Search NASA⌕ Search

SEARCH · Search NASA

Results for “sample integrity”

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 775 records · Page 43
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Demonstration of an Automated System for Vertical Profiles of Volatile Organic Compounds

Volatile organic compounds (VOCs) play important roles throughout the atmosphere, many of which are altitude dependent. This highlights the need for easily deployable devices to sample VOCs across different atmospheric layers. To address this, we present the design and initial application of a Time Resolved Automated Volatile organIc compounds Sampling system (TRAVIS). VOCs are collected on sorbent tubes, which are subsequently analyzed by a thermal desorption gas chromatography mass spectrometry pipeline. TRAVIS leverages a piezoelectric pump with an integrated pressure sensor for precise (0.1% flow rate relative standard deviation) and accurate (−3 ± 2% error in VOC quantitation) measurements. Via deployment on a tethered balloon system over an agricultural area, TRAVIS is used to show consistent vertically resolved VOC concentrations in a well-mixed (i.e., turbulent) atmosphere (e.g., 5% relative standard deviation for isoprene) and vertically dependent concentrations for a stratified atmosphere (e.g., prior to boundary layer development). Furthermore, we also show VOC information from an intermittent plume via both targeted and untargeted analysis, highlighting future applications for spurious events in agriculture, air quality monitoring, and environmental impact. Overall, the development of TRAVIS represents a lightweight, accurate, sensitive, and precise VOC sampling module for the scientific community.

Aerosols↗
DOE OSTI2022

A Multi-Branch Decoder Network Approach to Adaptive Temporal Data Selection and Reconstruction for Big Scientific Simulation Data

A key challenge in scientific simulation is that the simulation outputs often require intensive I/O and storage space to store the results for effective post hoc analysis. This article focuses on a quality-aware adaptive temporal data selection and reconstruction problem where the goal is to adaptively select simulation data samples at certain key timesteps in situ and reconstruct the discarded samples with quality assurance during post hoc analysis. This problem is motivated by the limitation of current solutions that a significant amount of simulation data samples are either discarded or aggregated during the sampling process, leading to inaccurate modeling of the simulated phenomena. Two unique challenges exist: 1) the sampling decisions have to be made in situ and adapted to the dynamics of the complex scientific simulation data; 2) the reconstruction error must be strictly bounded to meet the application requirement. To address the above challenges, we develop DeepSample , an error-controlled convolutional neural network framework, that jointly integrates a set of coherent multi-branch deep decoders to effectively reconstruct the simulation data with rigorous quality assurance. The results on two real-world scientific simulation applications show that DeepSample significantly outperforms other state-of-the-art methods on both sampling efficiency and reconstructed simulation data quality.

Zhang, Yang↗
DOE OSTI2022

SDSS-IV MaNGA: pyPipe3D Analysis Release for 10,000 Galaxies

We present here the analysis performed using the pyPipe3D pipeline for the final MaNGA data set included in the Sloan Digital Sky Survey data release 17. This data set comprises more than 10,000 individual data cubes, being the integral field spectroscopic (IFS) galaxy survey with the largest number of galaxies. pyPipe3D processes the IFS data cubes to extract spatially resolved spectroscopic properties of both the stellar population and the ionized gas emission lines. A brief summary of the properties of the sample and the characteristics of the analyzed data are included. The article provides details of: (i) the analysis performed; (ii) a description of the pipeline; (iii) the adopted stellar population library; (iv) the morphological and photometric analysis; (v) the adopted data model for the spatially resolved properties derived; and (vi) the individual integrated and characteristic galaxy properties included in the final catalog. Comparisons with the results from a previous version of the pipeline for earlier data releases and from other tools using this data set are included. A practical example of how to use the full data set and the final catalog illustrates how to handle the delivered product. Our full analysis can be accessed and downloaded from our web page.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Preliminary Design of a New Pulsed-Neutron Die--Away Experiments with Absorbers (IER-552 CED-1)

This report presents the preliminary design (CED-1) of IER-552, pulsed-neutron die-away (PNDA) experiments with absorbers, to be conducted by Lawrence Livermore National Laboratory (LLNL). The goal of the PNDA experiments with absorbing materials is to produce high-quality, low-cost integral benchmarks that can be used to validate cross sections that are high priority nuclear data for the Department of Energy’s Nuclear Criticality Safety Program (NCSP). PNDA experiments can be used to generate nuclear data validation benchmarks quickly and cheaply, as they do not require fissile material or a nuclear facility to conduct them. The original PNDA experiments were set up at LLNL under IER-501. Extending the PNDA experiments to include absorbing materials leverages existing facilities and resources at LLNL to validate cross sections, with a basic experiment design that uses a neutron generator to impinge a short, mono-energetic neutron pulse on a poisoned, moderating target sample. The neutron pulse lasts 100 µs, after which the neutron population in the sample reaches thermal equilibrium and later spatial equilibrium. In this fully equilibrized state, the neutron population has a characteristic decay-time eigenvalue. The eigenvalue can be used as an integral parameter to validate nuclear data involved with neutron scattering and absorption. Two observables can be measured: the neutron population and the population of secondary γ-rays produced by inelastic and capture reactions. This report explores the feasibility of both approaches. For γ-ray measurements, this allows for direct measurement of reactions of interest for validation. During the neutron burst time, inelastic (prompt) γ-rays dominate, and these data can be isolated for validation of neutron inelastic scattering data. After the pulse ends, neutron capture (delayed) γ-rays dominate and are produced by many reaction channels as neutrons slow down to thermal energies. Two sets of γ-ray spectra (inelastic, capture) can be produced for validation. Additionally, the time-dependent decay of the capture γ-ray population when the neutron population is thermalized and in its fundamental mode can be used for validation. Two experiments can be imagined where pulsed neutrons are used to validate the nuclear data of absorbing materials. The first integrates well into the existing PNDA testbed at LLNL. The absorber material would be introduced into a moderating target and the same neutron generator, neutron detectors, electronics, and shielding box used for IER-501 would be employed. From the same experimental setup, many absorbing materials can be validated by simply changing the neutron poison in the moderator. The only significant modification would be the introduction of a γ-ray detector. Such an experiment would primarily serve to validate nuclear data in the thermal energy range. The second kind of experiment would validate the fast and epithermal energy range. It would involve pulsing neutron into a bare assembly of the absorber material and measuring the die-away of neutrons. The fast leakage of such an assembly requires nanosecond neutron-generator pulse widths and fast detectors/electronics that could measure ~200 ns die-away curves. This kind of experiment would ideally be performed at a cyclotron, like that available at Lawrence Berkeley National Laboratory.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2023

Embedding of Fiber Optic Sensors in Stainless Steel for Sensing in Harsh Environments

Structural health monitoring is critical for components working in harsh environments to assess their safety and integrity. This can be realized with advanced sensors, such as fiber optic sensors, embedded in these components. Recent studies have used ultrasonic consolidation (UC) and laser-based additive manufacturing (AM) techniques to embed fiber optic sensors in metals. However, sensor embedding using these techniques has been challenging, with gaps and voids around the embedded fiber. Successful embedding of fiber optic sensors requires good contact between the fiber and the matrix to eliminate potential gas/liquid leaks in critical systems such as in nuclear and aerospace services. Defect-free fiber/matrix bond is also essential for accurate measurement of parameters of interest, such as temperature and strain. This study developed a novel fiber embedding technique based on electric field-assisted sintering (EFAS). Sapphire optical fibers with diameters of 125 µm and 250 µm were embedded in stainless steel (SS) 316L using EFAS. The embedding was completed by sintering SS 316L powders at a pressure of 50MPa, hold time of 5 min, and temperatures of 1000°C, 1100°C, and 1200°C. Joule heating mechanism during EFAS enabled rapid sintering of the SS 316L powder with resulting sample densities over 99%. Microscopy analysis performed on the cross-sections revealed a defect-free bond between the embedded fiber and SS 316L. The good integrity of the embedded fiber was evidenced by optical transmittance measurement. Optical attenuation of the fiber before and after embedment was examined in the near-infrared region. Scanning electron microscopy was employed to investigate the grain morphology of SS 316L near the embedded fiber. Hardness distribution in the sintered part was also investigated to study the effect of the embedded fiber on the mechanical properties of the as-fabricated component. The results indicated that EFAS was able to embed fiber in SS with good fiber functionality and part quality.

36 - MATERIALS SCIENCE↗
DOE OSTI2025

Uncertainty quantification and sensitivity analysis of a nuclear thermal propulsion reactor startup sequence

The research presented in this article describes progress in applying stochastic methods, uncertainty quantification, parametric studies, and variance-based sensitivity analysis (also known as Sobol sensitivity analysis) to a full-core model of a nuclear thermal propulsion (NTP) system simulated via the radiation transport code Griffin to simulate neutronics. Our goal is to develop a reduced-order (surrogate) model that can be rapidly sampled with perturbations to multiple input parameters. In this NTP system, reactivity and power feedback affect the rotation of control drums (CDs), which is itself controlled by a hybrid proportional-integral-derivative (PID) controller actuated by the power demand and reactivity feedback from the numerical model. This model uses reactor kinetic feedback (mean generation time [Λ] and effective delayed neutron fraction [ β eff ] from a transient Griffin simulation executed via Griffin’s improved quasi-static solver to provide the kinetic parameters) as inputs to functions that control the CD rotation angle. By investigating numerous stochastic approaches, we developed a dual-purpose surrogate model of the NTP system, using polynomial regression in the Multiphysics Object-Oriented Simulation Environment (MOOSE) Stochastic Tools Module (STM). The trained model can be rapidly sampled while simultaneously perturbing various input parameters, such as coefficients on the PID control or temperature (directly affecting the neutron cross section). The surrogate model delivers accurate (within 5%) results at speeds orders of magnitude faster (minutes, not days of computational time) than the base model. Once the surrogate model has been trained, distributions of the uncertain parameters can be changed at will to investigate the effects of perturbing multiple inputs as well as the effects of these inputs on the model output. For example, coefficients used in the PID control system may vary due to some type of physical interference, or uncertainty may exist in the temperature of the neutron cross sections in various regions of the reactor. A distribution can be placed on these parameters, and operational boundaries can be determined. The goal of this work is to support development of an advanced control system for operating CDs in a functioning NTP system. This work is a scoping study of the MOOSE STM.

21 - SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLAN↗
DOE OSTI2025

Open Specy 1.0: Automated (Hyper)spectroscopy for Microplastics

Microplastic spectral analysis is one of the most time-consuming processes in studying microplastic pollution, often requiring days per sample. Researchers are transitioning to automated batch and hyperspectral image analysis techniques to enhance efficiency. Open Specy, initially aimed at manual single-spectrum analysis, has now integrated automated methods. This updated version, Open Specy 1.0, introduces several new features, including two algorithms for automated processing (smoothing and particle compression), an extensive library containing over 40,000 open-source Raman and FTIR spectra, and two machine learning classifiers (logistic regression and k medoids) developed from this library. Furthermore, it includes a revamped user interface, an R package, and a benchmark data set for testing future advancements in automated techniques. Researchers evaluated various configurations for hyperspectral smoothing, particle identification, compression, and splitting, to achieve combined recovery rates between 50 and 150% particle counts, identities, and sizes with a coefficient of variation (CV) of less than 40% (the accredited standard). Mean absorbance times the standard deviation provided a consistent particle identification. Hyperspectral smoothing led to a 96% combined recovery rate and reduced variability (CV = 38%) compared to the 86% recovery (CV = 83%) of nonsmoothed controls. Additionally, compressing spectra for particles was significantly faster (>3x) and showed similar accuracy but with reduced variability than processing each pixel individually. Key challenges persist in automating spectral analysis, particularly in refining particle splitting algorithms, and improving identification routines to minimize false positives and negatives. In conclusion, new methods in sample preparation for better stabilization and dispersion of particles could overcome some of these issues.

13 HYDRO ENERGY↗
DOE OSTI2022

Mass–metallicity and star formation rate in galaxies: A complex relation tuned to stellar age

In this work we study the stellar mass–metallicity relation (MZR) of an extended sample of star-forming galaxies in the local Universe and its possible dependence on the star formation rate (SFR). We selected a sample of approximately 195 000 Sloan Digital Sky Survey (SDSS) star-forming galaxies up to z = 0.22 with the aim of analysing the behaviour of the MZR with respect to SFR whilst taking into account the age of their stellar populations. For the first time, with this sample, we obtained aperture corrected oxygen and nitrogen-to-oxygen abundances (O/H and N/O, respectively) and SFR using the empirical prescriptions from the Calar Alto Legacy Integral Field Area (CALIFA) survey. To perform this study we also make use of the stellar mass of the galaxies and the parameter Dn(4000) as a proxy for the age of the stellar population. We derive a robust MZR locus, which is found to be fully consistent with the ‘anchoring’ points of a selected set of well-studied nearby galaxies for which the chemical abundance has been derived using the direct method. We observe a complex relation between MZR and SFR across the whole range of galaxy mass and metallicity, where the slope changes seen in the O/H–SFR plane present a pattern that seems to be tuned to the stellar age of the galaxies, and therefore stellar age has to be taken into account in the stellar mass–metallicity–SFR relation. In order to provide an answer to the question of whether or not the MZR depends on the SFR, it is essential to take into account the age of the stellar populations of galaxies. A strong dependence of the MZR on SFR is observed mainly for star-forming galaxies with strong SFR values and low D n (4000). The youngest galaxies of our SDSS sample show the highest SFR measured for their stellar mass.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Microstructure and water absorption of ancient concrete from Pompeii: An integrated synchrotron microtomography and neutron radiography characterization

There is renewed interest in using advanced techniques to characterize ancient Roman concrete due to its exceptional durability and low-carbon footprint. In the present work, samples were drilled from the “Hospitium” in Pompeii and were analyzed by synchrotron microtomography (μCT) and neutron radiography to study how the microstructure, including the presence of induced cracks, affects their water adsorption. The water distribution and absorptivity were quantified by neutron radiography. The 3D crack propagation, pore size distribution and orientation, tortuosity, and connectivity were analyzed from μCT results using advanced imaging methods. Porosity was also measured by mercury intrusion porosimetry (MIP) as a reference. Ductile fracture patterns were observed once cracks were introduced. Compared to Portland cement mortar/concrete, the Pompeii samples had relatively high porosity, low connectivity, and a similar coefficient of capillary penetration. In addition, permeability was predicted from models based on percolation theory and pore structure data to evaluate the fluid transport properties. Understanding the microstructure of ancient Pompeii concrete is important because it could inspire the development of modern concrete with high durability.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Replication Data for: Multitask Machine Learning of Collective Variables for Enhanced Sampling of Rare Events

The data underlying this published work have been made publicly available in this repository as part of the IMASC Data Management Plan. This work was supported as part of the Integrated Mesoscale Architectures for Sustainable Catalysis (IMASC), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award # DE-SC0012573.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Characterizing electron-collecting CdTe for use in a 77 ns burst-rate imager

The Keck-PAD (pixel array detector) was developed at Cornell as a burst-rate imager capable of recording images from successive electron bunches (153 ns period) from the Advanced Photon Source (APS). Both Si and hole-collecting Schottky CdTe have been successfully bonded to this ASIC (application-specific integrated circuit) and used with this frame rate. The facility upgrades at the APS will lower the bunch period to 77 ns, which will require modifications to the Keck-PAD electronics to image properly at this reduced period. In addition, operation at high X-ray energies will require a different sensor material having a shorter charge collection time. For the target energy of 40 keV for this project, simulations have shown that electron-collecting CdTe should allow >90% charge collection within 35 ns. This collection time will be sufficient to sample the signal from one frame and prepare for the next. 750 µm-thick electron-collecting Schottky CdTe has been obtained from Acrorad and bonded to two different charge-integrating ASICs developed at Cornell, the Keck-PAD and the CU-APS-PAD. Carrier mobility has been investigated using the detector response to single X-ray bunches at the Cornell High Energy Synchrotron Source and to a pulsed optical laser. The tests indicate that the collection time will meet the requirements for 77 ns imaging.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Deep Spectroscopy with DESI for Photometric Redshift Training and Calibration

Deep spectroscopic samples can improve photometric redshift (photo-z) estimates and reduce uncertainties on redshift distributions. Such improvements can increase the cosmological constraining power of large imaging-based experiments such as the Vera C. Rubin Observatory’s Legacy Survey of Space and Time (LSST) and mitigate what may be a limiting systematic effect. We present results from the “DESI-Deep pilot” program, which was designed to assess the capability of the Dark Energy Spectroscopic Instrument (DESI) on the 4m Mayall telescope to measure redshifts of galaxies as faint as expected lensing samples for early LSST data (m i ≤ 24.5). We find that DESI is remarkably efficient at this task, with redshift success rates comparable to the results of observations from 10 m class telescopes with only ∼2 × longer integration time (rather than ∼8 × longer as would be expected from aperture-area scaling), while simultaneously achieving ∼30 times larger multiplexing. We also find that the signal-to-noise ratio of the spectra scales as expected for background-limited observations even for the longest exposure times (∼7 hr) and faintest targets in the program. These results demonstrate that DESI could provide the definitive redshift sample for the early years of LSST with a modest investment of observing time. Based upon the results of this program, we provide updated predictions for the time required to collect benchmark samples for photo-z training and calibration using a variety of spectroscopic facilities. Finally, we describe a potential “DESI-Deep” survey designed to train and calibrate photo-z’s for imaging experiments, and provide forecasts of its impact on cosmological inference.

Dey, Biprateep [University of Toronto; University ↗
DOE OSTI2026

An Integrated Framework for Memory-Centric Analysis: From Trace Collection to Co-Design

The memory wall phenomenon—where advances in processor performance significantly outpace those in memory subsystems—poses a fundamental challenge for contemporary computing systems. In memory-bound applications, memory subsystem behavior dominates performance, yet existing analysis approaches present significant limitations: detailed microarchitectural simulators require days to weeks to simulate modest workloads; hardware performance counters provide only aggregate statistics that obscure temporal and spatial access patterns; and scaled simulation approaches face challenges in capturing certain behaviors that emerge at larger scales. These limitations reflect a processor-centric design philosophy increasingly misaligned with memory-bound workloads where detailed understanding of memory access patterns, cache hierarchy interactions, and contention is critical for effective optimization. This paper presents an integrated framework for memory-centric analysis that enables effective hardware-software co-design. We describe practical trace collection techniques, including hardware-assisted processor tracing with minimal overhead and portable software-based instrumentation with statistical sampling. We present multi-perspective analysis methods that examine memory behavior from temporal, sequential, spatial, and relational viewpoints, revealing distinct optimization opportunities invisible in aggregate metrics. We detail an architectural modeling framework that uses sampled traces with temporal interpolation and confidence-based filtering to evaluate cache and memory configurations. Evaluation on representative benchmarks demonstrates that this framework achieves practical accuracy (L2 cache errors of 2.64\%, confidence-filtered L3 errors of 9.92\%, bandwidth errors of 7.33\%) while providing substantial speedup (26.8×) over cycle-accurate simulation, enabling rapid design space exploration. We demonstrate how this integrated framework enables systematic identification of both hardware optimizations (memory controller tuning, bank partitioning, NUMA configuration) and software optimizations (data layout restructuring, prefetching strategies, memory-aware scheduling). Through this comprehensive treatment of the memory-centric analysis pipeline—from trace collection through architectural modeling to co-design application—we provide researchers and practitioners with practical techniques for addressing memory bottlenecks in contemporary computing systems.

Gajaria, Dhruv Mayur↗
DOE OSTI2020

Search for a heavy pseudoscalar Higgs boson decaying into a 125 GeV Higgs boson and a Z boson in final states with two tau and two light leptons at $\sqrt{s}=$ 13 TeV

A search is performed for a pseudoscalar Higgs boson, A, decaying into a 125 GeV Higgs boson h and a Z boson. The h boson is specifically targeted in its decay into a pair of tau leptons, while the Z boson decays into a pair of electrons or muons. A data sample of proton-proton collisions collected by the CMS experiment at the LHC at $ \sqrt{s} $ = 13 TeV is used, corresponding to an integrated luminosity of 35.9 fb$^{−1}$. No excess above the standard model background expectations is observed in data. A model-independent upper limit is set on the product of the gluon fusion production cross section for the A boson and the branching fraction to Zh → ℓℓττ. The observed upper limit at 95% confidence level ranges from 27 to 5 fb for A boson masses from 220 to 400 GeV, respectively. The results are used to constrain the extended Higgs sector parameters for two benchmark scenarios of the minimal supersymmetric standard model.[graphic not available: see fulltext]

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Search for the decay of the Higgs boson to a pair of light pseudoscalar bosons in the final state with four bottom quarks in proton-proton collisions at $ \sqrt{\textrm{s}} $ = 13 TeV

A search is presented for the decay of the 125 GeV Higgs boson (H) to a pair of new light pseudoscalar bosons (a), followed by the prompt decay of each a boson to a bottom quark-antiquark pair, $ \textrm{H}\to \textrm{aa}\to \textrm{b}\overline{\textrm{b}}\textrm{b}\overline{\textrm{b}} $. The analysis is performed using a data sample of proton-proton collisions collected with the CMS detector at a center-of-mass energy of 13 TeV, corresponding to an integrated luminosity of 138 fb$^{−1}$. To reduce the background from standard model processes, the search requires the Higgs boson to be produced in association with a leptonically decaying W or Z boson. The analysis probes the production of new light bosons in a 15 < m$_{a}$ < 60 GeV mass range. Assuming the standard model predictions for the Higgs boson production cross sections for pp → WH and ZH, model independent upper limits at 95% confidence level are derived for the branching fraction $ \mathcal{B}\left(\textrm{H}\to \textrm{aa}\to \textrm{b}\overline{\textrm{b}}\textrm{b}\overline{\textrm{b}}\right) $. The combined WH and ZH observed upper limit on the branching fraction ranges from 1.10 for m$_{a}$ = 20 GeV to 0.36 for m$_{a}$ = 60 GeV, complementing other measurements in the μμττ, ττττ and bbℓℓ (ℓ = μ, τ) channels.[graphic not available: see fulltext]

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

Search for a heavy Higgs boson decaying into two lighter Higgs bosons in the ττbb final state at 13 TeV

A search for a heavy Higgs boson H decaying into the observed Higgs boson h with a mass of 125 GeV and another Higgs boson hS is presented. The h and hS bosons are required to decay into a pair of tau leptons and a pair of b quarks, respectively. The search uses a sample of proton-proton collisions collected with the CMS detector at a center-of-mass energy of 13 TeV, corresponding to an integrated luminosity of 137 fb -1 . Mass ranges of 240–3000 GeV for m H and 60–2800 GeV for m h S are explored in the search. No signal has been observed. Model independent 95% confidence level upper limits on the product of the production cross section and the branching fractions of the signal process are set with a sensitivity ranging from 125 fb (for m H = 240 GeV) to 2.7 fb (for m H = 1000 GeV). These limits are compared to maximally allowed products of the production cross section and the branching fractions of the signal process in the next-to-minimal supersymmetric extension of the standard model.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2022

A Neural Network Approach to Predict Gibbs Free Energy of Ternary Solid Solutions

Here, we present a data-centric deep learning (DL) approach using neural networks (NNs) to predict the thermodynamics of ternary solid solutions. We explore how NNs can be trained with a dataset of Gibbs free energies computed from a CALPHAD database to predict ternary systems as a function of composition and temperature. We have chosen the energetics of the FCC solid solution phase in 226 binaries consisting of 23 elements at 11 different temperatures to demonstrate the feasibility. The number of binary data points included in the present study is 102,000. We select six ternaries to augment the binary dataset to investigate their influence on the NN prediction accuracy. We examine the sensitivity of data sampling on the prediction accuracy of NNs over selected ternary systems. It is anticipated that the current DL workflow can be further elevated by integrating advanced descriptors beyond the elemental composition and more curated training datasets to improve prediction accuracy and applicability.

42 ENGINEERING↗
DOE OSTI2025

Deciphering the Scattering of Mechanically Driven Polymers Using Deep Learning

Here, we present a deep learning approach for analyzing two-dimensional scattering data of semiflexible polymers under external forces. In our framework, scattering functions are compressed into a three-dimensional latent space using a Variational Autoencoder (VAE), and two converter networks establish a bidirectional mapping between the polymer parameters (bending modulus, stretching force, and steady shear) and the scattering functions. The training data are generated using off-lattice Monte Carlo simulations to avoid the orientational bias inherent in lattice models, ensuring robust sampling of polymer conformations. The feasibility of this bidirectional mapping is demonstrated by the organized distribution of polymer parameters in the latent space. By integrating the converter networks with the VAE, we obtain a generator that produces scattering functions from given polymer parameters and an inferrer that directly extracts polymer parameters from scattering data. While the generator can be utilized in a traditional least-squares fitting procedure, the inferrer produces comparable results in a single pass and operates 3 orders of magnitude faster. This approach offers a scalable automated tool for polymer scattering analysis and provides a promising foundation for extending the method to other scattering models, experimental validation, and the study of time-dependent scattering data.

Ding, Lijie [Oak Ridge National Laboratory (ORNL),↗