Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Model Inference”

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 1,675 records · Page 93
All sourcesNASA NTRSDOE OSTI
NASA NTRS1996

Parallelization and Visual Analysis of Multidimensional Fields: Application to Ozone Production, Destruction, and Transport in Three Dimensions

The three-dimensional, spectral transport model used in the current project was first successfully integrated over climatological time scales by Dr. Guang Ping Lou for the simulation of atmospheric N2O using the United Kingdom Meteorological Office (UKMO) 4-dimensional, assimilated wind and temperature data set. A non-parallel, FORTRAN version of this integration using a fairly simple N2O chemistry package containing only photo-chemical reactions was used to verify our initial parallel model results. The integrations reproduced the gross features of the observed stratospheric climatological N2O distributions but also simulated the structure of the stratospheric Antarctic vortex and its evolution. Subsequently, Dr. Thomas Kindler, who produced much of the parallel version of our model, enlarged the N2O model chemistry package to include N2O reactions involving O(D-1) and also introduced assimilated wind data from NASA as well as UKMO. Initially, transport calculations without chemistry were run using Carbon-14 as a non-reactive tracer gas with the result that large differences in the transport properties of the two assimilated wind data sets were apparent from the resultant Carbon-14 distributions. Subsequent calculations for N2O, including its chemistry, with the two input winds data sets with verification from UARS satellite observations have refined the transport differences between the two such that the model's steering capabilities could be used to infer the correct climatological vertical velocity fields required to support the N2O observations. During this process, it was also discovered that both the NASA and the UKMO data contained spurious values in some of the higher frequency wave components, leading to incorrect local transport calculations and ultimately affecting the large scale properties of the model's N2O distributions, particularly at tropical latitudes. Subsequent model runs with wind data that had been filtered to remove some of the high frequency components produced much more realistic N2O distributions. During the past few months, the UKMO wind data base for a complete two-year period was processed into spectral form for model use. This new version of the input transport data base now includes complete temperature fields as well as the necessary wind data. This was done to facilitate advanced chemical calculations in the parallel model which often depend upon temperature. Additional UKMO data is being added as it becomes available.

Schwan, Karsten↗
DOE OSTI2025

Learning with Adaptive Conservativeness for Distributionally Robust Optimization: Incentive Design for Voltage Regulation

Information asymmetry between the Distribution System Operator (DSO) and Distributed Energy Resource Aggregators (DERAs) obstructs designing effective incentives for voltage regulation. To capture this effect, we employ a Stackelberg game-theoretic framework, where the DSO seeks to overcome the information asymmetry and refine its incentive strategies by learning from DERA behavior over multiple iterations. We introduce a model-based online learning algorithm for the DSO, aimed at inferring the relationship between incentives and DERA responses. Given the uncertain nature of these responses, we also propose a distributionally robust incentive design model to control the probability of voltage regulation failure and then reformulate it into a convex problem. This model allows the DSO to periodically revise distribution assumptions on uncertain parameters in the decision model of the DERA. Finally, we present a gradient-based method that permits the DSO to adaptively modify its conservativeness level, measured by the size of a Wasserstein metric-based ambiguity set, according to historical voltage regulation performance. The effectiveness of our proposed method is demonstrated through numerical experiments.

adaptation models↗
DOE OSTI2024

Learning with Adaptive Conservativeness for Distributionally Robust Optimization: Incentive Design for Voltage Regulation: Preprint

Information asymmetry between the Distribution System Operator (DSO) and Distributed Energy Resource Aggregators (DERAs) obstructs designing effective incentives for voltage regulation. To capture this effect, we employ a Stackelberg game-theoretic framework, where the DSO seeks to overcome the information asymmetry and refine its incentive strategies by learning from DERA behavior over multiple iterations. We introduce a model-based online learning algorithm for the DSO, aimed at inferring the relationship between incentives and DERA responses. Given the uncertain nature of these responses, we also propose a distributionally robust incentive design model to control the probability of voltage regulation failure and then reformulate it into a convex problem. This model allows the DSO to periodically revise distribution assumptions on uncertain parameters in the decision model of the DERA. Finally, we present a gradient-based method that permits the DSO to adaptively modify its conservativeness level, measured by the size of a Wasserstein metric-based ambiguity set, according to historical voltage regulation performance. The effectiveness of our proposed method is demonstrated through numerical experiments.

distribution system operator↗
NASA NTRS2022

GBaTSv2: a revised synthesis of the likely basal thermal state of the Greenland Ice Sheet

The basal thermal state (frozen or thawed) of the Greenland Ice Sheet is under-constrained due to few direct measurements, yet knowledge of this state is becoming increasingly important to interpret modern changes in ice flow. The first synthesis of this state relied on inferences from widespread airborne and satellite observations and numerical models, for which most of the underlying datasets have since been updated. Further, new and independent constraints on the basal thermal state have been developed from analysis of basal and englacial reflections observed by airborne radar sounding. Here we synthesize constraints on the Greenland Ice Sheet's basal thermal state from boreholes, thermomechanical ice-flow models that participated in the Ice Sheet Model Intercomparison Project for CMIP6 (ISMIP6; Coupled Model Intercomparison Project Phase 6), IceBridge BedMachine Greenland v4 bed topography, Making Earth Science Data Records for Use in Research Environments (MEaSUREs) Multi-Year Greenland Ice Sheet Velocity Mosaic v1 and multiple inferences of a thawed bed from airborne radar sounding. Most constraints can only identify where the bed is likely thawed rather than where it is frozen. This revised synthesis of the Greenland likely Basal Thermal State version 2 (GBaTSv2) indicates that 33 % of the ice sheet's bed is likely thawed, 40 % is likely frozen and the remainder (28 %) is too uncertain to specify. The spatial pattern of GBaTSv2 is broadly similar to the previous synthesis, including a scalloped frozen core and thawed outlet-glacier systems. Although the likely basal thermal state of nearly half (46 %) of the ice sheet changed designation, the assigned state changed from likely frozen to likely thawed (or vice versa) for less than 6 % of the ice sheet. This revised synthesis suggests that more of northern Greenland is likely thawed at its bed and conversely that more of southern Greenland is likely frozen, both of which influence interpretation of the ice sheet's present subglacial hydrology and models of its future evolution. The GBaTSv2 dataset, including both code that performed the analysis and the resulting datasets, is freely available at https://doi.org/10.5281/zenodo.6759384 (MacGregor, 2022).

Joseph A. MacGregor↗
DOE OSTI2025

Laser powder bed fusion parameter estimation with k-NN

Abstract Laser powder bed fusion (L-PBF) is a technique within additive manufacturing that uses a high power density laser to build parts from fused powdered metal alloy. This technology is well equipped to produce complex parts with otherwise impossible features, such as hidden voids or lattice structures. Alongside capability, reliability and quality are key characteristics considered when choosing a manufacturing method, and these are gaining attention as this method becomes more prevalent in industry. One main indicator of a stable L-PBF process is consistent melt pool geometry, and the properties of which are likely to determine the quality of the part produced. As computing power and sensing technologies become more advanced, this melt pool geometry could be studied in real time. This work addresses the challenge by leveraging a k-nearest neighbor (k-NN) model to identify key features within melt pool imagery and predict the energy density. The k-NN model was trained on data provided by the National Institute of Standards and Technology (NIST). Data preprocessing was performed on the images to extract features that were used in the k-NN model. This approach was used to accurately infer the energy density of unseen layers within the same part. The algorithm was subsequently tested with unique scan strategies and found to reasonably estimate the energy density of different parts. A fivefold cross validation found the algorithm to be consistently predicting the class of 91.4% of the in situ melt pool images.

Jung, Patrick (ORCID:0000000267890859)↗
DOE OSTI2022

AtomAI framework for deep learning analysis of image and spectroscopy data in electron and scanning probe microscopy

Over the past several decades, electron and scanning probe microscopes have become critical components of condensed matter physics, materials science and chemistry research. At the same time, the infrastructure for establishing a connection between microscopy observations and materials behaviour over a broader parameter space is lacking. In this work, we introduce AtomAI, an open-source software package bridging instrument-specific Python libraries, deep learning and simulation tools into a single ecosystem. AtomAI allows direct applications of deep neural networks for atomic and mesoscopic image segmentation converting image and spectroscopy data into class-based local descriptors for downstream tasks such as statistical and graph analysis. For atomically resolved imaging data, the output is types and positions of atomic species, with an option for subsequent refinement. AtomAI further allows the implementation of a broad range of image and spectrum analysis functions, including invariant variational autoencoders for disentangling structural factors of variation and im2spec type of encoder–decoder models for mapping structure–property relationships. Finally, our framework allows seamless connection to the first principles modelling with a Python interface on the inferred atomic positions.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Improving nuclear data evaluations with predictive reaction theory and indirect measurements

Nuclear reaction data required for astrophysics and applications is incomplete, as not all nuclear reactions can be measured or reliably predicted. Neutron-induced reactions involving unstable targets are particularly challenging, but often critical for simulations. In response to this need, indirect approaches, such as the surrogate reaction method, have been developed. Nuclear theory is key to extract reliable cross sections from such indirect measurements. We describe ongoing efforts to expand the theoretical capabilities that enable surrogate reaction measurements. We focus on microscopic predictions for charged-particle inelastic scattering, uncertainty-quantified optical nucleon-nucleus models, and neural-network enhanced parameter inference.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2016

Fault detection and analysis in nuclear research facility using artificial intelligence methods

In this article, an online detection of transducer and actuator condition is discussed. A case study is on the reading of area radiation monitor (ARM) installed at the chimney of PUSPATI TRIGA nuclear reactor building, located at Bangi, Malaysia. There are at least five categories of abnormal ARM reading that could happen during the transducer failure, namely either the reading becomes very high, or very low/ zero, or with high fluctuation and noise. Moreover, the reading may be significantly higher or significantly lower as compared to the normal reading. An artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) are good methods for modeling this plant dynamics. The failure of equipment is based on ARM reading so it is then to compare with the estimated ARM data from ANN/ ANFIS function. The failure categories in either ‘yes’ or ‘no’ state are obtained from a comparison between the actual online data and the estimated output from ANN/ ANFIS function. It is found that this system design can correctly report the condition of ARM equipment in a simulated environment and later be implemented for online monitoring. This approach can also be extended to other transducers, such as the temperature profile of reactor core and also to include other critical actuator conditions such as the valves and pumps in the reactor facility provided that the failure symptoms are clearly defined.

42 ENGINEERING↗
DOE OSTI2022

Comprehensive characterization of irradiation induced defects in ceria: Impact of point defects on vibrational and optical properties

Validation of multiscale microstructure evolution models can be improved when standard microstructure characterization tools are coupled with methods sensitive to individual point defects. We demonstrate how electronic and vibrational properties of defects revealed by optical absorption and Raman spectroscopies can be used to compliment transmission electron microscopy (TEM) and x-ray diffraction (XRD) in the characterization of microstructure evolution in ceria under non-equilibrium conditions. Experimental manifestation of non-equilibrium conditions was realized by exposing cerium dioxide (CeO2) to energetic protons at elevated temperature. Two sintered polycrystalline CeO2 samples were bombarded with protons accelerated to a few MeVs. These irradiation conditions produced a microstructure with resolvable extended defects and a significant concentration of point defects. A rate theory (RT) model was parametrized using the results of TEM, XRD, and thermal conductivity measurements to infer point defect concentrations. An abundance of cerium sublattice defects suggested by the RT model is supported by Raman spectroscopy measurements, which show peak shift and broadening of the intrinsic T2g peak and emergence of new defect peaks. Additionally, spectroscopic ellipsometry measurements performed in lieu of optical absorption reveals the presence of Ce3+ ions associated with oxygen vacancies. This work lays the foundation for a coupled approach that considers a multimodal characterization of microstructures to guide and validate complex defect evolution models.

Chauhan, Vinay S.↗
DOE OSTI2014

Three wide planetary-mass companions to FW Tau, ROXs 12, and ROXs 42B

We report the discovery of three planetary-mass companions (M = 6-20 M {sub Jup}) in wide orbits (ρ ∼ 150-300 AU) around the young stars FW Tau (Taurus-Auriga), ROXs 12 (Ophiuchus), and ROXs 42B (Ophiuchus). All three wide planetary-mass companions (PMCs) were reported as candidate companions in previous binary survey programs, but then were neglected for >10 yr. We therefore obtained followup observations that demonstrate that each candidate is comoving with its host star. Based on the absolute M{sub K{sup ′}} magnitudes, we infer masses (from hot-start evolutionary models) and projected separations of 10 ± 4 M {sub Jup} and 330 ± 30 AU for FW Tau b, 16 ± 4 M {sub Jup} and 210 ± 20 AU for ROXs 12, and 10 ± 4 M {sub Jup} and 140 ± 10 AU for ROXs 42B b. We also present similar observations for 10 other candidates that show that they are unassociated field stars, as well as multicolor JHK'L' near-infrared photometry for our new PMCs and for five previously identified substellar or planetary-mass companions. The near-infrared photometry for our sample of eight known and new companions generally parallels the properties of free-floating, low-mass brown dwarfs in these star-forming regions. However, five of the seven objects with M < 30 M {sub Jup} are redder in K' – L' than the distribution of young free-floating counterparts of similar J – K' color. We speculate that this distinction could indicate a structural difference in circumplanetary disks, perhaps tied to higher disk mass since at least two of the objects in our sample are known to be accreting more vigorously than typical free-floating counterparts.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2014

New evidence for a substellar luminosity problem: Dynamical mass for the brown dwarf binary Gl 417BC

We present new evidence for a problem with cooling rates predicted by substellar evolutionary models that implies that model-derived masses in the literature for brown dwarfs and directly imaged planets may be too high. Based on our dynamical mass for Gl 417BC (L4.5+L6) and a gyrochronology system age from its young, solar-type host star, commonly used models predict luminosities 0.2-0.4 dex lower than we observe. This corroborates a similar luminosity-age discrepancy identified in our previous work on the L4+L4 binary HD 130948BC, which coincidentally has nearly identical component masses (≈50-55 M{sub Jup}) and age (≈800 Myr) as Gl 417BC. Such a luminosity offset would cause systematic errors of 15%-25% in model-derived masses at this age. After comparing different models, including cloudless models that should not be appropriate for mid-L dwarfs like Gl 417BC and HD 130948BC but actually match their luminosities better, we speculate the observed overluminosity could be caused by opacity holes (i.e., patchy clouds) in these objects. Moreover, from hybrid substellar evolutionary models that account for cloud disappearance, we infer the corresponding phase of overluminosity may extend from a few hundred million years up to a few gigayears and cause masses to be overestimated by up to 25%, even well after clouds disappear from view entirely. Thus, the range of ages and spectral types affected by this potential systematic shift in luminosity evolution would encompass most known directly imaged gas-giants and field brown dwarfs.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2014

BRIGHT HOT IMPACTS BY ERUPTED FRAGMENTS FALLING BACK ON THE SUN: UV REDSHIFTS IN STELLAR ACCRETION

A solar eruption after a flare on 2011 June 7 produced EUV-bright impacts of fallbacks far from the eruption site, observed with the Solar Dynamics Observatory. These impacts can be taken as a template for the impact of stellar accretion flows. Broad redshifted UV lines have been commonly observed in young accreting stars. Here we study the emission from the impacts in the Atmospheric Imaging Assembly's UV channels and compare the inferred velocity distribution to stellar observations. We model the impacts with two-dimensional hydrodynamic simulations. We find that the localized UV 1600 Å emission and its timing with respect to the EUV emission can be explained by the impact of a cloud of fragments. The first impacts produce strong initial upflows. The following fragments are hit and shocked by these upflows. The UV emission comes mostly from the shocked front shell of the fragments while they are still falling, and is therefore redshifted when observed from above. The EUV emission instead continues from the hot surface layer that is fed by the impacts. Fragmented accretion can therefore explain broad redshifted UV lines (e.g., C IV 1550 Å) to speeds around 400 km s{sup –1} observed in accreting young stellar objects.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

RF-Net 2: fast inference of virus reassortment and hybridization networks

Abstract Motivation A phylogenetic network is a powerful model to represent entangled evolutionary histories with both divergent (speciation) and convergent (e.g. hybridization, reassortment, recombination) evolution. The standard approach to inference of hybridization networks is to (i) reconstruct rooted gene trees and (ii) leverage gene tree discordance for network inference. Recently, we introduced a method called RF-Net for accurate inference of virus reassortment and hybridization networks from input gene trees in the presence of errors commonly found in phylogenetic trees. While RF-Net demonstrated the ability to accurately infer networks with up to four reticulations from erroneous input gene trees, its application was limited by the number of reticulations it could handle in a reasonable amount of time. This limitation is particularly restrictive in the inference of the evolutionary history of segmented RNA viruses such as influenza A virus (IAV), where reassortment is one of the major mechanisms shaping the evolution of these pathogens. Results Here, we expand the functionality of RF-Net that makes it significantly more applicable in practice. Crucially, we introduce a fast extension to RF-Net, called Fast-RF-Net, that can handle large numbers of reticulations without sacrificing accuracy. In addition, we develop automatic stopping criteria to select the appropriate number of reticulations heuristically and implement a feature for RF-Net to output error-corrected input gene trees. We then conduct a comprehensive study of the original method and its novel extensions and confirm their efficacy in practice using extensive simulation and empirical IAV evolutionary analyses. Availability and implementation RF-Net 2 is available at https://github.com/flu-crew/rf-net-2. Supplementary information Supplementary data are available at Bioinformatics online.

Markin, Alexey (ORCID:0000000332809050)↗
DOE OSTI2022

Why are we still using 3D masses for cluster cosmology?

ABSTRACT The abundance of clusters of galaxies is highly sensitive to the late-time evolution of the matter distribution, since clusters form at the highest density peaks. However, the 3D cluster mass cannot be inferred without deprojecting the observations, introducing model-dependent biases and uncertainties due to the mismatch between the assumed and the true cluster density profile and the neglected matter along the sightline. Since projected aperture masses can be measured directly in simulations and observationally through weak lensing, we argue that they are better suited for cluster cosmology. Using the Mira–Titan suite of gravity-only simulations, we show that aperture masses correlate strongly with 3D halo masses, albeit with large intrinsic scatter due to the varying matter distribution along the sightline. Nonetheless, aperture masses can be measured ≈2–3 times more precisely from observations, since they do not require assumptions about the density profile and are only affected by the shape noise in the weak lensing measurements. We emulate the cosmology dependence of the aperture mass function directly with a Gaussian process. Comparing the cosmology sensitivity of the aperture mass function and the 3D halo mass function for a fixed survey solid angle and redshift interval, we find the aperture mass sensitivity is higher for Ωm and $w_a$, similar for σ8, ns, and $w_0$, and slightly lower for h. With a carefully calibrated aperture mass function emulator, cluster cosmology analyses can use cluster aperture masses directly, reducing the sensitivity to model-dependent mass calibration biases and uncertainties.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

Dark Energy Survey year 1 results: the relationship between mass and light around cosmic voids

What are the mass and galaxy profiles of cosmic voids? In this paper, we use two methods to extract voids in the Dark Energy Survey (DES) Year 1 redMaGiC galaxy sample to address this question. We use either 2D slices in projection, or the 3D distribution of galaxies based on photometric redshifts to identify voids. For the mass profile, we measure the tangential shear profiles of background galaxies to infer the excess surface mass density. The signal-to-noise ratio for our lensing measurement ranges between 10.7 and 14.0 for the two void samples. We infer their 3D density profiles by fitting models based on N-body simulations and find good agreement for void radii in the range 15–85 Mpc. Comparison with their galaxy profiles then allows us to test the relation between mass and light at the 10 per cent level, the most stringent test to date. We find very similar shapes for the two profiles, consistent with a linear relationship between mass and light both within and outside the void radius. We validate our analysis with the help of simulated mock catalogues and estimate the impact of photometric redshift uncertainties on the measurement. Our methodology can be used for cosmological applications, including tests of gravity with voids. Finally, this is especially promising when the lensing profiles are combined with spectroscopic measurements of void dynamics via redshift-space distortions.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

New constraints on warm dark matter from the Lyman- α forest power spectrum

The forest of Lyman-α absorption lines detected in the spectra of distant quasars encodes information on the nature and properties of dark matter and the thermodynamics of diffuse baryonic material. Its main observable—the 1D flux power spectrum (FPS)—should exhibit a suppression on small scales and an enhancement on large scales in warm dark matter (WDM) cosmologies compared to standard Λ⁢CDM. Here, we present an unprecedented suite of 1080 high-resolution cosmological hydrodynamical simulations run with the graphics processing unit-accelerated code cholla to study the evolution of the Lyman-α forest under a wide range of physically motivated gas thermal histories along with different free-streaming lengths of WDM thermal relics in the early Universe. A statistical comparison of synthetic data with the forest FPS measured down to the smallest velocity scales ever probed at redshifts 4.0≲z≲5.2 [E. Boera et al., Revealing reionization with the thermal history of the intergalactic medium: New constraints from the Ly⁢α flux power spectrum, Astrophys. J. 872, 101 (2019)] yields a lower-limit m WDM >3.1 keV (95% C.L.) for the WDM particle mass and constrains the amplitude and spectrum of the photoheating and photoionizing background produced by star-forming galaxies and active galactic nuclei at these redshifts. Interestingly, our Bayesian inference analysis appears to weakly favor WDM models with a peak likelihood value at the thermal relic mass of m WDM =4.5 keV. In conclusion, we find that the suppression of the FPS from free-streaming saturates at k≳0.1 s km -1 because of peculiar velocity smearing, and this saturated suppression combined with a slightly lower gas temperature provides a moderately better fit to the observed small-scale FPS for WDM cosmologies.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Cognitive IoT and Edge Computing for Intrusion Detection with Federated TinyML

Internet of Things (IoT) and Edge Computing (EC) are rapidly becoming an integral part of the modern society. By 2030, there is estimated to be over 40 billion active and connected IoT devices [1]. This rapid progress also comes with a significant implication on cybersecurity. Back-end infrastructure and systems have a much broader attack than they did previously due to vulnerable IoT/EC devices being connected to wireless networks. This expanding attack surface is a growing concern because IoT/EC are increasingly being used in critical systems such as power grids, health care, and smart homes. To effectively address a problem of this scale, cognitive cyber methods—which can autonomously detect and react to cyber attacks as they develop—are needed. To address this, we bring Artificial Intelligence (AI) and Machine Learning (ML) to IoT/EC devices, using tinyML to monitor voluminous IoT data against cyber threats, and using Federated Learning (FL) to share local detection knowledge across the system while preserving privacy. We propose a novel three-layer architecture: (1) an IoT layer for tinyML-based inference, (2) an edge layer for ML model training, and (3) a cloud layer for FL operations. Using the publicly available 11-class N-BaIoT dataset [2], we demonstrate that this architecture mitigates resource constraints at the IoT layer while improving detection accuracy over standard two-layer designs. An outlier-resistant scaler, feature reduction, and quantization enable the tinyML model to maintain detection accuracy with a reduced model size. Additionally, federated learning that only utilizes the intersection (across heterogenous devices) of the reduced feature set achieves superior detection accuracy compared to locally trained models.

Li, Mingyan [ORNL] (ORCID:0009000569532640)↗
DOE OSTI2025

Self-Supervised T-GCN for Detection of Disturbance and Propagation in Power Grid

Urban power systems increasingly rely on dense sensing to monitor grid reliability, yet disturbance labels are scarce and events are rare. We present a self-supervised spatio-temporal method that detects, localizes, and characterizes grid frequency disturbances across urban areas using only unlabeled data. Our approach trains a tiny Temporal Graph Convolutional Network (T-GCN) to forecast per-site frequency residuals (deviation from 60 Hz). The sensor graph is constructed directly from signals using pre-event Pearson correlation with a cross-correlation lag penalty without geocoding. At inference, node-level anomalies are the model's forecast errors; region-level alarms arise from connected components of high-score nodes. We estimate disturbance propagation by computing per-node arrival times (first persistent exceedance), then fit a planar or time-of-arrival model to obtain direction, speed, and an epicenter proxy. With only three real events collected at decisecond resolution across U.S. cities, we evaluate the T-GCN and report time-to-detect, footprint size, and propagation consistency. We further show that short-window embeddings from the T-GCN's hidden states enable few-shot event-vs-background recognition via a simple prototypical classifier. Despite minimal data and no labels, our system yields fast, spatially coherent detection and interpretable propagation maps, offering a practical, lightweight pathway to city-scale grid resilience analytics.

Niu, Haoran [ORNL] (ORCID:0000000155228297)↗