Engineering Papers⌕ Search

SEARCH · Engineering Papers

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 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Using Ultrasound Image Augmentation and Ensemble Predictions to Prevent Machine-Learning Model Overfitting

Deep learning predictive models have the potential to simplify and automate medical imaging diagnostics by lowering the skill threshold for image interpretation. However, this requires predictive models that are generalized to handle subject variability as seen clinically. Here, we highlight methods to improve test accuracy of an image classifier model for shrapnel identification using tissue phantom image sets. Using a previously developed image classifier neural network—termed ShrapML—blind test accuracy was less than 70% and was variable depending on the training/test data setup, as determined by a leave one subject out (LOSO) holdout methodology. Introduction of affine transformations for image augmentation or MixUp methodologies to generate additional training sets improved model performance and overall accuracy improved to 75%. Further improvements were made by aggregating predictions across five LOSO holdouts. This was done by bagging confidences or predictions from all LOSOs or the top-3 LOSO confidence models for each image prediction. Top-3 LOSO confidence bagging performed best, with test accuracy improved to greater than 85% accuracy for two different blind tissue phantoms. This was confirmed by gradient-weighted class activation mapping to highlight that the image classifier was tracking shrapnel in the image sets. Overall, data augmentation and ensemble prediction approaches were suitable for creating more generalized predictive models for ultrasound image analysis, a critical step for real-time diagnostic deployment.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2019

Abiotic synthesis of graphite in hydrothermal vents

Deciphering the origin, age, and composition of deep marine organic carbon remains a challenge in understanding the dynamics of the marine carbon cycle. In particular, the composition of aged organic carbon and what allows its persistence in the deep ocean and in sediment is unresolved. Here, we observe that both high and low temperature hydrothermal vents at the 9° 50' N; 104° 17.5W East Pacific Rise (EPR) vent field are a source for (sub) micron-sized graphite particles. We demonstrate that commonly applied analytical techniques for quantification of organic carbon detect graphite. These analyses thereby classify graphite as either dissolved or particulate organic carbon, depending on the particle size and filtration method, and overlook its relevance as a carbon source to the deep ocean. Settling velocity calculations indicate the potential for these (sub)micron particles to become entrained in the buoyant plume and distributed far from the vent fields. Furthermore, our observations provide direct evidence for hydrothermal vents acting as a source of old carbon to the deep ocean.

58 GEOSCIENCES↗
NASA NTRS2012

Concurrent System Engineering and Risk Reduction for Dual-Band (RF/optical) Spacecraft Communications

This paper describes a system engineering approach to examining the potential for combining elements of a deep-space RF and optical communications payload, for the purpose of reducing the size, weight and power burden on the spacecraft and the mission. Figures of merit and analytical methodologies are discussed to conduct trade studies, and several potential technology integration strategies are presented. Finally, the NASA Integrated Radio and Optical Communications (iROC) project is described, which directly addresses the combined RF and optical approach.

Fielhauer, Karl, B.↗
NASA NTRSDate not supplied

Applications of Nuclear Thermal Propulsion Systems for Deep Space Science Missions

Nuclear thermal propulsion (NTP) systems occupy a unique area in the space propulsion technology landscape, due to their ability to combine moderate-to-high thrust systems normally seen in chemical propulsion systems with specific impulses that are higher than those of more traditional chemical propulsion systems. Thus, NTP systems have the potential to greatly expand our access to deep space and can enable or enhance capabilities for a variety of missions that achieve science goals as outlined in NASA’s decadal surveys. This paper leverages previous work performed on applying NTP systems for multiple science missions. It also expands the mission portfolio to additional NTP-powered science mission profiles, evaluating the system performance benefits delivered by this propulsion option. This paper will also outline efforts to improve the fidelity of the existing NTP design concepts and vehicles that are utilized, leveraging previous work on nuclear propulsion systems for human-Mars missions to enhance the fidelity of results obtained from previous science mission concept studies – specifically Triton lander, solar polar orbiter, and the interstellar medium probe missions. The work outlined in this paper will also examine additional missions that have been highlighted in the recent decadal surveys, which may include robotic missions to the ice giant Uranus, robotic missions to Venus, and missions to various Kuiper belt objects (KBOs).

K A Polzin↗
DOE OSTI2013

Oxygen vacancies and ordering of d-levels control voltage suppression in oxide cathodes: the case of spinel LiNi0.5Mn1.5O4-δ

Spinel Li-Mn rich oxides form one of the most promising classes of high voltage cathode materials for next generation Li-ion batteries for electric vehicle applications. Our simulations for spinel LiNi0.5Mn1.5O4 (LNMO) show that neutral oxygen vacancies promote formation of Ni-rich regions, which are negatively charged with respect to the lattice. This makes the electrons associated with these vacancies to localize on Mn3+ eg states of two types: shallow states in the Ni-rich regions and deep states in the Ni-poor regions. The positive electrostatic potential produced by the oxygen vacancies and the existence of the shallow and deep Mn3+ states result in appearance of the low-voltage region at high Li content and high-voltage region at low Li content. This is consistent with characteristic changes in the voltage capacity curves observed experimentally during electrochemical cycling. We propose that doping LNMO with judiciously selected cations can help to remedy voltage suppression effects. This approach may also be used to enhance the electrochemical stability of Li-Mn rich oxides, which tend to experience continuous voltage fade.

Sushko, Petr V.↗
DOE OSTI2021

Impact of improved energy resolution on DUNE sensitivity to neutrino non-standard interactions

The full physics potential of the next-generation Deep Underground Neutrino Experiment (DUNE) is still being explored. In particular, there have been some recent studies on the possibility of improving DUNE’s neutrino energy reconstruction. The main motivation is that a better determination of the neutrino energy in an event-by-event basis will translate into an improved measurement of the Dirac C P phase and other neutrino oscillation parameters. To further motivate studies and improvements on the neutrino energy reconstruction, we evaluate the impact of energy resolution at DUNE on an illustrative new physics scenario, viz. non-standard interactions (NSI) of neutrinos with matter. We show that a better energy resolution in comparison to the ones given in the DUNE conceptual and technical design reports may significantly enhance the experimental sensitivity to NSI, particularly when degeneracies are present. While a better reconstruction of the first oscillation peak helps disentangling standard C P effects from those coming from NSIs, we find that the second oscillation peak also plays a nontrivial role in improving DUNE’s sensitivity.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2014

PbO networks composed of single crystalline nanosheets synthesized by a facile chemical precipitation method

Graphical abstract: - Highlights: • Synthesis of PbO networks through a simple chemical precipitation route. • The synthesis method is rapid and low-cost. • Each network is composed of single crystalline PbO nanosheets. • A possible growth mechanism is proposed for synthesized PbO networks. - Abstract: For the field of energy storage, nanostructured lead oxide (PbO) shows immense potential for increased specific energy and deep discharge for lead acid battery technologies. In this work, PbO networks composed of single crystalline nanosheets were synthesized utilizing a simple, low cost and rapid chemical precipitation method. The PbO networks were prepared in a single reaction vessel from starting reagents of lead acetate dehydrate, ammonium hydroxide and deionized water. Lead acetate dehydrate was chosen as a reagent, as opposed to lead nitrate, to eliminate the possibility of nitrate contamination of the final product. X-ray diffraction (XRD) analysis, high resolution scanning electron microscopy (HRSEM) and high resolution transmission electron microscopy (HRTEM) analysis were used to characterize the synthesized PbO networks. The reproducible method described herein synthesized pure β-PbO (massicot) powders, with no byproducts. A possible formation mechanism for these PbO networks is proposed. The growth is found to proceed predominately in the 〈1 1 1〉 and 〈2 0 0〉 directions while being limited in the 〈0 1 1〉 direction.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Hexagonal boron nitride: Epitaxial growth and device applications

As a newest family member of the III-nitrides, BN is considered amongst the remaining frontiers in wide energy bandgap semiconductors with potentials for technologically significant applications in deep UV (DUV) optoelectronics, solid-state neutron detectors, electron emitters, single photon emitters, switching/memory devices, and super-capacitors. It was shown that it is possible to produce h-BN epilayers with high hexagonal phase purity, UV transparency, and film stoichiometry by employing nitrogen-rich growth conditions. The quasi-2D nature of h-BN supports unusually strong optical transitions near the band edge and a large exciton binding energy on the order of 0.7 eV. Since the isotope of B-10 has a large capture cross-section for thermal neutrons, h-BN is an ideal material for the fabrication of solid-state neutron detectors for special nuclear materials detection, well and geothermal logging, and medical imaging applications. Freestanding B-10 enriched h-BN (h- 10 BN) epilayers with varying thicknesses up to 200 μm have been successfully synthesized by metal organic chemical vapor deposition (MOCVD) as of this writing. By utilizing the conductivity anisotropy nature of h-BN, 1 cm 2 lateral detectors fabricated from 100 μm thick h-10BN epilayers have demonstrated a detection efficiency of 59% for thermal neutrons, which is the highest on record among all solid-state neutron detectors as of today. It was noted that high growth temperatures, long growth times and the use of sapphire substrate tend to incorporate oxygen related impurities into h- 10 BN epilayers, which strongly impacted the carrier mobility-lifetime (μτ) products and charge collection efficiencies of h- 10 BN neutron detectors. As the h-BN material technology further develops, improved carrier mobilities and μτ products will allow the fabrication of h-BN devices with enhanced performance.

42 ENGINEERING↗
DOE OSTI2026

Short‐Term Hourly Weather Forecasting Using PredRNN With Image Preprocessing

Global weather forecast models are vital tools with numerous applications, including public safety, agriculture, and transportation. Recent advancements in artificial intelligence (AI) and deep learning (DL) have shown the potential to enhance weather forecasting accuracy and speed. In this study, we developed a short-term hourly weather forecast framework with a wavelet transform function for data preprocessing and a spatiotemporal DL model, PredRNN, for predicting five surface atmospheric variables, including wind speed and direction, mean sea level pressure (MSLP), temperature, and precipitation. The framework demonstrated promising results. It produces global forecasts at 0.25° (∼25 km) with a 1-day lead time RMSE of 1.8 m/s for wind components, 180 Pa for MSLP, and 1.8 K for temperature. Although our model does not surpass state-of-the-art AI weather forecast models across all metrics, it outperforms these models in precipitation forecasting and wind prediction at short lead times and achieves comparable accuracy for MSLP. Its native hourly forecasting capability, together with training on widely accessible GPU hardware, contributes meaningfully to the advancement of accessible DL weather forecasting methods. Our work highlights the importance of integrating temporal components and data transformation techniques to improve the predictability and accuracy of weather forecasts.

Tran, Hoang [Pacific Northwest National Laboratory↗
DOE OSTI2021

Machine learning for deep elastic strain engineering of semiconductor electronic band structure and effective mass

Abstract The controlled introduction of elastic strains is an appealing strategy for modulating the physical properties of semiconductor materials. With the recent discovery of large elastic deformation in nanoscale specimens as diverse as silicon and diamond, employing this strategy to improve device performance necessitates first-principles computations of the fundamental electronic band structure and target figures-of-merit, through the design of an optimal straining pathway. Such simulations, however, call for approaches that combine deep learning algorithms and physics of deformation with band structure calculations to custom-design electronic and optical properties. Motivated by this challenge, we present here details of a machine learning framework involving convolutional neural networks to represent the topology and curvature of band structures in k -space. These calculations enable us to identify ways in which the physical properties can be altered through “deep” elastic strain engineering up to a large fraction of the ideal strain. Algorithms capable of active learning and informed by the underlying physics were presented here for predicting the bandgap and the band structure. By training a surrogate model with ab initio computational data, our method can identify the most efficient strain energy pathway to realize physical property changes. The power of this method is further demonstrated with results from the prediction of strain states that influence the effective electron mass. We illustrate the applications of the method with specific results for diamonds, although the general deep learning technique presented here is potentially useful for optimizing the physical properties of a wide variety of semiconductor materials.

Tsymbalov, Evgenii↗
DOE OSTI2024

Reactive capture and electrochemical conversion of CO 2 with ionic liquids and deep eutectic solvents

Ionic liquids (ILs) and deep eutectic solvents (DESs) have tremendous potential for reactive capture and conversion (RCC) of CO 2 due to their wide electrochemical stability window, low volatility, and high CO 2 solubility. There is environmental and economic interest in the direct utilization of the captured CO 2 using electrified and modular processes that forgo the thermal- or pressure-swing regeneration steps to concentrate CO 2 , eliminating the need to compress, transport, or store the gas. The conventional electrochemical conversion of CO 2 with aqueous electrolytes presents limited CO 2 solubility and high energy requirement to achieve industrially relevant products. Additionally, aqueous systems have competitive hydrogen evolution. In the past decade, there has been significant progress toward the design of ILs and DESs, and their composites to separate CO 2 from dilute streams. In parallel, but not necessarily in synergy, there have been studies focused on a few select ILs and DESs for electrochemical reduction of CO 2 , often diluting them with aqueous or non-aqueous solvents. The resulting electrode–electrolyte interfaces present a complex speciation for RCC. In this review, we describe how the ILs and DESs are tuned for RCC and specifically address the CO 2 chemisorption and electroreduction mechanisms. Critical bulk and interfacial properties of ILs and DESs are discussed in the context of RCC, and the potential of these electrolytes are presented through a techno-economic evaluation.

36 MATERIALS SCIENCE↗
DOE OSTI2018

Weak Neutral Current Studies with Positrons

Weak neutral current interactions with charged leptons have offered unique opportunities to study novel aspects of hadronic structure and search for physics beyond the standard model. These studies in the medium energy community have been primarily through parity-violating processes with electron beams, but with the possibility of polarized positron beams, new and complementary observables can be considered in experiments analogous to their electron counterparts. Such studies include elastic proton, deep inelastic, and electron target scattering. Potential positron neutral current experiments along with their potential physics reach, requirements, and feasibility are presented.

Riordan, Seamus↗
DOE OSTI2023

An electron paramagnetic resonance study of the electron transport in heavily Si-doped high Al content AlxGa1−xN

High Al mole fraction AlGaN is an ultrawide bandgap semiconductor with potential applications in power electronics and deep UV detectors. Although n-type material is achievable with Si-doping, the role of Si is controversial, particularly for AlxGa1−xN with x > 0.8. For this paper, AlGaN films were grown by plasma-assisted molecular beam epitaxy onto bulk AlN substrates and doped with 1018–1020 cm−3 Si. We examine electron transport in heavily Si-doped AlxGa1−xN with x ≥ 0.65 using magnetic resonance, which allows us to probe the neutral donors directly rather than the free carriers and avoids complications due to electrical contacts. Transport was studied through the temperature-dependent linewidth of the electron paramagnetic resonance (EPR) signature for the neutral donor. Analysis shows evidence of hopping conductivity in the most lightly doped samples and impurity band formation in the most heavily doped ones. The EPR results, which are consistent with Hall measurements performed on the same samples, are promising for the development of highly conducting high Al content AlGaN.

Materials Science↗
DOE OSTI2013

A STRONGLY MAGNETIZED PULSAR WITHIN THE GRASP OF THE MILKY WAY'S SUPERMASSIVE BLACK HOLE

The center of our Galaxy hosts a supermassive black hole, Sagittarius (Sgr) A*. Young, massive stars within 0.5 pc of Sgr A* are evidence of an episode of intense star formation near the black hole a few million years ago, which might have left behind a young neutron star traveling deep into Sgr A*'s gravitational potential. On 2013 April 25, a short X-ray burst was observed from the direction of the Galactic center. With a series of observations with the Chandra and the Swift satellites, we pinpoint the associated magnetar at an angular distance of 2.4 ± 0.3 arcsec from Sgr A*, and refine the source spin period and its derivative (P = 3.7635537(2) s and P-dot = 6.61(4)×10{sup -12} s s{sup –1}), confirmed by quasi simultaneous radio observations performed with the Green Bank Telescope and Parkes Radio Telescope, which also constrain a dispersion measure of DM = 1750 ± 50 pc cm{sup –3}, the highest ever observed for a radio pulsar. We have found that this X-ray source is a young magnetar at ≈0.07-2 pc from Sgr A*. Simulations of its possible motion around Sgr A* show that it is likely (∼90% probability) in a bound orbit around the black hole. The radiation front produced by the past activity from the magnetar passing through the molecular clouds surrounding the Galactic center region might be responsible for a large fraction of the light echoes observed in the Fe fluorescence features.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Deep medical image analysis with representation learning and neuromorphic computing

Deep learning is increasingly used in medical imaging, improving many steps of the processing chain, from acquisition to segmentation and anomaly detection to outcome prediction. Yet significant challenges remain: (i) image-based diagnosis depends on the spatial relationships between local patterns, something convolution and pooling often do not capture adequately; (ii) data augmentation, the de facto method for learning three-dimensional pose invariance, requires exponentially many points to achieve robust improvement; (iii) labelled medical images are much less abundant than unlabelled ones, especially for heterogeneous pathological cases; and (iv) scanning technologies such as magnetic resonance imaging can be slow and costly, generally without online learning abilities to focus on regions of clinical interest. To address these challenges, novel algorithmic and hardware approaches are needed for deep learning to reach its full potential in medical imaging.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2013

DeepCwind Consortium Research Program: January 15, 2010 - March 31, 2013 (Final Technical Report)

This is the final technical report for the U.S. Department of Energy-funded program, DE-0002981: DeepCwind Consortium Research Program. The project objective was the partial validation of coupled models and optimization of materials for offshore wind structures. The United States has a great opportunity to harness an indigenous abundant renewable energy resource: offshore wind. In 2010, the National Renewable Energy Laboratory (NREL) estimated there to be over 4,000 GW of potential offshore wind energy found within 50 nautical miles of the US coastlines (Musial and Ram, 2010). The US Energy Information Administration reported the total annual US electric energy generation in 2010 was 4,120 billion kilowatt-hours (equivalent to 470 GW) (US EIA, 2011), slightly more than 10% of the potential offshore wind resource. In addition, deep water offshore wind is the dominant US ocean energy resource available comprising 75% of the total assessed ocean energy resource as compared to wave and tidal resources (Musial, 2008). Through these assessments it is clear offshore wind can be a major contributor to US energy supplies. The caveat to capturing offshore wind along many parts of the US coast is deep water. Nearly 60%, or 2,450 GW, of the estimated US offshore wind resource is located in water depths of 60 m or more (Musial and Ram, 2010). At water depths over 60 m building fixed offshore wind turbine foundations, such as those found in Europe, is likely economically infeasible (Musial et al., 2006). Therefore floating wind turbine technology is seen as the best option for extracting a majority of the US offshore wind energy resource. Volume 1 - Test Site; Volume 2 - Coupled Models; and Volume 3 - Composite Materials.

17 WIND ENERGY↗
DOE OSTI2018

Testing the possibility of magnetic domain imaging based on circular & linear dichroism using photoemission electron microscopy

In recent years, an increasing number of memory and spintronic devices have been developed exploiting the combination of ferromagnetic (FM) and anti-ferromagnetic (aFM) materials. Consequently, magnetic imaging based on continuous-wave (CW) ultraviolet (UV) (λ = 266nm and longer wavelength) photoemission electron microscopy (PEEM) is gaining considerable attention due to the possibility of determining magnetizations for FM and aFM materials with 10 nm lateral resolution at video rate image acquisition. This PEEM-based approach exploits the polarization-dependent photoemission yield, which is subject to the polarization vector and the FM or aFM magnetization direction. Because of this unique attribute, magnetic imaging using PEEM when coupled to a laser with multiple illumination geometries allows for characterizing in-plane and out-of-plane magnetizations. This concept, however, has not been tested using a deep-UV laser (λ = 213nm), which has a much broader application space than the longer wavelength excitation used in previous reports. The purpose of this project in FY17 was to show the proof-of-concept of magnetic circular dichroism (MCD)-PEEM imaging using a λ = 210nm pulsed laser. Our results demonstrated the feasibility of in-plane and out-of-plane magnetic imaging with the limitations in the lateral resolution, data acquisition time, and signal-to-noise ratio anticipated for using a pulsed laser of moderate power. The project goal for FY18 is to construct the automated polarization-controlled data acquisition, and to establish the new lab facility in anticipation of acquiring a state-of-the-art high-power 213nm CW laser, planned to be installed in FY19. We successfully demonstrate the former by measuring dielectric stacks with polarization-dependent photoemission yield. Extrapolating from our result, we conclude that the capability of PEEM-based magnetic imaging using a CW deep UV laser could be a potential game-changer for scientific investigations and technological developments of magnetic materials and spintronic devices. In addition, polarization controlled PEEM imaging shows the potential for ellipsometry imaging of embedded nanomaterials exploiting their subtle differences in optical constants with respect to their surrounding dielectrics.

36 MATERIALS SCIENCE↗