Search NASA⌕ Search

SEARCH · Search NASA

Results for “Distributed training”

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 1,585 records · Page 88

Adverse Events Involving Radiation Oncology Medical Devices: Comprehensive Analysis of US Food and Drug Administration Data, 1991 to 2015

Purpose: Radiation oncology relies on rapidly evolving technology and highly complex processes. The US Food and Drug Administration collects reports of adverse events related to medical devices. We sought to characterize all events involving radiation oncology devices (RODs) from the US Food and Drug Administration's postmarket surveillance Manufacturer and User Facility Device Experience (MAUDE) database, comparing these with non–radiation oncology devices. Methods and Materials: MAUDE data on RODs from 1991 to 2015 were sorted into 4 product categories (external beam, brachytherapy, planning systems, and simulation systems) and 5 device problem categories (software, mechanical, electrical, user error, and dose delivery impact). Outcomes included whether the device was evaluated by the manufacturer, adverse event type, remedial action, problem code, device age, and time since 510(k) approval. Descriptive statistics were performed with linear regression of time-series data. Results for RODs were compared with those for other devices by the Pearson χ{sup 2} test for categorical data and 2-sample Kolmogorov-Smirnov test for distributions. Results: There were 4234 ROD and 4,985,698 other device adverse event reports. Adverse event reports increased over time, and events involving RODs peaked in 2011. Most ROD reports involved external beam therapy (50.8%), followed by brachytherapy (24.9%) and treatment planning systems (21.6%). The top problem types were software (30.4%), mechanical (20.9%), and user error (20.4%). RODs differed significantly from other devices in each outcome (P<.001). RODs were more likely to be evaluated by the manufacturer after an event (46.9% vs 33.0%) but less likely to be recalled (10.5% vs 37.9%) (P<.001). Device age and time since 510(k) approval were shorter among RODs (P<.001). Conclusions: Compared with other devices, RODs may experience adverse events sooner after manufacture and market approval. Close postmarket surveillance, improved software design, and manufacturer–user training may help mitigate these events.

07 ISOTOPE AND RADIATION SOURCES↗

Recent increases in annual, seasonal, and extreme methane fluxes driven by changes in climate and vegetation in boreal and temperate wetland ecosystems

Climate warming is expected to increase global methane (CH 4 ) emissions from wetland ecosystems. Although in situ eddy covariance (EC) measurements at ecosystem scales can potentially detect CH 4 flux changes, most EC systems have only a few years of data collected, so temporal trends in CH 4 remain uncertain. Here, we use established drivers to hindcast changes in CH 4 fluxes (FCH 4 ) since the early 1980s. We trained a machine learning (ML) model on CH 4 flux measurements from 22 [methane-producing sites] in wetland, upland, and lake sites of the FLUXNET-CH 4 database with at least two full years of measurements across temperate and boreal biomes. The gradient boosting decision tree ML model then hindcasted daily FCH 4 over 1981-2018 using meteorological reanalysis data. We found that, mainly driven by rising temperature, half of the sites (n = 11) showed significant increases in annual, seasonal, and extreme FCH 4 , with increases in FCH 4 of ca. 10% or higher found in the fall from 1981–1989 to 2010–2018. The annual trends were driven by increases during summer and fall, particularly at high-CH 4 -emitting fen sites dominated by aerenchymatous plants. We also found that the distribution of days of extremely high FCH 4 (defined according to the 95th percentile of the daily FCH 4 values over a reference period) have become more frequent during the last four decades and currently account for 10–40% of the total seasonal fluxes. The share of extreme FCH 4 days in the total seasonal fluxes was greatest in winter for boreal/taiga sites and in spring for temperate sites, which highlights the increasing importance of the non-growing seasons in annual budgets. Our results shed light on the effects of climate warming on wetlands, which appears to be extending the CH 4 emission seasons and boosting extreme emissions.

54 ENVIRONMENTAL SCIENCES↗

Explosion Detection Using Smartphones: Ensemble Learning with the Smartphone High-Explosive Audio Recordings Dataset and the ESC-50 Dataset

Explosion monitoring is performed by infrasound and seismoacoustic sensor networks that are distributed globally, regionally, and locally. However, these networks are unevenly and sparsely distributed, especially at the local scale, as maintaining and deploying networks is costly. With increasing interest in smaller-yield explosions, the need for more dense networks has increased. To address this issue, we propose using smartphone sensors for explosion detection as they are cost-effective and easy to deploy. Although there are studies using smartphone sensors for explosion detection, the field is still in its infancy and new technologies need to be developed. We applied a machine learning model for explosion detection using smartphone microphones. The data used were from the Smartphone High-explosive Audio Recordings Dataset (SHAReD), a collection of 326 waveforms from 70 high-explosive (HE) events recorded on smartphones, and the ESC-50 dataset, a benchmarking dataset commonly used for environmental sound classification. Two machine learning models were trained and combined into an ensemble model for explosion detection. The resulting ensemble model classified audio signals as either “explosion”, “ambient”, or “other” with true positive rates (recall) greater than 96% for all three categories.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

What you get is not always what you see—pitfalls in solar array assessment using overhead imagery

Effective integration planning for small, distributed solar photovoltaic (PV) arrays into electric power grids requires access to high quality data: the location and power capacity of individual solar PV arrays. Unfortunately, national databases of small-scale solar PV do not exist; those that do are limited in their spatial resolution, typically aggregated up to state or national levels. While several promising approaches for solar PV detection have been published, strategies for evaluating the performance of these models are often highly heterogeneous from study to study. The resulting comparison of these methods for practical applications for energy assessments becomes challenging and may imply that the reported performance evaluations overly optimistic. The heterogeneity comes in many forms, each of which we explore in this work: the degree of diversity of the locations and sensors (e.g. different satellites, aerial photography) from which the training and validation data originate, the validation of ground truth (manual annotation of imagery vs known solar PV locations), the level of spatial aggregation (e.g. array-level vs regional estimates), and inconsistencies in the training and validation datasets (e.g. different datasets are used for each study and those data are not always made accessible). For each, we discuss emerging practices from the literature to address them or suggest directions of future research. As part of our investigation, we evaluate solar PV identification performance in two large regions: the entire state of Connecticut and the city of San Diego, CA. In Connecticut, we also use 33,114 known parcel-level solar PV installations from Berkeley Lab’s Tracking the Sun dataset to evaluate parcel-level performance and evaluate capacity estimates using 169 municipalities. We also make our code (which we call SolarMapper), pre-trained models, training data, and predictions publicly available and provide a web portal for interactively inspecting each prediction that was made. Here our findings suggest that traditional performance evaluation of the automated identification of solar PV from satellite imagery may be optimistic due to common limitations in the validation process. The takeaways from this work are intended to inform and catalyze the large-scale practical application of automated solar PV assessment techniques by energy researchers and professionals.

14 SOLAR ENERGY↗

Online MCMC Thinning with Kernelized Stein Discrepancy

A fundamental challenge in Bayesian inference is efficient representation of a target distribution. Many nonparametric approaches do so by sampling a large number of points using variants of Markov chain Monte Carlo (MCMC). Here, we propose an MCMC variant that retains only those posterior samples which exceed a kernelized Stein discrepancy (KSD) threshold, which we call KSD thinning. We establish the convergence and complexity trade-offs for several settings of KSD thinning as a function of the KSD threshold parameter, sample size, and other problem parameters. We provide experimental comparisons against other online nonparametric Bayesian methods that generate low-complexity posterior representations. We observe superior consistency/complexity trade-offs across a range of settings including MCMC sampling on two Bayesian inference problems from the biological sciences, and 10 × inference speedup and storage reduction for Bayesian neural networks with no loss of accuracy and no increase in training time. Our code is available at https://github.com/colehawkins/KSD-Thinning.

Bayesian inference↗

Physics-informed heterogeneous graph neural networks for DC blocker placement

The threat of geomagnetic disturbances (GMDs) to the reliable operation of the bulk energy system has spurred the development of effective strategies for mitigating their impacts. One such approach involves placing transformer neutral blocking devices, which interrupt the path of geomagnetically induced currents (GICs) to limit their impact. The high cost of these devices and the sparsity of transformers that experience high GICs during GMD events, however, calls for a sparse placement strategy that involves high computational cost. To address this challenge, we developed a physics-informed heterogeneous graph neural network (PIHGNN) for solving the graph-based dc-blocker placement problem. Our approach combines a heterogeneous graph neural network (HGNN) with a physics-informed neural network (PINN) to capture the diverse types of nodes and edges in ac/dc networks and incorporates the physical laws of the power grid. We train the PIHGNN model using a surrogate power flow model and validate it using case studies. Results demonstrate that PIHGNN can effectively and efficiently support the deployment of GIC dc-current blockers, ensuring the continued supply of electricity to meet societal demands. Furthermore, our approach has the potential to contribute to the development of more reliable and resilient power grids capable of withstanding the growing threat that GMDs pose.

24 POWER TRANSMISSION AND DISTRIBUTION↗

From Vehicles to Systems: Understanding Freight Transportation as a Connected Energy, Infrastructure, and Operations System

The U.S. freight system may need to handle 50% more cargo by 2050. Upgrading our freight system requires modernizing capital-intensive, long-lived assets including freight trains, ports, and terminal infrastructure. NLR is advancing freight system solutions spanning ALTRIOS, the first digital twin for the full freight rail system; ALTRIOS-LIFTS, which can create digital twins of freight terminals; INFORMES, the first national model of the intermodal freight system; MARINESim, used to simulate and optimize ocean-going vessel operations; and more. These modeling and simulation tools enable data-driven decision-making across freight modes and systems.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Regional Medium-Term Hourly Electricity Demand Forecasting Based on LSTM

This paper aims to forecast high-resolution (hourly) aggregated load for a certain region in the medium term (a few days to over a year). One region is defined as some places with similar climate characteristics because the climate influences people's daily lifestyles and hence the electric usage. We decompose the electric usage records into two parts: base load and seasonal load. Considering both temperature and time factors, different deep-learning methods are adopted to characterize them. The first goal of our approach is to predict the peak load which is critical for power system planning. Furthermore, our proposed forecast method can provide the depiction of the hourly load profile to provide customized load curves for high-level real-time applications. The proposed method is tested on real-world historical data collected by CAISO, BPA, and PACW. The experimental results show that trained by three years of data, our method could reduce the prediction error for a one-year lead hourly load below $5\%$ MAPE, and predict the occurrence of the peak load for next year in CAISO with an error within three days. Furthermore, as a byproduct, an interesting observation on the impact of COVID-19 on human life was made and discussed based on these case studies.

deep learning↗

Are your employees sick of hearing about safety? Ways to improve how safety is communicated at your company

Companies that care about their employees care about their employees’ safety and will go to great lengths to communicate the importance of working safely. Monthly safety meetings, creative safety contests, safety websites, sharing lessons learned—safety communicators tend to use a variety of methods to distribute procedures and critical safety information to help employees plan and perform work. However, the safety message falls on deaf ears in some cases, especially when employees feel they’re being overloaded with safety information to the point where they are sick of hearing about it. This poses a conundrum for safety communicators: Should they keep pouring on the safety, or should they lighten up? How much is the right amount?

99 GENERAL AND MISCELLANEOUS↗

Universal Monte Carlo Event Generator

With the Jefferson Lab 12 GeV physics program underway and plans for the future Electron-Ion Collider (EIC), the nuclear physics community is entering a new era of exploration of QCD phenomena involving extensive data taking and event-level processing. This brings with it the potential for unprecedented access to multidimensional particle momentum distributions (PMDs) that can be connected to various theoretical frameworks by unfolding the emergent quantum mechanical properties of QCD using the PMDs. In practice, the PMDs are rendered as discretized histograms (typically one- or two-dimensional projections), and detector effects must be taken into account to unfold the pure detector effect-free PMDs that can be connected with theory. One of the challenges in this new era is obtaining faithful reconstructions of the multidimensional PMDs that preserve all of the inherent particle correlations. In this LDRD project we developed a novel approach using machine learning (ML) that solves this challenge, by avoiding entirely the need to use histograms as the main numerical technique to obtain the detector effect-free PMDs. The new approach is, moreover, scalable to higher dimensional PMDs. The central idea involves training neural networks (NNs) to generate synthetic event-level data (momenta 4-vectors of final state particles) to preserve all correlations among the particles. This is achieved by converting the trial synthetic vertex-level events to detector-level events using detector simulators. The NNs are then tuned using a specialized distance metric between the synthetic detector events and the real detector events.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Celebrating Notable Advances in Compound Semiconductors: A Tribute to Dr. Wladyslaw Walukiewicz

The Special Topic "Native Defects, Impurities and the Electronic Structure of Compound Semiconductors" is dedicated to Dr. Wladyslaw (Wladek) Walukiewicz, our esteemed colleague who passed away on November 9, 2022, at age 76. It is not often that a scientist is able to profoundly impact more than one to two scientific topics in their lifetime. Dr. Walukiewicz, a condensed matter theorist by training, arguably did that for at least four. While his scientific accomplishments tell the story of a successful career and valuable contributions, it does not convey the entire story. For the rest, one has only to look at the community of researchers he brought together across the globe. Dr. Walukiewicz conducted his work with curiosity, integrity, and respect for others that elevated those around him. Here, we pay tribute to Dr. Walukiewicz's ideas and the legacy he leaves behind.

defects↗

Development of message passing-based graph convolutional networks for classifying cancer pathology reports

Abstract Background Applying graph convolutional networks (GCN) to the classification of free-form natural language texts leveraged by graph-of-words features (TextGCN) was studied and confirmed to be an effective means of describing complex natural language texts. However, the text classification models based on the TextGCN possess weaknesses in terms of memory consumption and model dissemination and distribution. In this paper, we present a fast message passing network (FastMPN), implementing a GCN with message passing architecture that provides versatility and flexibility by allowing trainable node embedding and edge weights, helping the GCN model find the better solution. We applied the FastMPN model to the task of clinical information extraction from cancer pathology reports, extracting the following six properties: main site, subsite, laterality, histology, behavior, and grade. Results We evaluated the clinical task performance of the FastMPN models in terms of micro- and macro-averaged F1 scores. A comparison was performed with the multi-task convolutional neural network (MT-CNN) model. Results show that the FastMPN model is equivalent to or better than the MT-CNN. Conclusions Our implementation revealed that our FastMPN model, which is based on the PyTorch platform, can train a large corpus (667,290 training samples) with 202,373 unique words in less than 3 minutes per epoch using one NVIDIA V100 hardware accelerator. Our experiments demonstrated that using this implementation, the clinical task performance scores of information extraction related to tumors from cancer pathology reports were highly competitive.

59 BASIC BIOLOGICAL SCIENCES↗

Mississippi State University Cooling, Heating, and Power (Micro-CHP) and Bio-Fuel Center

Between 2008 and 2014, the U.S. Department of Energy funded the MSU Micro-CHP and Bio-Fuel Center located at Mississippi State University. The overall objective of this project was to enable micro-CHP (micro-combined heat and power) utilization, to facilitate and promote the use of CHP systems and to educate architects, engineers, and agricultural producers and scientists on the benefits of CHP systems. Therefore, the work of the Center focused on the three areas: CHP system modeling and optimization, outreach, and research. In general, the results obtained from this project demonstrated that CHP systems are attractive because they can provide energy, environmental, and economic benefits. Some of these benefits include the potential to reduce operational cost, carbon dioxide emissions, primary energy consumption, and power reliability during electric grid disruptions. The knowledge disseminated in numerous journal and conference papers from the outcomes of this project is beneficial to engineers, architects, agricultural producers, scientists and the public in general who are interested in CHP technology and applications. In addition, more than 48 graduate students and 23 undergraduate students, benefited from the training and research performed in the MSU Micro-CHP and Bio-Fuel Center. The project scope consisted of six separate, but related areas: 1. Micro-CHP modeling - Computational simulations and verification/validation studies of selected models for technical and economic viability. 2. Fuel flexibility in micro-CHP applications – Demonstrating and quantifying the operation of micro-CHP systems on alternative fuels. 3. Outreach activities – Providing provide educational materials for engineers and architects and conducting outreach seminars, workshops, and demonstrations for engineers and architects as well as for more general audiences. 4. Biofuel and opportunity fuel processing and utilization – Determining the processing requirements and feasibility to utilize biofuels and disaster debris for micro-CHP systems. 5. CHP Systems Optimization – Optimization of internal combustion and Stirling engines integrated with heat recovery systems for CHP operation. 6. Distributed Generation and Grid Interconnection – Developing requirements and control systems for interconnection of micro-CHP systems with the electrical grid.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Automating the Analysis of Large Language Models Responses through Zero-Shot Question Answering

Recent advancements in Large Language Models (LLMs) have shown significant potential in various applications, yet their evaluation, particularly in zero-shot question answering scenarios, remains a challenging task. In this study, our objective was to explore precision metrics for Large Language Models (LLM) and design and implement a software pipeline to automatically evaluate LLMs' outputs under zero-shot question answering. Zero-shot question answering involves a model providing answers to questions about topics it hasn't seen during training. It leverages the principles of zero-shot learning by relying on semantic understanding and generalization from related knowledge. The data used was metadata from medical databases on congenital heart disease. We explored eleven LLM metrics and selected three for our evaluation: BLEU, BERTScore, and MoverScore. BLEU calculates a score based on the overlap of n-grams (contiguous sequences of n items, typically words) between the machine-generated translation and the reference translations. Higher BLEU scores indicate better correspondence between the machine-generated and human-generated translations. BERTScore is a metric used to evaluate the quality of machine-generated text by measuring the similarity of token embeddings produced by BERT (Bidirectional Encoder Representations from Transformers) between the generated text and reference text. MoverScore is a metric that quantifies the dissimilarity between the distributions of word embeddings from machine-generated text and reference text, emphasizing semantic similarity over exact token overlap. We also introduced HBKI, a composite metric summarizing these approaches. We tested five models —GPT-3, Llama-2, Gemini 1.5 Pro, Solar 10.7B, and Mixtral-8x7b. Our software pipeline, designed and implemented using Object-Oriented Programming principles, allows users to customize the selection and extraction of features for topics of interest in their own research. Our results show that MoverScore delivered the most precise evaluation of the LLM's outputs, while Mixtral-8x7b achieved the best overall performance in extracting metadata from the databases.

97 MATHEMATICS AND COMPUTING↗

SuperCDMS Event Reconstruction Using Convolutional Neural Networks [Thesis]

The SuperCDMS experiment uses cryogenic silicon and germanium detectors to search for dark matter candidates such as WIMPs (Weakly Interacting Massive Particles) streaming through the Earth. Collisions in the silicon and germanium crystals are expected to produce phonons whose thermal signatures can be measured. This thesis first describes the integration of a new Signal Distribution Unit (SDU) to the SuperCDMS data acquisition system, which allows for synchronization of multiple detectors and electronic/mechanical noise characterization via accelerometer, antenna, and AC phase measurements. From SuperCDMS detector data it is necessary to reconstruct the energies of the particle events. This thesis explores the use of Convolutional Neural Networks (CNNs) to perform this reconstruction and finds that, although they perform well, changing the noise model breaks the model and requires the neural network to be retrained. In order to mitigate this issue, a new CNN model is proposed which includes the noise Power Spectral Density (PSD) of the data as an additional input to the CNN. While it proves to be effective as a denoising algorithm, it still fails for data with a different noise model. However, including data from multiple PSDs in the neural network training sample allows it to handle data with different types of noise while still maintaining the quality of the reconstruction. Nevertheless, neural networks trained even on multiple PSDs do not robustly handle data taken with PSDs dissimilar to those in the training sample, suggesting that CNNs may need to be retrained whenever the noise environment changes in a significant way.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Signal-preserving CMB component separation with machine learning

Analysis of microwave sky signals, such as the cosmic microwave background, often requires component separation using multifrequency methods, whereby different signals are isolated according to their different frequency behaviors. Many so-called blind methods, such as the internal linear combination (ILC), make minimal assumptions about the spatial distribution of the signal or contaminants, and only assume knowledge of the frequency dependence of the signal. The ILC produces a minimum-variance linear combination of the measured frequency maps. In the case of Gaussian, statistically isotropic fields, this is the optimal linear combination, as the variance is the only statistic of interest. However, in many cases the signal we wish to isolate, or the foregrounds we wish to remove, are non-Gaussian and/or statistically anisotropic (in particular for the case of Galactic foregrounds). In such cases, it is possible that machine learning (ML) techniques can be used to exploit the non-Gaussian features of the foregrounds and thereby improve component separation. However, many ML techniques require the use of complex, difficult-to-interpret operations on the data. We propose a hybrid method whereby we train an ML model using only combinations of the data that , and combine the resulting ML-predicted foreground estimate with the ILC solution to reduce the error from the ILC. We demonstrate our methods on simulations of extragalactic temperature and Galactic polarization foregrounds and show that our ML model can exploit non-Gaussian features, such as point sources and spatially varying spectral indices, to produce lower-variance maps than ILC—e.g., reducing the variance of the B-mode residual by factors of up to 5—while preserving the signal of interest in an unbiased manner. Moreover, we often find improved performance even when applying our ML technique to foreground models on which it was not trained. Published by the American Physical Society 2025

McCarthy, Fiona (ORCID:0000000253893565)↗

Photometric redshift requirements for self-calibration of cluster dark energy studies

The ability to constrain dark energy from the evolution of galaxy cluster counts is limited by the imperfect knowledge of cluster redshifts. Ongoing and upcoming surveys will mostly rely on redshifts estimated from broadband photometry (photo-z's). For a Gaussian distribution for the cluster photo-z errors and a high cluster yield cosmology defined by the WMAP 1 year results, the photo-z bias, and scatter needs to be known better than 0.003 and 0.03, respectively, in order not to degrade dark energy constrains by more than 10% for a survey with specifications similar to the South Pole Telescope. Smaller surveys and cosmologies with lower cluster yields produce weaker photo-z requirements, though relative to worse baseline constraints. Comparable photo-z requirements are necessary in order to employ self-calibration techniques when solving for dark energy and observable-mass parameters simultaneously. On the other hand, self-calibration in combination with external mass inferences helps reduce photo-z requirements and provides important consistency checks for future cluster surveys. In our fiducial model, training sets with spectroscopic redshifts for {approx}5%-15% of the detected clusters are required in order to keep degradations in the dark energy equation of state lower than 20%.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

The physics experimental study for in-hospital neutron irradiator

MNSRs (Miniature Neutron Source Reactor) are low power research reactors designed and manufactured by China Institute of Atomic Energy (CIAE). MNSRs are mainly used for NAA, training and teaching, testing of nuclear instrumentation. The first MNSR, the prototype MNSR, was put into operation in 1984, later, eight other MNSRs had been built both at home and abroad. For MNSRs, highly enriched uranium (90%) is used as the fuel material. The In-Hospital Neutron Irradiator (IHNI) is designed for Boron Neutron Capture Therapy (BNCT) based on Miniature Neutron Source Reactor(MNSR). On both sides of the reactor core, there are two neutron beams, one is thermal neutron beam, and the other opposite to the thermal beam, is epithermal neutron beam. A small thermal neutron beam is specially designed for the measurement of blood boron concentration by the prompt gamma neutron activation analysis (PGNAA). In this paper, the experimental results of critical mass worth of the top Be reflectors worth of the control rod, neutron flux distribution and other components worth were measured, the experiment was done on the Zero Power Experiment equipment of MNSR. (author)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗