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 379 records · Page 21

All sources NASA NTRS DOE OSTI

NASA NTRS1984

Dynamic response of shock waves in transonic diffuser and supersonic inlet - An analysis with the Navier-Stokes equations and adaptive grid

An existing method which solves the multi-dimensional ensemble-averaged compressible time-dependent Navier-Stokes equations in conjunction with mixing length turbulence model and shock capturing technique has been extended to include the shock-tracking adaptive grid systems. The numerical scheme for solving the governing equations is based on a linearized block implicit approach. The effects of grid-motion and grid-distribution on the calculated flow solutions have been studied in relative detail and this is carried out in the context of physically steady, shocked flows computed with non-stationary grids. Subsequently, the unsteady dynamics of the flows occurring in a supercritically operated transonic diffuser and a mixed compression supersonic inlet have been investigated with the adaptive grid systems by solving the Navier-Stokes equations.

Liu, N.-S.↗

NASA NTRS1989

Light scattering instrumentation for micro gravity research

The analysis of light scattered from an ensemble of particles has long been a preferred method for characterizing their physical properties. Instrumentation to perform the measurements which forms the basis for such analysis is available in many forms based upon a variety of different experimental techniques. A system is presented which is singularly applicable for making many types of measurements in a microgravity environment. The commercial version of this device, the DAWN-F, has been used in many labs throughout the world to perform analyses of particular importance for both research and production. Light scattering theory is reviewed and the structure and function of the system is described.

Wyatt, Philip J.↗

DOE OSTI2023

A Stochastic Covariance Shrinkage Approach in Ensemble Transform Kalman Filtering

The Ensemble Kalman Filters (EnKF) employ a Monte-Carlo approach to represent covariance information, and are affected by sampling errors in operational settings where the number of model realizations is much smaller than the model state dimension. To alleviate the effects of these errors EnKF relies on model-specific heuristics such as covariance localization, which takes advantage of the spatial locality of correlations among the model variables. This work proposes an approach to alleviate sampling errors that utilizes a locally averaged-in-time dynamics of the model, described in terms of a climatological covariance of the dynamical system. We use this covariance as the target matrix in covariance shrinkage methods, and develop a stochastic covariance shrinkage approach where synthetic ensemble members are drawn to enrich both the ensemble subspace and the ensemble transformation. We additionally provide for a way in which this methodology can be localized similar to the state-of-the-art LETKF method, and that for a certain model setup, our methodology significantly outperforms it.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRS2013

A Single-column Model Ensemble Approach Applied to the TWP-ICE Experiment

Single-column models (SCM) are useful test beds for investigating the parameterization schemes of numerical weather prediction and climate models. The usefulness of SCM simulations are limited, however, by the accuracy of the best estimate large-scale observations prescribed. Errors estimating the observations will result in uncertainty in modeled simulations. One method to address the modeled uncertainty is to simulate an ensemble where the ensemble members span observational uncertainty. This study first derives an ensemble of large-scale data for the Tropical Warm Pool International Cloud Experiment (TWP-ICE) based on an estimate of a possible source of error in the best estimate product. These data are then used to carry out simulations with 11 SCM and two cloud-resolving models (CRM). Best estimate simulations are also performed. All models show that moisture-related variables are close to observations and there are limited differences between the best estimate and ensemble mean values. The models, however, show different sensitivities to changes in the forcing particularly when weakly forced. The ensemble simulations highlight important differences in the surface evaporation term of the moisture budget between the SCM and CRM. Differences are also apparent between the models in the ensemble mean vertical structure of cloud variables, while for each model, cloud properties are relatively insensitive to forcing. The ensemble is further used to investigate cloud variables and precipitation and identifies differences between CRM and SCM particularly for relationships involving ice. This study highlights the additional analysis that can be performed using ensemble simulations and hence enables a more complete model investigation compared to using the more traditional single best estimate simulation only.

numerical weather forecasting↗

DOE OSTI2020

Capturing multireference excited states by constrained-density-functional theory

The computation of excited electronic states with commonly employed (approximate) methods is challenging, typically yielding states of lower quality than the corresponding ground state for a higher computational cost. In this work, we present a mean-field method that extends the previously proposed excited constrained-density-functional theory (XCDFT) from single Slater determinants to ensemble one-body reduced density matrices for computing low-lying excited states. The method still retains an associated computational complexity comparable to a semilocal density functional theory (DFT) calculation while at the same time it is capable of approaching states with multireference character. Further, we benchmark the quality of this method on well-established test sets, finding good descriptions of the electronic structure of multireference states and maintaining an overall accuracy for the predicted excitation energies comparable to semilocal time-dependent DFT.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2020

3D diffractive imaging of nanoparticle ensembles using an x-ray laser

Single particle imaging at x-ray free electron lasers (XFELs) has the potential to determine the structure and dynamics of single biomolecules at room temperature. Two major hurdles have prevented this potential from being reached, namely, the collection of sufficient high-quality diffraction patterns and robust computational purification to overcome structural heterogeneity. We report the breaking of both of these barriers using gold nanoparticle test samples, recording around 10 million diffraction patterns at the European XFEL and structurally and orientationally sorting the patterns to obtain better than 3-nm-resolution 3D reconstructions for each of four samples. With these new developments, integrating advancements in x-ray sources, fast-framing detectors, efficient sample delivery, and data analysis algorithms, we illuminate the path towards sub-nanometer biomolecular imaging. The methods developed here can also be extended to characterize ensembles that are inherently diverse to obtain their full structural landscape.

47 OTHER INSTRUMENTATION↗

NASA NTRS2002

Goddard Cumulus Ensemble (GCE) Model: Application for Understanding Precipitation Processes

One of the most promising methods to test the representation of cloud processes used in climate models is to use observations together with Cloud Resolving Models (CRMs). The CRMs use more sophisticated and realistic representations of cloud microphysical processes, and they can reasonably well resolve the time evolution, structure, and life cycles of clouds and cloud systems (size about 2-200 km). The CRMs also allow explicit interaction between out-going longwave (cooling) and incoming solar (heating) radiation with clouds. Observations can provide the initial conditions and validation for CRM results. The Goddard Cumulus Ensemble (GCE) Model, a cloud-resolving model, has been developed and improved at NASA/Goddard Space Flight Center over the past two decades. Dr. Joanne Simpson played a central role in GCE modeling developments and applications. She was the lead author or co-author on more than forty GCE modeling papers. In this paper, a brief discussion and review of the application of the GCE model to (1) cloud interactions and mergers, (2) convective and stratiform interaction, (3) mechanisms of cloud-radiation interaction, (4) latent heating profiles and TRMM, and (5) responses of cloud systems to large-scale processes are provided. Comparisons between the GCE model's results, other cloud-resolving model results and observations are also examined.

Tao, Wei-Kuo↗

DOE OSTI2022

Exploring nonequilibrium phases of photo-doped Mott insulators with generalized Gibbs ensembles

Many experiments show that strong excitations of correlated quantum materials can cause non-thermal phases without equilibrium analogues. Understanding the origin and properties of these nonequilibrium states has been challenging due to the limitations of theoretical methods for nonequilibrium strongly correlated systems. In this work, we introduce a generalized Gibbs ensemble description that enables a systematic analysis of the long-time behavior of photo-doped states in Mott insulators based on equilibrium methods. We demonstrate the power of the method by mapping out the nonequilibrium phase diagram of the one-dimensional extended Hubbard model, which features η -pairing and charge density wave phases in a wide photo-doping range. We furthermore clarify that the peculiar kinematics of photo-doped carriers, and the interaction between them, play an essential role in the formation of these non-thermal phases. Our results establish a new path for the systematic analysis of nonequilibrium strongly correlated systems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2025

Scalable Bayesian Physics-Informed Kolmogorov-Arnold Networks

Uncertainty quantification (UQ) plays a pivotal role in scientific machine learning, especially when surrogate models are used to approximate complex systems. Although multilayer perceptions (MLPs) are commonly employed as surrogates, they often suffer from overfitting due to their large number of parameters. Kolmogorov-Arnold networks (KANs) offer an alternative solution with fewer parameters. However, gradient-based inference methods, such as Hamiltonian Monte Carlo (HMC), may result in computational inefficiency when applied to KANs, especially for large-scale datasets, due to the high cost of back-propagation. To address these challenges, we propose a novel approach, combining the dropout Tikhonov ensemble Kalman inversion (DTEKI) with Chebyshev KANs. This gradient-free method effectively mitigates overfitting and enhances numerical stability. In addition, we incorporate the active subspace method to reduce the parameter-space dimensionality, allowing us to improve the accuracy of predictions and obtain more reliable uncertainty estimates. Extensive experiments demonstrate the efficacy of our approach in various test cases, including scenarios with large datasets and high noise levels. Our results show that the new method achieves comparable or better accuracy, much higher efficiency as well as stability compared to HMC, in addition to scalability. Moreover, by leveraging the low-dimensional parameter subspace, our method preserves prediction accuracy while substantially reducing further the computational cost.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2022

Uncertainty Models for the Hybrid Parametric Variation Method of Uncertainty Quantification; Analysis

There is some level of uncertainty in every finite element model (FEM), which flows to a level of uncertainty in predicted results. The purpose of uncertainty quantification (UQ) is to provide statistical bounds on prediction accuracy based on model uncertainty. This is distinct from model updating, which attempts to modify models to improve their accuracy. UQ does not improve the accuracy of models, but accepts that the models are inherently inaccurate and attempts to quantify the impact of that inaccuracy on predicted results. Previously, an alternate method for UQ, called the Hybrid Parametric Variation (HPV) method, was applied to Space Launch System (SLS) Hurty/Craig-Bampton (HCB) components to predict system-level statistics for launch vehicle attitude control transfer functions and core stage section loads due to buffet. The HPV method combines a parametric variation of the HCB fixed-interface (FI) modal frequencies with a nonparametric variation (NPV) method that randomly varies the HCB mass and stiffness matrices as Wishart random matrix distributions using random matrix theory (RMT). Alternatively, the most common method for modeling uncertainty in the structural dynamics community is a parametric approach, which varies physical parameters in the model. However, there are several disadvantages associated with the parametric method. Determining a reduced set of parameters that have a significant impact on the system response can be time consuming, and the selected parameter probability distributions are rarely reliably known. Therefore, in practice, the parameters are surrogates for the actual errors, and the link to parameter uncertainty is unknown. Another major drawback is that the uncertainty that can be represented is limited to the form of the nominal FEM. It is the experience of the authors that based on numerous aerospace programs, almost all FEM errors are in form rather than parameter values. This hypothesis is supported by the observation of the authors that it is almost never possible to ‘tune’ a FEM to match modal test results by only modifying model parameters. Model-form uncertainty cannot be directly represented by FEM input parameters nor included in a parametric approach. However, model-form uncertainty can be modeled using RMT, where a probability distribution is developed for the matrix ensemble of interest. The major advantage of the NPV method is that it covers errors in model form. The HPV method anchors uncertainty at the HCB component level to component modal test results by matching the HCB and test modes based on mode descriptions or other methods, and then applying differing levels of frequency variation. The specific variations depend on the confidence to which a component FEM has been validated through modal testing. The NPV method is layered on the frequency variation to match modal test self-orthogonality and cross-orthogonality (XO) results. Once the component uncertainty models are identified, they are assembled, and the uncertainty is propagated to the system level using a Monte Carlo (MC) analysis approach that generates statistics for system-level predictions This provides a UQ method that can be traced to test data, which can be updated as additional data and improved correlated models become available. The purpose of this paper is to collect and present all of the theory for HPV that has been previously published in reports and papers and to present examples of its application. Specifically, component uncertainty models based on the dispersion of corresponding mass and stiffness matrices using proposed test/analysis correlation metrics are investigated. The first example is purely academic so that the true answers are known, and the validity of the HPV method and the corresponding uncertainty models can be determined. The purpose of this paper is to collect and present all of the theory for HPV that has been previously published in reports and papers and to present examples of its application. Specifically, component uncertainty models based on the dispersion of corresponding mass and stiffness matrices using proposed test/analysis correlation metrics are investigated. The first example is purely academic so that the true answers are known, and the validity of the HPV method and the corresponding uncertainty models can be determined. The second example is an application to a component that is design specific to the SLS. Based on this work and other assessments, the HPV method provides another tool to the toolset used for complex system UQ analysis. From experience gathered to date using the HPV method, additional design specific applications must be investigated to provide further confidence in the validity of the HPV method of UQ analysis.

Uncertainty quantification↗

DOE OSTI2020

Deeply uncertain: comparing methods of uncertainty quantification in deep learning algorithms

We present a comparison of methods for uncertainty quantification (UQ) in deep learning algorithms in the context of a simple physical system. Three of the most common uncertainty quantification methods - Bayesian Neural Networks (BNN), Concrete Dropout (CD), and Deep Ensembles (DE) - are compared to the standard analytic error propagation. We discuss this comparison in terms endemic to both machine learning ("epistemic" and "aleatoric") and the physical sciences ("statistical" and "systematic"). The comparisons are presented in terms of simulated experimental measurements of a single pendulum - a prototypical physical system for studying measurement and analysis techniques. Our results highlight some pitfalls that may occur when using these UQ methods. For example, when the variation of noise in the training set is small, all methods predicted the same relative uncertainty independently of the inputs. This issue is particularly hard to avoid in BNN. On the other hand, when the test set contains samples far from the training distribution, we found that no methods sufficiently increased the uncertainties associated to their predictions. This problem was particularly clear for CD. In light of these results, we make some recommendations for usage and interpretation of UQ methods.

59 BASIC BIOLOGICAL SCIENCES↗

NASA NTRS2004

ENSO Bred Vectors in Coupled Ocean-Atmosphere General Circulation Models

The breeding method has been implemented in the NASA Seasonal-to-Interannual Prediction Project (NSIPP) Coupled General Circulation Model (CGCM) with the goal of improving operational seasonal to interannual climate predictions through ensemble forecasting and data assimilation. The coupled instability as cap'tured by the breeding method is the first attempt to isolate the evolving ENSO instability and its corresponding global atmospheric response in a fully coupled ocean-atmosphere GCM. Our results show that the growth rate of the coupled bred vectors (BV) peaks at about 3 months before a background ENSO event. The dominant growing BV modes are reminiscent of the background ENSO anomalies and show a strong tropical response with wind/SST/thermocline interrelated in a manner similar to the background ENSO mode. They exhibit larger amplitudes in the eastern tropical Pacific, reflecting the natural dynamical sensitivity associated with the presence of the shallow thermocline. Moreover, the extratropical perturbations associated with these coupled BV modes reveal the variations related to the atmospheric teleconnection patterns associated with background ENSO variability, e.g. over the North Pacific and North America. A similar experiment was carried out with the NCEP/CFS03 CGCM. Comparisons between bred vectors from the NSIPP CGCM and NCEP/CFS03 CGCM demonstrate the robustness of the results. Our results strongly suggest that the breeding method can serve as a natural filter to identify the slowly varying, coupled instabilities in a coupled GCM, which can be used to construct ensemble perturbations for ensemble forecasts and to estimate the coupled background error covariance for coupled data assimilation.

Yang, S. C.↗

NASA NTRS2005

SVD analysis of Aura TES spectral residuals

Singular Value Decomposition (SVD) analysis is both a powerful diagnostic tool and an effective method of noise filtering. We present the results of an SVD analysis of an ensemble of spectral residuals acquired in September 2004 from a 16-orbit Aura Tropospheric Emission Spectrometer (TES) Global Survey and compare them to alternative methods such as zonal averages. In particular, the technique highlights issues such as the orbital variation of instrument response and incompletely modeled effects of surface emissivity and atmospheric composition.

spectra↗

NASA NTRS2013

Evaluating Downscaling Methods for Seasonal Climate Forecasts over East Africa

The U.S. National Multi-Model Ensemble seasonal forecasting system is providing hindcast and real-time data streams to be used in assessing and improving seasonal predictive capacity. The NASA / USAID SERVIR project, which leverages satellite and modeling-based resources for environmental decision making in developing nations, is focusing on the evaluation of NMME forecasts specifically for use in impact modeling within hub regions including East Africa, the Hindu Kush-Himalayan (HKH) region and Mesoamerica. One of the participating models in NMME is the NASA Goddard Earth Observing System (GEOS5). This work will present an intercomparison of downscaling methods using the GEOS5 seasonal forecasts of temperature and precipitation over East Africa. The current seasonal forecasting system provides monthly averaged forecast anomalies. These anomalies must be spatially downscaled and temporally disaggregated for use in application modeling (e.g. hydrology, agriculture). There are several available downscaling methodologies that can be implemented to accomplish this goal. Selected methods include both a non-homogenous hidden Markov model and an analogue based approach. A particular emphasis will be placed on quantifying the ability of different methods to capture the intermittency of precipitation within both the short and long rain seasons. Further, the ability to capture spatial covariances will be assessed. Both probabilistic and deterministic skill measures will be evaluated over the hindcast period.

Robertson, Franklin R.↗

NASA NTRS2013

Evaluating Downscaling Methods for Seasonal Climate Forecasts over East Africa

Roberts, J. Brent↗

DOE OSTI2021

Utilizing ensemble learning for performance and power modeling and improvement of parallel cancer deep learning CANDLE benchmarks

Abstract Machine learning (ML) continues to grow in importance across nearly all domains in modeling to learn from data. Often a tradeoff exists between a model's ability to minimize bias and variance. In this article, we utilize ensemble learning to combine linear, nonlinear, and tree‐/rule‐based ML methods to cope with the bias‐variance tradeoff and result in more accurate models. We use the datasets collected for two parallel cancer deep learning CANDLE benchmarks, NT3 and P1B2, to build performance and power models based on hardware performance counters using single‐object and multiple‐objects ensemble learning to identify the most important counters for improvement on the Cray XC40 Theta at Argonne National Laboratory. Based on the insights from these models, we improve the performance and energy of P1B2 and NT3 by optimizing the deep learning environments TensorFlow, Keras, Horovod, and Python under the huge page size of 8 MB. Experimental results show that ensemble learning not only produces more accurate models but also provides more robust performance counter ranking. We achieve up to 61.15% performance improvement and up to 62.58% energy saving for P1B2 and up to 55.81% performance improvement and up to 52.60% energy saving for NT3 on up to 24,576 cores.

Wu, Xingfu↗

DOE OSTI2023

Optimizing Seasonal‐To‐Decadal Analog Forecasts With a Learned Spatially‐Weighted Mask

Abstract Seasonal‐to‐decadal climate prediction is crucial for decision‐making in a number of industries, but forecasts on these timescales have limited skill. Here, we develop a data‐driven method for selecting optimal analogs for seasonal‐to‐decadal analog forecasting. Using an interpretable neural network, we learn a spatially‐weighted mask that quantifies how important each grid point is for determining whether two climate states will evolve similarly. We show that analogs selected using this weighted mask provide more skillful forecasts than analogs that are selected using traditional spatially‐uniform methods. This method is tested on two prediction problems using the Max Planck Institute for Meteorology Grand Ensemble: multi‐year prediction of North Atlantic sea surface temperatures, and seasonal prediction of El Niño Southern Oscillation. This work demonstrates a methodical approach to selecting analogs that may be useful for improving seasonal‐to‐decadal forecasts and understanding their sources of skill.

Geology↗

DOE OSTI2020

Deeply Uncertain: Comparing Methods of Uncertainty Quantification in Deep Learning Algorithms [Slides]

Deep learning is used in many applications in the physical sciences. In those sciences we are used to having uncertainty on every measurement or prediction. In the last years, many methods have been put forth for uncertainty quantification: Bayesian Neural Networks, Concrete Dropout, and Deep Ensembles. Uncertainty in deep learning is often divided into aleatoric or irreducible and epistemic or reducible. The problem remains: Which uncertainty quantification method should be chosen? How are those results interpreted?

97 MATHEMATICS AND COMPUTING↗