Search NASA⌕ Search

SEARCH · Search NASA

Results for “Deep uncertainty”

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 415 records · Page 23
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

A Review of Coupled Geochemical–Geomechanical Impacts in Subsurface CO 2 , H 2 , and Air Storage Systems

Increased demand for decarbonization and renewable energy has led to increasing interest in engineered subsurface storage systems for large-scale carbon reduction and energy storage. In these applications, a working fluid (CO 2 , H 2 , air, etc.) is injected into a deep formation for permanent sequestration or seasonal energy storage. The heterogeneous nature of the porous formation and the fluid–rock interactions introduce complexity and uncertainty in the fate of the injected component and host formations in these applications. Interactions between the working gas, native brine, and formation mineralogy must be adequately assessed to evaluate the efficiency, risk, and viability of a particular storage site and operational regime. This study reviews the current state of knowledge about coupled geochemical–geomechanical impacts in geologic carbon sequestration (GCS), underground hydrogen storage (UHS), and compressed air energy storage (CAES) systems involving the injection of CO 2 , H 2 , and air. Specific review topics include (1) existing injection induced geochemical reactions in these systems; (2) the impact of these reactions on the porosity and permeability of host formation; (3) the impact of these reactions on the mechanical properties of host formation; and (4) the investigation of geochemical-geomechanical process in pilot scale GCS. This study helps to facilitate an understanding of the potential geochemical–geomechanical risks involved in different subsurface energy storage systems and highlights future research needs.

08 HYDROGEN↗
DOE OSTI2023

Bayesian Inference with Latent Hamiltonian Neural Networks (L-HNNs)

When sampling for Bayesian inference, one popular approach is to use Hamiltonian Monte Carlo (HMC) and the No-U-Turn Sampler (NUTS). However, HMC and NUTS can require numerous numerical gradients of the target density and can prove slow in practice. We propose Hamiltonian neural networks (HNNs) with HMC and NUTS for solving Bayesian inference problems [1, 2]. Once trained, HNNs do not require gradients of the target density while sampling. Moreover, they satisfy important properties such as perfect time reversibility and Hamiltonian conservation, making them well suited for use within HMC and NUTS because stationarity can be shown. We also propose an HNN extension called latent HNNs (L-HNNs), which predict latent variable outputs. Compared to HNNs, L-HNNs offer improved expressivity and a reduction in integration errors. Finally, we propose employing L-HNNs in NUTS with an online error monitoring scheme to prevent degeneracy of the sampling in regions of low probability density. We demonstrate L-HNNs in NUTS with online error monitoring by using several example cases involving complex, heavy-tailed, and high local curvature probability densities. Overall, L-HNNs in NUTS with online error monitoring satisfactorily inferred these probability densities. Compared to traditional NUTS, L-HNNs in NUTS with online error monitoring improved the effective sample size (ESS) per gradient by an order of magnitude.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Self-adaptive weights based on balanced residual decay rate for physics-informed neural networks and deep operator networks

Physics-informed deep learning has emerged as a promising alternative for solving partial differential equations. However, for complex problems, training these networks can still be challenging, often resulting in unsatisfactory accuracy and efficiency. In this work, we demonstrate that the failure of plain physics-informed neural networks arises from the significant discrepancy in the convergence rate of residuals at different training points, where the slowest convergence rate dominates the overall solution convergence. Based on these observations, we propose a pointwise adaptive weighting method that balances the residual decay rate across different training points. The performance of our proposed adaptive weighting method is compared with current state-of-the-art adaptive weighting methods on benchmark problems for both physics-informed neural networks and physics-informed deep operator networks. In conclusion, through extensive numerical results we demonstrate that our proposed approach of balanced residual decay rates offers several advantages, including bounded weights, high prediction accuracy, fast convergence rate, low training uncertainty, low computational cost, and ease of hyperparameter tuning.

Balanced convergence rate↗
DOE OSTI2024

Deep energy-pressure regression for a thermodynamically consistent EOS model

Abstract In this paper, we aim to explore novel machine learning (ML) techniques to facilitate and accelerate the construction of universal equation-Of-State (EOS) models with a high accuracy while ensuring important thermodynamic consistency. When applying ML to fit a universal EOS model, there are two key requirements: (1) a high prediction accuracy to ensure precise estimation of relevant physics properties and (2) physical interpretability to support important physics-related downstream applications. We first identify a set of fundamental challenges from the accuracy perspective, including an extremely wide range of input/output space and highly sparse training data. We demonstrate that while a neural network (NN) model may fit the EOS data well, the black-box nature makes it difficult to provide physically interpretable results, leading to weak accountability of prediction results outside the training range and lack of guarantee to meet important thermodynamic consistency constraints. To this end, we propose a principled deep regression model that can be trained following a meta-learning style to predict the desired quantities with a high accuracy using scarce training data. We further introduce a uniquely designed kernel-based regularizer for accurate uncertainty quantification. An ensemble technique is leveraged to battle model overfitting with improved prediction stability. Auto-differentiation is conducted to verify that necessary thermodynamic consistency conditions are maintained. Our evaluation results show an excellent fit of the EOS table and the predicted values are ready to use for important physics-related tasks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Global QCD analysis and dark photons

We perform a global QCD analysis of high energy scattering data within the JAM Monte Carlo framework, including a coupling to a dark photon that augments the Standard Model (SM) electroweak coupling via kinetic mixing with the hypercharge B boson. We first set limits on the dark photon mass and mixing parameter assuming that the SM is the true theory of Nature, taking into account also the effect on g – 2 of the muon. If instead we entertain the possibility that the dark photon may play a role in deep-inelastic scattering (DIS), we find that the best fit is preferred over the SM at 6.5σ, even after accounting for missing higher order uncertainties. The improvement in χ 2 with the dark photon is stable against all the tests we have applied, with the improvements in the theoretical predictions spread across a wide range of x and Q 2 . The largest improvement corresponds to the fixed target and HERA DIS data, while the best fit yields a value of g – 2 which significantly reduces the disagreement with the latest experimental determination.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2019

Degradation Mechanism Detection in Photovoltaic Backsheets by Fully Convolutional Neural Network

Materials and devices age with time. Material aging and degradation has important implications for lifetime performance of materials and systems. While consensus exists that materials should be studied and designed for degradation, materials inspection during operation is typically performed manually by technicians. The manual inspection makes studies prone to errors and uncertainties due to human subjectivity. In this work, we focus on automating the process of degradation mechanism detection through the use of a fully convolutional deep neural network architecture (F-CNN). We demonstrate that F-CNN architecture allows for automated inspection of cracks in polymer backsheets from photovoltaic (PV) modules. The developed F-CNN architecture enabled an end-to-end semantic inspection of the PV module backsheets by applying a contracting path of convolutional blocks (encoders) followed by an expansive path of decoding blocks (decoders). First, the hierarchy of contextual features is learned from the input images by encoders. Next, these features are reconstructed to the pixel-level prediction of the input by decoders. The structure of the encoder and the decoder networks are thoroughly investigated for the multi-class pixel-level degradation type prediction for PV module backsheets. The developed F-CNN framework is validated by reporting degradation type prediction accuracy for the pixel level prediction at the level of 92.8%.

14 SOLAR ENERGY↗
DOE OSTI2022

Giant kelp genetic monitoring before and after disturbance reveals stable genetic diversity in Southern California

Given the impacts of climate change and other anthropogenic stressors on marine systems, there is a need to accurately predict how species respond to changing environments and disturbance regimes. The use of genetic tools to monitor temporal trends in populations gives ecologists the ability to estimate changes in genetic diversity and effective population size that may be undetectable by traditional census methods. Although multiple studies have used temporal genetic analysis, they usually involve commercially important species, and rarely sample before and after disturbance. In this study, we run a temporal analysis of giant kelp, Macrocystis pyrifera , genetic diversity over the scope of 10 years (2008-2018) using the same microsatellite marker panel to assess the genetic consequences of disturbance in several populations of giant kelp ( Macrocystis pyrifera ) in the Southern California Bight. The study is a rare pre- and post-disturbance microsatellite analysis that included declines to giant kelp caused by the 2015/16 El Nino Southern Oscillation event. We used canopy biomass estimated by remote sensing (Landsat) to quantify the extent of disturbance to kelp beds, and sea surface temperature data to understand how kelp was pushed towards its temperature limits during this period. Despite prolonged periods with decreased canopy at several sites, no changes in genetic structure and allelic richness were observed. We argue that giant kelp in the region is best described as a “patchy population” system where true extinctions are rare. We discuss how deep refugia of subsurface sporophytes and cryptic microscopic life stages could have kept genetic diversity through disturbance. Given the increasing effects of climate change and uncertainty in modeling impacts of species with cryptic life history stages, we suggest further investigation to reveal the role such stages play in species resilience. Genetic monitoring studies of sites selected by remote census demographic and climate surveys should be continued in the future given the predicted impacts of climate change.

Klingbeil, III, William H.↗
NASA NTRS2006

The Deep Space Network Array

This document is a viewgraph presentation that reviews the costs, and technological processing required to replace the current network of Deep Space Antennas. The concept of using an array for space communications is much less of a concern than the cost of implementing and operating such an array. Within the cost question, the cost uncertainty of the front-end components (repeated n-times) is of most importance. The activities at JPL have focused on both these aspects of the cost. A breadboard array of three antennas at JPL has been the vehicle to perform many investigations into the development of the new DSN. Several pictures of the antennas at JPL are shown.

Deep Space Network (DSN)↗
DOE OSTI2024

Flavor dependence of unpolarized quark transverse momentum distributions from a global fit

We present an extraction of the unpolarized transverse-momentum-dependent parton distribution and fragmentation functions that takes into account possible differences between quark flavors and final-state hadrons. The extraction is based on experimental measurements from Drell-Yan processes and semi-inclusive deep-inelastic scattering, whose combination is essential to distinguish flavor differences. The analysis is carried out at N 3 LL accuracy. The extracted flavor-dependent distributions give a very good description of the data (χ 2 /N dat = 1.08). The resulting error bands take fully into account also the uncertainties in the determination of the corresponding collinear distributions.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

Failure Analysis–Informed Risk Assessment Framework for Geological Carbon Storage Using Numerical Simulation and Machine Learning

Geological carbon storage (GCS) is recognized as a critical technology for achieving large-scale reductions in anthropogenic carbon dioxide (CO 2 ) emissions. Ensuring long-term containment and safety requires robust risk assessment frameworks that account for geological uncertainty and identify potential failure scenarios. Among various indicators, the area of review (AoR) serves as a key metric for evaluating storage performance, regulatory compliance, and monitoring design, as it delineates the spatial extent impacted by pressure buildup and plume migration. However, conventional AoR-based risk assessments typically perturb parameters within narrow uncertainty bounds, potentially overlooking rare but high-impact events arising from extreme geological conditions. In this study, we present a failure analysis–informed risk assessment framework for large-scale GCS projects to improve site prescreening and monitoring design. A suite of 300 numerical simulations was generated using stochastic geological models that vary five key parameters: net-to-gross ratio, anisotropy azimuth, porosity multiplier, permeability multiplier, and vertical-to-horizontal permeability ratio. Among these, 200 realizations represent normal geological uncertainty, while 100 additional cases explore extreme yet plausible conditions for failure-case analysis. The AoR was simulated and computed from pressure and CO 2 saturation fields, where the baseline AoR boundary, representing the extent predicted under typical geological uncertainty, was defined as the union of 200 normal-range simulations, and failure was identified when extreme-range cases exceeded this baseline. Results show that incorporating broader parameter uncertainty produces significantly larger AoR extents, underscoring the potential underestimation of risk under conventional uncertainty ranges. Furthermore, spatial probability maps derived from failure-induced AoR exceedance identify regions requiring enhanced monitoring attention. Various machine learning (ML)–based classifiers were developed to predict failure occurrence from geological parameters, with the random forest model achieving the highest performance (F1-score of 0.986). Consistent findings from correlation coefficient, feature importance, and Sobol sensitivity analyses reveal that low net-to-gross ratios and permeability multipliers are the dominant risk drivers, reflecting reduced reservoir connectivity and limited pressure dissipation. Altogether, these results provide a novel framework for risk-informed site prescreening and monitoring design that explicitly considers rare but high-impact geological scenarios in GCS projects.

25 ENERGY STORAGE↗
DOE OSTI2023

A comparison of machine learning surrogate models of street-scale flooding in Norfolk, Virginia

Low-lying coastal cities, exemplified by Norfolk, Virginia, face the challenge of street flooding caused by rainfall and tides, which strain transportation and sewer systems and can lead to personal and property damage. While high-fidelity, physics-based simulations provide accurate predictions of urban pluvial flooding, their computational complexity renders them unsuitable for real-time applications. Using data from Norfolk rainfall events between 2016 and 2018, this study compares the performance of a previous surrogate model based on a random forest algorithm with two deep learning models: Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). The comparison of deep learning to the random forest algorithm is motivated by the desire to utilize a machine learning architecture that allows for the future inclusion of common uncertainty quantification techniques and the effective integration of relevant, multi-modal features.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Observation of the singly Cabibbo-suppressed decay Λ c + → p π 0

Utilizing 4.5 fb − 1 of e + e − annihilation data collected with the BESIII detector at the BEPCII collider at center-of-mass energies between 4.600 and 4.699 GeV, the first observation of the singly Cabibbo-suppressed decay Λ c + → p π 0 is presented, with a statistical significance of 5.4 σ . The ratio of the branching fractions of Λ c + → p π 0 and Λ c + → p η is measured as B ( Λ c + → p π 0 ) / B ( Λ c + → p η ) = ( 0.120 ± 0.02 6 stat ± 0.00 7 syst ) . This result resolves the longstanding discrepancy between earlier experimental searches, providing both a decisive conclusion and valuable input for QCD-inspired theoretical models. A sophisticated deep learning approach using a Transformer-based architecture is employed to distinguish the signal from the prevalent hadronic backgrounds, complemented by thorough validation and systematic uncertainty quantification. Published by the American Physical Society 2025

Ablikim, M.↗
DOE OSTI2023

Bi-Level Adaptive Storage Expansion Strategy for Microgrids Using Deep Reinforcement Learning

Battery energy storage (BES) is a versatile resource for the secure and economic operation of microgrids (MGs). Prevailing stochastic optimization-based approaches for BES expansion planning for MGs are computationally complicated. This work proposes a data-driven bi-level multi-period BES expansion planning framework to determine the siting, sizing, and timing of BES installations. The proposed planning framework unifies deep reinforcement learning (DRL) and linear programming, thereby decoupling the determinations for the integer and continuous decision variables in two time scales, respectively. In the upper level, a rainbow DRL agent with quantile regression is trained to provide dynamic planning policies to accommodate stochastic renewable energy resources (RESs), load, and battery price changes efficiently. Further, the lower level computes the optimal operation of MGs with frequency constraints to hedge the islanding contingency. The two levels communicate with one another by exchanging storage configuration and operating expenses in order to accomplish the shared goal of minimizing investment and operation costs. Comparative case studies on an MG are carried out to demonstrate the superiority of the proposed DRL-based solution to the mixed-integer linear programming counterpart on efficiency, scalability, and adaptability.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Designing observables for measurements with deep learning

Many analyses in particle and nuclear physics use simulations to infer fundamental, effective, or phenomenological parameters of the underlying physics models. When the inference is performed with unfolded cross sections, the observables are designed using physics intuition and heuristics. We propose to design targeted observables with machine learning. Unfolded, differential cross sections in a neural network output contain the most information about parameters of interest and can be well-measured by construction. The networks are trained using a custom loss function that rewards outputs that are sensitive to the parameter(s) of interest while simultaneously penalizing outputs that are different between particle-level and detector-level (to minimize detector distortions). We demonstrate this idea in simulation using two physics models for inclusive measurements in deep inelastic scattering. We find that the new approach is more sensitive than classical observables at distinguishing the two models and also has a reduced unfolding uncertainty due to the reduced detector distortions.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Yet Another Discriminant Analysis (YADA): A Probabilistic Model for Machine Learning Applications

This paper presents a probabilistic model for various machine learning (ML) applications. While deep learning (DL) has produced state-of-the-art results in many domains, DL models are complex and over-parameterized, which leads to high uncertainty about what the model has learned, as well as its decision process. Further, DL models are not probabilistic, making reasoning about their output challenging. In contrast, the proposed model, referred to as Yet Another Discriminate Analysis(YADA), is less complex than other methods, is based on a mathematically rigorous foundation, and can be utilized for a wide variety of ML tasks including classification, explainability, and uncertainty quantification. YADA is thus competitive in most cases with many state-of-the-art DL models. Ideally, a probabilistic model would represent the full joint probability distribution of its features, but doing so is often computationally expensive and intractable. Hence, many probabilistic models assume that the features are either normally distributed, mutually independent, or both, which can severely limit their performance. YADA is an intermediate model that (1) captures the marginal distributions of each variable and the pairwise correlations between variables and (2) explicitly maps features to the space of multivariate Gaussian variables. Numerous mathematical properties of the YADA model can be derived, thereby improving the theoretic underpinnings of ML. Validation of the model can be statistically verified on new or held-out data using native properties of YADA. However, there are some engineering and practical challenges that we enumerate to make YADA more useful.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1989

Effect of earth albedo variation on the performance of spatial acquisition subsystem aboard a planetary spacecraft

Deep-space optical communication systems generally require the use of the sunlit earth as the pointing reference. For simple single-frame spatial acquisition processors, the uncertainties in earth albedo can result in an irreducible error in acquiring the spatial location of the receiver. Multiple frame processing algorithms can be used to estimate the earth albedo. The prior knowledge of the image shape and orientation can then be used to derive the location of the receiver.

Chen, Chien-Chung↗
NASA NTRS1997

Improved Passive Microwave Algorithms for North America and Eurasia

Microwave algorithms simplify complex physical processes in order to estimate geophysical parameters such as snow cover and snow depth. The microwave radiances received at the satellite sensor and expressed as brightness temperatures are a composite of contributions from the Earth's surface, the Earth's atmosphere and from space. Owing to the coarse resolution inherent to passive microwave sensors, each pixel value represents a mixture of contributions from different surface types including deep snow, shallow snow, forests and open areas. Algorithms are generated in order to resolve these mixtures. The accuracy of the retrieved information is affected by uncertainties in the assumptions used in the radiative transfer equation (Steffen et al., 1992). One such uncertainty in the Chang et al., (1987) snow algorithm is that the snow grain radius is 0.3 mm for all layers of the snowpack and for all physiographic regions. However, this is not usually the case. The influence of larger grain sizes appears to be of more importance for deeper snowpacks in the interior of Eurasia. Based on this consideration and the effects of forests, a revised SMMR snow algorithm produces more realistic snow mass values. The purpose of this study is to present results of the revised algorithm (referred to for the remainder of this paper as the GSFC 94 snow algorithm) which incorporates differences in both fractional forest cover and snow grain size. Results from the GSFC 94 algorithm will be compared to the original Chang et al. (1987) algorithm and to climatological snow depth data as well.

Foster, James↗
NASA NTRS2012

Sensor Calibration Inter-Comparison Methodologies and Applications TO AVHRR, MODIS, AND VIIRS Observations

As more and more satellite observations become available to the science and user community, their on-orbit calibration accuracy and consistency over time continue to be an important and challenge issue, especially in the reflective solar spectral regions. In recent years, many sensor calibration inter-comparison methodologies have been developed by different groups and applied to a range of satellite observations, aiming to the improvement of satellite instrument calibration accuracy and data quality. This paper provides an overview of different methodologies developed for inter-comparisons of A VHRR and MODIS observations, and extends their applications to the Visible-Infrared Imaging Radiometer Suite (VIIRS) instrument. The first VIIRS was launched on-board the NPP spacecraft on October 28, 2011. The VIIRS, designed with MODIS heritage, collects data in 22 spectral bands from visible (VIS) to long-wave infrared (LWIR). Like both Terra and Aqua MODIS, the VIIRS on-orbit calibration is performed using a set of on-board calibrators (OBC), Methodologies discussed in this paper include the use of well-characterized ground reference targets, near simultaneous nadir overpasses (SNO), lunar observations, and deep convective clouds (DeC). Results from long-term A VHRR and MODIS observations and initial assessment of VIIRS on-orbit calibration are presented. Current uncertainties of different methodologies and potential improvements are also discussed in this paper.

Xiong, Xiaoxiong↗