Search NASA⌕ Search

SEARCH · Search NASA

Results for “map interpretation”

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 1,693 records · Page 94
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Exchange of Ions across the TiN/TaO x Interface during Electroformation of TaO x -Based Resistive Switching Devices

The valence change model describes the resistive switching in metal oxide-based devices as due to electroreduction of the oxide and subsequent electromigration of oxygen vacancies. Here, we present cross-sectional X-ray energy-dispersive spectroscopy elemental maps of Ta, O, N, and Ti in electroformed TiN/TaO 2.0 /TiN structures. O, N, and Ti were exchanged between the anode and the functional oxide in devices formed at high power (~1 mW), but the exchange was below the detection limit at low power (<0.5 mW). All structures exhibit a similar Ta-enriched and O-depleted filament formed by the elemental segregation in the functional oxide by the temperature gradient. The elemental interchange is interpreted as due to Fick’s diffusion caused by high temperatures in the gap of the filament and is not an essential part of electroformation.

36 MATERIALS SCIENCE↗
DOE OSTI2024

DS-GL: Advancing Graph Learning via Harnessing the Power of Nature within Dynamic Systems

With the rapid digitization of the world, an increasing number of real-world applications are turning to nonEuclidean data, modeled as graphs. Due to their intrinsic high complexity and irregularity, learning from graph data demands tremendous computational power. Recently, CMOS-compatible Ising machines, i.e., dynamic systems composed of CMOS components, have emerged as a new approach that harnesses the inherent power of natural annealing within dynamic systems to efficiently resolve binary optimization problems and have been adopted for traditional graph computation, such as max-cut. However, when performing complex Graph Learning (GL) tasks, Ising machines face significant hurdles: (i) they are inherently binary and thus ill-suited for real-valued problems; (ii) their expensive all-to-all coupling network that guarantees effective natural annealing poses daunting scalability concerns. To address these challenges, this paper proposes a nature-powered graph learning framework dubbed DS-GL, which is the first effort to transform the process of solving graph learning problems into the natural annealing process within a parameterized dynamic system embodied as a CMOS chip. To tackle the two major hurdles, DS-GL first augments the Ising machine architecture to modify the self-reaction term of its Hamiltonian function from linear to quadratic, effectively serving as an energy regulator. This adjustment maintains the system’s original physical interpretation while enabling it to process continuous, real-valued data. Second, to address the scaling issue, DS-GL further upgrades the real-valued dense Ising machine by decomposing it into a mesh-based multi-PE dynamic system that supports efficient distributed spatial-temporal co-annealing across different PEs through sparse interconnects. By exploiting the inherent sparsity and component structures in real-world graphs, DS-GL is able to map complex graph learning tasks onto the scalable dynamic system while maintaining high accuracy. Evaluations with three diverse GL applications across six real-world datasets, including traffic flow and COVID-19 prediction, show that DS-GL can deliver from 102× to 106× speedups and 500× energy reduction over Graph Neural Networks on GPUs, with 5% - 20% accuracy enhancement.

Song, Ruibing↗
DOE OSTI2023

Convolutional Neural Networks for Problems in Transport Phenomena: A Theoretical Minimum

Convolutional neural network (CNN), a deep learning algorithm, has gained popularity in technological applications that rely on interpreting images (typically, an image is a 2D field of pixels). Transport phenomena is the science of studying different fields representing mass, momentum, or heat transfer. Some of the common fields are species concentration, fluid velocity, pressure, and temperature. Each of these fields can be expressed as an image(s). Consequently, CNNs can be leveraged to solve specific scientific problems in transport phenomena. Herein, we show that such problems can be grouped into three basic categories: (a) mapping a field to a descriptor (b) mapping a field to another field, and (c) mapping a descriptor to a field. After reviewing the representative transport phenomena literature for each of these categories, we illustrate the necessary steps for constructing appropriate CNN solutions using sessile liquid drops as an exemplar problem. If sufficient training data is available, CNNs can considerably speed up the solution of the corresponding problems. Finally, the present discussion is meant to be minimalistic such that readers can easily identify the transport phenomena problems where CNNs can be useful as well as construct and/or assess such solutions.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1972

Study of the application of ERTS-A imagery to fracture-related mine safety hazards in the coal mining industry

The author has identified the following significant results. Various data compilation and analysis activities in support of ERTS-1 imagery interpretation are in progress or are completed. These include the compilation of mine accident data, areas of mine roof instability and the analysis of high altitude color infrared photography and low altitude color and color infrared photography which was acquired by NASA in support of the project. The photography reveals that many fracture lineaments are detectable through a varied thickness of glacial till. These data will be compiled on a series of 1:250,000 scale base maps and evaluated for a correlation between fracture zones and mine accidents and rooffalls. Due to high occurrence of cloud cover in the project area and to the delay in imagery shipments, little progress has been made in the analysis of ERTS-1 imagery.

Wier, C. E.↗
NASA NTRS1989

Cloud cover determination in polar regions from satellite imagery

The objectives are to develop a suitable validation data set for evaluating the effectiveness of the International Satellite Cloud Climatology Project (ISCCP) algorithm for cloud retrieval in polar regions, to identify limitations of current procedures and to explore potential means to remedy them using textural classifiers, and to compare synoptic cloud data from model runs with observations. Toward the first goal, a polar data set consisting of visible, thermal, and passive microwave data was developed. The AVHRR and SMMR data were digitally merged to a polar stereographic projection with an effective pixel size of 5 sq km. With this data set, two unconventional methods of classifying the imagery for the analysis of polar clouds and surfaces were examined: one based on fuzzy sets theory and another based on a trained neural network. An algorithm for cloud detection was developed from an early test version of the ISCCP algorithm. This algorithm includes the identification of surface types with passive microwave, then temporal tests at each pixel location in the cloud detection phase. Cloud maps and clear sky radiance composites for 5 day periods are produced. Algorithm testing and validation was done with both actural AVHRR/SMMR data, and simulated imagery. From this point in the algorithm, groups of cloud pixels are examined for their spectral and textural characteristics, and a procedure is developed for the analysis of cloud patterns utilizing albedo, IR temperature, and texture. In a completion of earlier work, empirical analyses of arctic cloud cover were explored through manual interpretations of DMSP imagery and compared to U.S. Air Force 3D-nephanalysis. Comparisons of observed cloudiness from existing climatologies to patterns computed by the GISS climate model were also made.

Barry, R. G.↗
NASA NTRS2009

Chapter 13. Atmospheric Dynamics and Meteorology

Titan, after Venus, is the second example in the solar system of an atmosphere with a global cyclostrophic circulation, but in this case a circulation that has a strong seasonal modulation in the middle atmosphere. Direct measurement of Titan's winds, particularly observations tracking the Huygens probe at 10 deg S, indicate that the zonal winds are mostly in the sense of the satellite's rotation. They generally increase with altitude and become cyclostrophic near 35 km above the surface. An exception to this is a sharp minimum centered near 75 km, where the wind velocity decreases to nearly zero. Zonal winds derived from temperatures retrieved from Cassini orbiter measurements, using the thermal wind equation, indicate a strong winter circumpolar vortex, with maximum winds of 190 m/s at mid northern latitudes near 300 km. Above this level, the vortex decays. Curiously, the stratospheric zonal winds and temperatures in both hemispheres are symmetric about a pole that is offset from the surface pole by about 4 deg. The cause of this is not well understood, but it may reflect the response of a cyclostrophic circulation to the onset between the equator, where the distance to the rotation axis is greatest, and the seasonally varying subsolar latitude. The mean meridional circulation can be inferred from the temperature field and the meridional distribution of organic molecules and condensates and hazes. Both the warm temperatures near 400 km and the enhanced concentration of several organic molecules suggest subsidence in the north polar region during winter and early spring. Stratospheric condensates are localized at high northern latitudes, with a sharp cut-off near 50 deg N. Titan's winter polar vortex appears to share many of the same characteristics of isolating high and low-latitude air masses as do the winter polar vortices on Earth that envelop the ozone holes. Global mapping of temperatures, winds, and composition in the troposphere, by contrast, is incomplete. The few suitable discrete clouds that have been found for tracking indicate smaller velocities than aloft, consistent: with the Huygens measurements, Along the descent trajectory, the Huygens measurements indicate eastward zonal winds down to 7 km, where they shift westward, and then eastward again below 1 km dawn to the surface. The low-latitude dune fields seen in Cassini RADAR images have been interpreted as longitudinal dunes occurring in a mean eastward zonal wind. This is not like Earth, where the low-latitude winds are westward above the surface. Because the net zonal-mean time-averaged torque exerted by the surface on the atmosphere should vanish, there must be westward flow over part of the surface; the question is where and when. The meridional contrast in tropospheric temperatures deduced from radio occultations at low, mid, and high latitudes. is small, approximately 5 K at the tropopause and approximately 3 K at the surface. This implies efficient heat transport, probably by axisymmetric meridional circulations. The effect of the methane "hydrological" cycle on the atmospheric circulation is not well constrained by existing measurements, Understanding the mature of the surface-atmosphere coupling will be critical to elucidating the atmospheric transports of momentum, heat, and volatiles.

Flasar, F. M.↗
NASA NTRS1975

CARETS: A prototype regional environmental information system. Volume 3: Toward a national land use information system

It is recommended that a national land use information system be established by an agency of the Federal Government. This recommendation comes at a time of increasing demand for scientific information in support of environmentally relevant land use planning and management at all levels of government. It is also a time when new airborne and spaceborne remote sensors, tested in cooperation with the National Aeronautics and Space Administration (NASA) and the Earth Resources Observation Systems (EROS) Program of the Department of the Interior, make possible the gathering of land use information rapidly and on an unprecedented scale. Furthermore, information handling technology is developing toward a capability to receive, store, and disseminate the huge quantities of data that would be involved. The recommendation for the national land use information system is based upon careful analysis of the results of remote sensing experiments funded by NASA, EROS, and the Geography Program of the Geological Survey, with specific examples drawn from the demonstration project known as the Central Atlantic Regional Ecological Test Site (CARETS). CARETS is cast in the framework of a regional land use information system, channeling the flow bf information generated in response to users' declaration of their needs, through stages dealing with remote sensing data gathering systems, data processing and land mensuration, calibration in-terms of environmental impact, and evaluation with feedback from users. The proposed system would develop and implement a unified approach to the description and interpretation of the changing uses of the nation's land resources, building upon the base of interagency and intergovernmental cooperation already achieved in the experimental work to date. The land use data base that is being derived from high-altitude aerial color infrared photography would be the initial component of the recommended system. High-altitude photographic coverage would immediately be extended to as much of the nation as possible as technological developments and economic considerations permit. The system would later expand to include multiple-sensor, multiple-platform data sources. Six system characteristics are recommended: (1) High capacity storage of data available for quick retrieval, inexpensive processing, and update, (2) provision of accuracy appropriate to the scale of survey or to the level of detail dictated by different types of management and decision requirements; (3) permanent, publicly accessible sensor records for historical interpretation; (4) compatibility of the recording, storage, and retrieval system with all types of inputs, from ground observer to satellite; (5) products of diverse formats and scales, responsive to user feedback; (6) and standardization of formats, scales, and storage inputs to permit nationwide comparability.

Land use mapping↗
NASA NTRS1988

Spectral characteristics and the extent of paleosols of the Palouse formation

The objective of this study is to test the hypothesis that TM data is adequate in band selection and width and in spatial resolution to distinguish soil organic matter, iron oxide, and lime-silica contents to map several severity classes of erosion in soils of the Palouse region. The methodology used is as follows: (1) To develop spectral relationships from TM data that define the spatial distribution of soil areas by levels of (1) organic matter in the surface soil, (2) iron oxide and clay in exposed paleosol B horizons, and (3) lime-silica accumulations in exposed paleosol B horizons; (2) To compare areas determined by the method outlined in 1 to patterns interpreted from color aerial photos, and to ground observations on bare-soil fields; and (3) To define, on the basis of results of 1 and 2 to the extent possible, where exposed paleosols exist within fields that are not bare, but have a crop cover, and the distribution of desirable and undesirable soil properties in each field.

Frazier, B. E.↗
NASA NTRS1995

Curation of Federally Owned Archeological Collections at NASA Langley Research Center

As a Federal agency, NASA has a moral and legal obligation to the public to manage the archeological heritage resources under its control. Archeological sites are unique, nonrenewable resources that must be preserved so that future generations may experience and interpret the material remains of the past. These sites are protected by a wide array of federal regulations. These regulations are intended to ensure that our nation's cultural heritage is preserved for the study and enjoyment of future generations. Once a site has been excavated, all that remains of it are the artifacts and associated records which, taken together, allow researchers to reconstruct the past. With the contextual information provided by associated records such as field notes, maps and photographs, archeological collections can provide important information about life in the past. An integral component of the federal archeology program is the curation of these databases so that qualified scholars will have access to them in years to come. Standards for the maintenance of archeological collections have been codified by various professional organizations and by the federal government. These guidelines focus on providing secure, climate-controlled archival storage conditions for the collections and an adequate study area in which researchers can examine the artifacts and documents. In the 1970's and early 1980's, a group of NASA employees formed the LRC Historical and Archeological Society (LRCHAS) in order to pursue studies of the colonial plantations that ha been displaced by Langley Research Center (LaRC). They collected data on family histories and land ownership as well as conducting archeological surveys and excavations at two important 17th-20th century plantation sites in LaRC, Cloverdale and Chesterville. The excavations produced a wealth of information in the form of artifacts, photographs, maps and other documents. Unfortunately, interest on the part of the LRCHAS membership waned before a report was written, and since 1982 the artifacts have moldered in a flimsy trailer with no climate controls, which had once served as a field laboratory but which threatened to become a tomb for the collection. A recent analysis of Langley's cultural resources by Gray & Pape, Inc. recommended that the collection be organized, cataloged, and placed in a proper curation facility in accordance with Federal regulations. The project for the LARSS program was to research curation standards, organize the collection, catalog it, and prepare it for transfer to a facility which could provide adequate long-term curation conditions for the artifacts and documents. The first phase was to organize the artifacts, which were lying about the lab in various stages of cleaning, analysis, and conservation. Once all of the artifacts from the various excavation units and levels had been regrouped, they were cleaned and/or repackaged in archivally-stable materials. A basic catalog was prepared which will provide interested parties with a rough idea of what we have and where it can be found. Another aspect of the project was to organize the records left by the LRCHAS. Bundles of papers, photographs, and field data found in every corner and drawer of the laboratory trailer were put into order and, where appropriate, copies were made on acid-free Permabond paper for long term storage. Finally, the entire collection and most of the lab equipment was transferred into a secure, climate controlled room which will serve as an archive and study space for qualified scholars interested in exploring LaRC's rich historical heritage.

Eastman, John Arnold↗
DOE OSTI2016

Determining crystal structures through crowdsourcing and coursework

We show here that computer game players can build high-quality crystal structures. Introduction of a new feature into the computer game Foldit allows players to build and real-space refine structures into electron density maps. To assess the usefulness of this feature, we held a crystallographic model-building competition between trained crystallographers, undergraduate students, Foldit players and automatic model-building algorithms. After removal of disordered residues, a team of Foldit players achieved the most accurate structure. Analysing the target protein of the competition, YPL067C, uncovered a new family of histidine triad proteins apparently involved in the prevention of amyloid toxicity. From this study, we conclude that crystallographers can utilize crowdsourcing to interpret electron density information and to produce structure solutions of the highest quality.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2019

Imaging material functionality through three-dimensional nanoscale tracking of energy flow

The ability of energy carriers to move between atoms and molecules underlies biochemical and material function. Understanding and controlling energy flow, however, requires observing it on ultrasmall and ultrafast spatio-temporal scales, where energetic and structural roadblocks dictate the fate of energy carriers. Here, we developed a non-invasive optical scheme that leverages non-resonant interferometric scattering to track tiny changes in material polarizability created by energy carriers. We thus map evolving energy carrier distributions in four dimensions of spacetime with few-nanometre lateral precision and directly correlate them with material morphology. We visualize exciton, charge and heat transport in polyacene, silicon and perovskite semiconductors and elucidate how disorder affects energy flow in three dimensions. For example, we show that morphological boundaries in polycrystalline metal halide perovskites possess lateral- and depth-dependent resistivities, blocking lateral transport for surface but not bulk carriers. We also reveal strategies for interpreting energy transport in disordered environments that will direct the design of defect-tolerant materials for the semiconductor industry of tomorrow.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Reprocessing Models for the Optical Light Curves of Hypervariable Quasars from the Sloan Digital Sky Survey Reverberation Mapping Project

We explore reprocessing models for a sample of 17 hypervariable quasars, taken from the Sloan Digital Sky Survey Reverberation Mapping project, which all show coordinated optical luminosity hypervariability with amplitudes of factors ≳2 between 2014 and 2020. We develop and apply reprocessing models for quasar light curves in simple geometries that are likely to be representative of quasar inner environments. In addition to the commonly investigated thin-disk model, we include the thick-disk and hemisphere geometries. The thick-disk geometry could, for instance, represent a magnetically elevated disk, whereas the hemisphere model can be interpreted as a first-order approximation for any optically thick out-of-plane material caused by outflows/winds, warped/tilted disks, and so on. Of the 17 quasars in our sample, 11 are best-fitted by a hemisphere geometry, five are classified as thick disks, and both models fail for just one object. We highlight the successes and shortcomings of our thermal reprocessing models in case studies of four quasars that are representative of the sample. While reprocessing is unlikely to explain all of the variability that we observe in quasars, we present our classification scheme as a starting point for revealing the likely geometries of reprocessing for quasars in our sample and hypervariable quasars in general.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1986

Transient medium-scale wave activity in the summer stratosphere

Observations of medium-scale transient wave activity in the middle-latitude Southern Hemisphere summer stratosphere are interpreted using isentropic potential vorticity (IPV) distributions inferred from temperature data from the Nimbus-7 Limb Infrared Monitor of the Stratosphere (LIMS) instrument. Despite problems with errors in reference-level synoptic height information, the derived IPV distributions clearly exhibit dynamically coherent signatures of eddy mixing due to medium-scale disturbances in the lower-middle stratosphere during early January 1979. The IPV distributions are shown to be strongly correlated with those of a contemporaneous set of isentropic maps of quasi-conservative tracers such as ozone and nitric acid. The results suggest that IPV maps inferred from satellite temperature data provide a useful diagnostic for studies of large-scale dynamics and/or transport in the extratropical stratosphere. Furthermore, the comparatively high vertical resolution of the LIMS data has enabled the vertical attenuation of Southern Hemisphere wave activity to be diagnosed in greater detail than could be achieved from nadir-viewing satellite information.

Miles, T.↗
NASA NTRS1999

Implications of the Improved Understanding of the Gravitational and Topography Fields

The Mars Orbiter Laser Altimeter (MOLA) has been successfully mapping the shape and topography of Mars on an intermittent basis since the Mars Global Surveyor (MGS) spacecraft arrived at Mars in Sept. 1997. In early April, MGS reached it's mapping orbit and began the mapping phase of the mission. Since that time MOLA has been operating continuously. MOLA is a laser altimeter with a pulse rate of 10 Hz and a precision of about 30 to 40 cm over flat surfaces. The laser pulses are approximately 330 meters apart on the surface of Mars and each pulse illuminates an area approximately 150 meters in diameter. The orbital accuracy of MGS is a few meters in the radial direction enabling the topography to be interpreted at the few meter accuracy level. The initial data have provided a new geodetic description of the size and shape of Mars, including accurate estimates for the equatorial and polar radii, and the planetary flattening. The data have revealed the northern hemisphere to be very low in elevation, very flat, and almost devoid of craters. The north polar icecap lies in a basin about 5 km below the average altitude and reaches to a height of 3 km. The heavily cratered southern hemisphere is high and the terrain is rough. The south polar icecap is about 5 km above the average planetary altitude and 2 km above the surrounding terrain. The MOLA profile data show the gradual but steady increase in altitude of the surface of Mars from the north pole to the south pole. This is a consistent feature for all north-south profiles and reflects the 3 km separation of the center of figure for Mars from the center of mass along the rotation axis.

Smith, David E.↗
NASA NTRS2012

The Geology of the Marcia Quadrangle of Asteroid 4Vesta: An Integrated Mapping Study Using Dawn Spacecraft Data

We used geologic mapping applied to Dawn data as a tool to understand the geologic history of the Marcia quadrangle of Vesta. This region hosts a set of relatively fresh craters and surrounding ejecta field, an unusual dark hill named Arisia Tholus, and a orange (false color) diffuse material surrounding the crater Octavia. Stratigraphically, from oldest to youngest, three increasingly larger impact craters named Minucia, Calpurnia, and Marcia make up a snowmanlike feature, which is surrounded by a zone of dark material interpreted to consist of impact ejecta and possibly impact melts. The floor of Marcia contains a pitted terrain thought to be related to release of volatiles (1). The dark ejecta field has an enhanced signature of H, possibly derived from carbonaceous chondritic material that accumulated in Vesta s crust (2,3). The dark ejecta has a spectrally distinctive behavior with shallow pyroxenes band depths. Outside the ejecta field this quadrangle contains various cratered terrains, with increasing crater abundance moving south to north away from the Rheasilvia basin. Arisia Tholus, originally suggested as an ancient volcano, appears to be an impact-sculpted basin rim fragment with a superposed darkrayed impact crater. There remains no unequivocal evidence of volcanic features on Vesta s surface, likely because basaltic material of the HED meteorite suite demonstrates magmatism ended very early on Vesta (4). Ongoing work includes application of crater statistical techniques to obtain model ages of surface units, and more detailed estimates of the compositional variations among the surface units.

Williams, David A.↗
NASA NTRS2018

Solar Wind Implantation Into the Lunar Regolith: Monte Carlo Simulations of H Retention in a Surface With Defects and the H2 Exosphere

The solar wind implants protons into the top 20–30 nm of lunar regolith grains, and the implanted hydrogen will diffuse out of the regolith but also interact with oxygen in the regolith oxides. We apply a statistical approach to estimate the diffusion of hydrogen in the regolith hindered by forming temporary bonds with regolith oxygen atoms. A Monte Carlo simulation was used to track the temporal evolution of bound OH surface content and the H2 exosphere. The model results are consistent with the interpretation of the Chandrayaan-1 M3 observations of infrared absorption spectra by surface hydroxyls as discussed in Li and Milliken (2017). Themodel reproduced the latitudinal concentration of OH by using a Gaussian energy distribution of f(U(o) = 0.5eV, U(w) = 0.078 – 0.1 eV) to characterize the activation energy barrier to the diffusion of hydrogen in space weathered regolith. In addition, the model results of the exospheric content of H2 are consistent with observations by the Lyman Alpha Mapping Project on the Lunar Reconnaissance Orbiter. Therefore, we provide support for hydroxyl formation by chemically trapped solar wind protons.

O. J. Tucker↗
DOE OSTI2015

Smart align -- A new tool for robust non-rigid registration of scanning microscope data

Many microscopic investigations of materials may benefit from the recording of multiple successive images. This can include techniques common to several types of microscopy such as frame averaging to improve signal-to-noise ratios (SNR) or time series to study dynamic processes or more specific applications. In the scanning transmission electron microscope, this might include focal series for optical sectioning or aberration measurement, beam damage studies or camera-length series to study the effects of strain; whilst in the scanning tunnelling microscope, this might include bias voltage series to probe local electronic structure. Whatever the application, such investigations must begin with the careful alignment of these data stacks, an operation that is not always trivial. In addition, the presence of low-frequency scanning distortions can introduce intra-image shifts to the data. Here, we describe an improved automated method of performing non-rigid registration customised for the challenges unique to scanned microscope data specifically addressing the issues of low-SNR data, images containing a large proportion of crystalline material and/or local features of interest such as dislocations or edges. Careful attention has been paid to artefact testing of the non-rigid registration method used, and the importance of this registration for the quantitative interpretation of feature intensities and positions is evaluated.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Applying machine learning and quantum chemistry to predict the glass transition temperatures of polymers

Glass transition temperature (T g ) is important for understanding the physical and mechanical properties of a polymer material because it relates to the thermal energy required to transition between a hard glassy state and a soft rubbery one. Over the years, various models have been developed for predicting this thermal property from molecular structure to aid in designing novel polymers in selected classes. This work builds on those efforts by utilizing both machine learning (ML) and quantum chemistry (QC) techniques to develop models that can predict T g values from the molecular structure under different data availability scenarios and for a wide variety of polymer types. For the ML model, a graph convolutional network (GCN) was used to map topological polymer features; this model was trained against a dataset of more than 7500 T g values and resulted in a root mean square error (RMSE) of 38.1 °C. The QC-based regression model was trained on 83 T g values and produced an RMSE of 34.5 °C. In conclusion, this work demonstrated that while both model techniques produce accurate predictions and are suitable for different data availability scenarios, the QC-based regression model offered a more interpretable model framework with significantly less training data.

36 MATERIALS SCIENCE↗