SEARCH · Search NASA

Results for “ATTENTION”

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 181 records · Page 10

All sources NASA NTRS DOE OSTI

DOE OSTI2023

Mitigation of spatial nonstationarity with vision transformers

Spatial nonstationarity, the location variance of features’ statistical distributions, is ubiquitous in many natural settings. For example, in geological reservoirs rock matrix porosity varies vertically due to geomechanical compaction trends, in mineral deposits grades vary due to sedimentation and concentration processes, in hydrology rainfall varies due to the atmosphere and topography interactions, and in metallurgy crystalline structures vary due to differential cooling. Conventional geostatistical modeling workflows rely on the assumption of stationarity to be able to model spatial features for geostatistical inference. Nevertheless, this is often not a realistic assumption when dealing with nonstationary spatial data and this has motivated a variety of nonstationary spatial modeling workflows such as trend and residual decomposition, cosimulation with secondary features, and spatial segmentation and independent modeling over stationary subdomains. The advent of deep learning technologies has enabled new workflows for modeling spatial relationships. However, there is a paucity of demonstrated best practice and general guidance on mitigation of spatial nonstationarity with deep learning in the geospatial context. We demonstrate the impact of two common types of geostatistical spatial nonstationarity on deep learning model prediction performance and propose the mitigation of such impacts using self-attention (vision transformer) models. We demonstrate the utility of vision transformers for the mitigation of nonstationarity with relative errors as low as 10%, exceeding the performance of alternative deep learning methods such as convolutional neural networks. We establish best practice by demonstrating the ability of self-attention networks for modeling large-scale spatial relationships in the presence of commonly observed geospatial nonstationarity.

58 GEOSCIENCES↗

DOE OSTI2021

High-efficiency perovskite photovoltaic modules achieved via cesium doping

Perovskite solar modules have been attracting increasing attention due to their market potential, yet publications concerned with theintrinsic scale-up potential of different perovskite compositions remain relatively scarce. On the other hand, while great success is being made towards improving the power conversion efficiency (PCE) of perovskite solar cells (PSCs) by cesium cation (Cs + ) doping of the perovskite, more attention is being paid to the perovskite phase stabilization effect of Cs + doping, and less to other properties that are critical to understand and futher improve the PSC's. In this work, moderately-Cs-doped MAPbI 3 was employed as a model perovskite material in order to exclude the phase stabilization effect. Our systematic study revealed the influence of Cs + in organic-inorganic hybrid perovskites on the crystal structure, crystallization process, trap state density, band structure and charge (i.e., ions or photo-carriers) transport. Markedly, it has been observed that Cs + doping can greatly increase the carrier diffusion length in the perovskite films, thus improving the potential to scale-up PSC's.The PCE of small area devices (0.09 cm 2 ) was increased to 21.72% from 19.73%, with decreased hysteresis behavior and increased operational stability (T85 = 1000 h) after Cs + doped, where T85 refers to the retention of 85% of the initial PCE. Moreover, a PCE of 21.08% was obtained for a Cs + -containing perovskite module with an active area > 30 cm 2 , which demonstrates a better "reproducibility" than the reference sample (MAPbI3-based perovskite modules, PCE = 18.26%).

14 SOLAR ENERGY↗

DOE OSTI2020

Cathodic corrosion: 21st century insights into a 19th century phenomenon

Cathodic corrosion is an enigmatic electrochemical process that etches metallic electrodes at potentials below 0 V versus the normal hydrogen electrode. Although this phenomenon was discovered in the late 1800s by Fritz Haber, it remained mostly unnoticed during the 20th century and only attracted increased attention in the past decade. This recent attention has generated marked improvements in both the fundamental knowledge and the applications of cathodic corrosion. Fundamental new insights were gained into the effects and possible reaction intermediates of cathodic corrosion. Complementing these insights, recent advances involve applications of cathodic corrosion for nanoparticle synthesis and electrocatalyst modification. Both these applied and fundamental advances will be discussed in this short review on cathodic corrosion.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2022

Data-driven modeling of power generation for a coal power plant under cycling

Increased penetration of renewables for power generation has negatively impacted the dynamics of conventional fossil fuel-based power plants. The power plants operating on the base load are forced to cycle, to adjust to the fluctuating power demands. This results in an inefficient operation of the coal power plants, which leads up to higher operating losses. To overcome such operational challenge associated with cycling and to develop an optimal process control, this work analyzes a set of models for predicting power generation. Moreover, the power generation is intrinsically affected by the state of the power plant components, and therefore our model development also incorporates additional power plant process variables while forecasting the power generation. We present and compare multiple state-of-the-art forecasting data-driven methods for power generation to determine the most adequate and accurate model. We also develop an interpretable attention-based transformer model to explain the importance of process variables during training and forecasting. The trained deep neural network (DNN) LSTM model has good accuracy in predicting gross power generation under various prediction horizons with/without cycling events and outperforms the other models for long-term forecasting. The DNN memory-based models show significant superiority over other state-of-the-art machine learning models for short, medium and long range predictions. The transformer-based model with attention enhances the selection of historical data for multi-horizon forecasting, and also allows to interpret the significance of internal power plant components on the power generation. This newly gained insights can be used by operation engineers to anticipate and monitor the health of power plant equipment during high cycling periods.

01 COAL, LIGNITE, AND PEAT↗

DOE OSTI2025

Effects of dissolved cations, dissolved organic carbon, and exposure concentrations on per- and polyfluoroalkyl substances bioaccumulation in freshwater algae

Per- and polyfluoroalkyl substances (PFAS) have attracted global attention because of their persistence, toxicity, bioaccumulation potential, and associated adverse effects. As important primary producers, freshwater algae constitute the base of the food web in freshwater aquatic ecosystems. However, the effects of key environmental factors on PFAS uptake and bioaccumulation in freshwater algae have not been thoroughly studied. In this study, three bioaccumulation experiments were conducted to evaluate the influence of dissolved cations, dissolved organic carbon, and exposure concentrations on PFAS bioaccumulation in algae. Among the 14 studied PFAS, seven long-chain PFAS tended to bioaccumulate in algae. Elevated divalent cations (Ca 2+ and Mg 2+ ) and dissolved organic carbon did not significantly change the algal bioconcentration factors (BCFs) of PFAS, suggesting complexity of the interactions among PFAS, environmental factors, and biotic activities. Additionally, increasing exposure concentrations (0.5, 1, 5, and 10 μg/L of each PFAS) increased PFAS concentrations in algae but decreased the BCF values. This indicated that attention should be paid to the application of BCFs in future studies, including ecological risk assessment. Moreover, fluorotelomer sulfonic acids (FTSs) were incompletely recovered, suggesting that biotransformation occurred. Further studies should be conducted to evaluate whether algae play a role in FTSs biotransformation and to determine the mechanisms. In conclusion, studying the impacts of key environmental factors on PFAS bioaccumulation in algae is crucial for understanding the bioaccumulation processes that occur at the lowest trophic level and that eventually affect the dynamics of entire aquatic ecosystems.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗

DOE OSTI2021

Assessing the predictive impact of factor fixing with an adaptive uncertainty-based approach .

Despite widespread use of factor fixing in environmental modeling, its effect on model predictions has received little attention and is instead commonly presumed to be negligible. We propose a proof-of-concept adaptive method for systematically investigating the impact of factor fixing. The method uses Global Sensitivity Analysis methods to identify groups of sensitive parameters, then quantifies which groups can be safely fixed at nominal values without exceeding a maximum acceptable error, demonstrated using the 21-dimensional Sobol’ G-function. Furthermore, three error measures are considered for quantities of interest, namely Relative Mean Absolute Error, Pearson Product-Moment Correlation and Relative Variance. Results demonstrate that factor fixing may cause large errors in the model results unexpectedly, when preliminary analysis suggests otherwise, and that the default value selected affects the number of factors to fix. To improve the applicability and methodological development of factor fixing, a new research agenda encompassing five opportunities is discussed for further attention.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2026

A Decomposition-Based Learn-To-Optimize Approach with Feasibility Layer Assistance for Sub-Hourly Unit Commitment

Sub-hourly unit commitment (UC) with 15-min intervals is gaining significant attention as a way to respond rapidly to the fluctuations in electricity supply and demand introduced by renewable resources. However, the increased temporal resolution and complex inter-temporal dependencies pose substantial computational challenges for traditional optimization methods. To this end, this paper explores a decomposition-based learn-to-optimize approach. Building on recent advances in machine learning, our method revisits the long- overlooked Lagrangian relaxation framework, which is a classical decomposition technique that enables tractable subproblem solving. These smaller subproblems are inherently well-suited for machine learning, as their reduced dimensionality and structural regularity allow predictive models to efficiently learn and generalize solution patterns. We thus propose a generic predictive model, which embeds Gated Recurrent Units (GRUs) and Attention in the encoder-decoder structure, and integrate a rule-based feasibility layer to capture temporal dependencies, reduce training effort, and improve feasibility w.r.t. unit-level constraints. Our method has been validated on the IEEE 118-bus system, demonstrating promising performance in solving sub-hourly UC problems efficiently and feasibly.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Critical minerals lists for low-carbon transitions: Reviewing their structure, objectives, and limitations

Critical minerals lists have flourished in the past decade, in particular linked to the importance of critical minerals for low-carbon transitions. We identified 27 critical minerals or materials lists across 15 countries and the European Union (EU). These lists are designed to attract public and private attention and investments to secure both domestic and foreign supplies. This review article fills a gap in the existing literature by analyzing the ways in which these lists are defined and utilized by countries engaged in a mineral rush. We focus our attention on three categories of minerals – battery minerals, platinum-group metals (PGMs), and rare earth elements (REEs) that are particularly important to energy transitions. We situate this research in the broader legal and administrative developments that have driven critical minerals policies in the past decade. We provide an in-depth analysis of the commonalities and variations in the raw materials included in these lists, and identify six core limitations of critical minerals lists: (1) unclear links between criticality assessments and mineral prioritization (2) failure to account for the full mineral value-chain; (3) limited strategic alignment between allied nations; (4) limited flexibility in dynamic environments (5) limited consideration for recycling and by-product sourcing; and (6) reliance on incomplete reserve and resource data.

Battery minerals↗

DOE OSTI2022

Risk matrix for legacy wells within the Area of Review (AoR) of Carbon Capture & Storage (CCS) projects

The success of CCS depends on the capacity, injectivity, and confinement by the storage medium. Thousands of wells drilled over the past century with the intention to find the trapped oil and gas may penetrate the containment seals. These wells may provide leakage pathways for the CO 2 to escape and contaminate the underground sources of drinking water (USDW) or reach the surface in the worst-case scenario. Identifying the risky wells penetrating the containment seals and predicting their current as well as future well integrity is the most challenging task when limited data is available. Here, this paper proposes a unique methodology for risk assessment of the wells penetrating the containment seals based on the proximity of these wells from the proposed injection location, mechanical integrity, and accessibility of these wells over the lifecycle of the CCS project. This helps in identifying the wells which need immediate attention from the wells that need little to no attention. It also highlights the corrective actions necessary for the success of the CCS project as well as help estimate the approximate cost required to perform the corrective actions. This methodology focuses on all wells (producers, injectors, orphan, abandoned, water, stratigraphic, etc) while the majority of the studies found in literature focused on wells with sustained casing pressure (SCP) reports and cement bond logs (CBL). The proposed risk matrix, if applied to future CCS projects across the globe, will uniformly categorize the wells within the Area of Review (AoR).

58 GEOSCIENCES↗

DOE OSTI2022

Battery calendar aging and machine learning

Batteries continue to attract immense attention due to the critical need for new energy technologies that can help humankind transition to a net-zero future. To meet global goals for implementation of new batteries, the research community needs to more coherently consider how to go about the process of innovation, development, and validation of new technologies. Current practices for bringing a new battery technology from the benchtop to deployment can be very time consuming and complex, given that batteries can have distinct aging pathways based on both how they are cycled and their storage conditions. In nearly all use cases, batteries experience periods of active cycling and also extended periods of rest. Thus, calendar time often occurs when a battery is close to fully charged. Currently, the combined cycle and calendar life aspects receive inconsistent attention during most stages of research and development. For batteries to fulfill the critical role envisioned to meet global energy demands, greater uniformity in practice is needed to alleviate potential delays caused by the inconsistent acquisition of aging data. In this study, we present our thoughts and recommendations on the need to further advance commonality and availability of calendar aging data for emerging high-energy and long-duration batteries. While there is a broad need to adopt best practices for both calendar and cycle life assessment, a push to greater uniformity in approach and reporting is sorely needed to allow calendar life predictions to harness many of the advances that have recently been seen for cycle life prediction using machine learning (ML) and other techniques.

25 ENERGY STORAGE↗

DOE OSTI2022

Numerical simulations of membrane distillation systems with actively heated membranes

Membrane distillation (MD) is a thermal desalination process that is gaining attention for treating hypersaline brines. One recent approach uses composite membranes with a thermally conductive layer that delivers heat to the membrane–feed interface, where it drives evaporation. Though this increases single-pass recovery, it comes with the challenge of promoting lateral heat conduction through the thin membrane. Herein, we develop a 3D, computational fluid dynamics (CFD) model that simulates conjugate heat, mass, and momentum transport in the feed channel and composite membrane. We then use the CFD to verify a simpler numerical model that approximates the feed velocity field analytically. We validate the numerical model experimentally, and use it to investigate the impacts of the conductive layer thickness, feed channel geometry, and operating conditions on temperature and concentration polarization, permeate production, and specific energy consumption. Overall, we find that lateral heating increases permeate production at the expense of increased concentration polarization. In extreme cases, the concentration increases more than four-fold along the membrane surface. Furthermore, we show however, that the feed flow rate and conductive layer thickness can be tailored to mitigate concentration polarization, for only a small reduction in permeate production.

42 ENGINEERING↗

DOE OSTI2024

A novel methodology for gamma-ray spectra dataset procurement over varying standoff distances and source activities

The adoption of machine learning approaches for gamma-ray spectroscopy has received considerable attention in the literature. Many studies have investigated the deployment of various algorithm architectures to a specific task. However, little attention has been afforded to the development of the datasets leveraged to train the models. Such training datasets typically span a set of environmental or detector parameters to encompass a problem space of interest to a user. Variations in these measurement parameters will also induce fluctuations in the detector response, including expected pile-up and ground scatter effects. Fundamental to this work is the understanding that 1) the underlying spectral shape varies as the measurement parameters change and 2) the statistical uncertainties associated with two spectra impact their level of similarity. While previous studies attribute some arbitrary discretization to the measurement parameters for the generation of their synthetic training data, this work introduces a principled methodology for efficient spectral-based discretization of a problem space. A signal-to-noise ratio (SNR) respective spectral comparison measure and a Gaussian Process Regression (GPR) model are used to predict the spectral similarity across a range of measurement parameters. This innovative approach effectively showcased its capability by dividing a problem space, ranging from 5 cm to 100 cm standoff distances and 5 μCi–100 μCi of 137 Cs, into three unique combinations of measurement parameters. The findings from this work will aid in creating more robust datasets, which incorporate many possible measurement scenarios, reduce the number of required experimental test set measurements, and possibly enable experimental training data collection for gamma-ray spectroscopy.

data science↗

DOE OSTI2026

Beyond interpolation: Physics-inspired gating transformers for extrapolating irradiation conditions to novel nuclear fuels

The qualification of advanced nuclear fuels relies on irradiation experiments in test reactors that emulate commercial conditions. Designing these tests requires accurate prediction of key irradiation quantities, particularly heat generation rate and burnup, yet obtaining them typically involves computationally expensive multi-step simulation workflows. We propose a physics-inspired gating transformer (PIGT) that integrates an inverse-square, distance-based attenuation into the encoder representation to bias attention toward physically relevant spatial relationships while retaining data-driven flexibility. Using MiniFuel irradiation data from the High Flux Isotope Reactor at Oak Ridge National Laboratory, we benchmark against ensemble methods, feedforward and recurrent networks, convolutional models, and standard transformers. While baseline models perform well under interpolation, they exhibit a pronounced generalization gap when evaluated on fuels not included in the training set. The proposed model consistently improves extrapolative accuracy and stability, yielding the strongest performance on unseen fuel configurations. These results indicate that a lightweight physics structure embedded within attention mechanisms can substantially improve robustness, enabling more reliable surrogate predictions to accelerate the design of nuclear fuel irradiation experiments.

Fuel qualification↗

DOE OSTI2020

The synergy between stakeholders for cellulosic biofuel development: Perspectives, opportunities, and barriers

While understanding individual stakeholders' perspectives on the adoption and conversion to a biofuel-based landscape has been a subject of many previous studies on biofuels, there has been relatively little attention given to understanding how the interaction between multiple stakeholders involved in biofuel development could influence the widespread adoption of biofuel production. Here, this paper analyzes the key stakeholder interactions utilizing various data sources including survey results, social media posts, and empirical and theoretical analyses. An intensive review is conducted for a number of surveys and research papers on different aspects of biofuel development such as land use choices, biorefinery and transportation, infrastructure development, consumer priorities, environmental impacts, etc. Following that, a stakeholder synergy approach is applied to synthesizing typical responses of stakeholders, such as producers, consumers, biorefineries, rural communities, and the government, and discussing how their responses influence each other's decisions and the overall system performance. Based on the findings of inadequate stakeholder synergy, it is recommended that new surveys and further research should be conducted to understand why synergy between stakeholders in biofuel development is absent. Additionally, this paper provides research perspectives, including (1) applying cutting-edge text-mining techniques to conduct sentiment analysis, and research and public attention analysis; (2) using an agent-based model to simulate stakeholder interactions and understand the factors that influence stakeholder synergy and the emergence of a bioeconomy.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

DOE OSTI2025

yProv4ML: Effortless provenance tracking for machine learning systems

The rapid growth in interest in deep learning and foundation models (FMs) in particular, has attracted the attention of a diverse range of researchers thanks to their generalization ability. However, the advent of these techniques has also brought to light the lack of transparency and rigor in the way development is pursued. In particular, the inability to determine the number of epochs and other hyperparameters in advance presents challenges in identifying the best model. To address this challenge, machine learning frameworks such as MLFlow can automate the collection of this type of information. However, these tools capture data using proprietary formats and pose little attention to lineage. This paper proposes yProv4ML, a framework that captures provenance information generated during machine learning processes in PROV-JSON format, with minimal code modification.

Machine learning↗

DOE OSTI2020

Modeling evacuation demand during no-notice emergency events: Tour formation behavior

Disastrous events have been drastically increasing – both in frequency and destructive capacity – over the past few years. While advance-notice events have received a great deal of attention in the literature of disaster management, not much attention so far has been given to the no-notice events mainly because of the scarcity of data. As an attempt to address this critical gap, the current study proposes a disaggregate evacuation demand framework to understand evacuees’ travel behavior in case of no-notice emergency events. Here, the proposed framework comprises four main steps of evacuation decision, evacuation planning, tour formation, and activity schedule update. This article is dedicated to the introduction of the framework structure and elaboration on the tour formation step. In this step, we first estimate the total number of intermediate stops, travel time, and distance of the evacuation tours for those who decide to evacuate through a joint modeling structure and then, determine the type of each intermediate stop (if any). It is found that a broad range of factors including evacuees’ demographic profiles, built-environment attributes, and characteristics of the disastrous event plays a significant role in people’s evacuation behavior during no-notice emergency events. The findings of this study can assist responsible agencies in understanding evacuees’ complex behavior, and consequently, in devising effective strategies to alleviate economic damages and casualties resulted by such events.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2024

Implications of stop-and-go traffic on training learning-based car-following control

Learning-based car-following control (LCC) of connected and autonomous vehicles (CAVs) is gaining significant attention with the advancement of computing power and data accessibility. While the flexibility and large model capacity of model-free architecture enable LCC to potentially outperform the model-based car-following (CF) model in improving traffic efficiency and mitigating congestion, the generalizability of LCC for traffic conditions different from the training environment/dataset is not well-understood. Herein, this study seeks to explore the impact of stop-and-go traffic in the training dataset on the generalizability of LCC. It uses the characteristics of lead vehicle trajectories to describe stop-and-go traffic, and links the theory of identifiability (i.e., obtaining a unique parameter estimation result using sensor measurements) to the generalizability of behavior cloning (BC) and policy-based deep reinforcement learning (DRL). Correspondingly, the study shows theoretically that: (i) stop-and-go traffic can enable the property of identifiability and enhance the control performance of BC-based LCC in different traffic conditions; (ii) stop-and-go traffic is not necessary for DRL-based LCC to generalize to different traffic conditions; (iii) DRL-based LCC trained with only constant-speed lead vehicle trajectories (not sufficient to ensure identifiability) can be generalized to different traffic conditions; and (iv) stop-and-go traffic increases variance in the training dataset, which improves the convergence of parameter estimation while negatively impacting the convergence of DRL to the optimal control policy. Numerical experiments validate the above findings, illustrating that BC-based LCC entails comprehensive training datasets for generalizing to different traffic conditions, while DRL-based LCC can achieve generalization with simple free-flow traffic training environments. This further suggests DRL as a more promising and cost-effective LCC approach to reduce operational costs, mitigate traffic congestion, and enhance safety and mobility, which can accelerate the deployment and acceptance of CAVs.

33 ADVANCED PROPULSION SYSTEMS↗

DOE OSTI2021

In Situ Wood Delignification toward Sustainable Applications

As one of the most abundant and versatile natural materials on Earth, recently wood has attracted tremendous attention from scientists and engineers due to its outstanding advantages, including hierarchically porous microstructure, high mechanical strength, environmental friendliness, renewability, and biodegradability. Wood’s hierarchically porous structure and chemical components (e.g., cellulose, hemicelluloses, and lignin) enable its mechanical, ionic, optical, and thermal properties to be tuned via physical, chemical, and/or thermal modifications. Among these various approaches, the chemical delignification of bulk wood is the most fascinating, in which the majority of lignin and hemicelluloses is removed while leaving the cellulose intact, maintaining wood’s physical integrity and hierarchical structure. This delignified structure is unique, composed of hollow, aligned channels made up of cellulose microfibrils, and particularly attractive given its origin from a sustainable and renewable resource. As a result, delignified wood has attracted increasing attention for applications that go far beyond traditional wood utilization, such as lightweight yet strong structural materials, energy storage and conversion, environmental remediation, flexible electronics, and bioengineering. Here, we review recent developments in bulk wood delignification strategies toward the achievement of such advanced wood technologies for sustainable applications, with a focus on the research in our group. Similar to chemical pulping and bleaching, wood delignification involves a series of nucleophilic reactions based on alkaline Na 2 SO 3 or Na 2 S systems (i.e., chemical pulping) or electrophilic, radical, and oxidation reactions based on H 2 O 2 , ClO 2 , or NaClO systems (i.e., chemical bleaching) to deconstruct, fragment, and promote the hydrophilicity of lignin macromolecules, which finally make lignin easier to be removed. We discuss the structure and properties of partially and near-completely delignified wood, with a focus on process-structure–property relationships. The resulting delignified wood materials, with tunable structure and properties, demonstrate various advanced functions, in a wide range of advanced applications, such as building and construction, green energy, and electronics. Finally, the potential challenges and appealing perspectives of in situ wood delignification are discussed. In situ wood delignification, as a powerful modification strategy, has speeded up the development of advanced wood technologies and wood-based functional materials and products.

36 MATERIALS SCIENCE↗