Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Automated Testing”

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 937 records · Page 52
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

The Zwicky Transient Facility Bright Transient Survey. III. BTSbot: Automated Identification and Follow-up of Bright Transients with Deep Learning

Abstract The Bright Transient Survey (BTS) aims to obtain a classification spectrum for all bright ( m peak ≤ 18.5 mag) extragalactic transients found in the Zwicky Transient Facility (ZTF) public survey. BTS critically relies on visual inspection (“scanning”) to select targets for spectroscopic follow-up, which, while effective, has required a significant time investment over the past ∼5 yr of ZTF operations. We present BTSbot , a multimodal convolutional neural network, which provides a bright transient score to individual ZTF detections using their image data and 25 extracted features. BTSbot is able to eliminate the need for daily human scanning by automatically identifying and requesting spectroscopic follow-up observations of new bright transient candidates. BTSbot recovers all bright transients in our test split and performs on par with scanners in terms of identification speed (on average, ∼1 hr quicker than scanners). We also find that BTSbot is not significantly impacted by any data shift by comparing performance across a concealed test split and a sample of very recent BTS candidates. BTSbot has been integrated into Fritz and Kowalski , ZTF’s first-party marshal and alert broker, and now sends automatic spectroscopic follow-up requests for the new transients it identifies. Between 2023 December and 2024 May, BTSbot selected 609 sources in real time, 96% of which were real extragalactic transients. With BTSbot and other automation tools, the BTS workflow has produced the first fully automatic end-to-end discovery and classification of a transient, representing a significant reduction in the human time needed to scan.

Rehemtulla, Nabeel (ORCID:0000000256832389)↗
DOE OSTI2012

Automated design synthesis of robotic/human workcells for improved manufacturing system design in hazardous environments [Slides]

Manufacturing tasks that are deemed too hazardous for workers require the use of automation, robotics, and/or other remote handling tools. The associated hazards may be radiological or non-radiological, and based on the characteristics of the environment and processing, a design may necessitate robotic labor, human labor, or both. There are also other factors such as cost, ergonomics, maintenance, and efficiency that also effect task allocation and other design choices. Handling the tradeoffs of these factors can be complex, and lack of experience can be an issue when trying to determine if and what feasible automation/robotics options exist. To address this problem, we utilize common engineering design approaches adapted more for manufacturing system design in hazardous environments. We limit our scope to the conceptual and embodiment design stages, specifically a computational algorithm for concept generation and early design evaluation. In regard to concept generation, we first develop the functional model or function structure for the process, using the common 'verb-noun' format for describing function. A common language or functional basis for manufacturing was developed and utilized to formalize function descriptions and guide rules for function decomposition. Potential components for embodiment are also grouped in terms of this functional language and are stored in a database. The properties of each component are given as quantitative and qualitative criteria. Operators are also rated for task-relevant criteria which are used to address task compatibility. Through the gathering of process requirements/constraints, construction of the component database, and development of the manufacturing basis and rule set, design knowledge is stored and available for computer use. Thus, once the higher level process functions are defined, the computer can automate the synthesis of new design concepts through alternating steps of embodiment and function structure updates/decomposition. In the process, criteria guide function allocation of components/operators and help ensure compatibility and feasibility. Through multiple function assignment options and varied function structures, multiple design concepts are created. All of the generated designs are then evaluated based on a number of relevant evaluation criteria: cost, dose, ergonomics, hazards, efficiency, etc. These criteria are computed using physical properties/parameters of each system based on the qualities an engineer would use to make evaluations. Nuclear processes such as oxide conversion and electrorefining are utilized to aid algorithm development and provide test cases for the completed program. Through our approach, we capture design knowledge related to manufacturing and other operations in hazardous environments to enable a computational program to automatically generate and evaluate system design concepts.

42 ENGINEERING↗
DOE OSTI2018

Fully Robotic Hot Cell Crane for Target Hall Operations at Isotope Research Facility - 18329

There are many types of equipment used in high radiation environments, but the super-sized and high precision lifting requirements for an isotope research facility in the USA called for one of the largest fully robotic cranes in the US. The 20 Ton Remote Handling Bridge Crane provided by PaR Systems met many challenging requirements for accuracy, repeatability and unique lifting motions in the radiation-shielded target hall of this facility. This robotic system was designed to address four main issues. The first is that the crane needed to be operated remotely due to high radioactivity during the time when it is needed most. Second is that high precision is needed due to the complex fit of multiple heavy shielding blocks. Third problem is that automated routines and exclusion areas were needed due to the remote operation of the system around high capital cost equipment. Lastly is the survivability of the equipment due to the high radiation environment. The facility robotic crane is an overhead top running multiple girder hot cell bridge crane (hereafter referred to as the system). The system is intended to be used in either manually operated or automated operation under computer control. Under automated control, the system provides highly accurate motion control in three linear axes as well as hook rotation. Using motion control programming, this highly accurate system ensures safe, remote movement of in-cell components. The system was designed to be remotely operated by a technician from the remote handling control room in the facility. This is facilitated by computer control using the System Graphical User Interface with feedback from a facility supplied video system. It can also be operated using a wireless hand held controller. The system will be used to place sensitive components with high precision in locations that will be difficult to observe directly. In order to accomplish this, it is critical that the system be capable of moving its payload with minimal swing, and with high accuracy, repeatability and reliability. The system will be operating in an environment where it will be exposed to both high energy neutrons and gamma radiation. Under beam-off conditions, the target hall will be hands-on accessible to accommodate crane maintenance. The system was installed and tested in early 2017 in advance of the live operations of the facility. (authors)

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

A Run-Time Verification Framework for Smart Grid Applications Implemented on Simulation Frameworks

Smart grid applications are implemented and tested with simulation frameworks as the developers usually do not have access to large sensor networks to be used as a test bed. The developers are forced to map the implementation onto these frameworks which results in a deviation between the architecture and the code. On its turn this deviation makes it hard to verify behavioral constraints that are de- scribed at the architectural level. We have developed the ConArch toolset to support the automated verification of architecture-level behavioral constraints. A key feature of ConArch is programmable mapping for architecture to the implementation. Here, developers implement queries to identify the points in the target program that correspond to architectural interactions. ConArch generates run- time observers that monitor the flow of execution between these points and verifies whether this flow conforms to the behavioral constraints. We illustrate how the programmable mappings can be exploited for verifying behavioral constraints of a smart grid appli- cation that is implemented with two simulation frameworks.

Architecture baed testing↗
DOE OSTI2023

High-throughput mass spectrometry analysis using immediate drop-on-demand technology coupled with an open port sampling interface

Rationale: The sampling throughput of immediate drop-on-demand technology (I.DOT) coupled with an open port sampling interface (OPSI) is limited by software communication. To enable much-needed high-throughput mass spectrometry (MS) analysis capabilities, a novel software was developed that allows for flexible sample selection from a 96-well plate and for maximized analysis throughput using I.DOT/OPSI-MS coupling. Methods: Wells of a 96-well I.DOT plate were filled with propranolol solution and were used to test maximum sampling throughput strategies to minimize analysis time. Demonstration of chemical reaction monitoring was done using acid-catalyzed ring closure of 2,3-diaminonaphthalene (DAN) with nitrite to form 2,3-naphthotriazole (NAT). Analytes were detected in positive electrospray ionization mode using selected reaction monitoring. Results: A maximum throughput of 1.54 s/sample (7.41 min/96-well plate with three technical replicates) was achieved, and it was limited by the peak width of the MS signal resulting in an occasional slight overlap between the peaks. Further, relative standard deviation was 10 ± 1% with all tested sampling strategies. Chemical reaction monitoring of DAN to NAT using nitrite was successfully accomplished with 2 s/sample throughout showing almost complete transformation in 10 min with no signal overlap. Conclusions: This work illustrates the development of a noncontact, automated I.DOT/OPSI-MS system with improved throughput achieved through an optimized software interface. Its achievable analysis time and precision make it a viable approach for drug discovery and in situ reaction monitoring studies.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Consistent performance of large language models in rare disease diagnosis across ten languages and 4917 cases

Background Large language models (LLMs) are increasingly used medicine for diverse applications including differential diagnostic support. The training data used to create LLMs such as the Generative Pretrained Transformer (GPT) predominantly consist of English-language texts, but LLMs could be used across the globe to support diagnostics if language barriers could be overcome. Initial pilot studies on the utility of LLMs for differential diagnosis in languages other than English have shown promise, but a large-scale assessment on the relative performance of these models in a variety of European and non-European languages on a comprehensive corpus of challenging rare-disease cases is lacking. Methods We created 4917 clinical vignettes using structured data captured with Human Phenotype Ontology (HPO) terms with the Global Alliance for Genomics and Health (GA4GH) Phenopacket Schema. These clinical vignettes span a total of 360 distinct genetic diseases with 2525 associated phenotypic features. We used translations of the Human Phenotype Ontology together with language-specific templates to generate prompts in English, Chinese, Czech, Dutch, French, German, Italian, Japanese, Spanish, and Turkish. We applied GPT-4o, version gpt-4o-2024-08-06, and the medically fine-tuned Meditron3-70B to the task of delivering a ranked differential diagnosis using a zero-shot prompt. An ontology-based approach with the Mondo disease ontology was used to map synonyms and to map disease subtypes to clinical diagnoses in order to automate evaluation of LLM responses. Findings For English, GPT-4o placed the correct diagnosis at the first rank 19.9% and within the top-3 ranks 27.0% of the time. In comparison, for the nine non-English languages tested here the correct diagnosis was placed at rank 1 between 16.9% and 20.6%, within top-3 between 25.4% and 28.6% of cases. The Meditron3 model placed the correct diagnosis within the first 3 ranks for 20.9% of cases in English and between 19.9% and 24.0% for the other nine languages. Interpretation The differential diagnostic performance of LLMs across a comprehensive corpus of rare-disease cases was largely consistent across the ten languages tested. This suggests that the utility of LLMs in clinical settings may extend to non-English clinical settings.

Artificial intelligence↗
DOE OSTI2026

High-throughput oxidation screening and down-selection of refractory high entropy alloys in the Al-Cr-Mo-Nb-Ta-Ti system

Rapid experimentation and characterization are ever-present needs in the discovery of high entropy alloys. High entropy alloy systems are difficult to survey with systematic composition sweeps using traditional synthesis methods. The number of distinct compositions in even a four-element system is experimentally intractable. Exploration of these, and higher-element systems, necessitates thermodynamic prediction coupled with an automated sample creation method and a rapid screening methodology to effectively down-select alloys with targeted properties. As a result, a high-throughput method for evaluating the oxidation performance of refractory high entropy alloys was developed and tested. The six-element system of aluminum, chromium, molybdenum, niobium, tantalum, and titanium was evaluated for single phase stability and short-duration oxidation resistance. Target compositions were initially determined via thermodynamic predictions of single-phase stability across a wide temperature range. A twenty-five-sample build plate was produced using directed energy deposition additive manufacturing. After fabrication, the twenty-five 1 cm 3 samples were heat treated and characterized for composition and phase identification. The build plate was exposed to a high temperature oxidizing environment at 1000 °C for three hours. After oxidation, the composition, morphology, and chemistry of the oxides formed were characterized. Of the twenty-five samples produced, nine exhibited a favorable oxidation response, from which a single-phase BCC alloy at a composition of Al 13 Cr 7 Mo 19 Nb 18 Ta 26 Ti 17 was identified as the alloy with the most protective oxidation coating with a thin, adherent oxide scale. Finally, the complete experimental down-selection—from machine setup to final alloy identification—required approximately 45 labor hours, demonstrating a rapid validation for alloy discovery.

Additive manufacturing↗
DOE OSTI2022

EDGE COVID-19: a web platform to generate submission-ready genomes from SARS-CoV-2 sequencing efforts

Abstract Summary Genomics has become an essential technology for surveilling emerging infectious disease outbreaks. A range of technologies and strategies for pathogen genome enrichment and sequencing are being used by laboratories worldwide, together with different and sometimes ad hoc, analytical procedures for generating genome sequences. A fully integrated analytical process for raw sequence to consensus genome determination, suited to outbreaks such as the ongoing COVID-19 pandemic, is critical to provide a solid genomic basis for epidemiological analyses and well-informed decision making. We have developed a web-based platform and integrated bioinformatic workflows that help to provide consistent high-quality analysis of SARS-CoV-2 sequencing data generated with either the Illumina or Oxford Nanopore Technologies (ONT). Using an intuitive web-based interface, this workflow automates data quality control, SARS-CoV-2 reference-based genome variant and consensus calling, lineage determination and provides the ability to submit the consensus sequence and necessary metadata to GenBank, GISAID and INSDC raw data repositories. We tested workflow usability using real world data and validated the accuracy of variant and lineage analysis using several test datasets, and further performed detailed comparisons with results from the COVID-19 Galaxy Project workflow. Our analyses indicate that EC-19 workflows generate high-quality SARS-CoV-2 genomes. Finally, we share a perspective on patterns and impact observed with Illumina versus ONT technologies on workflow congruence and differences. Availability and implementation https://edge-covid19.edgebioinformatics.org, and https://github.com/LANL-Bioinformatics/EDGE/tree/SARS-CoV2. Supplementary information Supplementary data are available at Bioinformatics online.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2013

Investigation of processing advanced waste forms in Cold Crucible Induction Melter systems

The Cold Crucible Induction Melter (CCIM) technology has been identified in several independent reviews and studies as a potential alternative or supplemental vitrification technology for specific U.S. Department of Energy (DOE) high level radioactive waste (HLW) streams. These inventories have unique chemistries, such as high alumina, high iron, high chromium, etc., that are not amenable to efficient immobilization in conventional borosilicate glass (BSG) matrices. Advanced waste forms, such as iron phosphate glass (FePG), high crystalline BSG (HC-BSG), and alumino-silicate glass-ceramic (GC) have been shown to provide significantly improved immobilization matrices for these challenging waste streams, including dramatically improved waste loadings; however, they are difficult to efficiently process in the baseline joule-heated ceramic melters (JHCMs). The CCIM technology can effectively process these advanced waste forms at the desired high waste loadings. This paper describes the most recent collaborative activities between the DOE Office of Environmental Management (DOE EM) and various Russian institutions in modeling, development and testing of advanced waste forms in CCIMs, including FePG and HC-BSG formulations. Additionally, past interactions with other international partners, including France and Korea, are described. Finally, recent enhancements to CCIM processing related to modeling, draining techniques and automated control, which were collaboratively developed, will be discussed. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2023

Use of Graph Theory and Neural Networks for Microstructural Classification

Recent advances in materials data analytics have provided new avenues for determining process-structure-property (PSP) linkages in a variety of materials. Machine learning techniques including few-shot learning have increased the efficiency of classifying microscopy images for the purposes of material characterization. Modifications in segmentation also show potential in improving the accuracy of our current pyCHIP classifier. Replacing previous encoders trained on ImageNet with those trained on microscopy images like MicroNet has initially shown better performance at classifying images of irradiated samples. Additionally, different normalization approaches were tested to show no discernable effect on classification. The Louvain method for community detection is analyzed on a set of irradiated samples with different parameters to determine which proved beneficial under what circumstances. We suggest that microscopy experiments be automated in the future using a combination of these techniques to enable high-throughput analyses.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Impact of Lateral Alignment on the Energy Savings of a Truck Platoon

A truck platooning system was tested using two heavy-duty tractor-trailer trucks on a closed test track to investigate the sensitivity of intentional lateral offsets over a range of intervehicle spacings. The fuel consumption for both trucks in the platoon was measured using the SAE J1321 gravimetric procedure while travelling at 65 mph and loaded to a gross weight of 65,000 lb. In addition, the SAE J1939 instantaneous fuel rate was calibrated against the gravimetric measurements and used as proxy for additional analyses. The testing campaign demonstrated the effects of intervehicle gaps, following-vehicle longitudinal control, and manual lateral control. The new results are compared to previous truck-platooning studies to reinforce the value of the new information and demonstrate similarity to past trends. Fuel savings for the following vehicle was observed to exceed 10% at closer following distances. The results showed that energy savings generally increased in a non-linear fashion as the gap was reduced. The impact of different following-truck lateral offsets had a measurable impact, with up to 4% reduction in total fuel-savings (relative to an isolated vehicle condition) observed for offsets up to 1.3 m. The fuel-consumption savings on the straight segments of the track exceeded those on the curved segments by upwards of 6% and highlight some potential differences expected between close-track testing and on-highway use.

adaptive cruise control↗
DOE OSTI2024

Testing of Turner Manufacturing Wastewater Remediation Products in Advanced Bioreactor Systems

LANL received three types of simulated wastewater (oil-contaminated soil within a water matrix; commercial laundry; cheese production), which were treated with a corresponding biological agent (enzyme and/or microbial blend) provided by the company. Experiments were set up according to their instruction within an automated bioreactor/fermenter and monitored over time. Aliquots were obtained for total organic carbon (TOC), flow cytometry, Biological Oxygen Demand, Chemical Oxygen Demand.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2022

Automated System-wide Event Detection and Classification Using Machine Learning on Synchrophasor Data

As the number of phasor measurement units (PMUs) deployed in a power system increases, and their data volume streamed to the control canter intensifies, operators are facing challenges related to the analysis of such data, which need to be observed and responded to as the measurements are displayed in the Control Room. Humans are generally unable to process such large amount of data efficiently and rapidly. There is an apparent need for automated ways to analyze the data, extract actionable information about occurrence of specific events, and characterize the events quickly and cost effectively. This paper discusses the use of machine learning (ML) to facilitate such tasks by providing automated, highly computationally efficient, and cost-effective ways of extracting actionable information from synchrophasor big data in real-time. We developed Big Data Smart (BDSmart) ML-based prototype tool for the Control Room use that automatically analyses data properties from synchrophasor system measurements taken across the three grid Interconnections in the USA (Western, Eastern and ERCOT). The data collected from several hundreds of PMUs located across the Interconnections over a period of two years have been made available for our extensive study. As a result, we were able to identify a number of big data properties that influence how ML methodology is applied to select, develop, train and test the data models that can eventually be used for the tool implementation. The resulting set of candidate algorithms spans unsupervised, supervised, semi-supervised and transfer-learning approaches. Many ML techniques, such as decision trees, multinomial logistic regression, feed-forward neural networks, K-nearest neighbor, multiclass support vector machine, and single and multi-channel convolutional neural networks, are implemented, and their performance is examined. We offer the results from testing the data models. The novelty of our study is in the approaches for bad data detection and mitigation, selection of a simplified feature for event detection, and data label improvements. As a result, we came up with a list of recommendations for the utilities on how to improve the PMU recording practices to cater to the future ML applications aimed at automating the analysis of synchrophasor data.

Synchrophasors, Machine Learning, System-wide Even↗
DOE OSTI2023

Automated detection of double nuclei galaxies using GOTHIC and the discovery of a large sample of dual AGN

We present a novel algorithm to detect double nuclei galaxies (DNG) called Gothic (Graph-bOosTed iterated HIll Climbing) – that detects whether a given image of a galaxy has two or more closely separated nuclei. Our aim is to test for the presence of dual/multiple active galactic nuclei (AGN) in galaxies that visually represent a DNG. Although galaxy mergers are common, the detection of dual AGN is rare. Their detection is very important as they help us understand the formation of supermassive black hole (SMBH) binaries, SMBH growth and AGN feedback effects in multiple nuclei systems. There is thus a need for an algorithm to do a systematic survey of existing imaging data for the discovery of DNGs and dual AGNs. We have tested Gothic on an established sample of DNGs with a 100 per cent detection rate and subsequently conducted a blind search of 1 million SDSS DR16 galaxies (with spectroscopic data available) lying in the redshift range of z = 0 to 0.75. From the list of candidate DNGs found, we have detected 159 dual AGNs, of which 2 are triple AGN systems. Our results show that dual AGNs are not common, and triple AGN even rarer. The colour (u–r) magnitude plots of the DNGs indicate that star formation is quenched as the nuclei come closer and as the AGN fraction increases. The quenching is especially prominent for dual/triple AGN galaxies that lie at the extreme end of the red sequence.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

An algorithm for detecting and quantifying disturbance and recovery in high-frequency time series

Determining when a disturbance has occurred, its severity, and when the system recovered, is important to numerous questions in the aquatic sciences. This problem can be conceptualized as the timing and degree of perturbation from a typical state, and when the system returns to that typical state. We present an algorithm for detecting disturbance and recovery designed for high-frequency time series, e.g., data produced by automated sampling devices in instrumented buoys and flux towers. The algorithm quantifies differences in the empirical cumulative distribution functions of moving windows over reference and evaluation periods, and is sensitive to changes in the mean, variance, and higher statistical moments. Tests on simulated data show it accurately identifies disturbance and recovery. Three case studies illustrate the application of our algorithm in different empirical settings. A case study on dissolved oxygen in a Florida, USA estuary following a hurricane identified the disturbance and recovery 73 d later. A case study on air temperature and net ecosystem exchange in the Florida everglades identified cold snaps coinciding with periods of reduced carbon uptake. A case study on rotifer abundance following zebra mussel invasion in the Hudson River, NY showed rotifer collapse following invasion and recovery over a decade later. Methods such as ours can improve understanding response to disturbance and facilitate comparative and synthetic study of disturbance impacts across ecosystems.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Fractal Scaling of Explosively Driven Product Gases

ABSTRACT Characterization of the interface between explosive product gases and ambient air in an explosion is a complicated task due to the turbulent mixing and inherently three‐dimensional expansion of the interface. This study aims to quantify the evolution of the interface as a temporally varying Hausdorff dimension. Two test series were conducted with Composition C‐4 charges with masses of 105 and 880 g. Imaging data were collected from the time of detonation until shock wave detachment using ultra‐high‐speed cameras. Gas cloud profiles were extracted using automated image processing algorithms, and the Hausdorff dimension of these two‐dimensional slices of the gas cloud was then estimated using boxcounting algorithms. When scaled with standard gas dynamic nondimensional scalings, the Hausdorff dimension of all explosive events appears to collapse towards a single curve. The fireball was initially nonfractal and began to develop fractal properties as the shock wave separated from the detonation products. Artificial perturbation of the charged surface had no detectable impact on the evolution of the Hausdorff dimension in the early development of the fireball outside of error, despite visible phenomenological differences in the early development of mixing on the fireball surface.

42 ENGINEERING↗
DOE OSTI2018

Advancing the Fork detector for quantitative spent nuclear fuel verification

The Fork detector is widely used by the safeguards inspectorate of the European Atomic Energy Community (EURATOM) and the International Atomic Energy Agency (IAEA) to verify spent nuclear fuel. Fork measurements are routinely performed for safeguards prior to dry storage cask loading. Additionally, spent fuel verification will be required at the facilities where encapsulation is performed for acceptance in the final repositories planned in Sweden and Finland. The use of the Fork detector as a quantitative instrument has not been prevalent due to the complexity of correlating the measured neutron and gamma ray signals with fuel inventories and operator declarations. A spent fuel data analysis module based on the ORIGEN burnup code was recently implemented to provide automated real-time analysis of Fork detector data. This module allows quantitative predictions of expected neutron count rates and gamma units as measured by the Fork detectors using safeguards declarations and available reactor operating data. This study describes field testing of the Fork data analysis module using data acquired from 339 assemblies measured during routine dry cask loading inspection campaigns in Europe. Assemblies include both uranium oxide and mixed-oxide fuel assemblies. More recent measurements of 50 spent fuel assemblies at the Swedish Central Interim Storage Facility for Spent Nuclear Fuel are also analyzed. An evaluation of uncertainties in the Fork measurement data is performed to quantify the ability of the data analysis module to verify operator declarations and to develop quantitative go/no-go criteria for safeguards verification measurements during cask loading or encapsulation operations. The goal of this approach is to provide safeguards inspectors with reliable real-time data analysis tools to rapidly identify discrepancies in operator declarations and to detect potential partial defects in spent fuel assemblies with improved reliability and minimal false positive alarms. Finally, the results are summarized, and sources and magnitudes of uncertainties are identified, and the impact of analysis uncertainties on the ability to confirm operator declarations is quantified.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2023

A rule-free workflow for the automated generation of databases from scientific literature

Abstract In recent times, transformer networks have achieved state-of-the-art performance in a wide range of natural language processing tasks. Here we present a workflow based on the fine-tuning of BERT models for different downstream tasks, which results in the automated extraction of structured information from unstructured natural language in scientific literature. Contrary to existing methods for the automated extraction of structured compound-property relations from similar sources, our workflow does not rely on the definition of intricate grammar rules. Hence, it can be adapted to a new task without requiring extensive implementation efforts and knowledge. We test our data-extraction workflow by automatically generating a database for Curie temperatures and one for band gaps. These are then compared with manually curated datasets and with those obtained with a state-of-the-art rule-based method. Furthermore, in order to showcase the practical utility of the automatically extracted data in a material-design workflow, we employ them to construct machine-learning models to predict Curie temperatures and band gaps. In general, we find that, although more noisy, automatically extracted datasets can grow fast in volume and that such volume partially compensates for the inaccuracy in downstream tasks.

36 MATERIALS SCIENCE↗