SEARCH · Engineering Papers

Results for “ensemble methods”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 397 records · Page 22

All sources NASA NTRS DOE OSTI

DOE OSTI2020

Probabilistic predictions and uncertainty estimation for radiological and nuclear effects modeling

Better characterization of meteorological uncertainty is needed to predict the effects of radiological and nuclear releases in the atmosphere. Multi-physics ensembles of atmospheric simulations can estimate uncertainty of physical processes, but they are often limited for computational reasons. Because weather uncertainty is important for emergency response applications, but the conventional way of running multi-physics ensembles is infeasible, we have developed new machine learning-based methods that quickly estimate and identify key sources of meteorological uncertainty. Machine learning accelerates the analysis of weather uncertainty by iteratively learning about model physics options that affect plume predictions. We have demonstrated the power of machine learning by creating a large multi-physics weather and dispersion ensemble dataset containing 1,200 simulations for two hypothetical radiological release scenarios. By training on less than 10% of the ensemble, we can quickly and accurately predict different outcomes, including the mass of radiological material deposited and spatial distribution and extent of contaminated areas.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRS1996

Numerical Investigation of Two-Phase Flows With Charged Droplets in Electrostatic Field

A numerical method to solve two-phase turbulent flows with charged droplets in an electrostatic field is presented. The ensemble-averaged Navier-Stokes equations and the electrostatic potential equation are solved using a finite volume method. The transitional turbulence field is described using multiple-time-scale turbulence equations. The equations of motion of droplets are solved using a Lagrangian particle tracking scheme, and the inter-phase momentum exchange is described by the Particle-In-Cell scheme. The electrostatic force caused by an applied electrical potential is calculated using the electrostatic field obtained by solving a Laplacian equation and the force exerted by charged droplets is calculated using the Coulombic force equation. The method is applied to solve electro-hydrodynamic sprays. The calculated droplet velocity distributions for droplet dispersions occurring in a stagnant surrounding are in good agreement with the measured data. For droplet dispersions occurring in a two-phase flow, the droplet trajectories are influenced by aerodynamic forces, the Coulombic force, and the applied electrostatic potential field.

Kim, Sang-Wook↗

DOE OSTI2022

Predicting Oxygen Off-Stoichiometry and Hydrogen Incorporation in Complex Perovskite Oxides

Chemically and structurally complex solid compounds, including those with significant off-stoichiometry, are rapidly extending new material functionality across a variety of applications. Accelerated development of these compounds requires accurate predictions of material defect properties including effective defect formation energies and equilibrium defect concentrations. Traditional first-principles approaches typically examine dilute defect concentrations and relatively ordered atomic structures to identify the lowest energy defect sites. These approaches are rarely suitable for describing the disorder present in these systems and its influence on defect formation, which can lead to unphysically large predictions for defect concentrations. Here, we demonstrate a new method to accurately predict the temperature and pressure dependence of oxygen vacancy concentrations and proton interstitial concentrations in complex oxides. This method extends standard dilute defect calculations to incorporate atomic and magnetic disorder, employs the ensemble descriptions of defect sites resulting in improved predictions of defect formation energies, and accounts for effects beyond the dilute defect limit. To demonstrate our method, we show that the predicted defect concentrations in perovskites used as ceramic fuel cell cathodes, including Ba0.5Sr0.5Fe0.8Zn0.2O3-d, Ba0.5Sr0.5Co0.8Fe0.2O3-d, and BaCo1-x-y-zFexZryYzO3-d, are in good agreement with experimental values, thereby opening the door for predictive design of complex oxides by these applications.

DFT↗

DOE OSTI2024

STSR-INR: Spatiotemporal super-resolution for multivariate time-varying volumetric data via implicit neural representation

Implicit neural representation (INR) has surfaced as a promising direction for solving different scientific visualization tasks due to its continuous representation and flexible input and output settings. We present STSR-INR, an INR solution for generating simultaneous spatiotemporal super-resolution for multivariate time-varying volumetric data. Inheriting the benefits of the INR-based approach, STSR-INR supports unsupervised learning and permits data upscaling with arbitrary spatial and temporal scale factors. Unlike existing GAN- or INR-based super-resolution methods, STSR-INR focuses on tackling variables or ensembles and enabling joint training across datasets of various spatiotemporal resolutions. Here we achieve this capability via a variable embedding scheme that learns latent vectors for different variables. In conjunction with a modulated structure in the network design, we employ a variational auto-decoder to optimize the learnable latent vectors to enable latent-space interpolation. To combat the slow training of INR, we leverage a multi-head strategy to improve training and inference speed with significant speedup. We demonstrate the effectiveness of STSR-INR with multiple scalar field datasets and compare it with conventional tricubic+linear interpolation and state-of-the-art deep-learning-based solutions (STNet and CoordNet).

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Data Imbalance, Uncertainty Quantification, and Transfer Learning in Data‐Driven Parameterizations: Lessons From the Emulation of Gravity Wave Momentum Transport in WACCM

Abstract Neural networks (NNs) are increasingly used for data‐driven subgrid‐scale parameterizations in weather and climate models. While NNs are powerful tools for learning complex non‐linear relationships from data, there are several challenges in using them for parameterizations. Three of these challenges are (a) data imbalance related to learning rare, often large‐amplitude, samples; (b) uncertainty quantification (UQ) of the predictions to provide an accuracy indicator; and (c) generalization to other climates, for example, those with different radiative forcings. Here, we examine the performance of methods for addressing these challenges using NN‐based emulators of the Whole Atmosphere Community Climate Model (WACCM) physics‐based gravity wave (GW) parameterizations as a test case. WACCM has complex, state‐of‐the‐art parameterizations for orography‐, convection‐, and front‐driven GWs. Convection‐ and orography‐driven GWs have significant data imbalance due to the absence of convection or orography in most grid points. We address data imbalance using resampling and/or weighted loss functions, enabling the successful emulation of parameterizations for all three sources. We demonstrate that three UQ methods (Bayesian NNs, variational auto‐encoders, and dropouts) provide ensemble spreads that correspond to accuracy during testing, offering criteria for identifying when an NN gives inaccurate predictions. Finally, we show that the accuracy of these NNs decreases for a warmer climate (4 × CO 2 ). However, their performance is significantly improved by applying transfer learning, for example, re‐training only one layer using ∼1% new data from the warmer climate. The findings of this study offer insights for developing reliable and generalizable data‐driven parameterizations for various processes, including (but not limited to) GWs.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

Grafting nanometer metal/oxide interface towards enhanced low-temperature acetylene semi-hydrogenation

Metal/oxide interface is of fundamental significance to heterogeneous catalysis because the seemingly “inert” oxide support can modulate the morphology, atomic and electronic structures of the metal catalyst through the interface. The interfacial effects are well studied over a bulk oxide support but remain elusive for nanometer-sized systems like clusters, arising from the challenges associated with chemical synthesis and structural elucidation of such hybrid clusters. We hereby demonstrate the essential catalytic roles of a nanometer metal/oxide interface constructed by a hybrid Pd/Bi 2 O 3 cluster ensemble, which is fabricated by a facile stepwise photochemical method. The Pd/Bi 2 O 3 cluster, of which the hybrid structure is elucidated by combined electron microscopy and microanalysis, features a small Pd-Pd coordination number and more importantly a Pd-Bi spatial correlation ascribed to the heterografting between Pd and Bi terminated Bi 2 O 3 clusters. The intra-cluster electron transfer towards Pd across the as-formed nanometer metal/oxide interface significantly weakens the ethylene adsorption without compromising the hydrogen activation. As a result, a 91% selectivity of ethylene and 90% conversion of acetylene can be achieved in a front-end hydrogenation process with a temperature as low as 44 °C.

36 MATERIALS SCIENCE↗

DOE OSTI2026

Edge Reconstruction in a Quantum Spin Hall Insulator

We study interaction-driven edge reconstruction in a quantum spin Hall insulator described by the Bernevig-Hughes-Zhang model with Kanamori-Hubbard interactions using the real-space density matrix renormalization group method in both the grand-canonical and canonical ensembles. For a two-dimensional cylinder with a smooth edge, we identify discrete particle-number transitions that lead to a spin-polarized edge state stabilized by an emergent ferromagnetic exchange interaction. The reconstruction is orbital-selective, occurring predominantly in the 𝑠-orbital channel. Our results reveal a microscopic mechanism for emergent fluctuating moments at the edge that could compromise the topological protection of helical edge states by time-reversal symmetry.

Soni, Rahul [ORNL] (ORCID:0000000317714299)↗

DOE OSTI2025

MCCCS-MN

The MCCCS‒MN (Monte Carlo for Complex Chemical Systems‒Minnesota) software is developed by the Siepmann research group at the University of Minnesota. MCCCS‒MN allows for the simulation of multi-component molecular systems in the canonical, isobaric-isothermal (including constant stress for solids), grand-canonical, semi-grand, and Gibbs (NVT, NPT, and more than two simulation boxes) ensembles. It uses the configurational-bias Monte Carlo method to efficiently sample phase space for linear, branched and cyclic chain molecules, the adiabatic nuclear and electronic sampling Monte Carlo method to treat many-body polarization effects, and the aggregation-volume-bias Monte Carlo algorithm to efficiently sample the spatial distribution of associating molecules. MCCCS-MN employs a molecular representation of the system where force fields contain bonded and non-bonded terms. Funding for the development of MCCCS-MN through grants from the National Science Foundation (simulation of fluid phase equilibria and chromatography) and the Department of Energy (simulation of adsorption equilibria) is gratefully acknowledged.

Siepmann, J.Ilja [University of Minnesota - Twin C↗

NASA NTRS1980

Energy conservation - A test for scattering approximations

The roles of the extinction theorem and energy conservation in obtaining the scattering and absorption cross sections for several light scattering approximations are explored. It is shown that the Rayleigh, Rayleigh-Gans, anomalous diffraction, geometrical optics, and Shifrin approximations all lead to reasonable values of the cross sections, while the modified Mie approximation does not. Further examination of the modified Mie approximation for the ensembles of nonspherical particles reveals additional problems with that method.

Acquista, C.↗

NASA NTRS1991

Chaotic jumps in the generalized first adiabatic invariant in current sheets

The present study examines how the changes in the generalized first adiabatic invariant J derived from the separatrix crossing theory can be incorporated into the drift variable approach to generating distribution functions. A method is proposed for determining distribution functions for an ensemble of particles following interaction with the tail current sheet by treating the interaction as a scattering problem characterized by changes in the invariant. Generalized drift velocities are obtained for a 1D tail configuration by using the generalized first invariant. The invariant remained constant except for the discrete changes caused by chaotic scattering as the particles cross the separatrix.

Brittnacher, M. J.↗

DOE OSTI2020

Conservative Estimation of Tail Probabilities from Limited Sample Data

Several sparse-sample uncertainty quantification (UQ) methods are compared for conservative but not overly conservative estimation of small tail probabilities involving responses that lay beyond specified thresholds in the tails of probability distributions. Sixteen very differently shaped distributions (or probability density functions, PDFs) and tail probability magnitudes ranging from 10 -5 to 10 -1 are considered in order for the study to be relevant to a wide range of risk analysis and quantification of margins and uncertainty (QMU) problems. The emphasis of the study is on limited data regimes ranging from N = 2 to 20 samples, reflective of most experimental and some expensive computational situations. Relatively simple sparse-sample UQ methods tested for this regime involve statistical tolerance interval "Equivalent Normal and related "Ensemble of Normals" and "Superdistribution (SD) approaches. (The independently derived SD is effectively equivalent to the Bayesian posterior predictive distribution given the assumptions of the derivation.) The performance of the methods was generally improved for N ≥ 5 samples with a generalized Jackknife resampling technique, which determines a tail probability estimate by averaging estimates from smaller sub-samples. Several quantitative metrics for method conservatism and accuracy of tail probability estimation are used to assess and rank the methods' performance over many random trials for each test PDF and probability magnitude. A variant of Bootstrap resampling was also tried, but did not significantly improve tail probability estimates in most cases. Detailed results are presented from over 100-million tests over the above factors that provide useful granular information on which methods or combination of methods perform best in various areas of the factor space.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Data Science and Machine Learning for Genome Security

This report describes research conducted to use data science and machine learning methods to distinguish targeted genome editing versus natural mutation and sequencer machine noise. Genome editing capabilities have been around for more than 20 years, and the efficiencies of these techniques has improved dramatically in the last 5+ years, notably with the rise of CRISPR-Cas technology. Whether or not a specific genome has been the target of an edit is concern for U.S. national security. The research detailed in this report provides first steps to address this concern. A large amount of data is necessary in our research, thus we invested considerable time collecting and processing it. We use an ensemble of decision tree and deep neural network machine learning methods as well as anomaly detection to detect genome edits given either whole exome or genome DNA reads. The edit detection results we obtained with our algorithms tested against samples held out during training of our methods are significantly better than random guessing, achieving high F1 and recall scores as well as with precision overall.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2022

Gradient-Based Novelty Detection Boosted by Self-Supervised Binary Classification

Novelty detection aims to automatically identify out-of-distribution (OOD) data, without any prior knowledge of them. It is a critical step in data monitoring, behavior analysis and other applications, helping enable continual learning in the field. Conventional methods of OOD detection perform multi-variate analysis on an ensemble of data or features, and usually resort to the supervision with OOD data to improve the accuracy. In reality, such supervision is impractical as one cannot anticipate the anomalous data. In this paper, we propose a novel, self-supervised approach that does not rely on any pre-defined OOD data: (1) The new method evaluates the Mahalanobis distance of the gradients between the in-distribution and OOD data. (2) It is assisted by a self-supervised binary classifier to guide the label selection to generate the gradients, and maximize the Mahalanobis distance. In the evaluation with multiple datasets, such as CIFAR-10, CIFAR-100, SVHN and TinyImageNet, the proposed approach consistently outperforms state-of-the-art supervised and unsupervised methods in the area under the receiver operating characteristic (AUROC) and area under the precision-recall curve (AUPR) metrics. We further demonstrate that this detector is able to accurately learn one OOD class in continual learning.

Sun, Jingbo↗

DOE OSTI2021

Analytical derivatives of the individual state energies in ensemble density functional theory. II. Implementation on graphical processing units (GPUs)

Conical intersections control excited state reactivity, and thus, elucidating and predicting their geometric and energetic characteristics are crucial for understanding photochemistry. Locating these intersections requires accurate and efficient electronic structure methods. Unfortunately, the most accurate methods (e.g., multireference perturbation theories such as XMS-CASPT2) are computationally challenging for large molecules. The state-interaction state-averaged restricted ensemble referenced Kohn–Sham (SI-SA-REKS) method is a computationally efficient alternative. The application of SI-SA-REKS to photochemistry was previously hampered by a lack of analytical nuclear gradients and nonadiabatic coupling matrix elements. We have recently derived analytical energy derivatives for the SI-SA-REKS method and implemented the method effectively on graphical processing units. We demonstrate that our implementation gives the correct conical intersection topography and energetics for several examples. Furthermore, our implementation of SI-SA-REKS is computationally efficient, with observed sub-quadratic scaling as a function of molecular size. This demonstrates the promise of SI-SA-REKS for excited state dynamics of large molecular systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

NASA NTRS1994

En route noise levels from propfan test assessment airplane

The en route noise test was designed to characterize propagation of propfan noise from cruise altitudes to the ground. In-flight measurements of propfan source levels and directional patterns were made by a chase plane flying in formation with the propfan test assessment (PTA) airplane. Ground noise measurements were taken during repeated flights over a distributed microphone array. The microphone array on the ground was used to provide ensemble-averaged estimates of mean flyover noise levels, establish confidence limits for those means, and measure propagation-induced noise variability. Even for identical nominal cruise conditions, peak sound levels for individual overflights varied substantially about the average, particularly when overflights were performed on different days. Large day-to-day variations in peak level measurements appeared to be caused by large day-to-day differences in propagation conditions and tended to obscure small variations arising from operating conditions. A parametric evaluation of the sensitivity of this prediction method to weather measurement and source level uncertainties was also performed. In general, predictions showed good agreement with measurements. However, the method was unable to predict short-term variability of ensemble-averaged data within individual overflights. Although variations in absorption appear to be the dominant factor in variations of peak sound levels recorded on the ground, accurate predictions of those levels require that a complete description of operational conditions be taken into account. The comprehensive and integrated methods presented in this paper have adequately predicted ground-measured sound levels. On average, peak sound levels were predicted within 3 dB for each of the three different cruise conditions.

Garber, Donald P.↗

DOE OSTI2024

Global Carbon Budget 1800-2018 Loss of Additional Sink Capacity and Present Transient Difference

A dataset of two global carbon fluxes: the loss of additional sink capacity (LASC) and present versus transient difference (PTD). The loss of additional sink capacity is the loss of indirect anthropogenic sink capacity in land ecosystems caused by the loss of ecosystems that can increase carbon in response to changing environments, primarily forests. The present transient difference refers to the direct anthropogenic flux in land ecosystems (from land cover and land use change) and the difference in that flux when calculated assuming "present day" (actually more like 1980) ecosystem carbon stocks or when calculated assuming transient carbon stocks (i.e. caused by environmental change). In this dataset, the LASC and PTD fluxes were calculated using the TRENDY ensemble of Dynamic Global Vegetation Models (DGVMs; Sitch et al., 2024) following the methods described therein and in Obermeier et al. (2021). For further details see Obermeier et al. (2021), Friedlingstein et al. (2019), Sitch et al. (2024), and Walker et al. (2025). The dataset includes ensemble means and standard deviations for annual fluxes, fluxes cumulated annually over the whole time period, and cumulated since 1959. These data were originally collected to compare different approaches (DGVMs versus bookkeeping models) for estimating land use and land cover change (LULCC) emissions (Obermeier et al. 2021). More recently, and similarly, these data were used to harmonize the estimates of the direct and indirect anthropogenic fluxes in the global carbon cycle and to recalculate the global carbon budget (Walker et al., 2025).

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2024

coh3

CoH3 (CoH ver.3) is an optical model, exciton pre-equilibrium, and Hauser-Feshbach statistical model code, which calculates nuclear reaction cross sections for medium to heavy targets in the keV to MeV energy region. This program is written in standard C++, divided into approximately 200 source and header files. CoH solves the Schroedinger equation for optical potentials defined in the code, and calculates differential elastic scattering, reaction, and total cross sections, for neutron, proton, deuteron, triton, 3He, and alpha-particle. Deformed optical potentials are solved with the coupled-channels method, in which the ground state rotational band members, or vibrational phonon states are coupled. The optical model gives particle transmission coefficients that are fed into the statistical model calculations. CoH includes the pre-equilibrium model (exciton model), the direct/semidirect capture model, and the multi-stage Hauser-Feshbach statistical decay with width fluctuation correction based on the Gaussian orthogonal ensemble. For weakly coupled levels, the DWBA (distorted wave Born approximation) method is used to calculate the direct inelastic scattering process to the excited states.

Kawano, Toshihiko↗

DOE OSTI2021

Scalable statistical inference of photometric redshift via data subsampling

Handling big data has largely been a major bottleneck in traditional statistical models. Consequently, when accurate point prediction is the primary target, machine learning models are often preferred over their statistical counterparts for bigger problems. But full probabilistic statistical models often outperform other models in quantifying uncertainties associated with model predictions. We develop a data-driven statistical modeling framework that combines the uncertainties from an ensemble of statistical models learned on smaller subsets of data carefully chosen to account for imbalances in the input space. We demonstrate this method on a photometric redshift estimation problem in cosmology, which seeks to infer a distribution of the redshift—the stretching effect in observing the light of far-away galaxies—given multivariate color information observed for an object in the sky. Our proposed method performs balanced partitioning, graph-based data subsampling across the partitions, and training of an ensemble of Gaussian process models.

data subsampling↗