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 1,369 records · Page 76
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

Modeling Radiation Detection Using a Mixed-Effects Statistical Model for Background Variation

In this paper, we consider radiation source localization in an urban environment as an inverse problem and use Bayesian parameter estimation techniques to infer the source location and intensity. Since parameter estimation generally requires a large number of model evaluations, we employ a simplified photon transport model, as in a 2017 paper by Stefanescu, et al., to reduce computational requirements. However, the model in the 2017 paper does not account for the physical reality that background radiation varies with position within an urban locale. We incorporate statistical mixed-effects modeling into the simplified transport equation from the Stefanescu, et al. paper to reflect individual variation among the background radiation measurements at multiple detector locations. With our new model, we use a Bayesian DRAM-within-Gibbs technique, which incorporates the Delayed Rejection Adaptive Metropolis (DRAM) algorithm, to estimate the mixed-effects parameters. We note that without adequate prior information, we were unable to simultaneously estimate the set of parameters corresponding to the source intensity, source location, and individual background radiation counts. By calibrating the background in the absence of a source, we obtained narrow prior distributions, which permitted us to jointly infer the source intensity and location and the background radiation distribution.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2020

Data Assimilative Optimization of WSA Source Surface and Interface Radii using Particle Filtering

TheWang-Sheeley-Arge (WSA) model estimates solar wind speed and interplanetary magnetic field polarity in the inner heliosphere using global photospheric magnetic field maps. WSA employs the Potential Field Source Surface (PFSS) and Schatten Current Sheet (SCS) models to determine the Sun's global coronal magnetic field. The PFSS and SCS models are connected through two radial parameters, the source surface and interface radii, which specify the overlap region between the inner SCS and outer PFSS models. Though both radii values are adjustable, they have typically been fixed to 2.5 solar radii. Our work highlights how solar wind predictions improve when the radii are allowed to vary over time. Data assimilation using particle filtering (sequential Monte Carlo) is used to infer optimal values over a fixed time window. Solar wind model predictions and satellite observations are compared with a newly developed quality-of-agreement prediction metric. The agreement metric between the model and observations is assumed to correspond to the probability of the two key WSA model parameters, the source surface and interface radii, where the highest metric value implies the optimal radii.We find that the optimal particle filter values of solar radii can perform twice as well as standard values for an exploratory period during Carrington Rotation 1901, with these values also reducing nonphysical kinking effects seen in solar magnetic field lines. Data assimilation choices of input realization and time frame have implications for variation in the solar wind over time. We present this work's theoretical context and practical applications for prediction accuracy.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

NEW CONSTRAINTS ON THE ABUNDANCES OF SILICATE AND OXIDE STARDUST FROM SUPERNOVAE IN THE ACFER 094 METEORITE

We studied about 5000 μm{sup 2} of fine-grained matrix material in the Acfer 094 meteorite by high-resolution (nominal 50 nm) NanoSIMS ion imaging for the presence of O-rich presolar (stardust) grains. This approach permits identifying presolar grains down to <100 nm in size, compared to >150 nm in lower-resolution (nominal 100 nm) ion imaging surveys. The number density of identified presolar grains is a about a factor of two to three higher than what was found by lower-resolution ion imaging studies. The abundances of grains of O isotope Group 3 and 4 are higher than previously found. None of the presolar grains shows the strong enrichments in {sup 16}O expected from model predictions for the majority of supernova (SN) grains. Other potential O-rich SN grains, the Group 4 and some of the Group 3 grains, make up 33% by number and 19% by mass. This is clearly higher than the ∼10% (by number) inferred before and the 5% (by mass) estimated by a model for stellar dust in the interstellar medium. Our work shows that O-rich SN grains might be more abundant among the population of presolar grains in primitive solar system materials than currently thought, even without the {sup 16}O-rich grains as predominantly expected from SN models.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Vacuum energy density measured from cosmological data

Within the ΛCDM cosmological model, the absolute value of Einstein's cosmological constant Λ, sometimes expressed as the gravitating mass-energy density ρΛ of the physical vacuum, is a fundamental constant of nature, whose accurate measurement plays a central role in testing some proposed theories of quantum gravity. Several combinations of currently public cosmological data and an assumed flat ΛCDM cosmological model are used here to make a joint Bayesian inference on the combination of conventional parameters Ω Λ h 2 that corresponds to the absolute physical density ρΛ. In physical units, we obtain ρ Λ = (60.3 ± 1.3)× 10 -31 g/cm 3 , the most accurate constraint to date, with an absolute calibration of cosmological measurements based on CMB temperature. Significantly different values are obtained with calibrations that use a local distance scale, mainly connected to systematic differences in the value of the Hubble constant. Furthermore, it is suggested that future comprehensive cosmological parameter studies assuming the ΛCDM model include constraints on the vacuum density.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

A comparative analysis of YOLOv8 and U-Net image segmentation approaches for transmission electron micrographs of polycrystalline thin films

Metallic thin films offer a platform to experimentally study the dynamics of microstructural evolution, but the required transmission electron microscopy (TEM)-based imaging generates complex images that are challenging to segment and quantify. This work provides a comparative analysis of a new YOLOv8 model and an established U-Net model for bright-field TEM images of polycrystals, employing a framework leveraging physical observables to evaluate performance against two hand-traced benchmark datasets. This methodology obviates the comparison of large, diversely structured, and manually labeled datasets that are required to assess performance on a per-image/per-pixel basis. It is found that the YOLOv8 model, adapted for real-time instance segmentation, has up to 43× faster inferencing (NVIDIA GeForce RTX 4090) compared to U-Net and reconstructs hand-traced grain size distributions (GSDs) with excellent fidelity, finding mean diameter within 3% for grains near an optimal magnification; for grains that deviate from the optimal pixel-diameter, the size of small- (large)-diameter grains is systematically over- (under)-estimated. This is partially mitigated by including scale-aware augmentations during training. Moreover, when the bias is corrected post-inference by a rigid shift in distribution, the YOLOv8 model reproduces ground truth GSDs with exceptional fidelity, with statistical tests indicating <5% probability that the distributions are distinct. Based on ground truth data, calibration curves pertaining to this shift can be constructed for a given model. This issue is not present in the U-Net model’s results, indicating that for quantitative measurements where the true size of objects is of interest, special procedures must be implemented for YOLO-based models.

36 MATERIALS SCIENCE↗
DOE OSTI2018

Production of dark-matter bound states in the early universe by three-body recombination

The small-scale structure problems of the universe can be solved by self-interacting dark matter that becomes strongly interacting at low energy. A particularly predictive model for the self-interactions is resonant short-range interactions with an S-wave scattering length that is much larger than the range. The velocity dependence of the cross section in such a model provides an excellent fit to self-interaction cross sections inferred from dark-matter halos of galaxies and clusters of galaxies if the dark-matter mass is about 19 GeV and the scattering length is about 17 fm. Such a model makes definite predictions for the few-body physics of weakly bound clusters of the dark-matter particles. The formation of the two-body bound cluster is a bottleneck for the formation of larger bound clusters. We calculate the production of two-body bound clusters by three-body recombination in the early universe under the assumption that the dark matter particles are identical bosons, which is the most favorable case. If the dark-matter mass is 19 GeV and the scattering length is 17 fm, the fraction of dark matter in the form of two-body bound clusters can increase by as much as 4 orders of magnitude when the dark-matter temperature falls below the binding energy, but its present value remains less than 10 -6 . The present fraction can be increased to as large as 10 -3 by relaxing the constraints from small-scale structure and decreasing the mass of the dark matter particle.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

Inverse model based error detection in beamline optics

Optics tuning in transfer lines and LINACs can be challenging due to the fact that multiple combinations of machine settings can lead to the same diagnostic output. Moreover, the lack of a periodic solution can limit the ability to infer optics in the same way as rings from BPM signals. Model based approaches are often used to assist with the optics tuning in combination with optimization or parameter estimation. Here we have developed a novel approach using machine learning inverse models trained on a known configuration to detect variations in quadrupole settings without explicitly including them in the model. This paper shows a comparison of neural network models and linear models on both a simulation based study and experimental studies conducted at the AGS to RHIC transfer line at Brookhaven National Lab.

43 PARTICLE ACCELERATORS↗
DOE OSTI2026

Toward an AI-Powered Software Pipeline for Real-Time Tracking and Analysis of Wildfire and Smoke

Real-time tracking of wildfires and smoke is crucial for effective response, minimizing damage, protecting lives, and efficiently managing resources during fire emergencies. We develop a web-based AI-powered pipeline that detects wildfires in aerial video and estimates deployment-relevant behavior metrics, including cumulative burned area, burned-area growth rate, fire spread direction, and smoke dispersion. The system combines a YOLO-based detector with YCbCr-based fire segmentation, HSV-based smoke segmentation, Farneback optical flow, and centroid-based spatiotemporal tracking. Using ground sampling distance (GSD), pixel-level fire masks are converted to physical burned-area measurements by correlating fire pixel counts with camera altitude and tilt angle. We benchmark YOLO variants and non-YOLO baselines (GoogLeNet, CNN, DBN, Autoencoder, U-Net, and AlexNet) on the IEEE FLAME dataset and a newly created aerial frame dataset, Wildfire-DB. Cross-dataset evaluation uses a strict threshold-transfer protocol: decision thresholds are selected on FLAME validation and transferred unchanged to Wildfire-DB to quantify generalization under domain shift. YOLOv6 achieves the strongest cross-dataset frame-level fire detection on Wildfire-DB (ROC-AUC 0.8200, PR-AUC 0.8044, and transferred-threshold F1 0.7596). For tracking-oriented deployment requiring oriented localization, YOLO11-OBB provides the most reliable cross-dataset behavior among OBB-capable models while remaining computationally feasible. To analyze the feasibility of UAV deployment, we further measure inference efficiency using synchronized GPU and CPU power logs on a fixed workload of 1569 frames. YOLO-family models process the video in 5.73–12.47 seconds with net energy of 1247.28–1775.39 J, substantially lower latency and energy than heavier classification and reconstruction baselines. Overall, model optimality depends on operational objectives: YOLOv6 is best for cross-dataset detection robustness, whereas YOL...

Color segmentation↗
DOE OSTI2024

Uncertainty quantification of graph convolution neural network models of evolving processes

The application of neural network models to scientific machine learning tasks has proliferated in recent years. In particular, neural networks have proved to be adept at modeling processes with spatial–temporal complexity. Nevertheless, these highly parameterized models have garnered skepticism in their ability to produce outputs with quantified error bounds over the regimes of interest. Hence there is a need to find uncertainty quantification methods that are suitable for neural networks. In this work we present comparisons of the parametric uncertainty quantification of neural networks modeling complex spatial–temporal processes with Hamiltonian Monte Carlo and Stein variational gradient descent and its projected variant. Specifically we apply these methods to graph convolutional neural network models of evolving systems modeled with recurrent neural network and neural ordinary differential equations architectures. We show that Stein variational inference is a viable alternative to Monte Carlo methods with some clear advantages for complex neural network models. For our exemplars, Stein variational interference gave similar pushed forward uncertainty profiles through time compared to Hamiltonian Monte Carlo, albeit with generally more generous variance. As a result, projected Stein variational gradient descent also produced similar uncertainty profiles to the non-projected counterpart, but large reductions in the active weight space were confounded by the stability of the neural network predictions and the convoluted likelihood landscape.

36 MATERIALS SCIENCE↗
DOE OSTI2014

Purely dry mergers do not explain the observed evolution of massive early-type galaxies since z ∼ 1

Several studies have suggested that the observed size evolution of massive early-type galaxies (ETGs) can be explained as a combination of dry mergers and progenitor bias, at least since z ∼ 1. In this paper we carry out a new test of the dry-merger scenario based on recent lensing measurements of the evolution of the mass density profile of ETGs. We construct a theoretical model for the joint evolution of the size and mass density profile slope γ' driven by dry mergers occurring at rates given by cosmological simulations. Such dry-merger model predicts a strong decrease of γ' with cosmic time, inconsistent with the almost constant γ' inferred from observations in the redshift range 0 < z < 1. We then show with a simple toy model that a modest amount of cold gas in the mergers—consistent with the upper limits on recent star formation in ETGs—is sufficient to reconcile the model with measurements of γ'. By fitting for the amount of gas accreted during mergers, we find that models with dissipation are consistent with observations of the evolution in both size and density slope, if ∼4% of the total final stellar mass arises from the gas accreted since z ∼ 1. Purely dry merger models are ruled out at >99% CL. We thus suggest a scenario where the outer regions of massive ETGs grow by accretion of stars and dark matter, while small amounts of dissipation and nuclear star formation conspire to keep the mass density profile constant and approximately isothermal.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2013

Decay-phase cooling and inferred heating of M- and X-class solar flares

In this paper, the cooling of 72 M- and X-class flares is examined using GOES/XRS and SDO/EVE. The observed cooling rates are quantified and the observed total cooling times are compared with the predictions of an analytical zero-dimensional hydrodynamic model. We find that the model does not fit the observations well, but does provide a well-defined lower limit on a flare's total cooling time. The discrepancy between observations and the model is then assumed to be primarily due to heating during the decay phase. The decay-phase heating necessary to account for the discrepancy is quantified and found be ∼50% of the total thermally radiated energy, as calculated with GOES. This decay-phase heating is found to scale with the observed peak thermal energy. It is predicted that approximating the total thermal energy from the peak is minimally affected by the decay-phase heating in small flares. However, in the most energetic flares the decay-phase heating inferred from the model can be several times greater than the peak thermal energy.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Probabilistic Modeling of Commercial Building Occupancy Patterns Using Location-Based Map Data: Preprint

Considering occupancy patterns is crucial to simulate buildings' energy use. Current energy models use inputs that simplify the actual diversity in occupancy into static occupancy patterns and are not able to represent the numerous variations in occupancy patterns between buildings and across different locations. Recently, inferring occupancy schedules from metered electricity consumption data was used to model occupancy in commercial buildings. However, the translation from metered data to occupancy schedules requires many assumptions that might not capture the reality, and the process is hindered by the availability of data from advanced metering infrastructure. With the development of information technologies, occupancy modeling should not be limited to traditional approaches. The prevalence of social networks and location services with real-time user feedback provides publicly accessible data via Maps Application Programming Interfaces (APIs) such as Google Maps, SafeGraph, Mapbox, Foursquare, etc. This paper presents an automated framework for modeling parametric occupancy patterns using such APIs to calibrate commercial district buildings' energy models. This process includes three main steps: data extraction and processing, parametric schedules generation, and schedules integration. We demonstrated this framework in districts where we used maps API to generate more accurate behavioral patterns for operations and electric vehicle charging events. We used these patterns to determine differences in energy use across key sociodemographic and spatial parameters. The presented method has the potential for worldwide applications. Users can utilize this framework to extract data for selected locations of interest to create more realistic behavioral patterns for commercial facilities across different districts.

building energy modeling↗
DOE OSTI2018

Soil moisture estimation using tomographic ground penetrating radar in a MCMC–Bayesian framework

In this paper, we focus on a hydrogeological inverse problem specifically targeting monitoring soil moisture variations using tomographic ground penetrating radar (GPR) travel time data. Technical challenges exist in the inversion of GPR tomographic data for handling non-uniqueness, nonlinearity and high-dimensionality of unknowns. We have developed a new method for estimating soil moisture fields from crosshole GPR data. It uses a pilot-point method to provide a low-dimensional representation of the relative dielectric permittivity field of the soil, which is the primary object of inference: the field can be converted to soil moisture using a petrophysical model. We integrate a multi-chain Markov chain Monte Carlo (MCMC)–Bayesian inversion framework with the pilot point concept, a curved-ray GPR travel time model, and a sequential Gaussian simulation algorithm, for estimating the dielectric permittivity at pilot point locations distributed within the tomogram, as well as the corresponding geostatistical parameters (i.e., spatial correlation range). We infer the dielectric permittivity as a probability density function, thus capturing the uncertainty in the inference. The multi-chain MCMC enables addressing high-dimensional inverse problems as required in the inversion setup. The method is scalable in terms of number of chains and processors, and is useful for computationally demanding Bayesian model calibration in scientific and engineering problems. The proposed inversion approach can successfully approximate the posterior density distributions of the pilot points, and capture the true values. The computational efficiency, accuracy, and convergence behaviors of the inversion approach were also systematically evaluated, by comparing the inversion results obtained with different levels of noises in the observations, increased observational data, as well as increased number of pilot points.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Data-driven prediction of scaling and ignition of inertial confinement fusion experiments

Recent advances in inertial confinement fusion (ICF) at the National Ignition Facility (NIF), including ignition and energy gain, are enabled by a close coupling between experiments and high-fidelity simulations. Neither simulations nor experiments can fully constrain the behavior of ICF implosions on their own, meaning pre- and postshot simulation studies must incorporate experimental data to be reliable. Linking past data with simulations to make predictions for upcoming designs and quantifying the uncertainty in those predictions has been an ongoing challenge in ICF research. We have developed a data-driven approach to prediction and uncertainty quantification that combines large ensembles of simulations with Bayesian inference and deep learning. The approach builds a predictive model for the statistical distribution of key performance parameters, which is jointly informed by past experiments and physics simulations. The prediction distribution captures the impact of experimental uncertainty, expert priors, design changes, and shot-to-shot variations. We have used this new capability to predict a 10× increase in ignition probability between Hybrid-E shots driven with 2.05 MJ compared to 1.9 MJ, and validated our predictions against subsequent experiments. We describe our new Bayesian postshot and prediction capabilities, discuss their application to NIF ignition and validate the results, and finally investigate the impact of data sparsity on our prediction results.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Analysis of attenuation data from the decommissioned ZIon unit 1 reactor pressure vessel beltline weld

In order to examine the attenuation of radiation damage through the thickness of an irradiated reactor pressure vessel (RPV), four segments were acquired from the Zion Unit 1 power plant RPV after the plant was decommissioned. The Zion Unit 1 RPV Beltline Weld Segment 1 was cut into seven blocks, consisting of five base metal and two beltline welds from the high fluence region of the segment. Through-wall test specimens were machined and tested. Specimens included those used for Charpy impact, Master Curve fracture toughness testing, and chemical analysis. The observed through-thickness ductile-to-brittle transition temperatures in the beltline weld deviated significantly from the expected behavior based on the attenuation of fast fluence as a function of depth into the RPV. Beginning at the inside surface, the 41-J Charpy transition temperature was either flat or slightly increasing until the ¾ -T location. The results of a simple, model-based analysis of the Zion beltline weld material that included the irradiation conditions and material chemistry were generally consistent with industry trend curves and the standard attenuation model, rather than the observed data. Although there was no archive material from the RPV available to permit measurement of the unirradiated properties, fracture toughness specimens fabricated from archive surveillance weld were used to obtain an estimate of the initial through-thickness values of the Charpy transition temperature. The Charpy shifts obtained using this approach were similarly in disagreement with the predictions of the US NRC Regulatory Guide 1.99, Rev. 2. However, testing of irradiated Charpy specimens taken from the RPV following post-irradiation annealing (10 hr. at 500 °C) provided a quite different estimate of the unirradiated properties which improved the agreement between the inferred through-thickness Charpy shifts and exponential attenuation model included in Regulatory Guide 1.99/2. In conclusion, the analysis of the Zion data and data obtained in previous post-mortem examinations of decommissioned RPVs indicates that more work is needed to understand the through-thickness properties of RPV materials in order to properly assess through-wall damage attenuation.

Charpy impact↗
DOE OSTI2026

How to Build an Empirical Speed Distribution for Dark Matter in the Solar Neighborhood

The dark matter flux in a direct detection experiment depends on its local speed distribution. This distribution has been inferred from simulations of Milky Way–like galaxies, but such models serve only as proxies, given that no simulation directly captures the detailed evolution of our own Galaxy. This motivates alternative approaches that obtain this distribution directly from observations. In this work, we utilize 98 Milky Way analogues from the TNG50 simulation to develop and validate a procedure for inferring the dark matter speed distribution using the kinematics of nearby stars. We find that the dark matter that originated from old mergers, plus that from recent nonluminous accretions, is well described by a Maxwell–Boltzmann speed distribution centered at the local standard-of-rest velocity. Meanwhile, recently accreted dark matter from massive mergers has speeds that can be traced from the associated stellar debris of these events. The stellar populations systematically underestimate the velocity dispersion of their dark matter counterparts, but a simple kinematic boost brings the two into good alignment. Using the TNG50 host galaxies, we demonstrate that combining these two contributions provides an accurate reconstruction of the local dark matter speeds. As an application of the procedure to our own Galaxy, we utilize stellar kinematic data from Gaia to quantify how the dark matter remnants from the Milky Way’s last major merger impact its speed distribution in the solar neighborhood.

Shpigel, Tal [Princeton Univ., NJ (United States)]↗
DOE OSTI2014

TESTING MAGNETIC FIELD MODELS FOR THE CLASS 0 PROTOSTAR L1527

For the Class 0 protostar L1527 we compare 131 polarization vectors from SCUPOL/JCMT, SHARP/CSO, and TADPOL/CARMA observations with the corresponding model polarization vectors of four ideal-MHD, nonturbulent, cloud core collapse models. These four models differ by their initial magnetic fields before collapse; two initially have aligned fields (strong and weak) and two initially have orthogonal fields (strong and weak) with respect to the rotation axis of the L1527 core. Only the initial weak orthogonal field model produces the observed circumstellar disk within L1527. This is a characteristic of nearly all ideal-MHD, nonturbulent, core collapse models. In this paper we test whether this weak orthogonal model also has the best agreement between its magnetic field structure and that inferred from the polarimetry observations of L1527. We found that this is not the case; based on the polarimetry observations, the most favored model of the four is the weak aligned model. However, this model does not produce a circumstellar disk, so our result implies that a nonturbulent, ideal-MHD global collapse model probably does not represent the core collapse that has occurred in L1527. Our study also illustrates the importance of using polarization vectors covering a large area of a cloud core to determine the initial magnetic field orientation before collapse; the inner core magnetic field structure can be highly altered by a collapse, and so measurements from this region alone can give unreliable estimates of the initial field configuration before collapse.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

Parameter splitting in dark energy: is dark energy the same in the background and in the cosmic structures?

We perform an empirical consistency test of General Relativity/dark energy by disentangling expansion history and growth of structure constraints. We replace each late-universe parameter that describes the behavior of dark energy with two meta-parameters: one describing geometrical information in cosmological probes, and the other controlling the growth of structure. If the underlying model (a standard wCDM cosmology with General Relativity) is correct, that is under the null hypothesis, the two meta-parameters coincide. If they do not, it could indicate a failure of the model or systematics in the data. We present a global analysis using state-of-the-art cosmological data sets which points in the direction that cosmic structures prefer a weaker growth than that inferred by background probes. This result could signify inconsistencies of the model, the necessity of extensions to it or the presence of systematic errors in the data. We examine all these possibilities. The fact that the result is mostly driven by a specific sub-set of galaxy clusters abundance data, points to the need of a better understanding of this probe.

79 ASTRONOMY AND ASTROPHYSICS↗