Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Task Performance and Analysis”

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 1,117 records · Page 62
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Hydrogen Analysis by Gas Chromatography–Mass Spectrometry

The detection of hydrogen in complex gas mixtures is essential for many applications. Conventional approaches such as gas chromatography (GC) with thermal conductivity detection (TCD) and residual gas analyzers (RGAs) face significant limitations: TCD exhibits poor response when helium is used as a carrier gas, whereas RGAs lack chromatographic separation, preventing reliable quantification of hydrogen because of interference from other species. Traditional GC methods rely on dual-column configurations with packed or molecular-sieve porous layer open tubular (PLOT) columns to separate hydrogen from O2, N2, CO, CO2, CH4, and other hydrocarbons, increasing system complexity and limiting compatibility with mass spectrometry (MS) detection. In this work, we developed and validated a robust GC–MS method capable of directly detecting and quantifying hydrogen using electron ionization (EI) without dopants, reagent gases, or ion–molecule reaction schemes. By integrating a modified EI source and a cryogenically cooled single capillary column configuration, we achieved baseline separation of hydrogen from all major permanent gases and hydrocarbons in a refinery gas mixture using helium as the carrier gas. The method demonstrated excellent linearity, high sensitivity, and exceptional reproducibility. Adjustable sample-loop volumes and split ratios enabled optimization of peak shape and signal-to-noise performance for trace-level and percent-level hydrogen concentrations. Beyond hydrogen quantitation, the method provides simultaneous compositional profiling of other gases in a mixture in a single run, making it valuable for a wide range of tasks.

Lobodin, Vlad [ORNL]↗
DOE OSTI2020

Interpreting and Stabilizing Machine-Learning Parametrizations of Convection

Neural networks are a promising technique for parameterizing subgrid-scale physics (e.g., moist atmospheric convection) in coarse-resolution climate models, but their lack of interpretability and reliability prevents widespread adoption. For instance, it is not fully understood why neural network parameterizations often cause dramatic instability when coupled to atmospheric fluid dynamics. This paper introduces tools for interpreting their behavior that are customized to the parameterization task. First, we assess the nonlinear sensitivity of a neural network to lower-tropospheric stability and the midtropospheric moisture, two widely studied controls of moist convection. Second, we couple the linearized response functions of these neural networks to simplified gravity wave dynamics, and analytically diagnose the corresponding phase speeds, growth rates, wavelengths, and spatial structures. To demonstrate their versatility, these techniques are tested on two sets of neural networks, one trained with a super-parameterized version of the Community Atmosphere Model (SPCAM) and the second with a near-global cloud-resolving model (GCRM). Additionally, even though the SPCAM simulation has a warmer climate than the cloud-resolving model, both neural networks predict stronger heating/drying in moist and unstable environments, which is consistent with observations. Moreover, the spectral analysis can predict that instability occurs when GCMs are coupled to networks that support gravity waves that are unstable and have phase speeds larger than 5 m s -1 . In contrast, standing unstable modes do not cause catastrophic instability. Using these tools, differences between the SPCAM-trained versus GCRM-trained neural networks are analyzed, and strategies to incrementally improve both of their coupled online performance unveiled.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Deep kernel methods learn better: from cards to process optimization

Abstract The ability of deep learning methods to perform classification and regression tasks relies heavily on their capacity to uncover manifolds in high-dimensional data spaces and project them into low-dimensional representation spaces. In this study, we investigate the structure and character of the manifolds generated by classical variational autoencoder (VAE) approaches and deep kernel learning (DKL). In the former case, the structure of the latent space is determined by the properties of the input data alone, while in the latter, the latent manifold forms as a result of an active learning process that balances the data distribution and target functionalities. We show that DKL with active learning can produce a more compact and smooth latent space which is more conducive to optimization compared to previously reported methods, such as the VAE. We demonstrate this behavior using a simple cards dataset and extend it to the optimization of domain-generated trajectories in physical systems. Our findings suggest that latent manifolds constructed through active learning have a more beneficial structure for optimization problems, especially in feature-rich target-poor scenarios that are common in domain sciences, such as materials synthesis, energy storage, and molecular discovery. The Jupyter Notebooks that encapsulate the complete analysis accompany the article.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Benchmarking of Probabilistic Fracture Mechanics Models in Grizzly

In 2020, as part of contract/task order 31310019N0006/31310020F0060, the Nuclear Regulatory Commission tasked Idaho National Laboratory (INL) with benchmarking the MOOSE/GRIZZLY code against the NRC’s Fracture Analysis of Vessels - Oak Ridge (FAVOR), version 16.1 code. Both codes can predict large light-water reactor vessel integrity deterministically and probabilistically. In fact, such predictions are the primary purpose of the FAVOR code, and are a main focus of the GRIZZLY code, although GRIZZLY is capable of modeling a wider range of geometries and problems. The vessel integrity modeling capabilities of the GRIZZLY code are still under development and being gradually expanded, and the capability to model vessel fracture by crack growth has not yet been integrated into GRIZZLY (although it does exist in pre-release versions). Thus, the present benchmark study focused on comparing predictions for the conditional probability of crack growth initiation (CPI) between GRIZZLY and FAVOR v16.1. The present benchmarking study expands on a previous study completed in 2020, which focused on single plate regions, various flaw types, and full vessel studies. The previous study was limited in scope and revealed some relatively large discrepancies, particularly in the case of weld regions. In the present study, detailed investigations into the sources of discrepancies between GRIZZLY and FAVOR were performed, with a focus on weld regions.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2023

Design and implementation of I/O performance prediction scheme on HPC systems through large-scale log analysis

Abstract Large-scale high performance computing (HPC) systems typically consist of many thousands of CPUs and storage units used by hundreds to thousands of users simultaneously. Applications from large numbers of users have diverse characteristics, such as varying computation, communication, memory, and I/O intensity. A good understanding of the performance characteristics of each user application is important for job scheduling and resource provisioning. Among these performance characteristics, I/O performance is becoming increasingly important as data sizes rapidly increase and large-scale applications, such as simulation and model training, are widely adopted. However, predicting I/O performance is difficult because I/O systems are shared among all users and involve many layers of software and hardware stack, including the application, network interconnect, operating system, file system, and storage devices. Furthermore, updates to these layers and changes in system management policy can significantly alter the I/O behavior of applications and the entire system. To improve the prediction of the I/O performance on HPC systems, we propose integrating information from several different system logs and developing a regression-based approach to predict the I/O performance. Our proposed scheme can dynamically select the most relevant features from the log entries using various feature selection algorithms and scoring functions, and can automatically select the regression algorithm with the best accuracy for the prediction task. The evaluation results show that our proposed scheme can predict the write performance with up to 90% prediction accuracy and the read performance with up to 99% prediction accuracy using the real logs from the Cori supercomputer system at NERSC.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Shifting Left for Machine Learning: An Empirical Study of Security Weaknesses in Supervised Learning-based Projects

Context: Supervised learning-based projects (SLPs), i.e., software projects that use supervised learning algorithms, such as decision trees are useful for performing classification-related tasks. Yet, security weaknesses, such as the use of hard-coded passwords in SLPs, can make SLPs susceptible to security attacks. A characterization of security weaknesses in SLPs can help practitioners understand the security weaknesses that are frequent in SLPs and adopt adequate mitigation strategies. Objective: The goal of this paper is to help practitioners se-curely develop supervised learning-based projects by conducting an empirical study of security weaknesses in supervised learning-based projects. Methodology: We conduct an empirical study by quantifying the frequency of security weaknesses in 278 open source SLPs. Results: We identify 22 types of security weaknesses that occur in SLPs. We observe ‘use of potentially dangerous function’ to be the most frequently occurring security weakness in SLPs. Of the identified 3,964 security weaknesses, 23.79 % and 40.49 % respectively, appear for source code files used to train and test models. We also observe evidence of co-location, e.g., instances of command injection co-locates with instances of potentially dangerous function. Conclusion: Based on our findings, we advocate for a shift left approach for SLP development with security-focused code reviews, and application of security static analysis.

Bhuiyan, Farzana Ahamed↗
DOE OSTI2016

Angular Distributions of High-Mass Dilepton Production in Hadron Collisions

The SeaQuest experiment is a fixed-target dimuon experiment currently running at the Fermi National Accelerator Laboratory (FNAL). By utilizing the high-intensity, 120 GeV proton beam delivered by the FNAL Main Injector (MI), SeaQuest is able to measure proton-induced Drell-Yan dimuon production off of various nuclear targets in kinematic regions inaccessible to previous similar experiments. A suitably large fraction of the final dataset has been recorded, reconstructed, and analyzed. Very preliminary results from light-sea flavor asymmetry, nuclear dependence, and partonic energy loss analyses have been presented at numerous international conferences. A novel, FPGA-based trigger system has been designed, implemented, and optimized for the SeaQuest experiment. By implementing the trigger decision logic in FPGA firmware, it is more adaptable to changing experimental conditions. Additionally, the peripheral tasks of timing alignment, “trigger matrix” generation, and firmware uploading have been mostly automated, reducing the likelihood of user error in the maintenance and operation of the trigger system. Significant upgrades to hardware and firmware have greatly improved the performance of the trigger system since the 2012 commissioning run of SeaQuest. Four additional v1495 modules were added to facilitate thorough pulser testing of the firmware designs and in-situ pulser tests of all compiled firmware. These pulser tests proved crucial for diagnosing many errors that may have otherwise gone unnoticed. A significant change to the internal clocking of the trigger system eliminated a subtle source of rate-dependent trigger efficiency. With this upgrade, the trigger finally meets the “dead-time free” design specification. Drell-Yan dimuon data have been collected and analyzed for central θCS , with nearly flat acceptance in φCS , in the mass range 5.0 GeV < Mγ* < 10.0 GeV at forward xF with the SeaQuest spectrometer at FNAL. A very preliminary extraction of λ has been performed, and the remaining difficulties in extracting ν have been evaluated. Although the results are not yet publishable, significant progress has been made in developing this very challenging angular distributions analysis. A simple scheme for correcting for the angular acceptances of the spectrometer, trigger, and reconstruction has been developed and demonstrated. A generally applicable correction for the kinematically-dependent, rate-dependent reconstruction efficiency has been developed and applied to all current analyses on SeaQuest data. This rate-dependence correction was the first major hurdle in the path to publication of many preliminary SeaQuest results. The last remaining major correction for all analyses, but especially important for the angular parameter extraction, is the full characterization, rate-dependence correction, and subtraction of the combinatoric background contribution to the reconstructed dimuon sample. Independently, an intuitive, kinematic derivation of the single-event definitions of the Drell-Yan angular parameters has been developed under the assumption of unpolarized annihilating quarks within unpolarized nuclei. At O(αs), where the quarks remain co-planar with the hadrons in the photon rest frame, this kinematic method reproduces the Lam-Tung relation and derives an additional equality for µ2, which is only interpretable for single-event parameters. This method has been extended to the case of quark non- coplanarity, and the coplanar equalities become inequalities. A new equality was discovered, which should be obeyed by single-event parameters even in the case of a non-coplanar quark axis. The non-coplanar parameter relations have been used to derive constraints on the experimentally accessible values of λ and ν. These constraints are compared with existing data and have been found consistent, except in the cases where significant contributions from non-zero Boer-Mulders functions are expected. Finally, the kinematically- derived parameter definitions have been applied to high-precision CMS data. The relative contributions of the qq¯ and qg processes to the Z-boson “Drell-Yan” cross-section have been extracted. Further, an average measure of non-coplanarity, likely caused by O(α2) and higher processes, has been extracted.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2017

Gradient-based Optimization for Regression in the Functional Tensor-Train Format

We consider the task of low-multilinear-rank functional regression, i.e., learning a low-rank parametric representation of functions from scattered real-valued data. Our first contribution is the development and analysis of gradient-based optimization procedures, including stochastic gradient descent and quasi-Newton methods, for learning the parameters of a functional tensor-train (FT) that yields improved accuracy over standard alternating least squares methods. The functional tensor-train uses the tensor-train (TT) representation of low-rank arrays as an ansatz for a class of low-multilinear-rank functions. The FT is represented by a set of matrix-valued functions that contain a set of univariate functions, and the regression task is to learn the parameters of these univariate functions. Our second contribution demonstrates that using nonlinearly parameterized univariate functions, e.g., symmetric kernels with moving centers, within each core can outperform the standard approach of using a linear expansion of basis functions. Our final contributions are new rank adaptation and group-sparsity regularization procedures to minimize overfitting. We use several benchmark problems to demonstrate at least an order of magnitude lower accuracy with gradient-based optimization methods than ALS in the low-sample number regime. We also demonstrate an order of magnitude reduction in accuracy on a test problem resulting from nonlinear parameterizations over linear parameterizations. Finally we compare regression performance with 22 other nonparametric and parametric regression methods on 10 real-world data sets. We achieve top-five accuracy for seven of the data sets and best accuracy for two of the data sets. These rankings are the best amongst parametric models and competetive with the best non-parametric methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

Computerized detection of noncalcified plaques in coronary CT angiography: Evaluation of topological soft gradient prescreening method and luminal analysis

Purpose: The buildup of noncalcified plaques (NCPs) that are vulnerable to rupture in coronary arteries is a risk for myocardial infarction. Interpretation of coronary CT angiography (cCTA) to search for NCP is a challenging task for radiologists due to the low CT number of NCP, the large number of coronary arteries, and multiple phase CT acquisition. The authors conducted a preliminary study to develop machine learning method for automated detection of NCPs in cCTA. Methods: With IRB approval, a data set of 83 ECG-gated contrast enhanced cCTA scans with 120 NCPs was collected retrospectively from patient files. A multiscale coronary artery response and rolling balloon region growing (MSCAR-RBG) method was applied to each cCTA volume to extract the coronary arterial trees. Each extracted vessel was reformatted to a straightened volume composed of cCTA slices perpendicular to the vessel centerline. A topological soft-gradient (TSG) detection method was developed to prescreen for NCP candidates by analyzing the 2D topological features of the radial gradient field surface along the vessel wall. The NCP candidates were then characterized by a luminal analysis that used 3D geometric features to quantify the shape information and gray-level features to evaluate the density of the NCP candidates. With machine learning techniques, useful features were identified and combined into an NCP score to differentiate true NCPs from false positives (FPs). To evaluate the effectiveness of the image analysis methods, the authors performed tenfold cross-validation with the available data set. Receiver operating characteristic (ROC) analysis was used to assess the classification performance of individual features and the NCP score. The overall detection performance was estimated by free response ROC (FROC) analysis. Results: With our TSG prescreening method, a prescreening sensitivity of 92.5% (111/120) was achieved with a total of 1181 FPs (14.2 FPs/scan). On average, six features were selected during the tenfold cross-validation training. The average area under the ROC curve (AUC) value for training was 0.87 ± 0.01 and the AUC value for validation was 0.85 ± 0.01. Using the NCP score, FROC analysis of the validation set showed that the FP rates were reduced to 3.16, 1.90, and 1.39 FPs/scan at sensitivities of 90%, 80%, and 70%, respectively. Conclusions: The topological soft-gradient prescreening method in combination with the luminal analysis for FP reduction was effective for detection of NCPs in cCTA, including NCPs causing positive or negative vessel remodeling. The accuracy of vessel segmentation, tracking, and centerline identification has a strong impact on NCP detection. Studies are underway to further improve these techniques and reduce the FPs of the CADe system.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2025

Reimagining Disassembly Interfaces With Visualization: Combining Instruction Tracing and Control Flow With DisViz

In applications where efficiency is critical, developers may examine their compiled binaries, seeking to understand how the compiler transformed their source code and what performance implications that transformation may have. This analysis is challenging due to the vast number of disassembled binary instructions and the many-to-many mappings between them and the source code. These problems are exacerbated as source code size increases, giving the compiler more freedom to map and disperse binary instructions across the disassembly space. Interfaces for disassembly typically display instructions as an unstructured listing or sacrifice the order of execution. Here, we design a new visual interface for disassembly code that combines execution order with control flow structure, enabling analysts to both trace through code and identify familiar aspects of the computation. Central to our approach is a novel layout of instructions grouped into basic blocks that displays a looping structure in an intuitive way. We add to this disassembly representation a unique block-based mini-map that leverages our layout and shows context across thousands of disassembly instructions. Finally, we embed our disassembly visualization in a web-based tool, DisViz, which adds dynamic linking with source code across the entire application. DizViz was developed in collaboration with program analysis experts following design study methodology and was validated through evaluation sessions with ten participants from four institutions. Participants successfully completed the evaluation tasks, hypothesized about compiler optimizations, and noted the utility of our new disassembly view. Our evaluation suggests that our new integrated view helps application developers in understanding and navigating disassembly code.

Computer science↗
DOE OSTI2014

Cutting-edge issues of core-collapse supernova theory

Based on multi-dimensional neutrino-radiation hydrodynamic simulations, we report several cutting-edge issues about the long-veiled explosion mechanism of core-collapse supernovae (CCSNe). In this contribution, we pay particular attention to whether three-dimensional (3D) hydrodynamics and/or general relativity (GR) would or would not help the onset of explosions. By performing 3D simulations with spectral neutrino transport, we show that it is more difficult to obtain an explosion in 3D than in 2D. In addition, our results from the first generation of full general relativistic 3D simulations including approximate neutrino transport indicate that GR can foster the onset of neutrino-driven explosions. Based on our recent parametric studies using a light-bulb scheme, we discuss impacts of nuclear energy deposition behind the supernova shock and stellar rotation on the neutrino-driven mechanism, both of which have yet to be included in the self-consistent 3D supernova models. Finally we give an outlook with a summary of the most urgent tasks to extract the information about the explosion mechanisms from multi-messenger CCSN observables.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

Charge transfer excitation energies from ground state density functional theory calculations

Calculating charge transfer (CT) excitation energies with high accuracy and low computational cost is a challenging task. Kohn-Sham density functional theory (KS-DFT), due to its efficiency and accuracy, has achieved great success in describing ground state problems. To extend to excited state problems, our group recently demonstrated an approach with good numerical results to calculate low-lying and Rydberg excitation energies of an N-electron system from a ground state KS or generalized KS calculations of an (N-1)-electron system via its orbital energies. In the present work, we explore further the same methodology to describe CT excitations. Numerical results from this work show that performance of conventional density functional approximations (DFAs) is not as good for CT excitations as for other excitations due to the delocalization error. Applying localized orbital scaling correction (LOSC) to conventional DFAs, a recently developed method in our group to effectively reduce the delocalization error, can improve the results. Overall, the performance of this methodology is better than time dependent DFT (TDDFT) with conventional DFAs. In addition, it shows that results from LOSC-DFAs in this method reach similar accuracy to other methods, such as ΔSCF, G 0 W 0 with Bethe-Salpeter equations, particle-particle random phase approximation, and even high-level wavefunction methods like CC2. Our analysis shows that the correct 1/R trend for CT excitation can be captured from LOSC-DFA calculations, stressing that the application of DFAs with the minimal delocalization error is essential within this methodology. This work provides an efficient way to calculate CT excitation energies from ground state DFT.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2018

Characterization of Pulmonary Nodules Based on Features of Margin Sharpness and Texture

Lung cancer is the leading cause of cancer-related deaths in the world, and one of its manifestations occurs with the appearance of pulmonary nodules. The classification of pulmonary nodules may be a complex task to specialists due to temporal, subjective, and qualitative aspects. Therefore, it is important to integrate computational tools to the early pulmonary nodule classification process, since they have the potential to characterize objectively and quantitatively the lesions. In this context, the goal of this work is to perform the classification of pulmonary nodules based on image features of texture and margin sharpness. Computed tomography scans were obtained from a publicly available image database. Texture attributes were extracted from a co-occurrence matrix obtained from the nodule volume. Margin sharpness attributes were extracted from perpendicular lines drawn over the borders on all nodule slices. Feature selection was performed by different algorithms. Classification was performed by several machine learning classifiers and assessed by the area under the receiver operating characteristic curve, sensitivity, specificity, and accuracy. Highest classification performance was obtained by a random forest algorithm with all 48 extracted features. However, a decision tree using only two selected features obtained statistically equivalent performance on sensitivity and specificity.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2019

Rotating Detonation Combustion for Gas Turbines: Modeling and System Synthesis to Exceed 65% Efficiency Goal (Phase II)

As of May 2019, all RDE test hardware and test operations were completed for this program. Over 700 hot fire tests were conducted at 4 facilities with 7 RDE assemblies, with the data available for future analysis. In particular, measured static and dynamic pressure and computational fluid dynamics (CFD) provided basis for pressure gain combustion (PGC) analysis of the Engineering Scale RDE. It was demonstrated via measurements and CFD that a carefully designed diffuser can recover some total pressure of the highly oscillatory RDE exhaust. In general, the isolator/injector is largest contributor of pressure losses; however a low loss design was demonstrated that improves significantly on previous natural-gas/air RDEs. Also, a tailored injector and mixing technique was demonstrated that allowed NG-air RDE operation with overall equivalence ratio of less than 0.6. CFD models by AR, the UofM and Purdue University were extensively used throughout the program to refine diffuser design for improved pressure recovery, to simulate injection and mixing, to estimate turbine interaction, to estimate combustor pressure gain anchored with test data and to provide input to other design and analytical tasks. CFD analyses were completed for many of the configurations tested at Purdue with the 13.4” RDE including the shark fin and low area ratio isolators. Results were compared favorably to available hot fire test particle image velocimetry (PIV) measurements and wave speeds. The results of these CFD analyses suggest a modest pressure gain of 33% with a sufficiently low area ratio isolator. A thermodynamic power plant system model was set up which utilizes an RDE combustor. The performance of the RDE combustor module was scaled based on measurements from the Engineering Scale RDE tests with low loss injector and tailored fuel flow. This combined model indicates potential for significant plant efficiency improvement. The successful tailored fuel injection test result with lower than .6 equivalence ratio has implication that a power plant can use a RDE combustor configuration where all the air flows through the combustion zone thereby not requiring a bypass mixer. In this configuration, the results of the natural gas combined cycle (NGCC) plant system indicate net efficiency of 66.4%.

02 PETROLEUM↗
DOE OSTI2016

WTP Waste Feed Qualification: Hydrogen Generation Rate Measurement Apparatus Testing Report

The generation rate of hydrogen gas in the Hanford tank waste will be measured during the qualification of the staged tank waste for processing in the Hanford Tank Waste Treatment and Immobilization Plant. Based on a review of past practices in measurement of the hydrogen generation, an apparatus to perform this measurement has been designed and tested for use during waste feed qualification. The hydrogen generation rate measurement apparatus (HGRMA) described in this document utilized a 100 milliliter sample in a continuously-purged, continuously-stirred vessel, with measurement of hydrogen concentration in the vent gas. The vessel and lid had a combined 220 milliliters of headspace. The vent gas system included a small condenser to prevent excessive evaporative losses from the sample during the test, as well as a demister and filter to prevent particle migration from the sample to the gas chromatography system. The gas chromatograph was an on line automated instrument with a large-volume sample-injection system to allow measurement of very low hydrogen concentrations. This instrument automatically sampled the vent gas from the hydrogen generation rate measurement apparatus every five minutes and performed data regression in real time. The fabrication of the hydrogen generation rate measurement apparatus was in accordance with twenty three (23) design requirements documented in the conceptual design package, as well as seven (7) required developmental activities documented in the task plan associated with this work scope. The HGRMA was initially tested for proof of concept with physical simulants, and a remote demonstration of the system was performed in the Savannah River National Laboratory Shielded Cells Mockup Facility. Final verification testing was performed using non-radioactive simulants of the Hanford tank waste. Three different simulants were tested to bound the expected rheological properties expected during waste feed qualification testing. These simulants were tested at different temperatures using purge gas spiked with varying amounts of hydrogen to provide verification that the system could accurately measure the hydrogen in the vent gas at steady state.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2013

Analysis of Radiocarbon, Stable Isotopes and DNA in Teeth to Facilitate Identification of Unknown Decedents

The characterization of unidentified bodies or suspected human remains is a frequent and important task for forensic investigators. However, any identification method requires clues to the person’s identity to allow for comparisons with missing persons. If such clues are lacking, information about the year of birth, sex and geographic origin of the victim, is particularly helpful to aid in the identification casework and limit the search for possible matches. We present here results of stable isotope analysis of 13 C and 18 O, and bomb-pulse 14 C analyses that can help in the casework. The 14 C analysis of enamel provided information of the year of birth with an average absolute error of 1.8±1.3 years. We also found that analysis of enamel and root from the same tooth can be used to determine if the 14 C values match the rising or falling part of the bomb-curve. Enamel laydown times can be used to estimate the date of birth of individuals, but here we show that this detour is unnecessary when using a large set of crude 14 C data of tooth enamel as a reference. The levels of 13 C in tooth enamel were higher in North America than in teeth from Europe and Asia, and Mexican teeth showed even higher levels than those from USA. DNA analysis was performed on 28 teeth, and provided individual-specific profiles in most cases and sex determination in all cases. In conclusion, these analyses can dramatically limit the number of possible matches and hence facilitate person identification work.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2019

ASCR Workshop on In Situ Data Management: Enabling Scientific Discovery from Diverse Data Sources

In January 2019, the U.S. Department of Energy, Office of Science program in Advanced Scientific Computing Research, convened a workshop to identify priority research directions for in situ data management (ISDM). The workshop defined ISDM as the practices, capabilities, and procedures to control the organization of data and enable the coordination and communication among heterogeneous tasks, executing simultaneously in a high-performance computing system, cooperating toward a common objective. The workshop revealed two primary, interdependent motivations for processing and managing data in situ. The first motivation is that the in situ methodology enables scientific discovery from a broad range of data sources over a wide scale of computing platforms: leadership-class systems, clusters, clouds, workstations, and embedded devices at the edge. The successful development of ISDM capabilities will benefit real-time decision-making, design optimization, and data-driven scientific discovery. The second motivation is the need to decrease data volumes. ISDM can make critical contributions to managing large data volumes from computations and experiments to minimize data movement, save storage space, and boost resource efficiency, often while simultaneously increasing scientific precision. A fundamental finding of this workshop is that the methodologies used to manage data among a variety of tasks in situ can be used to facilitate scientific discovery from many different data sources—simulation, experiment, and sensors, for example—and that being able to do so at numerous computing scales will benefit real-time decision-making, design optimization, and data-driven scientific discovery across the Office of Science mission space. Applications wanting to use the in situ capabilities include those where data analysis feeds back to the simulation, decisions are made autonomously, big data or machine learning is among the tasks to be coordinated, and computations need to be completed in real time. The workshop identified six priority research directions that highlight the components and capabilities needed for ISDM to be successful for the wide variety of applications discussed: making ISDM capabilities more pervasive, controllable, composable, and transparent, with a focus on greater coordination with the software stack and a diversity of fundamentally new data algorithms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Automated Grain Boundary (GB) Segmentation and Microstructural Analysis in 347H Stainless Steel Using Deep Learning and Multimodal Microscopy

Austenitic 347H stainless steel offers superior mechanical properties and corrosion resistance required for extreme operating conditions such as high temperature. The change in microstructure due to composition and process variations is expected to impact material properties. Identifying microstructural features such as grain boundaries thus becomes an important task in the process-microstructure-properties loop. Applying convolutional neural network (CNN)-based deep learning models is a powerful technique to detect features from material micrographs in an automated manner. In contrast to microstructural classification, supervised CNN models for segmentation tasks require pixel-wise annotation labels. However, manual labeling of the images for the segmentation task poses a major bottleneck for generating training data and labels in a reliable and reproducible way within a reasonable timeframe. Microstructural characterization especially needs to be expedited for faster material discovery by changing alloy compositions. Here, in this study, we attempt to overcome such limitations by utilizing multimodal microscopy to generate labels directly instead of manual labeling. We combine scanning electron microscopy images of 347H stainless steel as training data and electron backscatter diffraction micrographs as pixel-wise labels for grain boundary detection as a semantic segmentation task. The viability of our method is evaluated by considering a set of deep CNN architectures. We demonstrate that despite producing instrumentation drift during data collection between two modes of microscopy, this method performs comparably to similar segmentation tasks that used manual labeling. Additionally, we find that naïve pixel-wise segmentation results in small gaps and missing boundaries in the predicted grain boundary map. By incorporating topological information during model training, the connectivity of the grain boundary network and segmentation performance is improved. Finally, our approach is validated by accurate computation on downstream tasks of predicting the underlying grain morphology distributions which are the ultimate quantities of interest for microstructural characterization.

36 MATERIALS SCIENCE↗