Search NASA⌕ Search

SEARCH · Search NASA

Results for “Deep potential”

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 721 records · Page 40
All sourcesNASA NTRSDOE OSTI
NASA NTRS2018

An Analysis of the Lightning Jump Algorithm Using Geostationary Lightning Mapper Flashes

Lightning's relation to severe weather has been studied since the 1980's. The invention of the Lightning Mapping Array allowed for total lightning measurements in a 125 km operational range. This brought forth an automated lightning Jump Algorithm (LIA) that predicted severe weather based on two-sigma increases in total lightning. The LIA's biggest downfall is being restrained to the limited field of view (FOV) of LMA's. The launch of the Geostationary Lightning Mapper (GLM) aboard the GOES-16 satellite now gives us hemispheric total lightning measurements. The wide FOV makes the GLM a good candidate to apply the LIA to. However the GLM and LMA have some differences. One being the coarser spatial resolution of GLM. Another being that LMA measures very high frequency (VHF) electromagnetic radiation while GLM measures optical radiation. These differences suggest an extensive study must be done on using the LIA with GLM to understand potential differences in the LIA and to maximize its operational skill. Four deep dive cases are conducted showcasing the differences between the GLM and LMA and their jumps.

Curtis, Nathan↗
NASA NTRSDate not supplied

Surface Systems and Interface Standardization

Space exploration on planetary surfaces will require the use of various surface systems which will likely need to interface with one another. These systems must communicate and share data, as well as distribute power and transfer fluids for sustainable surface operations. Such systems can range from landers, surface habitats, mobility systems, cargo, and In-Situ Resource Utilizations (ISRUs). Previous and current programs have developed space interoperability standards that help aid in reducing potential risks of interface integration. One example of this is the International Deep Space Interoperability Standards (IDSIS) which focuses on deep space exploration. A goal of the Artemis program is to develop a sustained human presence on the lunar surface that would be a training ground for future Mars exploration. Therefore, it will be beneficial to identify interface standards between the surface assets which will help mitigate risk and reduce complexity in the harsh environments of space exploration. A key contribution to sustainability is commonality between hardware and software interfaces. Generic interfaces for data, power, and fluids will reduce risk, promote interoperability, and define standard interfaces across surface exploration projects and programs. Standardized interfaces would be advantageous for improving efficiency and reducing overall complexity, which are critical considerations for future space exploration. Furthermore, it will provide cost reductions to the Artemis programs over their life cycle (in Operations & Maintenance (O&M) and Logistics). Unique proprietary interfaces if considered or allowed would not only increase complexity but also add cost to the programs. Of course, understanding what will work and not work effectively in these unique environments such as the lunar surface is important. These unique environments require much needed intelligent design, prototyping, comprehensive testing, and field experience, utilizing consensus on common interface solutions.

Jaime Gomez↗
DOE OSTI2023

On fast simulation of dynamical system with neural vector enhanced numerical solver

The large-scale simulation of dynamical systems is critical in numerous scientific and engineering disciplines. However, traditional numerical solvers are limited by the choice of step sizes when estimating integration, resulting in a trade-off between accuracy and computational efficiency. To address this challenge, we introduce a deep learning-based corrector called Neural Vector (NeurVec), which can compensate for integration errors and enable larger time step sizes in simulations. Our extensive experiments on a variety of complex dynamical system benchmarks demonstrate that NeurVec exhibits remarkable generalization capability on a continuous phase space, even when trained using limited and discrete data. NeurVec significantly accelerates traditional solvers, achieving speeds tens to hundreds of times faster while maintaining high levels of accuracy and stability. Moreover, NeurVec’s simple-yet-effective design, combined with its ease of implementation, has the potential to establish a new paradigm for fast-solving differential equations based on deep learning.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2024

Evaluation of Machine Learning and Deep Learning Algorithms for Fire Prediction in Southeast Asia

Vegetation fires are prevalent in South/Southeast Asian countries, making fire prediction crucial due to their potential environmental, economic, and social impacts. Accurate predictions of fires facilitate timely interventions, helping to mitigate uncontrolled fires that can lead to biodiversity loss and air quality issues. In this study, we utilize VIIRS satellite-derived fire data alongside six machine learning and deep learning models—Simple Persistence, Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), CNN-LSTM, and ConvLSTM—to determine the most effective fire prediction model, using Root Mean Square Error (RMSE) as the metric. Our results indicate that the CNN model is the most reliable in regions with spatial dependencies, such as Brunei, Indonesia, Malaysia, the Philippines, Timor-Leste, and Thailand. Conversely, the ConvLSTM model excels in countries with complex spatiotemporal dynamics like Laos, Myanmar, and Vietnam. The CNN-LSTM hybrid model also performed well in Cambodia, suggesting a need for a balanced approach in areas requiring both spatial and temporal feature extraction. Furthermore, simpler models like Persistence and MLP showed limitations in capturing dynamic patterns and temporal dependencies. Our findings highlight the importance of evaluating models before implementing any decision support systems (DSS) in fire management. By tailoring models to specific regional fire data, we can enhance prediction accuracy and responsiveness, ultimately improving fire risk management in Southeast Asia and beyond.

Deep learning↗
DOE OSTI2025

Recent Advances in the Use of Ionic Liquids and Deep Eutectic Solvents for Lignocellulosic Biorefineries and Biobased Chemical and Material Production

Biorefineries, which process biomass feedstocks into valuable (bio)products, aim to replace fossil fuel-based refineries to produce energy and chemicals, reducing environmental and health hazards, including climate change, and supporting a sustainable economy. In particular, lignocellulose-based biorefineries, utilizing the most abundant renewable feedstock on Earth, have significant potential to supply sustainable energy, chemicals and materials. Ionic liquids (ILs, organic salts with low melting temperatures) and deep eutectic solvents (DESs, mixtures with eutectic points lower than the ideal mixture) are capable of dissolving some of the key lignocellulose polymers, and even the whole biomass. Furthermore, they have intrinsic advantages over molecular solvents, including safer usage profiles and high tunability, which allow tailored physicochemical properties. Such properties provide unique opportunities for the development of new processes that could unlock the full potential of future biorefineries. Here, we review the current state of lignocellulosic biomass processing with ILs and DESs, with a specific focus on the pretreatment chemistry, process flow and products from each component; followed by discussions on sustainability assessments and technological challenges. We aim to inform the research community about the opportunities, challenges and perspectives in developing truly sustainable lignocellulose-based biorefineries.

09 BIOMASS FUELS↗
DOE OSTI2020

Evaluating the Potential and Challenges of an Uncertainty Quantification Method for Long Short–Term Memory Models for Soil Moisture Predictions

Recently, recurrent deep networks have shown promise to harness newly available satellite–sensed data for long–term soil moisture projections. However, to be useful in forecasting, deep networks must also provide uncertainty estimates. Here we evaluated Monte Carlo dropout with an input–dependent data noise term (MCD+N), an efficient uncertainty estimation framework originally developed in computer vision, for hydrologic time series predictions. MCD+N simultaneously estimates a heteroscedastic input–dependent data noise term (a trained error model attributable to observational noise) and a network weight uncertainty term (attributable to insufficiently constrained model parameters). Although MCD+N has appealing features, many heuristic approximations were employed during its derivation, and rigorous evaluations and evidence of its asserted capability to detect dissimilarity were lacking. To address this, we provided an in–depth evaluation of the scheme's potential and limitations. We showed that for reproducing soil moisture dynamics recorded by the Soil Moisture Active Passive (SMAP) mission, MCD+N indeed gave a good estimate of predictive error, provided that we tuned a hyperparameter and used a representative training data set. The input–dependent term responded strongly to observational noise, while the model term clearly acted as a detector for physiographic dissimilarity from the training data, behaving as intended. However, when the training and test data were characteristically different, the input–dependent term could be misled, undermining its reliability. Additionally, due to the data–driven nature of the model, data noise also influences network weight uncertainty, and therefore the two uncertainty terms are correlated. Altogether, this approach has promise, but care is needed to interpret the results.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Accelerating Resonance Searches via Signature-Oriented Pre-training

The search for heavy resonances beyond the Standard Model (BSM) is a key objective at the LHC. While the recent use of advanced deep neural networks for boosted-jet tagging significantly enhances the sensitivity of dedicated searches, it is limited to specific final states, leaving vast potential BSM phase space underexplored. We introduce a novel experimental method, Signature-Oriented Pre-training for Heavy-resonance ObservatioN (Sophon), which leverages deep learning to cover an extensive number of boosted final states. Pre-trained on the comprehensive JetClass-II dataset, the Sophon model learns intricate jet signatures, ensuring the optimal constructions of various jet tagging discriminates and enabling high-performance transfer learning capabilities. We show that the method can not only push widespread model-specific searches to their sensitivity frontier, but also greatly improve model-agnostic approaches, accelerating LHC resonance searches in a broad sense.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

DeepPhenoMem V1.0: deep learning modelling of canopy greenness dynamics accounting for multi-variate meteorological memory effects on vegetation phenology

Abstract. Vegetation phenology plays a key role in controlling the seasonality of ecosystem processes that modulate carbon, water and energy fluxes between the biosphere and atmosphere. Accurate modelling of vegetation phenology in the interplay of Earth's surface and the atmosphere is thus crucial to understand how the coupled system will respond to and shape climatic changes. Phenology is controlled by meteorological conditions at different timescales: on the one hand, changes in key meteorological variables (temperature, water, radiation) can have immediate effects on the vegetation development; on the other hand, phenological changes can be driven by past environmental conditions, known as memory effects. However, the processes governing meteorological memory effects on phenology are not completely understood, resulting in their limited performance of vegetation phenology represented in land surface models. A deep learning model, specifically a long short-term memory network (LSTM), has the potential to capture and model the meteorological memory effects on vegetation phenology. Here, we apply the LSTM to model the vegetation phenology using meteorological drivers and high-temporal-resolution canopy greenness observations through digital repeat photography by the PhenoCam network. We compare a multiple linear regression model, a no-memory-effect LSTM model and a full-memory-effect LSTM model to predict the whole seasonal greenness trajectory and the corresponding phenological transition dates across 50 sites and 317 site years during 2009–2018, covering deciduous broadleaf forests, evergreen needleleaf forests and grasslands. Results show that the deep learning model outperforms the multiple linear regression model, and the full-memory-effect LSTM model performs better than the no-memory-effect model for all three plant function types (median R2 of 0.878, 0.957 and 0.955 for broadleaf forests, evergreen needleleaf forests and grasslands). We also find that the full-memory-effect LSTM model is capable of predicting the seasonal dynamic variations of canopy greenness and reproducing trends in shifting phenological transition dates. We also performed a sensitivity analysis of the full-memory-effect LSTM model to assess its plausibility, revealing its coherence with established knowledge of vegetation phenology sensitivity to meteorological conditions, particularly changes in temperature. Our study highlights that (1) multi-variate meteorological memory effects play a crucial role in vegetation phenology, and (2) deep learning opens up new avenues for improving the representation of vegetation phenological processes in land surface models via a hybrid modelling approach.

Geology↗
DOE OSTI2024

Can Glacial Sea‐Level Drop‐Induced Gas Hydrate Dissociation Cause Submarine Landslides?

Abstract We conducted two‐dimensional numerical simulations to investigate the mechanisms underlying the strong spatiotemporal correlation observed between submarine landslides and gas hydrate dissociation due to glacial sea‐level drops. Our results suggest that potential plastic deformation or slip could occur at localized and small scales in the shallow‐water portion of the gas hydrate stability zone (GHSZ). This shallow‐water portion of the GHSZ typically lies within the area enclosed by three points: the BGHSZ–seafloor intersection, the seafloor at ∼600 m below sea level (mbsl), and the base of the GHSZ (BGHSZ) at ∼1,050 mbsl in low‐latitude regions. The deep BGHSZ (>1,050 mbsl) could not slip; therefore, the entire BGHSZ was not a complete slip surface. Glacial hydrate dissociation alone is unlikely to cause large‐scale submarine landslides. Observed deep‐water (much greater than 600 mbsl) turbidites containing geochemical evidence of glacial hydrate dissociation potentially formed from erosion or detachment in the GHSZ pinch‐out zone.

58 GEOSCIENCES↗
DOE OSTI2021

Enabling deeper learning on big data for materials informatics applications

Abstract The application of machine learning (ML) techniques in materials science has attracted significant attention in recent years, due to their impressive ability to efficiently extract data-driven linkages from various input materials representations to their output properties. While the application of traditional ML techniques has become quite ubiquitous, there have been limited applications of more advanced deep learning (DL) techniques, primarily because big materials datasets are relatively rare. Given the demonstrated potential and advantages of DL and the increasing availability of big materials datasets, it is attractive to go for deeper neural networks in a bid to boost model performance, but in reality, it leads to performance degradation due to the vanishing gradient problem. In this paper, we address the question of how to enable deeper learning for cases where big materials data is available. Here, we present a general deep learning framework based on Individual Residual learning (IRNet) composed of very deep neural networks that can work with any vector-based materials representation as input to build accurate property prediction models. We find that the proposed IRNet models can not only successfully alleviate the vanishing gradient problem and enable deeper learning, but also lead to significantly (up to 47%) better model accuracy as compared to plain deep neural networks and traditional ML techniques for a given input materials representation in the presence of big data.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Deep learning-enhanced design for functionally graded auxetic lattices

Auxetic materials with counterintuitive negative Poisson’s ratio have been of significant interest due to potential applications across diverse engineering fields. Functionally grading such auxetics further enables customization of the structural response and harnesses the potential for multi-functional applications. However, a critical challenge in designing functionally graded lattices is to efficiently determine the spatial variation of the functional gradient and the corresponding geometric designs to achieve the desired response. In this paper, a highly efficient deep learning-based inverse design framework for functionally graded tetra-petal auxetics with spatially tailored properties is presented. This framework significantly improves the efficiency of tailoring functionally graded auxetics where many unit cells need to be tailor-designed. The graded tetra-petal auxetics obtained from the inverse design framework are additively manufactured and subjected to impact tests. The results show superior impact performance compared with uniform designs, demonstrating the effectiveness of the proposed inverse design framework, which can be inspirable to promote advanced structures/materials with enhanced impact resistance.

42 ENGINEERING↗
DOE OSTI2024

Unifying Quantum Materials Modeling and Experiments: The Role of Machine Learning Interatomic Potentials

Computational experiments have emerged as a powerful complement to traditional experiments in the design of new materials. The development of machine learning (ML) and deep learning techniques, combined with database construction and data mining, has significantly enhanced traditional quantum mechanical methods. This synergy enables the rapid development of structure-property relationships. In this talk, I will discuss our recent efforts in applying Machine Learning Interatomic Potentials (MLIAPs) to accelerate materials modeling across various material classes and challenging applications where traditional methods fall short. First, I will highlight the success of MLIAPs in accurately modeling the melting behavior of complex materials. Our results demonstrate high fidelity with experimental observations and also with calculated reference melting temperatures. In the second application, I will discuss how MLIAPs are trained and applied to elucidate the interplay between segregation tendencies and surface reconstructions in CuNi alloys under oxidizing conditions. A key factor in the success of these MLIAP applications is the design of minimalistic yet flexible datasets along with a computational framework for training MLIAPs.

Saidi, Wissam↗
NASA NTRSDate not supplied

The BioSensor Instrument Beyond LEIA: a Versatile Platform for Lunar Biology

Introduction: The BioSensor is a deep-space-compatible automated microfluidic culturing instrument. While originally designed to measure the effects of deep space radiation on yeast growth for the BioSentinel mission, it has the potential to host a diverse range of life science experiments with single- and/or multi-celled organisms and can be adapted to interface with a diverse range of platforms in both crewed and uncrewed settings. It is therefore a leading candidate for hosting life sciences experimentation associated with a lunar surface habitat. BioSensor design: The function of the BioSensor is to monitor the growth and metabolic activity of samples in batch fluidic culture, without the need for crew involvement. The current configuration houses organisms in 16 wells within 16 microfluidic cards, accommodating a total of 256 samples, replicates, and controls. Each well has an optical system including three LEDs and a photodiode detector to measure absorbance at three wavelengths, enabling measurement of optical density, color change in dyes such as alamarBlue, and bioproduction of pigmented compounds. Organisms are loaded into fluidic wells and air-dried for storage during integration, launch, and transit, then activated by the introduction of culture medium from storage bags via manifolds that fill one card at a time. Temperature is controlled by individual card heaters, and timing of all activities (fluidics fills, optical measurements, temperatures) is directed by an experiment script. The self-contained BioSensor payload is roughly 4U in volume; with electrical/mechanical/thermal interface, e.g. for operation on ISS or a lunar lander, as well as a linear energy-transfer (LET) charged-particle radiation spectrometer, the entire system is closer to 6U. BioSentinel and LEIA: Flown on the ISS and in a deep-space free flyer for BioSentinel, the BioSensor has been modified for use in the LEIA mission, including improvements to reduce the sensitivity to lengthy launch delays. LEIA will monitor yeast growth in the radiation and reduced-gravity environment of the lunar surface no earlier than 2026, on a CLPS lander [4]. Changes include accommodating additional culture media and an additional LED color for a new biological assay (bioproduction of carotenoids-- dietary antioxidants), as well as modifications to the housing to allow late-load biology changeout and improved isolation between fluidics and electronics. Future prospects: Future work with the BioSensor, beyond LEIA, will include expanding the range and diversity of organisms and assays accommodated. Preliminary work has demonstrated the growth of Arabidopsis seedlings in BioSensor fluidic cards, including optical measurements of growth rate over time. Minor modifications could allow measurement of phenotypes related to photosynthetic capacity in both plants and cyanobacteria. The experimental capabilities of the BioSensor could be dramatically increased by introducing the capability for fluorescence measurements, and/or the design of novel biological assays using luminescence. The BioSensor can also be adapted for new platforms and experiment settings; in addition to free-flyer, ISS, and CLPS lander, a preliminary design concept has been developed for crewed deployment directly to the lunar surface. The instrument could be accommodated inside a lunar habitat, where its automated operation would make it an excellent candidate for experiments from fundamental investigations into the response of organisms to lunar surface conditions to applied-science purposes such as screening engineered strains of various organisms for bioproduction capability.

J A Lee↗
DOE OSTI2023

Domain Adaptation for Measurements of Strong Gravitational Lenses

Upcoming surveys are predicted to discover galaxy-scale strong lenses on the magnitude of 105, making deep learning methods necessary in lensing data analysis. Currently, there is insufficient real lensing data to train deep learning algorithms, but training only on simulated data results in poor performance on real data. Domain adaptation can bridge the gap between simulated and real datasets. We adopt domain adaptation on the estimation of Einstein radius in simulated galaxy-scale gravitational lensing images. We evaluate two domain adaptation techniques - domain adversarial neural networks (DANN) and maximum mean discrepancy (MMD). We train on a source domain of simulated lenses and apply it to a target domain with emulation of DES survey conditions. We show that both domain adaptation techniques can significantly improve the model performance on the more complex target domain datasets. Our results show the potential of using domain adaptation to perform analysis on future survey data with a deep neural network trained on simulated data.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS2004

High-Power Krypton Hall Thruster Technology Being Developed for Nuclear-Powered Applications

The NASA Glenn Research Center has been performing research and development of moderate specific impulse, xenon-fueled, high-power Hall thrusters for potential solar electric propulsion applications. These applications include Mars missions, reusable tugs for low-Earth-orbit to geosynchronous-Earth-orbit transportation, and missions that require transportation to libration points. This research and development effort resulted in the design and fabrication of the NASA-457M Hall thruster that has been tested at input powers up to 95 kW. During project year 2003, NASA established Project Prometheus to develop technology in the areas of nuclear power and propulsion, which are enabling for deep-space science missions. One of the Project-Prometheus-sponsored Nuclear Propulsion Research tasks is to investigate alternate propellants for high-power Hall thruster electric propulsion. The motivation for alternate propellants includes the disadvantageous cost and availability of xenon propellant for extremely large scale, xenon-fueled propulsion systems and the potential system performance benefits of using alternate propellants. The alternate propellant krypton was investigated because of its low cost relative to xenon. Krypton propellant also has potential performance benefits for deep-space missions because the theoretical specific impulse for a given voltage is 20 percent higher than for xenon because of krypton's lower molecular weight. During project year 2003, the performance of the high-power NASA-457M Hall thruster was measured using krypton as the propellant at power levels ranging from 6.4 to 72.5 kW. The thrust produced ranged from 0.3 to 2.5 N at a discharge specific impulse up to 4500 sec.

Jacobson, David T.↗
DOE OSTI2021

Deep Learning Based Approach to Integrate MyShake's Trigger Data with ShakeAlert for Faster and Robust EEW Alerts

Earthquake Early Warning (EEW) systems depend on the dense seismic network to make fast and accurate earthquake detections to issue useful early alerts. The recent development and implementation of the ShakeAlert system is relying on the west coast seismic network to detect and estimate the earthquake parameters in real-time. While working on making improvements on the current system, one potential approach is to include more seismic recordings from various sources, such as the low-cost smartphone seismic network, such as the MyShake network, or the Google Android smartphones, since these smartphone seismic networks have much more portable stations (phones) to potentially provide useful data to the system. This report summarizes the initial exploration of using a deep learning approach to combine the data from both traditional seismic stations and the smartphone data in California. Due to the heterogeneity nature of the data, we aggregate data from the phones (using simulation data), as well as that from traditional seismic stations to grid cells. By generating real-time triggering ratio grid cell maps, the designed deep learning algorithm can process the data from multiple sources and detect the earthquake faster than only using that from a traditional seismic network.

58 GEOSCIENCES↗
DOE OSTI2022

High Temperature, High Pressure Stability of Aqueous Foams for Potential Application in Enhanced Geothermal System (EGS)

Enhanced Geothermal Systems (EGS) are typically constructed by injecting high-pressure water into deep hot dry rocks (HDR) under carefully controlled conditions to create new or re-open existing fractures, which usually uses an immense quantity of water. In lieu, a more sustainable technique is to utilize aqueous foams as fracturing fluid to reduce water consumption and waste-water treatments of conventional hydraulic fracturing. Although foam-based fracturing has shown promising results in oil and gas industries, its feasibility is not demonstrated in EGS conditions that usually involve high temperature and high pressures. One potential barrier of utilizing foams in EGS applications is that foams are thermodynamically unstable and will become more unstable with increasing temperature due to liquid drainage, bubble coarsening, and coalescence. This work focuses on evaluation of the stability of selected aqueous foams under high-temperature and high-pressure conditions. Specifically, foams generated with surfactant alfa olefin sulfonate (AOS) were studied at temperature up to 200oC, while the pressure ranged between 100 psi and 1000 psi. The effect of additional stabilizing agents was also examined, including guar gum, bentonite clay, borate salt crosslinker, silicon dioxide nanoparticles (SiO2), and graphene oxide (GO) dispersion. Results showed that the stabilizing agents can enhance the foam thermal stability. Foams made with AOS and the borate salt crosslinker exhibited the longest half-life of 20 min at 200°C when the pressure was at 1000 psi. Data fitting showed that foam stability decreased exponentially as temperature increased. On the other hand, pressure increased the foam half-life which followed a power model. This study indicates that it may be possible to obtain highly stable foams at high-temperature and high-pressure conditions with appropriate stabilizing agents.

Thakore, Viren↗