Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “knowledge representation”

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 397 records · Page 22
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Heliostat Consortium: Update on Resource, Training, and Education Development and Women+ in Concentrating Solar

One of the major barriers to Concentrating Solar Power (CSP) industry growth in the U.S. is the lack of accessibility of expert knowledge, a lack of a stable workforce pipeline, and little public awareness of the technology area. Improved Resource, Training, and Education (RTE) is needed to address these gaps, to draw workers into the field from diverse technology areas and backgrounds, and to reduce the learning curves of newcomers into the industry. As part of the RTE topic under the NREL-led Heliostat Consortium (HelioCon), a public online resource database has been created to support information sharing across the heliostat and CSP community. Here, information will be presented on the existing and planned database resources, including introductory information about CSP and heliostats, a reference library of texts and publications, educational videos, a CSP power tower plant database and case studies, a contact list of component suppliers and developers, and a list of available software/metrology tools. Women+ in Concentrating Solar is a newly created organization to promote the education, professional development, and advancement of underrepresented genders in the Concentrating Solar Power (CSP) community in a way that includes diverse backgrounds, attitudes, and perspectives. The main objectives of the groups are to provide the following opportunities for underrepresented genders: 1) organize events to provide networking opportunities and community building, 2) boost partnerships and scientific collaboration, 3) create resources to facilitate knowledge and experience sharing, 4) provide mentorship programs to support those new to the industry, 5) bring visibility to underrepresented genders and attract more to the field, and 6) promote Diversity, Equity, and Inclusion (DEI) and engage minorities and underrepresented groups. Initial activities being pursued by the group include creating an expert database to support event organizers in recruiting speakers with representation of all gender identities, hosting online networking and webinar events, organizing conference events including seminars and networking/social events, and establishing a mentorship program to connect members in the CSP community across the globe. The group is open to all genders and has grown to over 60 members in its first few months.

concentrating solar↗
DOE OSTI2021

Tissue scale agent-based simulation of premalignant progressions in Barrett’s esophagus

Barrett’s esophagus (BE) is a benign condition of the distal esophagus that initiates a multistage pathway to esophageal adenocarcinoma (EAC). Short of frequent intrusive (and costly) surveillance, effective screening for neoplasia in BE populations is yet to be established since progressors are rare and virtually undetectable without routine biopsies, which often sample only a small portion of the BE tissue. As a result, reliable estimation of the true prevalence of dysplasia in a BE population and evidence-based optimization of screening for at-risk individuals is challenging. Data-driven microsimulations, i.e., model-generated instances of disease history in a predefined virtual population, have found utility in the EAC screening literature as low-overhead alternatives to real-world hypothesis testing of optimal interventions for dysplasia. Despite the successes, computational limitations, paucity of knowledge and data on Barrett’s dysplasia, and the complexities of disease progression as a multiscale multiphysics process have hindered the treatment of disease progression in BE as a spatial process. Agent-based modeling of nucleation and proliferation processes in dysplasia warrants exploration in this context as an approximation that operates at a trade-off between computational tractability and precise representation of the composition and physics of the substrate (tissue). In this study, we describe spatially resolved simulations of premalignant progression toward EAC in a coarse-grained model of Barrett’s tissue that resolves the metaplastic tissue at a length scale of 0.42 mm (~3300 crypts/mm 2 ). Finally, the model is calibrated to reproduce historical high-grade dysplasia prevalence when model-generated patients are screened using the Seattle protocol.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Quantifying model prediction sensitivity to model-form uncertainty

Computational and mathematical models are essential to understanding complex systems and phenomena. However, when developing such models, limited knowledge and/or resources necessitates the use of simplifying assumptions. It is therefore crucial to quantify the impact of such simplifying assumptions on the reliability and accuracy of resulting model predictions. This work develops a first-of-its-kind approach to quantify the impact of physics modeling assumptions on predictions. Here, we leverage the emerging field of model-form uncertainty (MFU) representations, which are parameterized modifications to modeling assumptions, in combination with grouped Sobol’ indices to quantitatively measure an assumption’s importance. Specifically, we compute the grouped Sobol’ index for the MFU representation’s parameters as a single importance measure of the assumption for which the MFU representation characterizes uncertainty. To ensure this approach is robust to the subjective choice of how to parameterize a MFU representation, we establish bounds for the difference between sensitivity results for two different MFU representations based on differences in model prediction statistics. The capabilities associated with this approach are demonstrated on three exemplar problems: an upscaled subsurface contaminant transport problem, ablation modeling for hypersonic flight, and nuclear waste repository modeling. We found that our grouped approach is able to assess the impact of modeling assumptions on predictions and offers computational advantages over classical Sobol’ index computation while providing more interpretable results.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Bridging molecular-scale interfacial science with continuum-scale models

Solid–water interfaces are crucial for clean water, conventional and renewable energy, and effective nuclear waste management. However, reflecting the complexity of reactive interfaces in continuum-scale models is a challenge, leading to oversimplified representations that often fail to predict real-world behavior. This is because these models use fixed parameters derived by averaging across a wide physicochemical range observed at the molecular scale. Recent studies have revealed the stochastic nature of molecular-level surface sites that define a variety of reaction mechanisms, rates, and products even across a single surface. To bridge the molecular knowledge and predictive continuum-scale models, we propose to represent surface properties with probability distributions rather than with discrete constant values derived by averaging across a heterogeneous surface. This conceptual shift in continuum-scale modeling requires exponentially rising computational power. By incorporating our molecular-scale understanding of solid–water interfaces into continuum-scale models we can pave the way for next generation critical technologies and novel environmental solutions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Semi-Automated, Object-Based Tomography of Dislocation Structures

The characterization of the three-dimensional arrangement of dislocations is important for many analyses in materials science. Dislocation tomography in transmission electron microscopy is conventionally accomplished through intensity-based reconstruction algorithms. Although such methods work successfully, a disadvantage is that they require many images to be collected over a large tilt range. Here, we present an alternative, semi-automated object-based approach that reduces the data collection requirements by drawing on the prior knowledge that dislocations are line objects. Our approach consists of three steps: (1) initial extraction of dislocation line objects from the individual frames, (2) alignment and matching of these objects across the frames in the tilt series, and (3) tomographic reconstruction to determine the full three-dimensional configuration of the dislocations. Drawing on innovations in graph theory, we employ a node-line segment representation for the dislocation lines and a novel arc-length mapping scheme to relate the dislocations to each other across the images in the tilt series. We demonstrate the method for a dataset collected from a dislocation network imaged by diffraction-contrast scanning transmission electron microscopy. Based on these results and a detailed uncertainty analysis for the algorithm, we discuss opportunities for optimizing data collection and further automating the method.

47 OTHER INSTRUMENTATION↗
DOE OSTI2022

Geometric prior of multi-resolution yielding manifolds and the local closest point projection for nearly non-smooth plasticity

Elastoplasticity models often introduce a scalar-valued yield function to implicitly represent the boundary between elastic and plastic material states. Here this paper introduces a new alternative where the yield envelope is represented by a manifold of which the topology and the geometry are learned from a set of data points in a parametric space (e.g. principal stress space, π-plane). Here, deep geometric learning enables us to reconstruct a highly complex yield envelope by breaking it down into multiple coordinate charts. The global atlas that consists of these coordinate charts in return allows us to represent the yield surface via multiple overlapping patches, each with a specific local parametrization. This setup provides several advantages over the classical implicit function representation approach. For instance, the availability of coordinate charts enables us to introduce an alternative stress integration algorithm where the trial stress may project directly on a local patch and hence circumvent the issues related to non-smoothness and the lack of convexity of yield surfaces. Meanwhile, the local parametric approach also enables us to predict hardening/softening locally in the parametric space, even without complete knowledge of the yield surface. Comparisons between the classical yield function approach on the non-smooth plasticity and anisotropic cam-clay plasticity model are provided to demonstrate the capacity of the models for highly precise yield surface and the feasibility of the implementation of the learned model in the local stress integration algorithm.

42 ENGINEERING↗
DOE OSTI2021

Phoenix: A Scalable Streaming Hypergraph Analysis Framework

We present Phoenix, a scalable hypergraph analytics framework for data analytics and knowledge discovery that was implemented on the leadership class computing platforms at Oak Ridge National Laboratory (ORNL). Our software framework comprises a distributed implementation of a streaming server architecture which acts as a gateway for various hypergraph generators/external sources to connect. Phoenix has the capability to utilize diverse hypergraph generators, including HyGen, a very large-scale hypergraph generator developed by ORNL. Phoenix incorporates specific algorithms for efficient data representation by exploiting hidden structures of the hypergraphs. Our experimental results demonstrate Phoenix’s scalable and stable performance on massively parallel computing platforms. Phoenix’s superior performance is due to the merging of high-performance computing with data analytic.

Kurte, Kuldeep↗
DOE OSTI2024

HDBind: encoding of molecular structure with hyperdimensional binary representations

Traditional methods for identifying “hit” molecules from a large collection of potential drug-like candidates rely on biophysical theory to compute approximations to the Gibbs free energy of the binding interaction between the drug and its protein target. These approaches have a significant limitation in that they require exceptional computing capabilities for even relatively small collections of molecules. Increasingly large and complex state-of-the-art deep learning approaches have gained popularity with the promise to improve the productivity of drug design, notorious for its numerous failures. However, as deep learning models increase in their size and complexity, their acceleration at the hardware level becomes more challenging. Hyperdimensional Computing (HDC) has recently gained attention in the computer hardware community due to its algorithmic simplicity relative to deep learning approaches. The HDC learning paradigm, which represents data with high-dimension binary vectors, allows the use of low-precision binary vector arithmetic to create models of the data that can be learned without the need for the gradient-based optimization required in many conventional machine learning and deep learning methods. This algorithmic simplicity allows for acceleration in hardware that has been previously demonstrated in a range of application areas (computer vision, bioinformatics, mass spectrometery, remote sensing, edge devices, etc.). To the best of our knowledge, our work is the first to consider HDC for the task of fast and efficient screening of modern drug-like compound libraries. We also propose the first HDC graph-based encoding methods for molecular data, demonstrating consistent and substantial improvement over previous work. We compare our approaches to alternative approaches on the well-studied MoleculeNet dataset and the recently proposed LIT-PCBA dataset derived from high quality PubChem assays. We demonstrate our methods on multiple target hardware platforms, including Graphics Processing Units (GPUs) and Field Programmable Gate Arrays (FPGAs), showing at least an order of magnitude improvement in energy efficiency versus even our smallest neural network baseline model with a single hidden layer. Our work thus motivates further investigation into molecular representation learning to develop ultra-efficient pre-screening tools. We make our code publicly available at https://github.com/LLNL/hdbind.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Exploring the impacts of unprecedented climate extremes on forest ecosystems: hypotheses to guide modeling and experimental studies

Abstract. Climatic extreme events are expected to occur more frequently in the future, increasing the likelihood of unprecedented climate extremes (UCEs) or record-breaking events. UCEs, such as extreme heatwaves and droughts, substantially affect ecosystem stability and carbon cycling by increasing plant mortality and delaying ecosystem recovery. Quantitative knowledge of such effects is limited due to the paucity of experiments focusing on extreme climatic events beyond the range of historical experience. Here, we present a road map of how dynamic vegetation demographic models (VDMs) can be used to investigate hypotheses surrounding ecosystem responses to one type of UCE: unprecedented droughts. As a result of nonlinear ecosystem responses to UCEs that are qualitatively different from responses to milder extremes, we consider both biomass loss and recovery rates over time by reporting a time-integrated carbon loss as a result of UCE, relative to the absence of drought. Additionally, we explore how unprecedented droughts in combination with increasing atmospheric CO2 and/or temperature may affect ecosystem stability and carbon cycling. We explored these questions using simulations of pre-drought and post-drought conditions at well-studied forest sites using well-tested models (ED2 and LPJ-GUESS). The severity and patterns of biomass losses differed substantially between models. For example, biomass loss could be sensitive to either drought duration or drought intensity depending on the model approach. This is due to the models having different, but also plausible, representations of processes and interactions, highlighting the complicated variability of UCE impacts that still need to be narrowed down in models. Elevated atmospheric CO2 concentrations (eCO2) alone did not completely buffer the ecosystems from carbon losses during UCEs in the majority of our simulations. Our findings highlight the consequences of differences in process formulations and uncertainties in models, most notably related to availability in plant carbohydrate storage and the diversity of plant hydraulic schemes, in projecting potential ecosystem responses to UCEs. We provide a summary of the current state and role of many model processes that give way to different underlying hypotheses of plant responses to UCEs, reflecting knowledge gaps which in future studies could be tested with targeted field experiments and an iterative modeling–experimental conceptual framework.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Generation and representation of synthetic smart meter data

Advanced energy algorithms running at big-data scale will be necessary to identify, realize, and verify energy savings to meet government and utility goals of building energy efficiency. Any algorithm must be well characterized and validated before it is trusted to run at these scales. Smart meter data from real buildings will ultimately be required for the development, testing, and validation of these energy algorithms and processes. However, for initial development and testing, smart meter data are difficult to work with due to privacy restrictions, noise from unknown sources, data accessibility, and other concerns which can complicate algorithm development and validation. This paper describes a new methodology to generate synthetic smart meter data of electricity use in buildings using detailed building energy modeling, which aims to capture the variability and stochastics of real energy use in buildings. The methodology can create datasets tailored to represent specific scenarios with known truth and controllable amounts of synthetic noise. Knowledge of ground truth also allows the development and validation of enhanced processes which leverage building metadata, such as building type or size (floor area), in addition to smart meter data. The methodology described in this paper includes the key influencing factors of real-world building energy use including weather data, occupant-driven loads, building operation and maintenance practices, and special events. Data formats to support workflows leveraging both synthetic meter data and associated metadata are proposed and discussed. Finally, example use cases of the synthetic meter data are described to illustrate potential applications.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

A successful start for mLife

mLife made its debut with the publication of the inaugurating issue on March 30, 2022. We are pleased to have published 42 high-quality articles, including reviews, research articles, and application notes, which cover microbial genetics, biochemistry and physiology, ecology and evolution, pathogenic and public health microbiology, and so forth. Nine of the articles concern technical innovations and applications in areas including genome editing, single-cell analysis, and database. To pay tribute to the great microbiologist Carl Woese on the 10th anniversary of his death, the Journal published a special section containing three thought-provoking articles in the last issue of the year. mLife and its collaborating publisher Wiley have made great efforts to increase the Journal's national and continental representation. As a result, the Journal received submissions from over 20 countries in five continents last year. The authors of the published articles are from China, United States, Great Britain, France, Canada, Ger many, Japan, Switzerland, and Austria.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

CRCNS US-France Research Proposal: Collaborative Research: Encoding reward expectation in Drosophilia

The fruit fly Drosophila melanogaster has been a valuable model for investigating the genetic and neural bases that underlie learning and memory. Early and most current studies use basic behavior conditioning protocols to study learning in controlled laboratory settings. More recently, the ability to transgenically manipulate many of the brain neurons in the fruit fly with exquisite specificity, and the recent knowledge of the synaptic ‘connectome’ of the fruit fly brain, makes these animals almost unique as a comprehensive model for studies of learning, memory and motivated behavior. In fact, the connectome has revealed many types of new connections that had until now been overlooked. Within this context, the thesis of this proposal is that studies of learning and memory will be greatly enhanced by using more sophisticated means for evaluating memory representations, such as have been developed in vertebrates, and combining those studies with information from the connectome guided by computational modelling. We propose to push beyond the boundaries of existing conditioning protocols for fruit flies to investigate more complex memory representations. In particular, we will investigate the function of reinforcement pathways in relation to the absence of expected reinforcement. More specifically, we propose a series of experiments designed to investigate the memory representations in fruit flies when an expected consequence of a Conditioned Stimulus (CS) fails to occur. Although studies have evaluated how this failure can establish extinction memory for the CS, our studies will go beyond studying extinction. Specifically, we predict that in Drosophila when a CS is associated with a failed expectation of an appetitive food reinforcement it will acquire aversive value, and vice versa for a failed expectation of an aversive reinforcer. We combine these studies with manipulations of reinforcement pathways in the CNS inspired from the connectome, iteratively knitted in with established computational models. Intellectual Merit: The concept of reinforcement expectation and incentive contrast have been influential in the development of studies of associative learning in mammals. These questions are particularly challenging to answer in vertebrates because they require exquisite cellular, temporal, and genetic specificity of experimental manipulations. The recent development of work with identified neurons and their connectomes makes the larval and adult fly brains ripe as models for pushing our understanding of neural bases for these higher- order conditioning phenomena. Broader Impacts: Public health: These analyses and the conceptual framework of prediction error processing underlying them have a profound impact on our understanding of reinforcement-related behavior in humans, including monetary rewards and the mnemonic consequences of traumatic experiences, and for pathologies of the dopamine reinforcement system. Educational: This project will provide interdisciplinary training for postdoctoral researchers, Ph.D. and undergraduate students. The PIs will act as co-supervisors or mentors of students working in the different labs via face-to-face and internet-based technologies. We will also work with ASU’s award-winning Ask- A-Biologist program. This is an online science program designed to enrich the learning experiences of students of all ages and to provide classroom material for use by K-12 teachers. We will develop an extension of a game developed under a prior NSF award, and the new game will include modules to teach K-12 students about how insects learn. We will also integrate into the AAB site a program developed by a collaborator (B Gerber) at the Leibniz Institut für Neurobiologie, Magdeburg, and now in use in schools in Germany, to teach K-12 students how to train animals using the fruit fly larval learning paradigm. Underrepresented groups: All PIs will work with their university offices of Academic Diversity and Equal Opportunity for reaching underrepresented students.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Confronting the Challenge of Modeling Cloud and Precipitation Microphysics

In the atmosphere, microphysics refers to the microscale processes that affect cloud and precipitation particles and is a key linkage among the various components of Earth’s atmospheric water and energy cycles. The representation of microphysical processes in models continues to pose a major challenge leading to uncertainty in numerical weather forecasts and climate simulations. In this paper, the problem of treating microphysics in models is divided into two parts: i) how to represent the population of cloud and precipitation particles, given the impossibility of simulating all particles individually within a cloud, and ii) uncertainties in the microphysical process rates owing to fundamental gaps in knowledge of cloud physics. The recently-developed Lagrangian particle-based method is advocated as a way to address several conceptual and practical challenges of representing particle populations using traditional bulk and bin microphysics parameterization schemes. For addressing critical gaps in cloud physics knowledge, sustained investment for observational advances from laboratory experiments, new probe development, and next-generation instruments in space is needed. Greater emphasis on laboratory work, which has apparently declined over the past several decades relative to other areas of cloud physics research, is argued to be an essential ingredient for improving process-level understanding. More systematic use of natural cloud and precipitation observations to constrain microphysics schemes is also advocated. Because it is generally difficult to quantify individual microphysical process rates from these observations directly, this presents an inverse problem that can be viewed from the standpoint of Bayesian statistics. Following this idea, a probabilistic framework is proposed that combines elements from statistical and physical modeling. Besides providing rigorous constraint of schemes, there is an added benefit of quantifying uncertainty systematically. Finally, a broader hierarchical approach is proposed to accelerate improvements in microphysics schemes, leveraging the advances described in this paper related to process modeling (using Lagrangian particle-based schemes), laboratory experimentation, cloud and precipitation observations, and statistical methods.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Towards automating structural discovery in scanning transmission electron microscopy *

Abstract Scanning transmission electron microscopy is now the primary tool for exploring functional materials on the atomic level. Often, features of interest are highly localized in specific regions in the material, such as ferroelectric domain walls, extended defects, or second phase inclusions. Selecting regions to image for structural and chemical discovery via atomically resolved imaging has traditionally proceeded via human operators making semi-informed judgements on sampling locations and parameters. Recent efforts at automation for structural and physical discovery have pointed towards the use of ‘active learning’ methods that utilize Bayesian optimization with surrogate models to quickly find relevant regions of interest. Yet despite the potential importance of this direction, there is a general lack of certainty in selecting relevant control algorithms and how to balance a priori knowledge of the material system with knowledge derived during experimentation. Here we address this gap by developing the automated experiment workflows with several combinations to both illustrate the effects of these choices and demonstrate the tradeoffs associated with each in terms of accuracy, robustness, and susceptibility to hyperparameters for structural discovery. We discuss possible methods to build descriptors using the raw image data and deep learning based semantic segmentation, as well as the implementation of variational autoencoder based representation. Furthermore, each workflow is applied to a range of feature sizes including NiO pillars within a La:SrMnO 3 matrix, ferroelectric domains in BiFeO 3 , and topological defects in graphene. The code developed in this manuscript is open sourced and will be released at github.com/nccreang/AE_Workflows .

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

Evaluating the impact of peat soils and snow schemes on simulated active layer thickness at pan-Arctic permafrost sites

Abstract Permafrost stability is significantly influenced by the thermal buffering effects of snow and active-layer peat soils. In the warm season, peat soils act as a barrier to downward heat transfer mainly due to their low thermal conductivity. In the cold season, the snowpack serves as a thermal insulator, retarding the release of heat from the soil to the atmosphere. Currently, many global land models overestimate permafrost soil temperature and active layer thickness (ALT), partially due to inaccurate representations of soil organic matter (SOM) density profiles and snow thermal insulation. In this study, we evaluated the impacts of SOM and snow schemes on ALT simulations at pan-Arctic permafrost sites using the Energy Exascale Earth System Model (E3SM) land model (ELM). We conducted simulations at the Circumpolar Active Layer Monitoring (CALM) sites across the pan-Arctic domain. We improved ELM-simulated site-level ALT using a knowledge-based hierarchical optimization procedure and examined the effects of precipitation-phase partitioning methods (PPMs), snow compaction schemes, and snow thermal conductivity schemes on simulated snow depth, soil temperature, ALT, and CO 2 fluxes. Results showed that the optimized ELM significantly improved agreement with observed ALT (e.g. RMSE decreased from 0.83 m to 0.15 m). Our sensitivity analysis revealed that snow-related schemes significantly impact simulated snow thermal insulation levels, soil temperature, and ALT. For example, one of the commonly used snow thermal conductivity schemes (quadratic Sturm or SturmQua) generally produced warmer soil temperatures and larger ALT compared to the other two tested schemes. The SturmQua scheme also amplified the model’s sensitivity to PPMs and predicted deeper ALTs than the other two snow schemes under both current and future climates. The study highlights the importance of accurately representing snow-related processes and peat soils in land models to enhance permafrost dynamics simulations.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Track Seeding and Labelling with Embedded-space Graph Neural Networks

To address the unprecedented scale of HL-LHC data, the Exa.TrkX project is investigating a variety of machine learning approaches to particle track reconstruction. The most promising of these solutions, graph neural networks (GNN), process the event as a graph that connects track measurements (detector hits corresponding to nodes) with candidate line segments between the hits (corresponding to edges). Detector information can be associated with nodes and edges, enabling a GNN to propagate the embedded parameters around the graph and predict node-, edge- and graph-level observables. Previously, message-passing GNNs have shown success in predicting doublet likelihood, and we here report updates on the state-of-the-art architectures for this task. In addition, the Exa.TrkX project has investigated innovations in both graph construction, and embedded representations, in an effort to achieve fully learned end-to-end track finding. Hence, we present a suite of extensions to the original model, with encouraging results for hitgraph classification. In addition, we explore increased performance by constructing graphs from learned representations which contain non-linear metric structure, allowing for efficient clustering and neighborhood queries of data points. We demonstrate how this framework fits in with both traditional clustering pipelines, and GNN approaches. The embedded graphs feed into high-accuracy doublet and triplet classifiers, or can be used as an end-to-end track classifier by clustering in an embedded space. A set of post-processing methods improve performance with knowledge of the detector physics. Finally, we present numerical results on the TrackML particle tracking challenge dataset, where our framework shows favorable results in both seeding and track finding.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2024

Estimates of Southern Hemispheric Gravity Wave Momentum Fluxes across Observations, Reanalyses, and Kilometer-Scale Numerical Weather Prediction Model

Abstract Gravity waves (GWs) are among the key drivers of the meridional overturning circulation in the mesosphere and upper stratosphere. Their representation in climate models suffers from insufficient resolution and limited observational constraints on their parameterizations. This obscures assessments of middle atmospheric circulation changes in a changing climate. This study presents a comprehensive analysis of stratospheric GW activity above and downstream of the Andes from 1 to 15 August 2019, with special focus on GW representation ranging from an unprecedented kilometer-scale global forecast model (1.4 km ECMWF IFS), ground-based Rayleigh lidar (CORAL) observations, modern reanalysis (ERA5), to a coarse-resolution climate model (EMAC). Resolved vertical flux of zonal GW momentum (GWMF) is found to be stronger by a factor of at least 2–2.5 in IFS compared to ERA5. Compared to resolved GWMF in IFS, parameterizations in ERA5 and EMAC continue to inaccurately generate excessive GWMF poleward of 60°S, yielding prominent differences between resolved and parameterized GWMFs. A like-to-like validation of GW profiles in IFS and ERA5 reveals similar wave structures. Still, even at ∼1 km resolution, the resolved waves in IFS are weaker than those observed by lidar. Further, GWMF estimates across datasets reveal that temperature-based proxies, based on midfrequency approximations for linear GWs, overestimate GWMF due to simplifications and uncertainties in GW wavelength estimation from data. Overall, the analysis provides GWMF benchmarks for parameterization validation and calls for three-dimensional GW parameterizations, better upper-boundary treatment, and vertical resolution increases commensurate with increases in horizontal resolution in models, for a more realistic GW analysis. Significance Statement Gravity wave–induced momentum forcing forms a key component of the middle atmospheric circulation. However, complete knowledge of gravity waves, their atmospheric effects, and their long-term trends are obscured due to limited global observations, and the inability of current climate models to fully resolve them. This study combines a kilometer-scale forecast model, modern reanalysis, and a coarse-resolution climate model to first compare the resolved and parameterized momentum fluxes by gravity waves generated over the Andes, and then evaluate the fluxes using a state-of-the-art ground-based Rayleigh lidar. Our analysis reveals shortcomings in current model parameterizations of gravity waves in the middle atmosphere and highlights the sensitivity of the estimated flux to the formulation used.

Meteorology & Atmospheric Sciences↗
DOE OSTI2025

A knowledge-informed large language model framework for U.S. nuclear power plant shutdown initiating event classification for probabilistic risk assessment

Identifying and classifying shutdown initiating events (SDIEs) is critical for developing shutdown probabilistic risk assessment for nuclear power plants. Existing computational approaches cannot achieve satisfactory performance due to the challenges of unavailable large, labeled datasets, imbalanced event types, and label noise. To address these challenges, we propose a hybrid pipeline that integrates a knowledge-informed machine learning model to prescreen non-SDIEs and a large language model (LLM) to classify SDIEs into four types. In the prescreening stage, we proposed a set of 44 SDIE text patterns that consist of the most salient keywords and phrases from six SDIE types. Text vectorization based on the SDIE patterns generates feature vectors that are highly separable by using a simple binary classifier. The second stage builds Bidirectional Encoder Representations from Transformers (BERT)-based LLM, which learns generic English language representations from self-supervised pretraining on a large dataset and adapts to SDIE classification by fine-tuning it on an SDIE dataset. The proposed approaches are evaluated on a dataset with 10,928 events using precision, recall ratio, F 1 score, and average accuracy. In conclusion, the results demonstrate that the prescreening stage can exclude more than 97% non-SDIEs, and the LLM achieves an average accuracy of 95.1% for SDIE classification.

99 - GENERAL AND MISCELLANEOUS↗