Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Deep”

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 1,315 records · Page 73
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

DeFault: DEep‐Learning‐Based FAULT Delineation Using the IBDP Passive Seismic Data at the Decatur CO2 Storage Site

Abstract The carbon capture, utilization, and storage (CCUS) framework is an essential component in reducing greenhouse gas emissions, with its success hinging on the comprehensive knowledge of subsurface geology and geomechanics. Passive seismic event relocation and fault detection offer vital insights into subsurface structures and the ability to monitor fluid migration pathways. Accurate identification and localization of seismic events, however, face significant challenges, including the necessity for high‐quality seismic data and advanced computational methods. To address these challenges, we introduce a novel deep learning method, , specifically designed for passive seismic source relocation and fault delineating for passive seismic monitoring projects. By leveraging data domain‐adaptation, allows us to train a neural network with labeled synthetic data and apply it directly to field data. Using , the passive seismic sources are automatically clustered based on their recording time and spatial locations, and subsequently, faults and fractures are delineated accordingly. We demonstrate the efficacy of on a field case study involving injection related microseismic data from Decatur, Illinois area. Our approach accurately and efficiently relocated passive seismic events, identified faults and could aid in potential damage induced by seismicity. Our results highlight the potential of as a valuable tool for passive seismic monitoring, emphasizing its role in ensuring CCUS project safety. This research bolsters the understanding of subsurface characterization in CCUS, illustrating machine learning’s capacity to refine these methods. Ultimately, our work has significant implications for CCUS technology deployment, an essential strategy in combating climate change. Plain Language Summary In our quest to tackle climate change, we use a strategy known as carbon capture, utilization, and storage (CCUS) to keep greenhouse gases out of the atmosphere. This strategy relies heavily on our ability to understand what's happening deep under the earth's surface. To make sure we store super critical safely, we need to accurately map out the geological structure, especially faults, but this is tough without high‐quality data and complex computer programs. We've developed a new tool called “DeFault,” which uses advanced machine learning to improve how we find and map these underground features. “DeFault” is smart enough to learn from numerically simulated data and then apply what it’s learned to real‐world situations. It groups together seismic activity—tiny tremors and shifts in the earth—based on when and where they happen, which helps us spot where there might be cracks or faults. We tested “DeFault” in Illinois, where CO 2 is injected underground, and it successfully pinpointed where these tremors occurred and mapped out the faults, helping to prevent accidents accurately in the future. Our study shows that “DeFault” will be a powerful ally in making CCUS safer and more effective, especially for the Illinois Basin Decatur Project. Key Points Faults and fractures introduced by carbon storage can be monitored by passive seismicity DeFault algorithm enables an automatic process for accurate and efficient passive seismic event locating and clustering

58 GEOSCIENCES↗
DOE OSTI2024

Urban Effect on Precipitation and Deep Convective Systems Over Dallas‐Fort Worth

A range of multi-year observational data sets are used to characterize the hydroclimate of the Dallas Fort-Worth area (DFW) and to investigate the impact of urban land cover on daily accumulated precipitation, RADAR composite reflectivity (cREF), and cloud top height (CTH) during the warm season. Analyses of observational data indicate rainfall rates (RR) in a 45° annulus sector 50–100 km downwind of the city are enhanced relative to an upwind area of comparable size. Enhancement of mean precipitation intensity in this annulus sector is not observed on days with spatially averaged RR > 6 mm/day. Under some flow directions, the probability of cREF >30 dBZ, occurrence of hail, and the probability of CTH >10,000 geopotential meters are also enhanced up to 200 km downwind of DFW. Two deep convection events that passed over DFW are simulated with the Weather Research and Forecasting model using a range of microphysical schemes and evaluated using RADAR observations. Model configurations that exhibit the highest fidelity in these control simulations are used in a series of perturbation experiments where the areal extent of the city is varied between zero (replacement with grassland) and eight times its current size. These perturbation experiments indicate a non-linear response of Mesoscale Convective System properties to the urban areal extent and a very strong sensitivity to the microphysical scheme used. The impact on precipitation from the urban area, even when it is expanded to eight-times the current extent, is much less marked for deep convection with stronger synoptic forcing.

58 GEOSCIENCES↗
DOE OSTI2025

Simulating Atmospheric Processes in Earth System Models and Quantifying Uncertainties With Deep Learning Multi‐Member and Stochastic Parameterizations

Abstract Deep learning is a powerful tool to represent subgrid processes in climate models, but many application cases have so far used idealized settings and deterministic approaches. Here, we develop stochastic parameterizations with calibrated uncertainty quantification to learn subgrid convective and turbulent processes and surface radiative fluxes of a superparameterization embedded in an Earth System Model (ESM). We explore three methods to construct stochastic parameterizations: (a) a single Deep Neural Network (DNN) with Monte Carlo Dropout; (b) a multi‐member parameterization; and (c) a Variational Encoder Decoder with latent space perturbation. We show that the multi‐member parameterization improves the representation of convective processes, especially in the planetary boundary layer, compared to individual DNNs. The respective uncertainty quantification illustrates that methods (b) and (c) are advantageous compared to a dropout‐based DNN parameterization regarding the spread of convective processes. Hybrid simulations with our best‐performing multi‐member parameterizations remained challenging and crash within the first days. Therefore, we develop a pragmatic partial coupling strategy relying on the superparameterization for condensate emulation. Partial coupling reduces the computational efficiency of hybrid Earth‐like simulations but enables model stability over 5 months with our multi‐member parameterizations. However, our hybrid simulations exhibit biases in thermodynamic fields and differences in precipitation patterns. Despite this, the multi‐member parameterizations enable improvements in reproducing tropical extreme precipitation compared to a traditional convection parameterization. Despite these challenges, our results indicate the potential of a new generation of multi‐member machine learning parameterizations leveraging uncertainty quantification to improve the representation of stochasticity of subgrid effects.

Behrens, Gunnar [Deutsches Zentrum für Luft‐ und R↗
DOE OSTI2024

Bridging Hydrological Ensemble Simulation and Learning Using Deep Neural Operators

Ensemble-based simulation and learning (ESnL) has long been used in hydrology for parameter inference, but computational demands of process-based ESnL can be quite high. To address this issue, we propose a deep neural operator learning approach. Neural operators are generic machine learning algorithms that can learn functional mappings between infinite-dimensional spaces, providing a highly flexible tool for scientific machine learning. Our approach is built upon DeepONet, a specific deep neural operator, and is designed to address several common problems in hydrology, namely, model parameter estimation, prediction at ungaged locations, and uncertainty quantification. Here we demonstrate the effectiveness of our DeepONet-based workflow using an existing large model ensemble created for an eastern U.S. watershed that is instrumented with 10 streamflow gages. Results suggest DeepONet achieves high efficiency in learning an ML surrogate model from the model ensemble, with the modified Kling-Gupta Efficiency exceeding 0.9 on holdout test sets. Parameter inference, carried out using the trained DeepONet surrogate model and genetic algorithm, also yields robust results. Additionally, we formulate and train a separate DeepONet model for physics-informed, seq-to-seq streamflow forecasting, which further reduces biases in the pre-trained DeepONet surrogate model. While this study focuses primarily on a single watershed, our approach is general and may be extended to enable learning from model ensembles across multiple basins or models. Thus, this research represents a significant contribution to the application of hybrid machine learning in hydrology.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Deep Convection‐Driven Downward Transport of Trace Gases and Aerosols From the Free Troposphere to the Boundary Layer

Deep convective clouds regulate Earth's energy and moisture budgets, yet their impact on the atmospheric boundary layer (BL) composition remains underexplored. Using long-term observations from three mid-latitude sites, we show that deep convection (DC) consistently enhances nighttime surface ozone and is often accompanied by modest increases in ultrafine particle concentrations. Within the BL, the condensational growth of these transported ultrafine particles may contribute up to 60% of total cloud condensation nuclei (CCN). Mass flux calculations suggest that short-lived convective cores (∼30 min) account for ∼2% of total vertical air mass transport relative to steady entrainment, increasing to ∼13% when the trailing stratiform regions are included. These results show that DC provides an episodic but efficient pathway linking the free troposphere and BL, influencing oxidant budgets, CCN variability, and climate forcing. Accurately representing this process in climate models may help reduce uncertainties in climate projections, under both preindustrial and present-day conditions.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Sound Velocity and Grüneisen Parameter in Shock‐Melted Silica at Deep Earth Pressures

Silica is a primary component of rocky planet interiors and its melt properties are important for understanding planetary formation and differentiation, magma oceans, and the deep mantle. Although well understood in the solid state, the high-pressure behavior of liquid silica is poorly constrained at lower mantle pressures. Using laser interferometry to measure shock wave profiles, we report measured stress-density states and longitudinal sound speeds in shock-synthesized stishovite, from fused silica staring material, and across the solid-liquid phase boundary up to 154 GPa. Our results constrain completion of melt at 80 GPa and show that at pressures relevant to the deep mantles of Earth-sized, rocky planets, the Grüneisen parameter for liquid silica increases with compression. This finding is consistent with a continuous increase in Si-O coordination above six for liquid silica at core-mantle boundary relevant pressures and temperatures.

Ocampo, I. K. [Princeton University, NJ (United St↗
DOE OSTI2015

Metabolic potential of fatty acid oxidation and anaerobic respiration by abundant members of Thaumarchaeota and Thermoplasmata in deep anoxic peat

To probe the metabolic potential of abundant Archaea in boreal peats, we reconstructed two near-complete archaeal genomes, affiliated with Thaumarchaeota group 1.1c (bin Fn1, 8% abundance), which was a genomically unrepresented group, and Thermoplasmata (bin Bg1, 26% abundance), from metagenomic data acquired from deep anoxic peat layers. Each of the near-complete genomes encodes the potential to degrade long-chain fatty acids (LCFA) via β-oxidation. Fn1 has the potential to oxidize LCFA either by syntrophic interaction with methanogens or by coupling oxidation with anaerobic respiration using fumarate as a terminal electron acceptor (TEA). Fn1 is the first Thaumarchaeota genome without an identifiable carbon fixation pathway, indicating that this mesophilic phylum encompasses more diverse metabolisms than previously thought. Furthermore, we report genetic evidence suggestive of sulfite and/or organosulfonate reduction by Thermoplasmata Bg1. In deep peat, inorganic TEAs are often depleted to extremely low levels, yet the anaerobic respiration predicted for two abundant archaeal members suggests organic electron acceptors such as fumarate and organosulfonate (enriched in humic substances) may be important for respiration and C mineralization in peatlands.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2014

Continental degassing of 4He by surficial discharge of deep groundwater

Radiogenic He-4 is produced by the decay of uranium and thorium in the Earths mantle and crust. From here, it is degassed to the atmosphere(1-5) and eventually escapes to space(1,5,6). Assuming that all of the He-4 produced is degassed, about 70% of the total He-4 degassed from Earth comes from the continental crust(2,-5,7). However, the outgoing flux of crustal He-4 has not been directly measured at the Earths surface(2) and the migration pathways are poorly understood(2-4,7,8). Here we present measurements of helium isotopes and the long-lived cosmogenic radio-isotope Kr-81 in the deep, continental-scale Guarani aquifer in Brazil and show that crustal He-4 reaches the atmosphere primarily by the surficial discharge of deep groundwater. We estimate that He-4 in Guarani groundwater discharge accounts for about 20% of the assumed global flux from continental crust, and that other large aquifers may account for about 33%. Old groundwater ages suggest that He-4 in the Guarani aquifer accumulates over half- to one-million-year timescales. We conclude that He-4 degassing from the continents is regulated by groundwater discharge, rather than episodic tectonic events, and suggest that the assumed steady state between crustal production and degassing of He-4, and its resulting atmospheric residence time, should be re-examined

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Detection and imaging of chemicals and hidden explosives using terahertz time-domain spectroscopy and deep learning

Detecting concealed chemicals and explosives remains a critical challenge in global security. Terahertz time-domain spectroscopy (THz-TDS) offers a promising non-invasive and stand-off detection technique owing to its ability to penetrate optically opaque materials without causing ionization damage. While many chemicals exhibit distinct spectral features in the terahertz range, conventional terahertz-based detection methods often struggle in real-world environments, where variations in sample geometry, thickness, and packaging can lead to inconsistent spectral responses. In this study, we present a chemical imaging system that integrates THz-TDS with deep learning to enable accurate pixel-level identification and classification of different explosives. Operating in reflection mode and enhanced with plasmonic nanoantenna arrays, our THz-TDS system achieves a peak dynamic range of 96 dB and a detection bandwidth of 4.5 THz, supporting practical, stand-off operation. By analyzing individual time-domain pulses with deep neural networks, the system exhibits strong resilience to environmental variations and sample inconsistencies. Blind testing across eight chemicals—including pharmaceutical excipients and explosive compounds—resulted in an average classification accuracy of 99.42% at the pixel level. Notably, the system maintained an average accuracy of 88.83% when detecting explosives concealed under opaque paper coverings, demonstrating its robust generalization capability. These results highlight the potential of combining advanced terahertz spectroscopy with neural networks for highly sensitive and specific chemical and explosive detection in diverse and operationally relevant scenarios.

Imaging and sensing↗
DOE OSTI2018

Metastable silica high pressure polymorphs as structural proxies of deep Earth silicate melts

Modelling of processes involving deep Earth liquids requires information on their structures and compression mechanisms. However, knowledge of the local structures of silicates and silica (SiO 2 ) melts at deep mantle conditions and of their densification mechanisms is still limited. Here we report the synthesis and characterization of metastable high-pressure silica phases, coesite-IV and coesite-V, using in situ single-crystal X-ray diffraction and ab initio simulations. Their crystal structures are drastically different from any previously considered models, but explain well features of pair-distribution functions of highly densified silica glass and molten basalt at high pressure. Built of four, five-, and six-coordinated silicon, coesite-IV and coesite-V contain SiO 6 octahedra, which, at odds with 3 rd Pauling’s rule, are connected through common faces. Our results suggest that possible silicate liquids in Earth’s lower mantle may have complex structures making them more compressible than previously supposed.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Altered chemistry of oxygen and iron under deep Earth conditions

A drastically altered chemistry was recently discovered in the Fe-O-H system under deep Earth conditions, involving the formation of iron superoxide (FeO 2 Hx with x =0 to 1), but the puzzling crystal chemistry of this system at high pressures is largely unknown. Here we present evidence that despite the high O/Fe ratio in FeO 2 Hx, iron remains in the ferrous, spin-paired and non-magnetic state at 60–133 GPa, while the presence of hydrogen has minimal effects on the valence of iron. The reduced iron is accompanied by oxidized oxygen due to oxygen-oxygen interactions. The valence of oxygen is not –2 as in all other major mantle minerals, instead it varies around –1. This result indicates that like iron, oxygen may have multiple valence states in our planet’s interior. Laslty, our study suggests a possible change in the chemical paradigm of how oxygen, iron, and hydrogen behave under deep Earth conditions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Two-photon excited deep-red and near-infrared emissive organic co-crystals

Abstract Two-photon excited near-infrared fluorescence materials have garnered considerable attention because of their superior optical penetration, higher spatial resolution, and lower optical scattering compared with other optical materials. Herein, a convenient and efficient supramolecular approach is used to synthesize a two-photon excited near-infrared emissive co-crystalline material. A naphthalenediimide-based triangular macrocycle and coronene form selectively two co-crystals. The triangle-shaped co-crystal emits deep-red fluorescence, while the quadrangle-shaped co-crystal displays deep-red and near-infrared emission centered on 668 nm, which represents a 162 nm red-shift compared with its precursors. Benefiting from intermolecular charge transfer interactions, the two co-crystals possess higher calculated two-photon absorption cross-sections than those of their individual constituents. Their two-photon absorption bands reach into the NIR-II region of the electromagnetic spectrum. The quadrangle-shaped co-crystal constitutes a unique material that exhibits two-photon absorption and near-infrared emission simultaneously. This co-crystallization strategy holds considerable promise for the future design and synthesis of more advanced optical materials.

36 MATERIALS SCIENCE↗
DOE OSTI2021

A data-independent acquisition-based global phosphoproteomics system enables deep profiling

Phosphoproteomics can provide insights into cellular signaling dynamics. To achieve deep and robust quantitative phosphoproteomics profiling for minute amounts of sample, we here develop a global phosphoproteomics strategy based on data-independent acquisition (DIA) mass spectrometry and hybrid spectral libraries derived from data-dependent acquisition (DDA) and DIA data. Benchmarking the method using 166 synthetic phosphopeptides shows high sensitivity (<0.1 ng), accurate site localization and reproducible quantification (~5% median coefficient of variation). As a proof-of-concept, we use lung cancer cell lines and patient-derived tissue to construct a hybrid phosphoproteome spectral library covering 159,524 phosphopeptides (88,107 phosphosites). Based on this library, our single-shot streamlined DIA workflow quantifies 36,350 phosphosites (19,755 class 1) in cell line samples within two hours. Application to drug-resistant cells and patient-derived lung cancer tissues delineates site-specific phosphorylation events associated with resistance and tumor progression, showing that our workflow enables the characterization of phosphorylation signaling with deep coverage, high sensitivity and low between-run missing values.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

ECNet is an evolutionary context-integrated deep learning framework for protein engineering

Abstract Machine learning has been increasingly used for protein engineering. However, because the general sequence contexts they capture are not specific to the protein being engineered, the accuracy of existing machine learning algorithms is rather limited. Here, we report ECNet (evolutionary context-integrated neural network), a deep-learning algorithm that exploits evolutionary contexts to predict functional fitness for protein engineering. This algorithm integrates local evolutionary context from homologous sequences that explicitly model residue-residue epistasis for the protein of interest with the global evolutionary context that encodes rich semantic and structural features from the enormous protein sequence universe. As such, it enables accurate mapping from sequence to function and provides generalization from low-order mutants to higher-order mutants. We show that ECNet predicts the sequence-function relationship more accurately as compared to existing machine learning algorithms by using ~50 deep mutational scanning and random mutagenesis datasets. Moreover, we used ECNet to guide the engineering of TEM-1 β-lactamase and identified variants with improved ampicillin resistance with high success rates.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Cross-property deep transfer learning framework for enhanced predictive analytics on small materials data

Abstract Artificial intelligence (AI) and machine learning (ML) have been increasingly used in materials science to build predictive models and accelerate discovery. For selected properties, availability of large databases has also facilitated application of deep learning (DL) and transfer learning (TL). However, unavailability of large datasets for a majority of properties prohibits widespread application of DL/TL. We present a cross-property deep-transfer-learning framework that leverages models trained on large datasets to build models on small datasets of different properties. We test the proposed framework on 39 computational and two experimental datasets and find that the TL models with only elemental fractions as input outperform ML/DL models trained from scratch even when they are allowed to use physical attributes as input, for 27/39 (≈ 69%) computational and both the experimental datasets. We believe that the proposed framework can be widely useful to tackle the small data challenge in applying AI/ML in materials science.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Inferring protein expression changes from mRNA in Alzheimer’s dementia using deep neural networks

Identifying the molecular systems and proteins that modify the progression of Alzheimer’s disease and related dementias (ADRD) is central to drug target selection. However, discordance between mRNA and protein abundance, and the scarcity of proteomic data, has limited our ability to advance candidate targets that are mainly based on gene expression. Therefore, by using a deep neural network that predicts protein abundance from mRNA expression, here we attempt to track the early protein drivers of ADRD. Specifically, by applying the clei2block deep learning model to 1192 brain RNA-seq samples, we identify protein modules and disease-associated expression changes that were not directly observed at the mRNA level. Moreover, pseudo-temporal trajectory inference based on the predicted proteome became more closely correlated with cognitive decline and hippocampal atrophy compared to RNA-based trajectories. This suggests that the predicted changes in protein expression could provide a better molecular representation of ADRD progression. Furthermore, overlaying clinical traits on protein pseudotime trajectory identifies protein modules altered before cognitive impairment. These results demonstrate how our method can be used to identify potential early protein drivers and possible drug targets for treating and/or preventing ADRD.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Prediction of inter-chain distance maps of protein complexes with 2D attention-based deep neural networks

Residue-residue distance information is useful for predicting tertiary structures of protein monomers or quaternary structures of protein complexes. Many deep learning methods have been developed to predict intra-chain residue-residue distances of monomers accurately, but few methods can accurately predict inter-chain residue-residue distances of complexes. We develop a deep learning method CDPred (i.e., Complex Distance Prediction) based on the 2D attention-powered residual network to address the gap. Tested on two homodimer datasets, CDPred achieves the precision of 60.94% and 42.93% for top L/5 inter-chain contact predictions (L: length of the monomer in homodimer), respectively, substantially higher than DeepHomo’s 37.40% and 23.08% and GLINTER’s 48.09% and 36.74%. Tested on the two heterodimer datasets, the top Ls/5 inter-chain contact prediction precision (Ls: length of the shorter monomer in heterodimer) of CDPred is 47.59% and 22.87% respectively, surpassing GLINTER’s 23.24% and 13.49%. Moreover, the prediction of CDPred is complementary with that of AlphaFold2-multimer.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2019

Deep learning analysis of defect and phase evolution during electron beam-induced transformations in WS2

Abstract Recent advances in scanning transmission electron microscopy (STEM) allow the real-time visualization of solid-state transformations in materials, including those induced by an electron beam and temperature, with atomic resolution. However, despite the ever-expanding capabilities for high-resolution data acquisition, the inferred information about kinetics and thermodynamics of the process, and single defect dynamics and interactions is minimal. This is due to the inherent limitations of manual ex situ analysis of the collected volumes of data. To circumvent this problem, we developed a deep-learning framework for dynamic STEM imaging that is trained to find the lattice defects and apply it for mapping solid state reactions and transformations in layered WS 2 . The trained deep-learning model allows extracting thousands of lattice defects from raw STEM data in a matter of seconds, which are then classified into different categories using unsupervised clustering methods. We further expanded our framework to extract parameters of diffusion for sulfur vacancies and analyzed transition probabilities associated with switching between different configurations of defect complexes consisting of Mo dopant and sulfur vacancy, providing insight into point-defect dynamics and reactions. This approach is universal and its application to beam-induced reactions allows mapping chemical transformation pathways in solids at the atomic level.

36 MATERIALS SCIENCE↗