Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “improvements”

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 685 records · Page 38
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Combining biomarker and virus phylogenetic models improves HIV-1 epidemiological source identification

To identify and stop active HIV transmission chains new epidemiological techniques are needed. Here, we describe the development of a multi-biomarker augmentation to phylogenetic inference of the underlying transmission history in a local population. HIV biomarkers are measurable biological quantities that have some relationship to the amount of time someone has been infected with HIV. To train our model, we used five biomarkers based on real data from serological assays, HIV sequence data, and target cell counts in longitudinally followed, untreated patients with known infection times. The biomarkers were modeled with a mixed effects framework to allow for patient specific variation and general trends, and fit to patient data using Markov Chain Monte Carlo (MCMC) methods. Subsequently, the density of the unobserved infection time conditional on observed biomarkers were obtained by integrating out the random effects from the model fit. This probabilistic information about infection times was incorporated into the likelihood function for the transmission history and phylogenetic tree reconstruction, informed by the HIV sequence data. To critically test our methodology, we developed a coalescent-based simulation framework that generates phylogenies and biomarkers given a specific or general transmission history. Testing on many epidemiological scenarios showed that biomarker augmented phylogenetics can reach 90% accuracy under idealized situations. Under realistic within-host HIV-1 evolution, involving substantial within-host diversification and frequent transmission of multiple lineages, the average accuracy was at about 50% in transmission clusters involving 5–50 hosts. Realistic biomarker data added on average 16 percentage points over using the phylogeny alone. Using more biomarkers improved the performance. Shorter temporal spacing between transmission events and increased transmission heterogeneity reduced reconstruction accuracy, but larger clusters were not harder to get right. More sequence data per infected host also improved accuracy. We show that the method is robust to incomplete sampling and that adding biomarkers improves reconstructions of real HIV-1 transmission histories. The technology presented here could allow for better prevention programs by providing data for locally informed and tailored strategies.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2026

An improved dataset for predicting mammal infecting viruses from genetic sequence information

There have been several attempts to develop machine learning (ML) models to identify human infecting viruses from their genomic sequences, with varying degrees of success. Direct comparison between models is problematic, because these models are typically trained and evaluated on different datasets with alternative data splitting schemes, features, and model performance metrics. In this paper we present a standardized dataset of mammal infecting and non-infecting viral pathogens, refined from the previous work of Mollentze et al. to include the latest literature evidence, roughly doubling the number of curated host-virus records available to the community, and new host target labels, primate and mammal. The new host labels were included for several reasons, including previous reports that classification performance is better at broader taxonomic ranks and the idea that there may be more data for primate infection that might serve as a suitable proxy for zoonotic potential and avoidance of false positives for human infection due to absence of evidence. On this dataset, we report the performance of eight machine learning models for predicting mammal-infecting viruses from their genomic sequences. We find that randomly assigning cases in our improved dataset to training/testing sets, when compared to the original assignments into training/testing in Mollentze et al., increases the overall average ROC AUC of prediction of human infection from 0.663 ± 0.070 to 0.784 ± 0.013, consistent with the reduction in phylogenetic distance between train and test sets (relative entropy change from 3.00 to 0.08). The broadest host category of mammal infection can be predicted most reliably at 0.850 ± 0.020. We share our improved dataset and code to enable standardized comparisons of machine learning methods to predict human host infections. Overall, we have presented preliminary evidence that classification of virus host infection is more tractable at higher taxonomic ranks, that unsurprisingly reducing the phylogenetic distance between training and test sets can improve predictive performance, that peptide kmer features appear to be harmful to out of sample model performance, and we are left with the question of whether models for virus host prediction can reasonably be expected to perform well in out of sample scenarios given the likelihood that viruses do not share a common ancestor. Consistent with this concern, when the data is resampled such that there is no overlap between viral families in training and test sets (relative entropy > 24), models perform no better than random chance at prediction of human infection regardless of whether kmers are included (ROC AUC 0.50 ± 0.08) or not (ROC AUC 0.50 ± 0.04).

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Multi-contrast machine learning improves schistosomiasis diagnostic performance

Schistosomiasis currently affects over 250 million people and remains a public health burden despite ongoing global control efforts. Conventional microscopy is a practical tool for diagnosis and screening ofSchistosoma haematobium, but identification of eggs requires a skilled microscopist. Here we present a machine learning (ML)-based strategy for automated detection ofS. haematobiumthat combines two imaging contrasts, brightfield (BF) and darkfield (DF), to improve diagnostic performance. We collected BF and DF images of urine samples, many of them containingS. haematobiumeggs, during two different field studies in Côte d’Ivoire using a mobile phone-based microscope, the SchistoScope. We then trained separate egg-detection ML models and compared the patient-level performance of BF and DF models alone to combinations of BF and DF models, using annotations from trained microscopists as the gold standard. We found that models trained on DF images, and almost all BF and DF combinations, performed significantly better than models trained on BF images only. When models were trained on images from the first field study (n = 349 patients, 748 images of each contrast), patient-level classification performance on patient images from the second study (n = 375 patients, 752 images of each contrast) met the WHO Diagnostic Target Product Profile (TPP) sensitivity and specificity for the monitoring and evaluation use case (sensitivity for all models and combinations was >75% when evaluated at a confidence score threshold that resulted in specificity >96.5%). When we used images from both field studies for the training set, performance of the models was improved. Overall, this work shows that the use of DF and BF increases the performance of ML models on images from devices with low-cost optics, while retaining the portability, power, and time-to-results of the WHO’s diagnostic TPP. DF requires no additional sample preparation and does not increase the complexity of the imaging system. It thus offers a practical means to improve performance of automated diagnostics forS. haematobiumas well as other microscopy-based diagnostics.

Infectious Diseases↗
DOE OSTI2020

Improving five-year survival prediction via multitask learning across HPV-related cancers

Oncology is a highly siloed field of research in which sub-disciplinary specialization has limited the amount of information shared between researchers of distinct cancer types. This can be attributed to legitimate differences in the physiology and carcinogenesis of cancers affecting distinct anatomical sites. However, underlying processes that are shared across seemingly disparate cancers probably affect prognosis. The objective of the current study is to investigate whether multitask learning improves 5-year survival cancer patient survival prediction by leveraging information across anatomically distinct HPV related cancers. Furthermore, data were obtained from the Surveillance, Epidemiology, and End Results (SEER) program database. The study cohort consisted of 29,768 primary cancer cases diagnosed in the United States between 2004 and 2015. Ten different cancer diagnoses were selected, all with a known association with HPV risk. In the analysis, the cancer diagnoses were categorized into three distinct topography groups of varying specificity. The most specific topography grouping consisted of 10 original cancer diagnoses differentiated by the first two digits of the ICD-O-3 topography code. The second topography grouping consisted of cancer diagnoses categorized into six distinct organ groups. Finally, the third topography grouping consisted of just two groups, head-neck cancers and ano-genital cancers. The tasks were to predict 5-year survival for patients within the different topography groups using 14 predictive features which were selected among descriptive variables available in the SEER database. The information from the predictive features was shared between tasks in three different ways, resulting in three distinct predictive models: 1) Information was not shared between patients assigned to different tasks (single task learning); 2) Information was shared between all patients, regardless of task (pooled model); 3) Only relevant information was shared between patients grouped to different tasks (multitask learning). Prediction performance was evaluated with Brier scores. All three models were evaluated against one another on each of the three distinct topography-defined tasks. The results showed that multitask classifiers achieved relative improvement for the majority of the scenarios studied compared to single task learning and pooled baseline methods. In this study, we have demonstrated that sharing information among anatomically distinct cancer types can lead to improved predictive survival models.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Feasibility and preliminary efficacy for morning bright light therapy to improve sleep and plasma biomarkers in US Veterans with TBI. A prospective, open-label, single-arm trial

Mild traumatic brain injury (TBI) is associated with persistent sleep-wake dysfunction, including insomnia and circadian rhythm disruption, which can exacerbate functional outcomes including mood, pain, and quality of life. Present therapies to treat sleep-wake disturbances in those with TBI (e.g., cognitive behavioral therapy for insomnia) are limited by marginal efficacy, poor patient acceptability, and/or high patient/provider burden. Thus, this study aimed to assess the feasibility and preliminary efficacy of morning bright light therapy, to improve sleep in Veterans with TBI (NCT03578003). Thirty-three Veterans with history of TBI were prospectively enrolled in a single-arm, open-label intervention using a lightbox (~10,000 lux at the eye) for 60-minutes every morning for 4-weeks. Pre- and post-intervention outcomes included questionnaires related to sleep, mood, TBI, post-traumatic stress disorder (PTSD), and pain; wrist actigraphy as a proxy for objective sleep; and blood-based biomarkers related to TBI/sleep. The protocol was rated favorably by ~75% of participants, with adherence to the lightbox and actigraphy being ~87% and 97%, respectively. Post-intervention improvements were observed in self-reported symptoms related to insomnia, mood, and pain; actigraphy-derived measures of sleep; and blood-based biomarkers related to peripheral inflammatory balance. The severity of comorbid PTSD was a significant positive predictor of response to treatment. Morning bright light therapy is a feasible and acceptable intervention that shows preliminary efficacy to treat disrupted sleep in Veterans with TBI. A full-scale randomized, placebo-controlled study with longitudinal follow-up is warranted to assess the efficacy of morning bright light therapy to improve sleep, biomarkers, and other TBI related symptoms.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2026

Dataset 3: A National Dataset on Actionable Items in Improving Pooled Rideshare, 2025.

Dataset 3: A National Dataset on Actionable Items in Improving Pooled Rideshare.” 2025. Dataset Description: Pooled Rideshare Acceptance Survey - Phase 3 (2025, N = 8,296). This dataset represents the third and final phase of a national survey aimed at understanding user acceptance and preferences related to pooled rideshare (PR) services in the United States. Building on insights from earlier phases, this phase expands both the sample size and the depth of analysis to support policymaking, transportation planning, and service design for sustainable mobility systems. The Phase 3 survey was administered online to a nationally representative sample of 8,296 U.S. adults. The sample includes a wide range of demographics. The survey retained core questions from previous phases while introducing 77 detailed service features (actionable items) to evaluate potential improvements to PR offerings. Each feature was designed to assess whether a specific improvement, such as enhanced safety measures, real-time ride tracking, or user training would increase participants’ willingness to adopt PR services. In addition, behavioral predictors, current rideshare habits, environmental attitudes, and perceived barriers (e.g., safety, privacy, and comfort) were captured. - Phase_3_Final - The dataset contains rows corresponding to individual respondents and columns representing survey items, demographic characteristics, and response values. The data is available in both .CSV and .SAV formats. - Phase_3_Final_MapFile - Accompanying this dataset is a data dictionary explaining each variable, value range, and coding schema. An .XLSX format of the full survey instrument is included to support interpretation and reuse of the dataset.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2022

A Spatiotemporal-Aware Weighting Scheme for Improving Climate Model Ensemble Predictions

Multimodel ensembling has been widely used to improve climate model predictions, and the improvement strongly depends on the ensembling scheme. In this work, we propose a Bayesian neural network (BNN) ensembling method, which combines climate models within a Bayesian model averaging framework, to improve the predictive capability of model ensembles. Our proposed BNN approach calculates spatiotemporally varying model weights and biases by leveraging individual models' simulation skill, calibrates the ensemble prediction against observations by considering observation data uncertainty, and quantifies epistemic uncertainty when extrapolating to new conditions. More importantly, the BNN method provides interpretability about which climate model contributes more to the ensemble prediction at which locations and times. Thus, beyond its predictive capability, the method also brings insights and understanding of the models to guide further model and data development. In this study, we design experiments using an ensemble of CMIP6 climate model simulations to illustrate the BNN ensembling method's capability with respect to prediction accuracy, interpretability, and uncertainty quantification (UQ). We demonstrate that BNN can correctly assign larger weights to the regions and seasons where the individual model fits the observation better. Moreover, its offered interpretability is consistent with our understanding of localized climate model performance. Additionally, BNN shows an increasing uncertainty when the prediction is farther away from the period with constrained data, which appropriately reflects our trustworthiness of the models in the changing climate.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

AmeriFlux FLUXNET-1F US-ABc Archbold grazed improved pasture

This is the AmeriFlux Management Project (AMP) created FLUXNET-1F version of the carbon flux data for the site US-ABc Archbold grazed improved pasture. This is the FLUXNET version of the carbon flux data for the site US-ABc Archbold grazed improved pasture produced by applying the standard ONEFlux (1F) software. Site Description - This site is an improved pasture, regularly fertilized (nitrogen, ~26 kg/ha every 2 years) and limed (~229 kg/ha every 5 years, and historically received phosphorus fertilizer until 1986; . The soil type is a fine sandy Spodosol

Bernacchi, Carl [Global Change and Photosynthesis ↗
DOE OSTI2026

AmeriFlux FLUXNET-1F US-ABd Archbold ungrazed improved pasture

This is the AmeriFlux Management Project (AMP) created FLUXNET-1F version of the carbon flux data for the site US-ABd Archbold ungrazed improved pasture. This is the FLUXNET version of the carbon flux data for the site US-ABd Archbold ungrazed improved pasture produced by applying the standard ONEFlux (1F) software. Site Description - This site is an improved pasture, regularly fertilized (nitrogen, ~26 kg/ha every 2 years) and limed (~229 kg/ha every 5 years; and historically received phosphorus fertilizer until 1986; . The soil type is a fine sandy Spodosol . Grazing was removed in 2013-2015 for the duration of the study

Bernacchi, Carl [Global Change and Photosynthesis ↗
DOE OSTI2026

Improved Regional Moment Tensor Inversion for Moderately Large Earthquakes in the Western United States Using a 3D Earth Model Based on Full Waveform Tomography

The nature of seismic sources for moderately large (moment magnitude, M w 5.0–6.5) events are commonly characterized by their moment tensor (MT) solutions and obtained by inversion of regional distance (200–1600 km) long‐period (20–50 s) waveforms. Regional MT estimates are often calculated from average plane‐layered, one‐dimensional (1D) velocity models. However, 1D model calculations can produce misfits in the arrival times and waveform shapes that introduce errors, particularly at longer distances or for shorter periods, which are necessary for analyzing lower magnitude events. Approximate Earth models (e.g., 1D) representing broad areas may be inadequate, particularly in the crust and uppermost mantle of tectonically complex regions. In this study, we show how a three‐dimensional (3D) Earth model obtained from full waveform inversion tomography can improve waveform fits and decrease phase errors. We developed a platform and workflow to perform routine 3D MT inversions and inverted MTs for 25 earthquakes in the western United States and seven nuclear explosions using an average 1D and a recent 3D Earth model, WUS256 (Rodgers et al., 2022). Using the 3D model improves waveform fits (variance reduction and phase time shifts) compared with the 1D model, and the 3D MT solutions are stable across large distances. This study shows that 3D models obtained from full waveform tomography can improve MTs and source characterization especially at far regional distances (>800 km).

Geosciences↗
DOE OSTI2025

Towards Improving luminosity using optics tuning and data-driven methods

The results of Run 24 experiments at Relativistic Heavy Ion Collider (RHIC) for improving luminosity using optics tuning are presented in this study. In the first experiment, MADx matching was used to output magnet strengths corresponding to specific s star movements around Interaction Region 8 (IR8). The corresponding Zero Degree Calorimeter (ZDC) signal was measured in place of luminosity, and Bayesian Optimization aids search of optimal movements. It was found that values retrieved from matching were inaccurate, resulting in negative feedback loops. The second experiment focused on calculating accurate s star movements. The matching method was replaced with a linear sensitivity matrix, directly relating optics to power supply, and its null space was used to fit constraints such as hysteresis effects. At the experiment, beam losses were observed at collimators around boundary of IR8, which were fixed for the third experiment. Dynamic mode decomposition was also introduced to improve quality of turn-by-turn (TBT) data as well as accuracy and consistency of optics measurements at IR8. These improvements will be tested in the experiment of next RHIC run for luminosity optimization.

Accelerator Physics↗
DOE OSTI2022

Improvements to the Recycler/Main Injector to Deliver 850 kW+

The Main Injector is used to deliver a 120 GeV high power proton beam for Neutrino experiments. The design power of 700 kW was reached in early 2017 but further improvements have seen a new sustained peak power of 893 kW. Two of the main improvements include the shortening of the Main Injector ramp length as well optimizing the slip-stacking procedure performed in the Recycler to reduce the amount of uncaptured beam making its way into the Main Injector. These improvements will be discussed in this paper as well future upgrades to reach higher beam powers.

43 PARTICLE ACCELERATORS↗
DOE OSTI2026

CHARACTERIZING AND CONTROLLING RECOVERY AND RECRYSTALLIZATION IN NIOBIUM FOR IMPROVED SRF CAVITY PERFORMANCE

Crystal defects, such as dislocations and low-angle boundaries, provide sources of magnetic flux trapping in the Nb materials used for superconducting radio frequency (SRF) resonating cavities. Improving the performance of SRF cavities, as measured through the quality factor, requires reducing these defects. SRF cavity production involves deformation processing, such as rolling and forming, and strategic annealing heat treatments. The resulting microstructures can be recovered, recrystallized, or both. Because recovery leaves many defects that can trap flux, recrystallization should improve cavity performance. Thus, processing schedules that produce complete recrystallization without excessive grain growth need to be designed. Solutions to this problem require understanding physical metallurgy and differentiating between recovered and recrystallized regions of microstructure. Backscattered electron microscopy techniques are applied to this end. We demonstrate that the conditions required to produce fully recrystallized microstructures depend on Nb impurity content, suggesting that processing schedules may need to be adjusted by material heat or lot. We also demonstrate that processing can be used to control growth of recrystallized grains to maintain mechanical strength in fully recrystallized materials. Forming cavities from cold-rolled Nb sheet material may provide strategic new routes to obtain microstructures that improve SRF cavity performance.

Taleff, E. [The University of Texas at Austin]↗
DOE OSTI2020

Improving radiograph analysis throughput through transfer learning and object detection

SIGN Fracture Care International partners with surgeons in low-resource hospitals worldwide to provide access to effective orthopedic care by donating educational materials and innovatively designed surgical implants. Over two decades, SIGN’s Online Surgical Database (SOSD) has grown to contain over 500,000 medical images, with radiographs holding the majority share. One challenge in working with hospitals worldwide is that both the radiographs uploaded to the SOSD and the data entry accompanying the uploads vary in quality. To improve the accuracy of data in the SOSD, we trained a model to detect surgical implants in radiographs. We first developed a tool to automatically detect radiographs, then trained an object detection model to determine the number and placement of surgical implants visible in the radiograph. Active learning was used to generate a training set containing 2,510 radiographs with screws, nails, and plates labeled by bounding boxes. Training a model to simultaneously recognize all three classes of implants gave a low average precision (AP) for the plate class, likely due to the low number of plate instances in our training set and the large variety of surgical plates used by SIGN-partnered surgeons. Applying standard image augmentation techniques to increase the plate count in our training set did not appreciably increase the AP of plate detection. To improve plate detection, we redrew the bounding boxes to account for correlations between the screw and plate classes. Training one model to detect nails and screws and a separate model to detect plates increased the AP of plate detection by 78.8 percentage points. The AP of each class was 80.7% for screws, 93.6% for nails, and 92.6% for plates; meanwhile, the sensitivity was 92% for screws, 86% for nails, and 81% for plates. We show that object detection methods can be used to detect surgical implants in radiographs of varying quality; however, the detection ability is dependent on the type of implant, and some implants, in our case plates, must be treated differently than others. Such tools can improve the throughput of radiograph analysis, assisting physicians and surgeons with the treatment of bone fractures.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Improved Radiative Recombination in AlInGaP LEDs

Direct-emitting red and amber LEDs based on the AlInGaP material system offer a higher efficacy potential than phosphor-converted (pc) red and amber LEDs due to the absence of Stokes losses. Direct red LEDs (~610-620nm) are increasingly used in color-mixed and color-tunable illumination applications, where they have a significant advantage over pc-red LEDs, with high power conversion efficiency (PCE) and narrower emission enabling higher luminous efficacy of radiation (LER). Still, adoption is limited as the complexities associated with mixing different emitters offset the value of the efficacy gain. Adoption of direct amber LEDs (~580-595nm) in illumination applications is virtually non-existent due to their lower PCE and the availability of very efficient pc-amber emitters, which suffer less from Stokes loss and LER penalties associated with the broader bandwidth than pc-red. Further improvement in PCE of direct red and amber LEDs is needed to achieve wider adoption and ultimately enable more energy savings than possible with pc-LEDs. In this project, Lumileds has developed improved high-power amber and red LEDs incorporating tensile-strain barriers into the active region to improve carrier confinement and thereby increase external quantum efficiency (EQE). This approach is most beneficial for the short (amber) end of the AlInGaP wavelength range where carrier confinement is lowest. An EQE of 22% was achieved for amber LEDs, exceeding the project target of 20%

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2020

New Approaches to Improved PEM Fuel Cell Catalyst Layers

Polymer-electrolyte membrane (PEM) fuel-cells are one of the most promising energy conversion technologies for renewable clean energy applications. A major challenge preventing their widespread commercialization is achieving high performance with lowloadings of platinum group metal (PGM) catalysts. One of the factors driving performance limitations in the cell is the mass transport losses within the cathode catalyst layers due to sluggish oxygen-reduction reactions occurring at the platinum-ionomer interface, which is believed to be linked to reduced transport of ions and oxygen. A viable solution to reduce the transport resistances in the catalyst layers is to create new ionomers that can provide good ion and oxygen transport needed to accomplish high-performing fuel cell catalysts. Characterization of transport properties of ionomers for various molecular architectures is the key step, in the effort to create and identify the optimized polymer structure with improved transport. Using this approach, Tetramer Technologies and LBNL propose improved fuel-cell catalyst ionomers based on Tetramers proprietary polymer chemistry, as highlighted under subtopic 17a Innovative Approaches Toward Discovery and Development of Improved Ionomers for Polymer Electrolyte Membrane Fuel Cell Catalyst Layer.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2020

Workshop on Improving Holdup Monitoring in the US

The characterization and quantification of nuclear deposits, or residual nuclear material retained in process equipment, generally referred to as holdup, continues to challenge nuclear processing facilities in DOE program areas such as nuclear criticality safety, material control and accountability, environmental management. The efficiency and effectiveness of nuclear operations in these facilities depends heavily on the results from nondestructive techniques designed to measure nuclear materials in situ. The traditional methods used to measure holdup were established more than 30 years ago and assume unrealistic conditions for most measurement cases. These blind, in situ measurements present unique challenges due to the following attributes: unique, non-ideal geometries; unknown deposit thicknesses; lack of representative calibration standards; large number of holdup deposit locations; and poor accessibility for many measurement locations. These difficulties and the poor assumptions made to handle them have resulted in measurement uncertainties that are often too large for many programs to effectively utilize. In an effort to address and identify solutions to the current holdup challenges facing the US nuclear industry, Oak Ridge National Laboratory hosted the Technical Workshop on Improving Holdup Monitoring in the US. The two-day workshop held August 21–22, 2019, was attended by 35 holdup measurement practitioners, program managers, stakeholders from US Department of Energy sites, the Nuclear Regulatory Commission, and representatives from commercial industry. The primary objectives were to identify critical elements of a successful and defensible holdup measurement program; identify technical challenges, needs, and potential improvements associated with measurement of holdup; and share best practices and lessons learned from recent accumulation events. Fourteen presentations were delivered on holdup topics including necessary elements for effective program management, measurement needs, emerging technologies and other measurement solutions, and lessons learned from recent events. These presentations highlighted best practices related to accumulation monitoring, considerations and proactive steps to facilitate holdup in new facility designs, imaging solutions to improve holdup assumptions, and new software developed in support of holdup measurements. Breakout sessions were also employed to determine additional needs and identify potential solutions to common holdup challenges. The primary discussions focused on causes and prevention of material accumulations, needs for uranium and plutonium measurements, and reduction of measurement uncertainty, all of which were explicitly identified as critical needs during the 2018 Workshop on Technical and Programmatic Needs for a Sustainable NDA Program for the US DOE.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2020

Improving High-Energy Particle Detectors with Machine Learning

Microseconds after the Big Bang, the universe existed in a state called the quark-gluon plasma (QGP). To experimentally study its properties, the QGP is recreated in high-energy nuclear collisions at the LHC, and the particles produced from the QGP are reconstructed from their energy deposition in the ATLAS calorimeter. This requires both classifying the particles and calibrating their deposited energy. The objective of this project is to improve the reconstruction by using machine learning techniques, where the energy depositions of clusters of cells, formed by ATLAS topo-clustering methods, are treated as three-dimensional images when inputted to neural networks. This approach significantly improves the calibration of deposited energies when cross-validating while training, and models trained on idealized data predict the calibrated energies of particles in more complex data sets well. Additionally, implementation of a data generator using uproot allows the program to load input data into memory as needed while training or predicting, significantly reducing the amount of memory used. The data generator also allows for use of multiprocessing to speed up training and evaluating. This work illustrates that using machine learning methods for both classification and calibration has the potential to significantly improve particle reconstruction.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗