MatFold: Cross-validation Protocols to Systematically Evaluate Generalization Errors in Materials Discovery Models
Explore the source record for details and available documents.
Schindler, Peter [Northeastern University]↗
SEARCH · Search NASA
Results for “Model generalization”
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.
A Simulator for Neyer Tests of Explosives
Explosives and explosive devices such as detonators are typically tested by applying a range of stimuli such as voltage or mechanical shock, and recording binary “detonated/did not detonate” responses. These are analyzed using maximum likelihood or generalized linear models to provide estimates of quantities such as the all-fire and no-fire points. Given that the true threshold for detonation is unknown a priori , sequential design methods are typically used to optimize the set of test points. One popular method, implemented in commercial software, is Neyer’s algorithm. To support simulation and experimental design, we have developed code in the R programming language to duplicate the functions of the Neyer software. We provide code for the simulator along with a description and examples of usage.
42 ENGINEERING↗
Barge Site - Avian Radar System / Derived Data
This is a combined data set of 67,410 bird/bat tracks from an avian radar system deployed on a research barge (MERLIN True3D, DeTect, Panama City, Florida, USA) and concurrent wind measurements from two scanning lidars (WindCube v2.1, Vaisala, Vantaa, Finland, and Halo XR+, Halo Photonics, Lannion, France). The research barge (16.5 m x 61 m) was deployed as part of the Wind Forecast Improvement Project (WFIP-3) off the northeast coast of the United States south of Massachusetts (40.9 deg N, 70.79 deg W). This data set comprises 5 weeks of data between August 27th 2024 and September 27th 2024. Radar data were provided by DeTect and Lidar data were accessed through the Wind Data Hub (wfip3/barg.WINDPROF.z01.a0) The data have been filtered and sorted into two size groups ("big" and "small") based on a clustering approach. See Snortland, A., Clerc, J., Hein, C., & Cotter, E. (2025). Wind as Driver of Bird and Bat Abundance, Flight Direction, Altitude, and Speed on the North Atlantic Shelf. arXiv preprint arXiv:2511.14983 for complete details. Data are provided in 2 files: "Birds" and "Birds_hourly" Birds: This file contains information about each of the 67,410 flying animal tracks detected by the radar during the data collection period, including parameters measured by the radar and wind information interpolated from the lidar wind measurements. We note that the raw radar dataset contained 301,618 tracks; tracks in this processed dataset were filtered based on the requirements described in Snortland et al. (2025). Birds_hourly: This file contains timeseries of the number of tracks detected per hour over the course of the data collection period, including wind conditions and sun position for each hour. These data were used for generalized additive modeling in Snortland et al. (2025).
17 WIND ENERGY↗
Principal Component Regression to Study the Impact of Economic Factors on Disadvantaged Communities
Our study employs principal component generalized linear models to comprehensively understand how economic factors such as income, employment status, and employment by industry affect the DAC status in Washington, USA.
Principal Component Regression, disadvantaged comm↗
Seroprevalence and risk factors for brucellosis amongst livestock and humans in a multi-herd ranch system in Kagera, Tanzania
Background Brucellosis remains a significant health and economic challenge for livestock and humans globally. Despite its public health implications, the factors driving the endemic persistence ofBrucellaat the human-livestock interface in Tanzania remain poorly elucidated. This study aimed to identify the seroprevalence ofBrucellainfection in livestock and humans within a ranching system and determine associated risk factors for disease endemicity. Methods A cross-sectional sero-epidemiological study was conducted in 2023 in Tanzania’s Karagwe District, involving 725 livestock (cattle, goats, sheep) from 10 herds and 112 humans from associated camps. Seroprevalence was assessed using competitive ELISA while epidemiological data were collected via questionnaires. Generalized Linear Models and Contrast Analysis were used to identify risk factors for infection. Results Overall seroprevalence was 34% in livestock and 41% in humans. Goats exhibited the highest prevalence (69.2%), while cattle had the lowest (22.6%). Mixed-species herds (Odds Ratio, OR = 2.96, CI [1.90–4.60]) and small ruminants-only herds (OR = 6.54, CI [3.65–11.72]) showed a significantly higher risk of seropositivity compared to cattle-only herds. Older cattle (OR = 5.23, CI [2.70–10.10]) and lactating females (OR = 2.87, CI [1.78–4.63]) represented significant risks for brucellosis in livestock. In humans, close contact with animals (OR = 7.20, CI [1.97–36.31]) and handling animals during parturition or aborted fetuses (OR = 2.37, CI [1.01–5.58]) were significant risk factors. Notably, no spatial association was found in seroprevalence between herds and nearby human communities. Conclusion The lack of spatial correlation between livestock and human seroprevalence suggests complex transmission dynamics, potentially involving endemic circulation in livestock and human infections from multiple sources of exposure to livestock. This study highlights the need for comprehensive zoonotic risk education and targeted intervention strategies. Further research is crucial to elucidate transmission pathways and improveBrucellainfection control. This includes developing robust methods for identifying infective species and implementing effective strategies to mitigateBrucellainfection in endemic regions.
Public, Environmental & Occupational Health↗
Anti-Ultralocality and Plateau Models of Inflation
Anti-ultralocality refers to the growth of spatial gradient terms relative to velocity terms in the coupled Einstein--scalar field equations. It is a characteristic feature of decelerated expansion before the onset of inflation. Previous numerical relativity studies have shown that anti-ultralocality prevents the onset of inflation in models with power-law inflaton potentials. In this paper, we show that models with plateau-shaped inflaton potentials, which are considered to be the simplest way to generate a tensor-to-scalar ratio below current observational upper limits, are especially vulnerable to anti-ultralocality effects. The reasons are the flatness of the plateau and the energy density gap of $\sim 10$ orders of magnitude between the Planck density and the plateau potential energy. To study the problem, we develop a protocol for assessing the viability of inflationary models in general, and we apply it to a plateau potential using a previously validated numerical relativity code. We find that, starting from generic initial conditions, the growth of gradient terms in the Einstein equations relative to non-gradient terms either prevents inflation from lasting for enough $e$-folds or triggers a phase of quantum runaway. We show that the fine-tuning of initial conditions necessary to avoid these issues becomes more severe as the energy scale of inflation is made smaller, disfavoring common approaches for reducing the tensor-to-scalar ratio.
FOS: Physical sciences↗
Towards Diverse and Representative Global Pretraining Datasets for Remote Sensing Foundation Models
The design of a pretraining dataset is emerging as a critical component for the generality of foundation models. In the remote sensing realm, large volumes of imagery and benchmark datasets exist that can be leveraged to pretrain foundation models, however using this imagery in absence of a well-crafted sampling strategy is inefficient and has the potential to create biased and less generalizable models. Here, we provide a discussion and vision for the curation and assessment of pretraining datasets for remote sensing geospatial foundation models. We highlight the importance of geographic, temporal, and image acquisition diversity and review possible strategies to enable such diversity at global scale. In addition to these characteristics, support for various spatial-temporal pretext tasks within the dataset is also critical. Ultimately, our primary objective is to place emphasis on and draw attention to the data curation stage of the foundation model development pipeline. By doing so, we think it is possible to reduce biases of geospatial foundation models, as well as enable broader generalization to downstream remote sensing tasks and applications.
Arndt, Jacob↗
Impact of mixed grain size sediment input on braided river Morphodynamics: Insights from experimental and numerical modeling
The rate of sediment supply has significant impacts on river morphology, making it crucial to understand the geomorphic changes and grain size distribution dynamics in rivers. However, the effects of varying grain size sediment input on morphological changes in braided channels remain poorly understood. This study is the first to investigate the bar development and sediment sorting processes in braided channels with non-uniform sediment inputs using both numerical and experimental approaches. We applied a two-dimensional numerical model, Nays2DH to confirm and generalize experimental results. The model reproduced key experiment results, including 1) stream elevation changes, and 2) grain size distribution. Using this validated model, we explored the morphological changes and sorting process in a braided river with sediment inputs. The numerical experiments demonstrate that sediment input controls the elevation of the stream bed and the grain size distribution. Notably, both the elevation and grain-size distribution become relatively stable in downstream of the channel. Additionally, the simulation results suggest that an increased sediment supply leads to greater channel complexity, with bed surface armoring decreasing.
Nays2DH↗
sf-24-103
pyzag is a library for efficiently training generic models defined with a recursive nonlinear function. We developed it primarily to provide an efficient means to train NEML2 models it can be used to train other, more general types of models.
Messner, Mark [Argonne National Laboratory (ANL), ↗
Deuterium and tritium anomalous transport in the generalized Hasegawa–Wakatani resistive drift wave turbulence model with finite ion Larmor radius
Anomalous transport of mixed deuterium–tritium plasma in the edge of magnetic fusion reactors is investigated using numerical solutions of resistive drift wave turbulence model equations, including finite Larmor radius effects, that are derived within the generalized Hasegawa–Wakatani framework. The anomalous cross field diffusivities of deuterium and tritium are compared in turbulence regimes with different values of the electron adiabaticity parameter controlling the existence of zonal flow. The dependence of the tritium-to-deuterium diffusivity ratio on the deuterium and tritium densities and the logarithmic density gradients is analyzed, and a scaling relation is obtained.
70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
Simulation of Channel Flow with Square Ribs for Blanket First-Wall Cooling: Geometry-Specific Tuning of k-ω Model Using Adjoint Method
Cooling of the plasma-facing first wall is challenging in the design of blanket components because of the high heat flux (on the order of 𝑀𝑊/𝑚2) from the plasma, especially when a low thermal mass medium like helium is chosen as the coolant. Therefore, heat transfer enhancement in which the convective heat transfer rate is augmented by the addition of turbulence-promoting structures becomes a key initiative for providing sufficient cooling capability with helium. Previously, computational fluid dynamics simulations had been performed on pipe flows with different transverse and longitudinal ribbed geometries at Oak Ridge National Laboratory to compare the enhancement performance among different ribbed geometries. Rib shape morphing had been conducted to obtain an optimized rib profile. In the work presented here, the adjoint method is adopted in the ANSYS Fluent solver for turbulence model augmentation, and the Generalized k-ω (GEKO) turbulence model is employed because of its ability of tuning the turbulence model. The Nusselt number and pressure drop obtained from the channel flow with bottom ribbed wall experiments are used as the targets. Sensitivity analysis provides information as guidance to improve the turbulence model accuracy. The augmented GEKO model is tuned for the studied ribbed channel geometry and flow conditions, providing improved predictive accuracy within this context. Extension to other configurations offers potential but may require additional tuning and validation.
Xu, Tracy [ORNL] (ORCID:0009000193700887)↗
Examining Cloud Feedback Components in the Simple Cloud-Resolving E3SM Atmosphere Model (SCREAM)
Cloud feedback remains the main source of uncertainty in climate sensitivity estimated by global climate models (GCMs), largely because subgrid cloud responses are parameterized in GCMs due to their coarse resolution. Here, this study examines cloud feedback in the global 3.25-km Simple Cloud-Resolving Energy Exascale Earth System Model (E3SM) Atmosphere Model (SCREAM 3 km) through a pair of 1-yr atmosphere-only simulations with control and +4-K sea surface temperature perturbations. SCREAM 3 km produces a positive cloud feedback that falls within but at the upper end of the range of Coupled Model Intercomparison Project phase 5 (CMIP5) and CMIP phase 6 (CMIP6) models and expert judgment. The positive cloud feedback arises from positive contributions from both high- and low-level clouds, with increases in high-cloud altitude and decreases in low-cloud amount and optical depth playing key roles. The stronger-than-CMIP-average feedback is mainly attributable to the high-cloud altitude feedback, owing to cloud tops rising nearly isothermally in SCREAM 3 km. The positive low-cloud amount feedback is weaker in SCREAM than in GCMs because estimated inversion strength (EIS) increases more dramatically with warming. A coarser 12-km resolution version of SCREAM exhibits a weaker positive cloud feedback than SCREAM 3 km, mainly because its low-cloud-radiative flux is more sensitive to EIS, leading to a stronger negative low-cloud amount feedback. With this process-level assessment of cloud feedback, this study reveals where SCREAM aligns with and diverges from conventional GCMs and expert assessment, providing insights to inform further model improvement and future expert assessment.
Cloud radiative effects↗
Developing a generalized ML-based NLTE spectral model for HED applications
Explore the source record for details and available documents.
Schaeuble, Marc-Andre↗
Interpretable Machine Learning for Characterizing Electric Vehicle Charging Behavior: Insights from Real-World Data
As electric vehicle (EV) adoption rises globally, concerns about the impact on aging electrical grids grow, particularly regarding the charging behavior of EV drivers. This study analyzes real-world driving and charging data from Ford battery electric vehicles (BEVs) collected between 2018 and 2019 to develop interpretable models that characterize charging behavior and quantify influencing factors. Prior research has relied on assumptions regarding driver behavior, often overlooking actual charging patterns. By employing generalized linear mixed models (GLMMs), this work offers insights into how various elements, such as next trip distance and state of charge (SOC), influence charging decisions. The dataset comprises over three million park-trip pairs from 1,997 vehicles, revealing that features related to driving behavior significantly dictate charging behavior, while infrastructure and regional factors have lesser impacts. The findings suggest that existing simulation models may oversimplify EV charging behavior assumptions. This work utilizes real-world EV driving and charging data to train interpretable models that describe charging behavior and quantify the factors most associated with how drivers use charging infrastructure. This research underscores the need for interpretable, data-driven methodologies to inform future EV infrastructure planning and grid management.
29 - ENERGY PLANNING, POLICY AND ECONOMY↗
Modeling and Simulation of Fuel Dispersal During the Loss-of-Coolant Accident
This document is the compilation of the milestone portion to a larger end of project NEUP report. The executive summary of the modeling portion is provided below: In the event of cladding rupture during a postulated LOCA in a pressurized water reactor, fuel particles, along with fission gases, can be expelled into the reactor core from the fractured fuel rod, a phenomenon referred to as fuel dispersal. The initial stage of fuel dispersal is strongly influenced by the high-pressure ejection of fuel fragments, the size and geometry of the ruptured cladding, and the depressurization history of the fuel rod during the postulated LOCA transient. Depending on the location of the burst orifice relative to the quench front, the dispersal event represents an intricate three-phase flow and heat transfer phenomenon, where high-temperature fuel particles carried by the fission gases interact with the coolant within the narrow subchannels of the fuel assemblies, inducing localized phase change. Given the unique multiphysics nature of this phenomena, the current study develops a dedicated computational framework to predict the mass distribution and cooling of dispersing fuel particles, facilitating post-accident assessment and management of the fuel assemblies. Considering the scale of nuclear reactor applications, a continuum three-fluid model is proposed for simulating the transport of solids within the reactor core. With high-temperature fuel fragments within the liquid media, nucleation sites inducing phase changes are dispersed within the flow domain. Coupled with the fact that the transient dispersal event occurs on different time scales than other three-phase flow applications, this study derives a time-averaged three-fluid flow model without losing generality. The assumptions regarding the continuum treatment of the solid phase and the modeling of fuel dispersal behavior are incorporated to simplify the governing equations and derive applicable closure relations. The computational validation of the model was conducted using adiabatic experimental results obtained from ongoing research at Oregon State University, focusing on characterizing fuel dispersal behavior during simulated LOCA conditions. Settlement characteristics of the solids, quantified by the probability distribution of equivalent particles, closely matched the probability density functions reported in experimental studies. The transport of fuel particles within a scaled 5 × 5 lattice of a pressurized-water reactor rod bundle geometry was modeled through a two-fluid Eulerian framework. The required boundary conditions were evaluated from the fuel performance code BISON in a postulated large-break LOCA scenario. The modeling framework considered solid fuel particles as granular matter, interacting with the gaseous dry steam phase and fission gases through the governing interfacial momentum exchange between the participating fluids. The simulation results provided the volume fraction of the solids obtained at the bottom surface of the enclosing tank geometry. Postulated LOCA leading to fuel dispersal phenomena involves the strong coupling between fuel thermomechanics, cladding deformation, thermal-hydraulics, and fuel particle transport. Incorporation of such a strong coupling in numerical simulation is performed by coupling the multiphysics solvers. In the case of fuel dispersal, a strong coupled simulation can be performed by coupling the BISON code for fuel performance, the TRACE code for system-level thermal hydraulics, and fuel particle transport in Multiphysics Object-Oriented Simulation Environment (MOOSE). For such intricate infrastructure, the MOOSE Framework eases the data transfer between codes. The recent version of MOOSE has incorporated the Navier-Stokes module for the fluid flow. An exploratory exercise was done to gain familiarity with finite volume capabilities in the MOOSE framework to incorporate the Spalart-Allmaras (SA) turbulence model. New finite-volume and auxiliary kernels were introduced to assemble the SA transport equation, compute turbulent viscosity, and evaluate wall distance and diagnostic turbulence terms, fully integrated with existing Navier-Stokes modules. A turbulent lid-driven cavity at a Reynolds number of approximately 10,000 is used for verification. MOOSE shows the robust solver convergence and produces the turbulent features. But it underpredicts the velocity profile and turbulent quantities, emphasizing the need to develop improved SA near-wall treatments (e.g., low-Re corrections or wall functions) as a key direction for future work.
11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
Boosted Dark Matter Search with the ICARUS Detector at the Gran Sasso Underground National Laboratory
The quest to discover dark matter is one of the most pressing topics in modern physics. Evidence of gravitational anomalies is present at different cosmological scales, spanning the stability and rotation of galaxies to the overall density profile of the entire observable Universe. Due to the discrepancy between the mass measured and the gravitational influence on the mass, if General Relativity is to correctly describe gravity at different scales, extra electromagnetically invisible matter termed dark matter must exist. In recent decades, many experiments have been performed to detect dark matter with masses that span many orders of magnitude, from Mass Compact Halo Objects (MACHOs) with masses several orders of magnitude that of our sun to very light particles on the order of $\sim\mu$eV. However, experiments have only been able to constrain the possible parameter space of dark matter models and, in general, probe dark matter models with single constituent dark matter like Weakly Interacting Massive Particles (WIMPs). In the Boosted Dark Matter (BDM) paradigm, there are four particles in the dark sector: $\chi_0$, $\chi_1$, $\chi_2$, and the dark photon $X$. WIMP-like properties are ascribed to $\chi_0$ to describe the physics of the gravitational anomalies, while relativistic components are assigned to $\chi_1$ and $\chi_2$. The inelastic boosted dark matter (iBDM) channel $\chi_1 e^-\rightarrow \chi_2 e^-\rightarrow \chi_1 Xe^-\rightarrow \chi_1 (e^+ e^- ) e^-$, where $\chi_2$ is an excited state of $\chi_1$, has a unique signature that is distinguishable from neutrino interactions, making neutrino experiments a viable search environment for BDM. Using data collected by the ICARUS collaboration with ICARUS T-600 Liquid Argon Time Projection Chamber (LArTPC) during its operation at the Gran Sasso Underground Laboratory (LNGS), the first-ever LArTPC dark matter search using the ICARUS detector is presented. With a total detector exposure of 0.13 kton$\cdot$year, a total of 4 iBDM candidate events were identified by several stages of data scanning. All were rejected due to defined selection criteria and a wire-by-wire $dE/dx$ evaluation, giving a null result. Exclusion plots at 90\%C.L. for the dark photon ($m_X$, $\epsilon$) parameter space for fixed ($m_0,m_1,m_2$) DM mass sets are presented, setting new limits on the dark photon visible decay parameter space.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
Urban morphology from a landscape perspective: How building morphology distribution land models (BMDLM) emulate pattern and process
Urban form (e.g., building morphology such as height or footprint) can be used to predict environmental footprints, such as energy/water consumption and carbon emissions. Although progress has been made in predicting building characteristics to fill gaps in observation or derive 3-D representations, the relationships between morphology and other variables such as land use and population are poorly understood. Understanding these relationships may enable projections for how cities will evolve with landscapes in the future. A suite of random forest models, the Building Morphology Distribution Land Models (BMDLM), was developed to determine how well building morphology for two distinct statistical measures (central tendency and frequency) can be predicted using land use (e.g., zoning) and population at different resolutions. Clark County, Nevada and Los Angeles County, California are explored as case studies. Generally, 1-km models outperformed 30-m models. Frequency distribution models had the best performance, especially in LA County. Frequency models significantly outperformed spatial autocorrelative models using inverse distance weighting (IDW). BMDLM offers a new take on modeling urban form in which generalized landscape patterns are characterized to understand the influence of population and zoning on urban development, as described by urban scaling theory.
Sturtevant, Jillian [Baylor Univ., Waco, TX (Unite↗
Generalizing synthetic data-trained acoustic predictive models to real-world measurements
Acoustic Resonance Spectroscopy (ARS) is highly sensitive to structural properties such as material, geometry, and environmental conditions; as a consequence, it can noninvasively measure internal properties that are unobservable by most other methods. Because of its sensing capabilities and low implementation cost and complexity, ARS has potential as a paradigm shift in noninvasive sensing, characterization, and monitoring applications. However, extracting specific properties from ARS measurements, comprising the vibration spectrum of a test object, is challenging due to the sensitivity of the spectra to other structural changes not being measured, e.g. manufacturing tolerances, component coupling, environmental variation, etc. Neural Networks are promising tools for identifying trends in ARS measurements, but their training typically requires large datasets, which are often impractical to obtain for real-world systems. Synthetic data can be simulated efficiently, but discrepancies between synthetic and real-world data frequently lead to poor generalization when testing on the real-world data. We propose a novel ARS model training framework that enables networks trained exclusively on synthetic ARS data to generalize effectively to real-world measurements. Our approach leverages the Correlation Alignment (CORAL) technique to enforce the extraction of features common to both synthetic and real-world domains. As a case study, we demonstrate noninvasive ARS-based pressure measurements in sealed systems. Finite element method (FEM) simulations were used to generate synthetic training data across diverse vessel configurations and pressure conditions, and model performance was then tested on real-world measurements. We demonstrate that robust machine learning models for ARS can be developed without large real-world datasets, significantly broadening the applicability of ARS for noninvasive sensing. Moreover, the approach is extensible to other sensing modalities where synthetic data are abundant but real-world data are limited.
36 MATERIALS SCIENCE↗