant potency differences, pose challenges for traditional models due to their reliance on shared structural features. By aligning model attributions with chemist-friendly interpretations, the ACES-GNN framework bridges the gap between prediction and explanation. Validated across 30 pharmacological targets, ACES-GNN consistently enhances both predictive accuracy and attribution quality for ACs compared to unsupervised GNNs. Our results demonstrate a positive correlation between improved predictions and accurate explanations, offering a robust and adaptable framework to better understand and interpret ACs. This work underscores the potential of explanation-guided learning to advance interpretable artificial intelligence in molecular modeling and drug discovery.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1991

Molecular gas in luminous infrared galaxies

Radio observations of 60 bright IRAS galaxies with redshifts of 1500-25,000 km/sec are reported. Data obtained in the 1-0 line of CO using the 12-m NRAO radio telescope during 1985-1988 are presented in extensive tables, graphs, and line profiles and analyzed along with similar data on 29 less distant IRAS bright galaxies (Tinney et al., 1990). The galaxies are found to have H2 masses of (1-60) x 10 to the 9th solar mass and a mean ratio of H2 to warm dust of 540 + or - 290, corresponding to a total gas/dust ratio of 900-1100. The discrepancy between this value and that for the Galaxy (about 150) is tentatively attributed to the presence of undetected cold dust or errors in interpreting the IR data. The mechanisms which might be responsible for the high ratios of IR luminosity to H2 mass (2-220 solar luminosity per solar mass) are discussed.

Sanders, D. B.↗
NASA NTRS1987

Uranus satellites - Densities and composition

Homogeneous and core-differentiated silicate/ice models of the Uranian satellites Miranda, Ariel, Umbriel, Titania, and Oberon are examined in the light of imaging observations and mass and density determinations obtained during the Voyager 2 encounter with Uranus in January 1986. The data and model predictions are compared in extensive tables and graphs and discussed in detail. The mass fractions of silicates in Oberon and Titania are found to be between 0.42 and 0.65, about the same as the average for the satellites of Jupiter and Saturn but significantly higher than that for the smaller Saturnian satellites or that predicted by current solar-nebula models. It is suggested that the satellites formed by accretion of material from their primary planets' outer envelopes. The observed rock/ice fractions are attributed to solar-nebula CO and solid-organics abundances and to preferential dissolution of H2O in outer-envelope planetesimals.

Johnson, Torrence V.↗
DOE OSTI2025

Efficient Hamiltonian encoding algorithms for extracting quantum control mechanism as interfering pathway amplitudes in the Dyson series

Hamiltonian encoding is a methodology for revealing the mechanism behind the dynamics governing controlled quantum systems. In this paper, following Mitra and Rabitz \cite{abhra_1}, we define mechanism via pathways of eigenstates that describe the evolution of the system, where each pathway is associated with a complex-valued amplitude corresponding to a term in the Dyson series. The evolution of the system is determined by the constructive and destructive interference of these pathway amplitudes. Pathways with similar attributes can be grouped together into pathway classes. The amplitudes of pathway classes are computed by modulating the Hamiltonian matrix elements and decoding the subsequent evolution of the system rather than by direct computation of the individual terms in the Dyson series. The original implementation of Hamiltonian encoding was computationally intensive and became prohibitively expensive in large quantum systems. This paper presents two new encoding algorithms that calculate the amplitudes of pathway classes by using techniques from graph theory and algebraic topology to exploit patterns in the set of allowed transitions, greatly reducing the number of matrix elements that need to be modulated. These new algorithms provide an exponential decrease in both computation time and memory utilization with respect to the Hilbert space dimension of the system. To demonstrate the use of these techniques, they are applied to two illustrative state-to-state transition problems.

Abrams, Erez [Princeton University, Massachusetts ↗
DOE OSTI2020

Determining flow directions in river channel networks using planform morphology and topology

Abstract. The abundance of global, remotely sensed surface water observations has accelerated efforts toward characterizing and modeling how water moves across the Earth's surface through complex channel networks. In particular, deltas and braided river channel networks may contain thousands of links that route water, sediment, and nutrients across landscapes. In order to model flows through channel networks and characterize network structure, the direction of flow for each link within the network must be known. In this work, we propose a rapid, automatic, and objective method to identify flow directions for all links of a channel network using only remotely sensed imagery and knowledge of the network's inlet and outlet locations. We designed a suite of direction-predicting algorithms (DPAs), each of which exploits a particular morphologic characteristic of the channel network to provide a prediction of a link's flow direction. DPAs were chained together to create “recipes”, or algorithms that set all the flow directions of a channel network. Separate recipes were built for deltas and braided rivers and applied to seven delta and two braided river channel networks. Across all nine channel networks, the recipe-predicted flow directions agreed with expert judgement for 97 % of all tested links, and most disagreements were attributed to unusual channel network topologies that can easily be accounted for by pre-seeding critical links with known flow directions. Our results highlight the (non)universality of process–form relationships across deltas and braided rivers.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Deciphering the Solvation Structure of Aqueous ZnCl 2 Solutions from X-ray Absorption Spectra Using the Interpretable Graph Neural Network

Machine learning (ML) provides powerful pathways for predicting spectroscopic observables from atomic structures, but its broader impact depends on making model predictions interpretable in terms of physical and chemical principles. Here, we introduce a physics-guided graph neural network (GNN) model that predicts Zn K-edge X-ray spectroscopy (XAS) spectra of aqueous ZnCl 2 solutions. Training data are generated from ab initio XAS calculations on molecular dynamics snapshots obtained using a machine learning interatomic potential. The GNN reproduces experimental spectra across concentrations from dilute (<0.1 m) to highly concentrated (30 m, “water-in-salt”) regimes and scales efficiently to large, disordered liquid systems beyond the reach of conventional ab initio approaches. Gradient-based attribution analysis reveals that the model learns physically meaningful structure-spectrum relationships. Ligand-specific attributions reflect orbital hybridization patterns and the origin of the excitations derived from the density functional theory. Bond-length attributions recover spectral shifts consistent with multiple-scattering theory. Finally, this work bridges data-driven prediction with electronic-structure theory, establishing a general paradigm for interpretable ML that links atomic structure, electronic structure, and spectroscopic observables.

25 ENERGY STORAGE↗
NASA NTRS1983

Results of thermal environment measurements on the thermal cannister experiment and get away special enclosure

Because the OSS-1 pallet contained a variety of instruments with irregular surface geometry and properties which limited predictability, the total absorbed flux on thermal canister radiators was measured to determine heat rejection capability. The instrumentation and sensor cup design and locations for the thermal canister experiment are illustrated. Graphs show flux sensor history in hot and coal orbits. Kapton erosion is also considered. Results show that the flux levels measured in all STS attitudes are higher than predictions. In cold and moderate attitudes, flight results are a factor of two to three higher than predicts. In hot attitude, much better agreement occurred. It is concluded that in cold or moderate attitudes other sources may be contributing to added inputs (albedo, Earth shine, shuttle background). In hot attitude, smaller differences could be attributed to coatings assumptions or calculation uncertainty.

Ollendorf, S.↗
NASA NTRS1986

Comment on the paper 'On the influx of small comets into the earth's upper atmosphere. I - Observations'

The observations of transient decreases or holes in the EUV dayglow reported by Frank et al. (1986) and attributed to an influx of small comets into the earth atmosphere are discussed critically. The techniques used in acquiring and analyzing the observational data are examined, and it is argued that the decreases are probably instrument artifacts. A critique of the geophysical basis of the comet hypothesis is also included. In a reply by Frank et al., the instrument-artifact argument is rejected, in part on the basis of the statistical properties of the holes observed. Additional observational data are presented in graphs and dynamics Explorer 1 images are analyzed in detail.

Chubb, T. A.↗
NASA NTRS1984

Far-infrared and submillimeter observations of the low-luminosity protostars L1455 FIR and L1551 IRS 5 - The confinement of bipolar outflows

Far-IR observations of L1455 FIR and L1551 IRS 5 obtained in December, 1981, and in September, 1982, and January, 1983, respectively, using the H-1 and G-2 bolometer systems on the 91-cm telescope of the NASA Kuiper Aiborne Observatory are reported. Additional observations at 400 microns were obtained with UKIRT in November, 1981, (L1551) and with NASA IRTF in November, 1982 (L1455). The results are presented in tables and graphs, and L1455 and L1551 are compared with other objects in terms of central-star parameters, clouds, and overflows. The characteristics calculated for L1455 and L1551 include luminosity 1.5 and 23 solar luminosity, size less than 7 x 10 to the 16th cm, H2 density greater than or equal to 4 x 10 to the 5th and 1 x 10 to the 6th/cu cm, and 35-K dust cores of 0.2 and 0.7 solar mass, respectively. The bipolar appearance of the L1551 outflow is attributed to circumstellar material only.

Davidson, J. A.↗
NASA NTRS1984

The depletion of pre-interstellar matter and an examination of Ni II oscillator strengths

IUE absorption-line data obtained at 110-320 nm with resolution 10-30 pm toward the B-star component of the multiple system of Alpha Sco are presented in graphs and tables and used to characterize the circumstellar shell around the M1.5 Iab component. Resonant lines of Ni II are identified at strengths in excess of 4 pm and fit to a growth curve constructed using the Fe II, Si II, and Mn II lines, and the results are found to confirm the oscillator strengths determined by Kurucz and Peytremann (1975) to within 50 percent. Shell depletions 2-20 times smaller than those for dense clouds such as the surrounding Oph complex are calculated and attributed to the formation of circumstellar dust. The implications of the results for the condensation of various elements in dense and diffuse clouds are explored.

Cardelli, J. A.↗
NASA NTRS2001

Evidence for yellow light suppression of lettuce growth

Researchers studying plant growth under different lamp types often attribute differences in growth to a blue light response. Lettuce plants were grown in six blue light treatments comprising five blue light fractions (0, 2, 6% from high-pressure sodium [HPS] lamps and 6, 12, 26% from metal halide [MH] lamps). Lettuce chlorophyll concentration, dry mass, leaf area and specific leaf area under the HPS and MH 6% blue were significantly different, suggesting wavelengths other than blue and red affected plant growth. Results were reproducible in two replicate studies at each of two photosynthetic photon fluxes, 200 and 500 mumol m-2 s-1. We graphed the data against absolute blue light, phytochrome photoequilibrium, phototropic blue, UV, red:far red, blue:red, blue: far red and 'yellow' light fraction. Only the 'yellow' wavelength range (580-600 nm) explained the differences between the two lamp types.

Non-NASA Center↗
NASA NTRS1988

Infrared observations of eclipses of Io, its thermophysical parameters, and the thermal radiation of the Loki volcano and environs

Observations of Io during eclipses by Jupiter in 1981-1984 are reported. Data obtained at 3.45-30 microns using bolometer system No. 1 on the 3-m IRTF telescope at Mauna Kea are presented in extensive tables and graphs and analyzed by means of least-squares fitting of thermophysical models to the eclipse cooling and heating curves, thermal-radiation calculations for the Io volcanoes, and comparison with Voyager data. Best fits are obtained for a model comprising (1) a bright region with a vertically inhomogeneous surface and (2) a dark vertically homogeneous region with thermal inertia only about 0.1 times that of (1). Little evidence of volcanic-flux variability during the period is found, and the majority (but not all) of the excess thermal IR radiation in the sub-Jovian hemisphere is attributed to the Loki volcano and its lava lake.

Sinton, William M.↗
DOE OSTI2026

Scaling kinetic Monte-Carlo simulations of grain growth with combined convolutional and graph neural networks

Graph neural networks (GNN) have emerged as a promising machine learning method for microstructure simulations such as grain growth. However, accurate modeling of realistic grain boundary networks requires large simulation cells, which GNN has difficulty scaling up to. To alleviate the computational costs and memory footprint of GNN, we suggest a hybrid architecture combining a convolutional neural network (CNN) based bijective autoencoder to compress the spatial dimensions, and a GNN that evolves the microstructure in the latent space of reduced spatial sizes. Our results demonstrate that the new design significantly reduces computational costs with using fewer message passing layer (from 12 down to 3) compared with GNN alone. The reduction in computational cost becomes more pronounced as the spatial size increases, indicating strong computational scalability. For the largest mesh evaluated (160 3 ), our method reduces memory usage and runtime in inference by 117× and 115×, respectively, compared with GNN-only baseline. More importantly, it shows higher accuracy and stronger spatiotemporal capability than the GNN-only baseline, especially in long-term testing. Such combination of scalability and accuracy is essential for simulating realistic material microstructures over extended time scales. The improvements can be attributed to the bijective autoencoder’s ability to compress information losslessly from spatial domain into a high dimensional feature space, thereby producing more expressive latent features for the GNN to learn from, while also contributing its own spatiotemporal modeling capability. Training data are generated from stochastic grain growth simulations, providing realistic variability for learning robust microstructure evolution. Comprehensive system validation confirms that the model is accurate, robust, and scalable.

36 MATERIALS SCIENCE↗
NASA NTRS1986

Transonic airfoil calculations including wind tunnel wall-interference effects

The results of Reynolds-averaged time-dependent inviscid and turbulent compressible Navier-Stokes computations using the implicit finite-difference approach of Steger (1978), modified by incorporating a pressure boundary condition, (PBC) to account for wall interference are compared with experimental data on a NACA 64A010 airfoil (Johnson and Bachalo, 1980) in graphs and briefly characterized. The computational approach is the same as that used by King and Johnson (1980), but a 137 x 50 mesh is used instead of a 97 x 35 mesh, and special care is taken in resolving the nose, shock, and trailing-edge regions. Imposition of PBC is shown to improve significantly the accuracy of the computations for the flowfield on the upper surface of the airfoil, shifting the shock forward to its experimentally measured position in the case of turbulent flow. The failure of the method, even with PBC, to match the experimental shock location in the case of a flow with a separation bubble is attributed to inadequacies in the algebraic turbulence model employed (Baldwin and Lomax, 1978).

King, L. S.↗
DOE OSTI2020

Machine Learning Approaches for Nuclear Material Accounting Data from Irradiation and Reprocessing

We are currently exploring data analysis methods for their ability to strengthen the synthesis and evaluation of information generated within domestic and international safeguard regimes. Safeguard data typically includes rich heterogenous datasets amenable to advanced data analytics methods, such as machine learning. We have converted transactional data containing both numerical and categorical attributes from a domestic nuclear material control and accountability (NMC&A) system into a low-dimensional numerical structure via linear principal component analysis. This data representation allows for global structure discovery via cluster analysis, which can characterize the typical behavior of each of the primary types of transaction events. Furthermore, the structure of the top principal components captures the “typical behavior” of the data and is thus amenable to anomaly detection through statistical hypothesis testing. We explored this capability by generating erroneous permutations of the data and computing the Q-residual quantity increase associated with the information loss when these data are projected into the low-dimensional principal component analysis space representative of typical transactions.Future work will focus on identifying data transformations (e.g., graph networks) that more closely align with the inherent structure of these transactions to explain more salient information and disambiguate the underlying nuclear process—in this case, irradiation and reprocessing—from artifacts of NMC&A system transactional record keeping data.

Drescher, Adam↗
NASA NTRS1984

Modulation of the cytosolic androgen receptor in striated muscle by sex steroids

The effects of orchiectomy (GDX) and of subsequent administration of testosterone propionate (TP) or 17(beta)-estradiol (E2) on the maximum binding (Bmax) and apparent Kd of the cytosolic androgen receptor in levator ani (LA) and skeletal muscles of adult male Sprague-Dawley rats are investigated experimentally. The results are presented in graphs and discussed. In LA, BMAX is found to rise from a control level of 2.5 fmol/mg protein to 280, 600, 478, and 133 percent of control at 12 h, 14 d, 30 d, and 44 d after GDX, respectively, while Kd increased only insignificantly (from 680 to 960 fM); Bmax is held at control levels for 6 h by cycloheximide given at GDX, is unaffected by TP given at 30 d, and is further increased (by 480 percent at 44 d) by administration of E2 at 30 d. Bmax in skeletal muscles is found to increase to 139, 212, 220, and 158 percent of control at 12 h, 14 d, 30 d, and 44 d, respectively; Bmax is returned to control at 44 d by TP at 30 d but is not affected by E2. The effect of E2 in LA is attributed to either induction of the cytosolic receptor or a decreased rate of receptor degradation.

Rance, N. E.↗
NASA NTRS2004

Effects of Various Wavelength Ranges of Vacuum Ultraviolet Radiation on Teflon FEP Film Investigated

Teflon Fluorinated Ethylene Propylene (FTP) films (DuPont) have been widely used for spacecraft thermal control and have been observed to become embrittled and cracked upon exposure to the space environment. This degradation has been attributed to a synergistic combination of radiation and thermal effects. A research study was undertaken at the NASA Glenn Research Center to examine the effects of different wavelength ranges of vacuum ultraviolet (VUV) radiation on the degradation of the mechanical properties of FEP. This will contribute to an overall understanding of space radiation effects on Teflon FEP, and will provide information necessary to determine appropriate techniques for using laboratory tests to estimate space VUV degradation. Research was conducted using inhouse facilities at Glenn and was carried out, in part, through a grant with the Cleveland State University. Samples of Teflon FEP film of 50.8 microns thickness were exposed to radiation from a VUV lamp from beneath different cover windows to provide different exposure wavelength ranges: MgF2 (115 to 400 nm), crystalline quartz (140 to 400 nm), and fused silica (FS, 155 to 400 nm). Following exposure, FEP film specimens were tensile tested to determine the ultimate tensile strength and elongation at failure as a function of the exposure duration for each wavelength range. The graphs show the effect of ultraviolet exposure on the mechanical properties of the FEP samples.

Dever, Joyce A.↗
NASA NTRS2001

Effect of Air and Vacuum Storage on the Degradation of X-Ray-Exposed Aluminized-Teflon Investigated

Metalized Teflon FEP (fluorinated ethylene propylene, DuPont), a common thermal control material, has been found to degrade in the low-Earth-orbit space environment. The aluminized-FEP (Al-FEP) exterior layer on the Hubble Space Telescope has become extremely embrittled, with extensive cracking occurring on all sides of the telescope. This embrittlement has been primarily attributed to radiation exposure (x-rays from solar flares, electron/proton radiation, and possibly near-ultraviolet radiation) combined with thermal cycling. Limited samples of FEP tested after long-term exposure to low Earth orbit on the Hubble Space Telescope and on the Long Duration Exposure Facility indicated that there might be continued degradation in tensile properties over time. An investigation was conducted at the NASA Glenn Research Center to evaluate the effect of air and vacuum storage on the mechanical properties of x-ray-exposed FEP. Aluminized-FEP (5-mil-thick) tensile samples were x-ray exposed with 15.3-kV copper xrays for 2 hr, reducing the percent elongation to failure by approximately 50 percent in comparison to that for pristine Al-FEP. X-ray-exposed samples were stored in air or under vacuum for various time periods to see the effect of storage on tensile properties. Tensile results indicated that samples stored in air had larger decreases in tensile properties than samples stored under vacuum had, as seen in the graph. Samples stored under vacuum (for up to 400 hr) showed no further decrease in tensile properties over time, whereas samples stored in air (for up to 900 hr) appeared to show decreases in tensile properties over time. X-ray-exposed samples stored in air developed a hazy appearance in the exposed area, as seen in the photographs. When the source of the haziness was evaluated using scanning electron microscopy and atomic force microscopy, it was found to reside at the Al/FEP interface as witnessed by an increased surface roughness of the aluminized side of the material and a dramatic decrease in the adhesion between the Al and FEP. Optical properties of air-stored irradiated samples showed an increase in the diffuse reflectance, which is consistent with observed roughening that was characterized by AFM. These findings indicate that air exposure helps degrade x-ray-irradiated FEP. These results indicate that proper sample handling and storage is necessary with space-retrieved materials and with those exposed to ground-based irradiation simulation exposures.

deGroh, Kim K.↗