Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep generative models”

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 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
DOE OSTI2011

THERMAL EVOLUTION AND STRUCTURE MODELS OF THE TRANSITING SUPER-EARTH GJ 1214b

The planet GJ 1214b is the second known super-Earth with a measured mass and radius. Orbiting a quiet M star, it receives considerably less mass-loss driving X-ray and UV radiation than CoRoT-7b, so that the interior may be quite dissimilar in composition, including the possibility of a large fraction of water. We model the interior of GJ 1214b assuming a two-layer (envelope+rock core) structure where the envelope material is either H/He, pure water, or a mixture of H/He and H{sub 2}O. Within this framework, we perform models of the thermal evolution and contraction of the planet. We discuss possible compositions that are consistent with M{sub p} = 6.55 M{sub +}, R{sub p} = 2.678 R{sub +}, an age {tau} = 3-10 Gyr, and the irradiation level of the atmosphere. These conditions require that if water exists in the interior, it must remain in a fluid state, with important consequences for magnetic field generation. These conditions also require the atmosphere to have a deep isothermal region extending down to 80-800 bar, depending on composition. Our results bolster the suggestion of a metal-enriched H/He atmosphere for the planet, as we find water-world models that lack an H/He atmosphere to require an implausibly large water-to-rock ratio of more than 6:1. We instead favor an H/He/H{sub 2}O envelope with high water mass fraction ({approx}0.5-0.85), similar to recent models of the deep envelope of Uranus and Neptune. Even with these high water mass fractions in the H/He envelope, generally the bulk composition of the planet can have subsolar water:rock ratios. Dry, water-enriched, and pure water envelope models differ to an observationally significant level in their tidal Love numbers k{sub 2} of, respectively, {approx}0.018, {approx}0.15, and {approx}0.7.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Neural-based time series forecasting of loss of coolant accidents in nuclear power plants

During the last few years, deep learning in neural networks has demonstrated impressive successes in the areas of computer vision, speech and image recognition, text generation, and many others. However, sensitive engineering areas such as nuclear engineering benefited less from these efficient techniques. In this work, deep learning expert systems are utilized to model and predict time series progression of a design-basis nuclear accident, featuring a loss of coolant accident. Two major findings are accomplished in this work. First, the ability to train expert systems with high accuracy, which could help nuclear power plant operators to figure out plant responses during the accident. Second, building fast, efficient, and accurate deep models to simulate nuclear phenomena, which could be valuable to nuclear computational science. In this work, large amount of time series data is obtained from simulation tools by simulating different conditions of the base-case/nominal accident scenario. Four critical outputs/responses are monitored during the accident (e.g. temperature, pressure, break flow rate, water level). Two approaches are adopted in this work. The first approach is to use feedforward deep neural networks (DNN) to fit all time steps and outputs in a single model. The second approach is to use long short-term memory (LSTM) to fit all time steps together for each reactor response separately. Both DNN and LSTM demonstrate very good performance in predicting the test and base-case scenarios, with accuracy as low as 92% and as high as 99%, where these test scenarios are unknown to the expert systems and are not included in the model training. In addition, both approaches demonstrate a significant reduction in computational costs, as the deep expert system is able to accurately predict the accident 100,000 times faster than the original simulation tool. Given sufficient data, the methodology adopted in this study demonstrates that DNN/LSTM expert systems can be used as a decision support system to model advanced time series phenomena within nuclear power plants with high accuracy and negligible computational costs.

42 ENGINEERING↗
DOE OSTI2023

Deep learning workflow for the inverse design of molecules with specific optoelectronic properties

The inverse design of novel molecules with a desirable optoelectronic property requires consideration of the vast chemical spaces associated with varying chemical composition and molecular size. First principles-based property predictions have become increasingly helpful for assisting the selection of promising candidate chemical species for subsequent experimental validation. However, a brute-force computational screening of the entire chemical space is decidedly impossible. To alleviate the computational burden and accelerate rational molecular design, we here present an iterative deep learning workflow that combines (i) the density-functional tight-binding method for dynamic generation of property training data, (ii) a graph convolutional neural network surrogate model for rapid and reliable predictions of chemical and physical properties, and (iii) a masked language model. As proof of principle, we employ our workflow in the iterative generation of novel molecules with a target energy gap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Minority carrier lifetime in iodine-doped molecular beam epitaxy-grown HgCdTe

The minority carrier lifetime in molecular beam epitaxy grown layers of iodine-doped Hg{sub 1−x}Cd{sub x}Te (x ∼ 0.3) on CdZnTe substrates has been studied. The samples demonstrated extrinsic donor behavior for carrier concentrations in the range from 2 × 10{sup 16} cm{sup −3} to 6 × 10{sup 17} cm{sup −3} without any post-growth annealing. At a temperature of 77 K, the electron mobility was found to vary from 10{sup 4} cm{sup 2}/V s to 7 × 10{sup 3} cm{sup 2}/V s and minority carrier lifetime from 1.6 μs to 790 ns, respectively, as the carrier concentration was increased from 2 × 10{sup 16} cm{sup −3} to 6 × 10{sup 17} cm{sup −3}. The diffusion of iodine is much lower than that of indium and hence a better alternative in heterostructures such as nBn devices. The influence of carrier concentration and temperature on the minority carrier lifetime was studied in order to characterize the carrier recombination mechanisms. Measured lifetimes were also analyzed and compared with the theoretical models of the various recombination processes occurring in these materials, indicating that Auger-1 recombination was predominant at higher doping levels. An increase in deep-level generation-recombination centers was observed with increasing doping level, which suggests that the increase in deep-level trap density is associated with the incorporation of higher concentrations of iodine into the HgCdTe.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2019

ExaGAN v1.0

This software provides a basic Generative Adversarial Network (GAN) template for modeling 2D slices of N-body dark matter simulations, as part of the Exalearn project. The underlying design is a standard deep convolutional GAN, but we augment it with a technique we are calling multi-channel rescaling. This technique adds a second image channel to the output of the GAN, which represents the data in a different normalization in order to improve the quality of the results.

Harrington, Peter↗
DOE OSTI2021

NEURAL NETWORK FOR COHERENT DIFFRACTION IMAGE INVERSION

A deep neural network model plus automatic differentiation is developed for retrieving phase information from 3D coherent diffraction images. The model is implemented using Tensorflow and the training dataset is generated using physics-based atomistic simulations. Custom codes are written to handle the resampling of diffraction images to oversampling ratios appropriate for the neural network model.

CHAN, HENRY↗
NASA NTRS2008

Rapid Turnaround of Costing/Designing of Space Missions Operations

The Ground Segment Team (GST), at NASA's Jet Propulsion Laboratory in Pasadena, California, provides high-level mission operations concepts and cost estimates for projects that are in the formulation phase. GST has developed a tool to track costs, assumptions, and mission requirements, and to rapidly turnaround estimates for mission operations, ground data systems, and tracking for deep space and near Earth missions. Estimates that would often take several weeks to generate are now generated in minutes through the use of an integrated suite of cost models. The models were developed through interviews with domain experts in areas of Mission Operations, including but not limited to: systems engineering, payload operations, tracking resources, mission planning, navigation, telemetry and command, and ground network infrastructure. Data collected during interviews were converted into parametric cost models and integrated into one tool suite. The tool has been used on a wide range of missions from small Earth orbiters, to flagship missions like Cassini. The tool is an aid to project managers and mission planners as they consider different scenarios during the proposal and early development stages of their missions. The tool is also used for gathering cost related requirements and assumptions and for conducting integrated analysis of multiple missions.

Cost estimation↗
DOE OSTI2023

Uncertainty quantification of a deep learning fuel property prediction model

Deep learning models are being widely used in the field of combustion. Given the black-box nature of typical neural network based models, uncertainty quantification (UQ) is critical to ensure the reliability of predictions as well as the training datasets, and for a principled quantification of noise and its various sources. Deep learning surrogate models for predicting properties of chemical compounds and mixtures have been recently shown to be promising for enabling data-driven fuel design and optimization, with the ultimate goal of improving efficiency and lowering emissions from combustion engines. In this study, UQ is performed for a multi-task deep learning model that simultaneously predicts the research octane number (RON), Motor Octane Number (MON), and Yield Sooting Index (YSI) of pure components and multicomponent blends. The deep learning model is comprised of three smaller networks: Extractor 1, Extractor 2, and Predictor, and a mixing operator. The molecular fingerprints of individual components are encoded via Extractor 1 and Extractor 2, the mixing operator generates fingerprints for mixtures/blends based on linear mixing operation, and the predictor maps the fingerprint to the target properties. Two different classes of UQ methods, Monte Carlo ensemble methods and Bayesian neural networks (BNNs), are employed for quantifying the epistemic uncertainty. Combinations of Bernoulli and Gaussian distributions with DropConnect and DropOut techniques are explored as ensemble methods. All the DropConnect, DropOut and Bayesian layers are applied to the predictor network. Aleatoric uncertainty is modeled by assuming that each data point has an independent uncertainty associated with it. The results of the UQ study are further analyzed to compare the performance of BNN and ensemble methods. Although this study is confined to UQ of fuel property prediction, the methodologies are applicable to other deep learning frameworks that are being widely used in the combustion community.

33 ADVANCED PROPULSION SYSTEMS↗
NASA NTRS1998

Ma'adim Vallis Estuarine Delta in Elysium Basin and Its Relevance as a Landing Site for Exobiology Exploration on Mars

The debouche of Ma'adim Vallis in the Elysium Basin generated a transitional transported sediment structure, which planimetric shape is controlled by the enclosing topography of a deep reentrant gulf of the Basin into the highland. We defined it as an estuarine delta. The location and the importance of this estuarine delta is supported by the theoretical model of graded profile constructed for Ma'adim Vallis, and by two approaches: (i) the reconstruction of Ma'adim Vallis downstream course from Gusev to Elysium Basin, and (ii) the survey of the sediment deposit in the alleged estuary. The longitudinal graded profile of Ma'adim Vallis finds its base-level in the Elysium Basin, at a about 1000 m elevation, which is in agreement with the observed Basin shoreline. This model is supported by observational evidence of flow between the northern rim of Gusev crater, and the Elysium Basin shoreline. This downstream course of Ma'adim Vallis can be divided into three hydrogeologic regions. into three hydrogeologic regions. (a) The first region is a flooded plain (Zephiria Mensae), consisting in chaotic terrain formed by highland rocks, and disintegrated lava of the western flank of Apollinaris. Morphologic indicators of the flood process are: (1) the sediment deposit over the Gusev crater northern rim that reflects the overspilling of the crater-lake water through a 40-km wide gap provided by an ancient impact crater, (2) the tear-drop shaped feature on the northeastern flank of Apollinaris Patera, and (3) the chaotic terrain that suggest the emergence of ground water generated by the seepage of the crater lake through high-permeable broken rampart material. This underground water circulation sustained by the hydrostatic pressure of the crater-lake has likely generated a hydrothermal system in the volcanic environment of Apollinaris Patera. The stratigraphy of the flooded area is identified as Hesperian age, with occurrences of Noachian hilly individual features, and as Amazonian flooded plain and chaotic material, (b) The second region is located on the western flank of Apollinaris Patera. It is surrounded by relics of deep valleys that suggest a former downstream course of Ma'adim Vallis. The geologic setting of this region (Lucus Planum) is interpreted to be an Amazonian formation composed by the middle and lower members of the Medusae Formation., c) The third region corresponds to the convergence of the west and east branches of Ma'adirn Vallis into a deep re-entrant wide gulf that penetrates about 100 km into the highland. This topographic depression is delineated by the 1000 in elevation contour. This gulf has formed an estuarine configuration centered at 3S/190W within the Elysium Basin. This configuration has favored the formation of a estuarine sedimentary delta, because of topographically controlled lateral migration. This estuarine structure is strongly dominated by the incoming supply of Ma'adim Vallis fluvial sediment extracted from Zephiria Mensae and Lucus Planum. The obtuse-angle geometry of the estuary increases the sedimentation rate, which is higher than in the course of the channel. The sediment deposition process is governed by the estuarine water circulation. The inflowing loaded fluvial water enters the estuary as a bottom current, and mixes with the relatively less-loaded water of the receiving basin. When they mixed. the inflowing fluvial material, and the landward basin circulating water generate an accumulation of highly-diversified estuarine deposit stratification. This accumulation of material is mostly centered in the transitional zone of the delta. The sediment trapping efficiency of the estuary is function of the energy balance between the inflowing fluvial water, and the ingoing basin current. The submergence of the delta by the rising of the water-level increases the estuary water-depth, and consequently the sediment entrapment is favored. The locus of sediment accumulation moves landward in the zone of inflowing fluvial water. This results in the rising of the channel base-level, thus in the increase of the length of the longitudinal graded-profile. The sediment deposit facies of the zone A shows a generally smooth surface. The longitudinal deposit is bordered by alluvial terraces that reflect the variations of the channel level. The waning of the Elysium Basin caused the erosion of the Basin estuarine zone by small channels, this episode being characterized by dissected tear-drop shaped mesa-like morphologies in the delta. Our estuarine delta model predicts a lithostratigraphic depositional sequence associated with the water submergence and the transgression of Elysium Basin. The thickness of the estuarine sediment corresponds to the Elysium Basin levels changes relatively to the bed floor of the estuary, The depositional sequence of Ma'adim Vallis are described: (1) a pro-current filled region (A), where fluvial are longitudinally accumulated by the inflowing water, (2) inverse current from Elysium Basin (B), where fluvial and lacustrine sediments are accumulated, and (3) zone of current equilibrium (C), where the sediments are distributed as a shoreline at the boundary of the estuarine delta. The estuary sedimentology dynamics collects and keeps the record of the geologic unit material crossed by Ma'adim Vallis, and those of the lakebed deposit of Elysium Basin. The predicted mixed stratigraphic sequence from fluvial and lacustrine sediment makes this site an exceptional environment to concentrate potential multi-origin biologic records. We envision four possible strategies to explore this sedimentologic record: (1) longitudinal surface and subsurface traverses in region A to investigate outcrop levees, (2) exploration of the mesa walls in region B, (3) deep drilling hole lodging of the sequential deposits in the zones A and B, and (4) surface and subsurface exploration of the shoreline delta. The expected results for each of these strategies are: (1) in the deepest layers of region A are predicted frequent and abundant coarse material, sandy lenses lamination grading downward from sand to cobbles. Volcanic debris from the Noachian crustal Plateau unit material, hydrothermal altered rocks, carbonates, Hesperian and possibly Amazonian volcanic material, from Apollinaris Patera, altered rocks and carbonates from Zephiria Mensae are expected. As a favorable environment for inception of life, possible biological records are expected in transported rock, (2) At the surface, and subsurface (<=100 m), large deposits sandy to silted material from Elysium paleolake basin mixed with fine-grained sediments from Ma'adim Vallis are expected mostly in the upstream part of region B, (3) on the shoreline of the estuarine delta, abundant fine material from Elysium paleolake basin (evaporites, carbonates), mostly Amazonian in age are expected. The Ma'adini estuary is a favorable landing site for all the above mentioned science aspects, and .for its location. The site lies near the equator, which is favorable for the rover solar power supply, and at 1000m elevation, which is a favorable configuration for the descent system braking. Another advantage is the extent of the area of high scientific interest (33,000 sq km), which provides a good ellipse, and potential long study traverses.

Grin, E. A.↗
DOE OSTI2020

On Transfer Learning Of Neural Networks Using Bi-Fidelity Data For Uncertainty Propagation

Due to their high degree of expressiveness, neural networks have recently been used as surrogate models for mapping inputs of an engineering system to outputs of interest. Once trained, neural networks are computationally inexpensive to evaluate and remove the need for repeated evaluations of computationally expensive models in uncertainty quantification applications. However, given the highly parameterized construction of neural networks, especially deep neural networks, accurate training often requires large amounts of simulation data that may not be available in the case of computationally expensive systems. In this paper, to alleviate this issue for uncertainty propagation, we explore the application of transfer learning techniques using training data generated from both high- and low-fidelity models. We explore two strategies for coupling these two datasets during the training procedure, namely, the standard transfer learning and the bi-fidelity-weighted learning. In the former approach, a neural network model mapping the inputs to the outputs of interest is trained based on the low-fidelity data. The high-fidelity data are then used to adapt the parameters of the upper layer(s) of the low-fidelity network, or train a simpler neural network to map the output of the low-fidelity network to that of the high-fidelity model. In the latter approach, the entire low-fidelity network parameters are updated using data generated via a Gaussian process model trained with a small high-fidelity dataset. The parameter updates are performed via a variant of stochastic gradient descent with learning rates given by the Gaussian process model. Using three numerical examples, we illustrate the utility of these bi-fidelity transfer learning methods where we focus on accuracy improvement achieved by transfer learning over standard training approaches.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Decomposing sources of value for electricity and negative emissions technologies in net-zero power systems

Deep decarbonization of the US power system would require rapid deployment of variable renewable energy (VRE) resources, which are projected to provide a substantial share of electricity generation at the time of net-zero emissions. However, the exact share of generation met by VRE and the roles of other technologies in supplying key electricity services—energy and firm capacity—remain uncertain. This study employs a detailed model of the US power sector to decompose the provision and value of electricity services, including negative emissions, by technology across a range of deep decarbonization scenarios. Results indicate that while technology deployment and the share of services provided by each technology vary significantly depending on future technological and market conditions, the value composition and future roles of individual technologies remain consistent. These findings offer guidance for research and development priorities and provide insights to inform electricity policy and planning.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2015

Hubble Tarantula Treasury Project. II. The star-formation history of the starburst region NGC 2070 in 30 Doradus

We present a study of the recent star formation (SF) of 30 Doradus in the Large Magellanic Cloud (LMC) using the panchromatic imaging survey Hubble Tarantula Treasury Project. In this paper we focus on the stars within 20 pc of the center of 30 Doradus, the starburst region NGC 2070. We recovered the SF history by comparing deep optical and near-infrared color–magnitude diagrams (CMDs) with state-of-the-art synthetic CMDs generated with the latest PAdova and TRieste Stellar Evolution Code (PARSEC) models, which include all stellar phases from pre-main-sequence to post-main-sequence. For the first time in this region we are able to measure the SF using intermediate- and low-mass stars simultaneously. Our results suggest that NGC 2070 experienced prolonged activity. In particular, we find that the SF in the region (1) exceeded the average LMC rate ≈ 20 Myr ago, (2) accelerated dramatically ≈ 7 Myr ago, and (3) reached a peak value 1–3 Myr ago. We did not find significant deviations from a Kroupa initial mass function down to 0.5 M{sub ⊙}. The average internal reddening E(B−V) is found to be between 0.3 and 0.4 mag.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

Computationally Efficient Training of Deep Neural Networks via Transfer Learning

Transfer learning is a highly successful approach to train deep neural networks for customized image recognition tasks. Transfer learning leverages deep learning models that are previously trained on massive datasets and re-trains the network for a novel image recognition dataset. Typically, the advantage of transfer learning has been measured in sample efficiency, but instead, we investigate the computational efficiency of transfer learning. A good pre-trained model provides features that can be used as input to a new classifier (usually the top layers of a neural network). We show that if a good pre-trained model is selected, that training a new classifier can be much more computationally efficient than training a deep neural network without transfer learning. The first step in transfer learning is to select a pre-trained model to use as a feed-forward network for generating features. Here this selection process either uses human intuition/convenience or a methodical but computationally intensive validation loop. Instead, we would prefer a method to select the pre-trained model that will produce the best transfer results for the least amount of computation. To this end, we provide a computationally efficient metric for the fit between a pre-trained model and a novel image recognition task. The better the fit, the less computation will be needed to re-train the pre-trained model for the novel task. As machine learning becomes ubiquitous, highly-accurate trained models will proliferate and computationally efficient transfer learning methods will enable rapid development of new image recognition models.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

A global DSCOVR/EPIC-based hourly/daily shortwave radiation/PAR dataset

Accurate characterization of the global downward shortwave (SW) and photosynthetically active radiation (PAR) is fundamental for Earth system modeling and global change research. Combined with a machine-learning method, we used the Earth Polychromatic Imaging Camera (EPIC) data onboard the Deep Space Climate Observatory (DSCOVR) to generate the global SW/PAR dataset at both hourly and daily temporal scales and 0.1degree×0.1degree spatial resolution. Covering a period from June, 2015 to June (about 4 years), 2019, the dataset consists of total SW/PAR and their direct and diffuse components. Our dataset will be valuable for understanding the diurnal cycles and seasonal patterns of ecosystem processes, as well as the solar energy utilizations

14 SOLAR ENERGY↗
NASA NTRS1980

Prospects for a breakthrough in field-dependent 'propulsion'

If sufficient justification exists to start the development of a field physics laboratory, attention might well be given to the development of two advanced types of field-independent propulsion systems. These are gravimagnetic systems and field resonance systems. The first are multipurpose propulsion systems employing the 'gravitational' effects of coherent electromagnetic energy configurations. The second are deep-space propulsion systems which artificially generate an energy pattern that matches precisely, or resonates with, a virtual pattern associated with a distant space-time point. A new model of space-time/field interactions is used in the present paper to describe the potential characteristics of electromagnetic/gravitational field interactions and the performance capabilities of these two propulsion systems.

Holt, A. C.↗
DOE OSTI2024

Accelerated CO2 Storage Optimization Using Multi-Resolution Fourier Neural Operator at the Illinois Basin Decatur Project (IBDP)

This paper presents a deep learning-based approach for optimizing CO2 injection in carbon capture and storage (CCS) operations. We developed a multi-resolution machine learning model to significantly reduce data generation costs. Utilizing this proxy model, we implemented a multi-objective genetic algorithm to optimize well control during the CO2 injection process. The proposed approach was applied to the Illinois Basin Decatur Project (IBDP), successfully optimizing the CO2 injection schedule based on three key objectives: maximizing the amount of CO2 stored, maximizing sweep efficiency, and minimizing pressure increase. The use of the proxy model accelerated the optimization workflow by two orders of magnitude, while the cost of data generation for the proxy model was reduced by 90% by utilizing a coarse-scale model.

accelerated CO2 storage optimization↗
DOE OSTI2022

Composition design of high-entropy alloys with deep sets learning

High entropy alloys (HEAs) are an important material class in the development of next-generation structural materials, but the astronomically large composition space cannot be efficiently explored by experiments or first-principles calculations. Machine learning (ML) methods might address this challenge, but ML of HEAs has been hindered by the scarcity of HEA property data. In this work, the EMTO-CPA method was used to generate a large HEA dataset (spanning a composition space of 14 elements) containing 7086 cubic HEA structures with structural properties, 1911 of which have the complete elastic tensor calculated. The elastic property dataset was used to train a ML model with the Deep Sets architecture. The Deep Sets model has better predictive performance and generalizability compared to other ML models. Association rule mining was applied to the model predictions to describe the compositional dependence of HEA elastic properties and to demonstrate the potential for data-driven alloy design.

36 MATERIALS SCIENCE↗