Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “ensemble prediction”

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 379 records · Page 21
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Assessing decadal variability of subseasonal forecasts of opportunity using explainable AI

Abstract Identifying predictable states of the climate system allows for enhanced prediction skill on the generally low-skill subseasonal timescale via forecasts with higher confidence and accuracy, known as forecasts of opportunity. This study takes a neural network approach to explore decadal variability of subseasonal predictability, particularly during forecasts of opportunity. Specifically, this work quantifies subseasonal prediction skill provided by the tropics within the Community Earth System Model Version 2 (CESM2) Large Ensemble and assesses how this skill evolves on decadal timescales. Utilizing the networks’ confidence and explainable artificial intelligence, physically meaningful sources of predictability associated with periods of enhanced skill are identified. Using these networks, we find that tropically-driven subseasonal predictability varies on decadal timescales during forecasts of opportunity. Further, we investigate the drivers of the low frequency modulation of the tropical-extratropical teleconnection and discuss the implications. Analysis is extended to ECMWF Reanalysis v5 data, revealing that the relationships learned within the CESM2-Large Ensemble holds in modern reanalysis data. These results indicate that the neural networks are capable of identifying predictable decadal states of the climate system within CESM2 that are useful for making confident, accurate subseasonal precipitation predictions in the real world.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2018

Seasonal Predictability of Cloud Droplet Number Concentration

Aerosol emissions modify the properties of clouds hence impacting climate. The aerosol indirect effect may have offset part of the global warming caused by anthropogenic greenhouse gas emissions during the industrial era. It however remains unclear whether the same effect is significant over time scales relevant for seasonal and weather climate prediction. Answering such a question has been difficult since most weather prediction systems lack a proper representation of the aerosol evolution and transport and their interaction with clouds. Even in advanced systems it is not clear to what extent cloud microphysical properties are predictable over subseasonal to seasonal time scales. Such an issue is addressed in this study. We use a set of 30 year, four ensemble member, 9 month lead hindcast simulations of the NASA GEOS seasonal prediction system (GEOS-S2S) to study the predictability of cloud droplet number concentration in warm stratocumulus clouds. The latest version GEOS-S2S system implements interactive aerosol as well as a two moment cloud microphysics scheme therefore it is suitable for studying the aerosol indirect effect on climate. Long term retrievals from the MODIS (Moderate Resolution Imaging Spectroradiometer) are used to validate the model predictions and assess its skill in predicting cloud droplet number concentration.

Barahona, Donifan↗
DOE OSTI2025

Catalytic Activity of an Ensemble of Sites for CO 2 Hydrogenation to Methanol on a ZrO 2 -on-Cu Inverse Catalyst

The significant increase in CO 2 emissions from heavy fossil fuel utilization has raised serious concerns, highlighting the need for effective methods to convert CO 2 into value-added chemicals. Here, in this work, we report a computational investigation on the catalytic activity of ZrO 2 -on-Cu inverse catalysts for CO 2 hydrogenation to methanol, considering highly dispersed ZrO 2 trimers on Cu (111). Such clusters present a large ensemble of formate-containing configurations, Zr 3 O n (OH) m (OCHO) l , making the evaluation of the catalytic activity very challenging. We found that the sites on the various catalyst configurations exhibit markedly different activities for formate hydrogenation, despite their similar free energy and composition. To understand these differences in reactivity, we examined the structural and electronic nature of the low free-energy catalyst configurations and identified that the energy of the lowest unoccupied orbital of the reacting formate, modified by its binding with the catalytic site, is a descriptor for the reaction energy of the formate hydrogenation step. From there, we screened an ensemble of catalyst structures using this descriptor to predict highly active metastable catalyst configurations and computed the reaction pathways and transition states for formate hydrogenation. From this investigation, we distinguished reactive from nonreactive sites and formate species on the ZrO 2 /Cu inverse catalyst based on structural and electronic features. We showed that rare metastable configurations control the activity. Additionally, an efficient method for examining the reactivity of a large number of coexisting catalyst structures was developed.

catalysts↗
DOE OSTI2024

Gas Transfer Across Air‐Water Interfaces in Inland Waters: From Micro‐Eddies to Super‐Statistics

In inland water covering lakes, reservoirs, and ponds, the gas exchange of slightly soluble gases such as carbon dioxide, dimethyl sulfide, methane, or oxygen across a clean and nearly flat air‐water interface is routinely described using a water‐side mean gas transfer velocity $\overline{k_{L}}$, where overline indicates time or ensemble averaging. The micro‐eddy surface renewal model predicts $\overline{k_{L}}$ = α o Sc -1/2 ($v\bar{ϵ}$) 1/4 , where Sc is the molecular Schmidt number, $v$ is the water kinematic viscosity, and $\bar{ϵ}$ is the waterside mean turbulent kinetic energy dissipation rate at or near the interface. While α o = 0.39 - 0.46 has been reported across a number of data sets, others report large scatter or variability around this value range. It is shown here that this scatter can be partly explained by high temporal variability in instantaneous ϵ around $\bar{ϵ}$, a mechanism that was not previously considered. As the coefficient of variation (CV e ) in ϵ increases, α o must be adjusted by a multiplier (1 = CV e 2 ) -3/32 that was derived from a log‐normal model for the probability density function of ϵ. Reported variations in α o with a macro‐scale Reynolds number can also be partly attributed to intermittency effects in ϵ. Such intermittency is characterized by the long‐range (i.e., power‐law decay) spatial auto‐correlation function of ϵ. That α o varies with a macro‐scale Reynolds number does not necessarily violate the micro‐eddy model. Instead, it points to a coordination between the macro‐ and micro‐scales arising from the transfer of energy across scales in the energy cascade.

Batchelor scale↗
DOE OSTI2020

Potential Impacts of Assimilating All-sky Satellite Infrared Radiances on Convection-Permitting Analysis and Prediction of Tropical Convection

Geostationary infrared satellite observations are spatially dense [< 1/(20 km)2] and temporally frequent (< 1/hour). These suggest the possibility of using these observations to constrain sub-synoptic features over data sparse regions, such as tropical oceans. In this study, the potential impacts of assimilating water vapor channel brightness temperature (WV-BT) observations from the geostationary Meteorological Satellite 7 (Meteosat-7) on tropical convection analysis and prediction were systematically examined through a series of ensemble data assimilation experiments. WV-BT observations were assimilated hourly into convection-permitting ensembles using Penn State’s ensemble square root filter (EnSRF). Comparisons against independently observed Meteosat-7 window channel brightness temperature (Window-BT) show that the assimilation of WV-BT generally improved the intensities and locations of large-scale cloud patterns at spatial scales larger than 100 km. However, comparisons against independent soundings indicate that the EnSRF analysis produced a much stronger dry bias than the no data assimilation experiment. This strong dry bias is associated with the use of the simulated WV-BT from the prior mean during the EnSRF analysis step. A stochastic variant of the ensemble Kalman filter (NoMeanSF) is proposed. The NoMeanSF algorithm was able to assimilate the WV-BT without causing such a strong dry bias and the quality of the analyses’ horizontal cloud pattern is similar to EnSRF’s analyses. Finally, deterministic forecasts initiated from the NoMeanSF analyses possess better horizontal cloud patterns above 500 km than those of the EnSRF. These results suggest that it might be better to assimilate all-sky WV-BT through the NoMeanSF algorithm than EnSRF algorithm.

Chan, Man-Yau↗
DOE OSTI2025

Precision calibration of calorimeter signals in the ATLAS experiment using an uncertainty-aware neural network

The ATLAS experiment at the Large Hadron Collider explores the use of modern neural networks for a multi-dimensional calibration of its calorimeter signal defined by clusters of topologically connected cells (topo-clusters). The Bayesian neural network (BNN) approach not only yields a continuous and smooth calibration function that improves performance relative to the standard calibration but also provides uncertainties on the calibrated energies for each topo-cluster. The results obtained by using a trained BNN are compared to the standard local hadronic calibration and to a calibration provided by training a deep neural network. The uncertainties predicted by the BNN are interpreted in the context of a fractional contribution to the systematic uncertainties of the trained calibration. They are also compared to uncertainty predictions obtained from an alternative estimator employing repulsive ensembles.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
NASA NTRS1996

Cosmological Implications of the Effects of X-Ray Clusters on the Cosmic Microwave Background

We have been carrying forward a program to confront X-ray observations of clusters and their evolution as derived from X-ray observatories with observations of the cosmic microwave background radiation (CMBR). In addition to the material covered in our previous reports (including three published papers), most recently we have explored the effects of a cosmological constant on the predicted Sunyaev-Zel'dovich effect from the ensemble of clusters. In this report we summarize that work from which a paper will be prepared.

Forman, William R.↗
NASA NTRS2011

Geospace Environment Modeling 2008-2009 Challenge: Ground Magnetic Field Perturbations

Acquiring quantitative metrics!based knowledge about the performance of various space physics modeling approaches is central for the space weather community. Quantification of the performance helps the users of the modeling products to better understand the capabilities of the models and to choose the approach that best suits their specific needs. Further, metrics!based analyses are important for addressing the differences between various modeling approaches and for measuring and guiding the progress in the field. In this paper, the metrics!based results of the ground magnetic field perturbation part of the Geospace Environment Modeling 2008 2009 Challenge are reported. Predictions made by 14 different models, including an ensemble model, are compared to geomagnetic observatory recordings from 12 different northern hemispheric locations. Five different metrics are used to quantify the model performances for four storm events. It is shown that the ranking of the models is strongly dependent on the type of metric used to evaluate the model performance. None of the models rank near or at the top systematically for all used metrics. Consequently, one cannot pick the absolute winner : the choice for the best model depends on the characteristics of the signal one is interested in. Model performances vary also from event to event. This is particularly clear for root!mean!square difference and utility metric!based analyses. Further, analyses indicate that for some of the models, increasing the global magnetohydrodynamic model spatial resolution and the inclusion of the ring current dynamics improve the models capability to generate more realistic ground magnetic field fluctuations.

Pulkkinen, A.↗
DOE OSTI2025

Impact of microkinetic modeling assumptions on predicted kinetics and mechanisms over undercoordinated sites

Accurate modeling of catalytic reactions on undercoordinated sites requires accounting for the structural and ensemble-specific nature of the active sites. This study examines how common microkinetic modeling (MKM) assumptions affect predicted kinetics and mechanisms on the stepped Pt(211) facet for the ethane dehydrogenation (EDH) and the ethane hydrogenolysis (EH). Six (211) MKMs were developed, differing in (i) the number of active sites represented, (ii) adsorbate site occupancy treatment, and (iii) inclusion of cross-facet interactions. These models are benchmarked against a particle-based microkinetic model (PB-MKM), which best represents step-edge behavior. MKM assumptions caused deviations in turnover frequencies exceeding ten orders of magnitude and led to contrasting mechanistic and selectivity predictions. Multi-site MKMs overestimate activity by inflating free site availability, single-site models underestimate activity, and uniform occupancy models overpredict coverage of multi-dentate intermediates, leading to reaction-specific artifacts. Overall, the Combined Site Edge Model (CSEM), a single-site MKM accounting for site occupancy and cross-facet interactions, most closely approximates PB-MKM predictions. All models predict similar kinetics when surfaces are clean or primarily occupied by monodentate species. This work provides practical guidance for selecting MKM frameworks for undercoordinated catalytic surfaces and highlights the critical role of modeling assumptions in catalytic predictions.

(211) facet↗
DOE OSTI2021

Machine Learning Emulation of Spatial Deposition from a Multi-Physics Ensemble of Weather and Atmospheric Transport Models

In the event of an accidental or intentional hazardous material release in the atmosphere, researchers often run physics-based atmospheric transport and dispersion models to predict the extent and variation of the contaminant spread. These predictions are imperfect due to propagated uncertainty from atmospheric model physics (or parameterizations) and weather data initial conditions. Ensembles of simulations can be used to estimate uncertainty, but running large ensembles is often very time consuming and resource intensive, even using large supercomputers. In this paper, we present a machine-learning-based method which can be used to quickly emulate spatial deposition patterns from a multi-physics ensemble of dispersion simulations. We use a hybrid linear and logistic regression method that can predict deposition in more than 100,000 grid cells with as few as fifty training examples. Logistic regression provides probabilistic predictions of the presence or absence of hazardous materials, while linear regression predicts the quantity of hazardous materials. The coefficients of the linear regressions also open avenues of exploration regarding interpretability—the presented model can be used to find which physics schemes are most important over different spatial areas. A single regression prediction is on the order of 10,000 times faster than running a weather and dispersion simulation. However, considering the number of weather and dispersion simulations needed to train the regressions, the speed-up achieved when considering the whole ensemble is about 24 times. Ultimately, this work will allow atmospheric researchers to produce potential contamination scenarios with uncertainty estimates faster than previously possible, aiding public servants and first responders.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Using ensembles and distillation to optimize the deployment of deep learning models for the classification of electronic cancer pathology reports

One of the goals of the Surveillance, Epidemiology, and End Results (SEER) program is to estimate incidence, prevalence, and mortality of all cancers. To that end, cancer registries across the country maintain a massive database of cancer pathology reports which contain rich information to understand cancer trends. However, these reports are stored in the form of unstructured text, and human annotators are required to read and extract relevant information. In this article, we show that existing deep learning models for automating information extraction from cancer pathology reports can be significantly improved by using ensemble model distillation. We found that by training multiple predictive models and transferring their knowledge to a single, low-resource model, we can reduce the number of highly confident wrong predictions. Our results show that our implemented methods could save 1000s of manual annotation hours.

60 APPLIED LIFE SCIENCES↗
NASA NTRS2021

Machine Learning the COSMO Model for Predicting Thermodynamics of Electrolyte Mixtures

Bottom-up design of electrolyte mixtures for battery systems requires predicting macro thermodynamic properties from molecular constituents. For instance, molten salt electrolyte batteries require conditions far above room temperature to operate. Therefore, discovering mixtures with increasingly lower eutectic melting points is desirable. A model that can approximate chemical activity is a valuable tool to search through the vast compositional design space. Machine learning can predict properties of materials such as vibrational free energies, electronic energy gaps, and thermal conductivities. Moreover, they can learn physical models such as interatomic potentials. The COSMO-SAC model uses theory and empirical parameterization to predict liquid-vapor and liquid-solid properties using first-principles calculations. However, obtaining activity coefficients required for parameterizing the COSMO-SAC model is costly and limited to a select chemical space. In this work, we explored if machine learning methods could improve the COSMO-SAC model and bridge density functional theory calculations to liquid phase thermodynamic properties. Our data-driven approach uses existing databases for sigma-profiles of organic solvents and reconciles their methodological differences via ensemble averaging. First, an optimal machine learning model is constructed for each dataset. Our machine learning algorithms use the sigma-profile as an input feature to predict binary mixtures' activity coefficients using multi-output regression. Each dataset uses different choices of functionals, methods, and basis sets. Therefore, our ensemble model attempts to predict corrected activity coefficients given the combination of all the model outputs. The activity coefficients used for training are generated using the COSMO-SAC model. This approach enables the extraction of meaningful information from the existing datasets to improve the COSMO-SAC model for obtaining thermodynamic properties of electrolyte mixtures. With the liquid phase activities, we can identify electrolyte mixtures that meet desired phase equilibria conditions.

Thermodynamics↗
DOE OSTI2026

Uncertainty guided online ensemble for non-stationary data streams in fusion science

Machine Learning (ML) is poised to play a pivotal role in the development and operation of next-generation fusion devices. Fusion data shows non-stationary behavior with distribution drifts, resulted by both experimental evolution and machine wear-and-tear. ML models assume stationary distribution and fail to maintain performance when encountered with such non-stationary data streams. Online learning techniques have been leveraged in other domains, however it has been largely unexplored for fusion applications. In this paper, we investigate online learning for continuous adaptation to drifting data streams in the prediction of Toroidal Field (TF) coils deflection at the DIII-D fusion facility. We further address the short-term performance degradation inherent to standard online learning, which arises because ground truth is unavailable at prediction time. To mitigate this issue, we propose an uncertainty-guided online ensemble framework. The method leverages the Deep Gaussian Process Approximation (DGPA) for calibrated uncertainty estimation and uses these uncertainty measures to guide a meta-algorithm that aggregates predictions from learners trained over different historical horizons. Our results show that online learning reduces prediction error by 80% compared to a static model. The online ensemble and the proposed uncertainty-guided ensemble further reduce error by approximately 6%, and 10% respectively, relative to standard single-model online learning, while also providing calibrated uncertainty estimates to support operational decision-making.

AI↗
DOE OSTI2025

AI Improves the Accuracy, Reliability, and Economic Value of Continental‐Scale Flood Predictions

Accurate flood early warnings are critical to minimize damage and loss of life. Current large‐scale operational forecasting systems, however, have limited accuracy, description of uncertainty, and computational efficiency. While Artificial intelligence (AI) can address these limitations in principle, the accuracy and reliability of AI forecasts have thus far proven insufficient. Here we present a novel hybrid framework that integrates AI‐based machinery termed Errorcastnet (ECN) with the National Water Model (NWM) to showcase the potential of ensemble AI flood forecasts over the contiguous U.S. ECN boosts prediction accuracy four‐ to six‐fold across lead times of 1–10 days, while providing uncertainty quantification. It also outperforms Google's state‐of‐the‐art global AI model. ECN‐based forecasts offer superior economic value (up to four‐fold) for decision‐making as compared to those from NWM alone. ECN performs well in varied ecoregions, physiography, and land management conditions. The framework is computationally efficient, enabling national‐scale ensemble forecasts in minutes.

artificial intelligence↗
DOE OSTI2023

Atomistic learning in the electronically grand-canonical ensemble

Abstract A strategy is presented for the machine-learning emulation of electronic structure calculations carried out in the electronically grand-canonical ensemble. The approach relies upon a dual-learning scheme, where both the system charge and the system energy are predicted for each image. The scheme is shown to be capable of emulating basic electrochemical reactions at a range of potentials, and coupling it with a bootstrap-ensemble approach gives reasonable estimates of the prediction uncertainty. The method is also demonstrated to accelerate saddle-point searches, and to extrapolate to systems with one to five water layers. We anticipate that this method will allow for larger length- and time-scale simulations necessary for electrochemical simulations.

36 MATERIALS SCIENCE↗
DOE OSTI2020

A Novel Modeling Framework for Computationally Efficient and Accurate Real-Time Ensemble Flood Forecasting With Uncertainty Quantification

A novel modeling framework that simultaneously improves accuracy, predictability, and computational efficiency is presented. It embraces the benefits of three modeling techniques integrated together for the first time: surrogate modeling, parameter inference, and data assimilation. The use of polynomial chaos expansion (PCE) surrogates significantly decreases computational time. Parameter inference allows for model faster convergence, reduced uncertainty, and superior accuracy of simulated results. Ensemble Kalman filters assimilate errors that occur during forecasting. To examine the applicability and effectiveness of the integrated framework, we developed 18 approaches according to how surrogate models are constructed, what type of parameter distributions are used as model inputs, and whether model parameters are updated during the data assimilation procedure. We conclude that (1) PCE must be built over various forcing and flow conditions, and in contrast to previous studies, it does not need to be rebuilt at each time step; (2) model parameter specification that relies on constrained, posterior information of parameters (so-called Selected specification) can significantly improve forecasting performance and reduce uncertainty bounds compared to Random specification using prior information of parameters; and (3) no substantial differences in results exist between single and dual ensemble Kalman filters, but the latter better simulates flood peaks. The use of PCE effectively compensates for the computational load added by the parameter inference and data assimilation (up to ~80 times faster). Therefore, the presented approach contributes to a shift in modeling paradigm arguing that complex, high-fidelity hydrologic and hydraulic models should be increasingly adopted for real-time and ensemble flood forecasting.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

An Ensemble of Neural Networks for Moist Physics Processes, Its Generalizability and Stable Integration

Abstract With the recent advances in data science, machine learning has been increasingly applied to convection and cloud parameterizations in global climate models (GCMs). This study extends the work of Han et al. (2020, https://doi.org/10.1029/2020MS002076 ) and uses an ensemble of 32‐layer deep convolutional residual neural networks, referred to as ResCu‐en, to emulate convection and cloud processes simulated by a superparameterized GCM, SPCAM. ResCu‐en predicts GCM grid‐scale temperature and moisture tendencies, and cloud liquid and ice water contents from moist physics processes. The surface rainfall is derived from the column‐integrated moisture tendency. The prediction uncertainty inherent in deep learning algorithms in emulating the moist physics is reduced by ensemble averaging. Results in 1‐year independent offline validation show that ResCu‐en has high prediction accuracy for all output variables, both in the current climate and in a warmer climate with +4K sea surface temperature. The analysis of different neural net configurations shows that the success to generalize in a warmer climate is attributed to convective memory and the 1‐dimensional convolution layers incorporated into ResCu‐en. We further implement a member of ResCu‐en into CAM5 with real world geography and run the neural‐network‐enabled CAM5 (NCAM) for 5 years without encountering any numerical integration instability. The simulation generally captures the global distribution of the mean precipitation, with a better simulation of precipitation intensity and diurnal cycle. However, there are large biases in temperature and moisture in high latitudes. These results highlight the importance of convective memory and demonstrate the potential for machine learning to enhance climate modeling.

Meteorology & Atmospheric Sciences↗
DOE OSTI2022

Generalizability analysis of tool condition monitoring ensemble machine learning models

Tool condition monitoring (TCM) is an essential research area for the optimization and automation of metal machining processes, and could help manufacturers reduce costs, production time, machine downtime, energy use, and part scrappage. However, TCM systems developed in prior studies have struggled to reach the high level of generalizability which is necessary for industrial applications. This study addresses TCM system generalizability to new machining conditions, how variations in machining and environmental conditions may be used to improve model generalizability, and ensemble machine learning techniques for TCM. Further, milling tool life experiments were conducted using various machining conditions, and the processes' sound, spindle power, and axial load signals were collected. Different machine learning models were evaluated for the prediction of tool wear levels, including four individual models and five ensemble models. Changes in cutting speed were found to display a large effect on model performance, while the chip load showed some effect, and the feed rate had little effect. A simulated noise data augmentation technique for model improvement is applied within TCM for the first time, and resulted in increased model generalizability and reduced overfitting. Across several performance metrics the extremely randomized trees ensemble machine learning model generally performed the best for this application, achieving a leave-one-group-out cross validation accuracy score of 92.4 %, a 10-fold cross validation score of 98.9 %, and an averaged accuracy across 11 generalizability tests of 87.3 %.

42 ENGINEERING↗