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 1,423 records · Page 79
All sourcesNASA NTRSDOE OSTI
NASA NTRS1988

The lambda point experiment in microgravity

The motivation and potential for performing very high resolution measurements of the heat capacity singularity at the lambda point of helium in microgravity conditions was briefly discussed. It is clear that tests extending deep into the asymptotic region can be performed, where the theoretical predictions take on their simplest form. This advantageous situation should lead to a major improvement in the understanding of the range of applicability of current theoretical ideas in this field. The lambda transition holds out the prospect of giving the maximum advance of any system, and with the application of cryogenic techniques, the potential of this system can be realized. The technology for the initial experiments is already developed, and results could be obtained in 1990.

Lipa, J. A.↗
DOE OSTI2013

The antibody mining toolbox: An open source tool for the rapid analysis of antibody repertoires

In vitro selection has been an essential tool in the development of recombinant antibodies against various antigen targets. Deep sequencing has recently been gaining ground as an alternative and valuable method to analyze such antibody selections. The analysis provides a novel and extremely detailed view of selected antibody populations, and allows the identification of specific antibodies using only sequencing data, potentially eliminating the need for expensive and laborious low-throughput screening methods such as enzyme-linked immunosorbant assay. The high cost and the need for bioinformatics experts and powerful computer clusters, however, have limited the general use of deep sequencing in antibody selections. Here, we describe the AbMining ToolBox, an open source software package for the straightforward analysis of antibody libraries sequenced by the three main next generation sequencing platforms (454, Ion Torrent, MiSeq). The ToolBox is able to identify heavy chain CDR3s as effectively as more computationally intense software, and can be easily adapted to analyze other portions of antibody variable genes, as well as the selection outputs of libraries based on different scaffolds. The software runs on all common operating systems (Microsoft Windows, Mac OS X, Linux), on standard personal computers, and sequence analysis of 1–2 million reads can be accomplished in 10–15 min, a fraction of the time of competing software. Use of the ToolBox will allow the average researcher to incorporate deep sequence analysis into routine selections from antibody display libraries.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

Deep Sequencing-guided Design of a High Affinity Dual Specificity Antibody to Target Two Angiogenic Factors in Neovascular Age-related Macular Degeneration

The development of dual targeting antibodies promises therapies with improved efficacy over mono-specific antibodies. Here, we engineered a Two-in-One VEGF/angiopoietin 2 antibody with dual action Fab (DAF) as a potential therapeutic for neovascular age-related macular degeneration. Crystal structures of the VEGF/angiopoietin 2 DAF in complex with its two antigens showed highly overlapping binding sites. To achieve sufficient affinity of the DAF to block both angiogenic factors, we turned to deep mutational scanning in the complementarity determining regions (CDRs). By mutating all three CDRs of each antibody chain simultaneously, we were able not only to identify affinity improving single mutations but also mutation pairs from different CDRs that synergistically improve both binding functions. Furthermore, insights into the cooperativity between mutations allowed us to identify fold-stabilizing mutations in the CDRs. The data obtained from deep mutational scanning reveal that the majority of the 52 CDR residues are utilized differently for the two antigen binding function and permit, for the first time, the engineering of several DAF variants with sub-nanomolar affinity against two structurally unrelated antigens. Here, the improved variants show similar blocking activity of receptor binding as the high affinity mono-specific antibodies against these two proteins, demonstrating the feasibility of generating a dual specificity binding surface with comparable properties to individual high affinity mono-specific antibodies.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2016

Evaluating an Exterior Insulation and Finish System for Deep Energy Retrofits

This report provides a case study of a deep energy retrofit using a site-applied exterior insulation and finish system (EIFS). Prefabricated EIFS panels are also discussed as a potential wall insulation retrofit solution. EIFSs are common in new and retrofit commercial construction. Such products typically consist of five layers installed over the top of a water resistive barrier as follows: (1) Adhesive (2) Foam insulation (3) Base coat (4) Reinforcing mesh embedded in the base coat (5) Top or finish coat. They can be applied to new or existing buildings that employ wood or masonry construction, and can utilize an insulation thickness of up to 16 in. However, such systems are rarely used in lowrise residential retrofits. Reasons for this might include a lack of demand for or knowledge of the systems in residential retrofit industry, high cost, and/or lack of suitable distribution channels serving the residential market. While the final appearance of an EIFS installation resembles stucco, it is a distinctly different system. Stucco is a generic cementitious-based material, whereas EIFSs are proprietary synthetic formulations distributed by manufacturers to a network of authorized applicators. An EIFS is composed of polymeric (organic) bonded aggregate and cement reinforced with a glass mesh. Stucco is made of inorganic cement, sand, and lime. In this project, a home was retrofitted using a site-applied 4-in.-thick EIFS. Site-specific details were developed as required for the residential retrofit application. Site work and the costs of the EIFS were documented. The case study home was modeled using version 2 of the National Renewable Energy Laboratory-developed Building Energy Optimization (BEopt) energy and cost analysis software. While a package utilizing the 4-in. EIFS is the optimal retrofit of the options analyzed in terms of annualized energy-related costs, it is higher cost than replacing the siding with midrange vinyl alone. However, compared to high-end vinyl, the EIFS package has a lower annualized energy cost. This held true for both the Islip, New York (climate zone 4) and Boston (climate zone 5) case studies. The report discusses the risks, selection criteria, interactions with other building systems, cost, performance, and other aspects of using an EIFS in a deep energy retrofit. EIFSs do not require special site safety precautions beyond general construction site precautions. Moisture design is important. Trapping moisture between two vapor/air barriers is a risk of all wall systems, including EIFS cladding applications if an interior vapor barrier exists and if there is not adequate ventilation through the drainage plane behind the insulation layer. Face-sealed approaches that rely on exposed sealants do not provide acceptable rain control or durability and are very risky. Generally, fire codes permit EIFSs for most building types and conditions.

airtightness↗
DOE OSTI2024

A Medium‐Sized Paleo‐Tsunami Reconstruction by a Deep Neural Network Processing Sedimentary Deposits

Abstract Reconstructing the magnitude and recurrence time of tsunamis, one of the most destructive and unpredictable natural hazards impacting coastal communities, is essential. While major tsunamis are the most studied due to their disastrous impact, small/medium tsunamis (SMTs) are much more frequent and can still significantly impact the coast. Therefore, SMTs potentially provide an extensive archive of information preserved in the geological record. Analyzing the deposits of small/medium paleo‐tsunamis (SMPTs) opens a window into when their direct observation was unavailable. However, deposits of SMPTs are often degraded, traditional sediment deposition inversion models might fail. Recent research has shown that Deep Neural Networks (DNN) can effectively reconstruct the flow conditions of major tsunamis from their deposits. We evaluate the effectiveness of this approach in reconstructing the characteristics of a recent medium size tsunami (2006 Java) and of a medium paleo‐tsunami (1929 Grand Banks). We successfully reconstruct the flow characteristics of the 2006 Java event and show that an inversion of comparable quality is possible for the 1929 Grand Banks tsunami, despite greater uncertainties due to the deposit degradation. Our research shows that Machine Learning has the potential to unseal the meaning of data of thousands SMPTs.

Batubo, P.↗
DOE OSTI2024

Self-Supervised Cloud Classification

Abstract Low-level marine clouds play a pivotal role in Earth’s weather and climate through their interactions with radiation, heat and moisture transport, and the hydrological cycle. These interactions depend on a range of dynamical and microphysical processes that result in a broad diversity of cloud types and spatial structures, and a comprehensive understanding of cloud morphology is critical for continued improvement of our atmospheric modeling and prediction capabilities moving forward. Deep learning has recently accelerated our ability to study clouds using satellite remote sensing, and machine learning classifiers have enabled detailed studies of cloud morphology. A major limitation of deep learning approaches to this problem, however, is the large number of hand-labeled samples that are required for training. This work applies a recently developed self-supervised learning scheme to train a deep convolutional neural network (CNN) to map marine cloud imagery to vector embeddings that capture information about mesoscale cloud morphology and can be used for satellite image classification. The model is evaluated against existing cloud classification datasets and several use cases are demonstrated, including training cloud classifiers with very few labeled samples, interrogation of the CNN’s learned internal feature representations, cross-instrument application, and resilience against sensor calibration drift and changing scene brightness. The self-supervised approach learns meaningful internal representations of cloud structures and achieves comparable classification accuracy to supervised deep learning methods without the expense of creating large hand-annotated training datasets. Significance Statement Marine clouds heavily influence Earth’s weather and climate, and improved understanding of marine clouds is required to improve our atmospheric modeling capabilities and physical understanding of the atmosphere. Recently, deep learning has emerged as a powerful research tool that can be used to identify and study specific marine cloud types in the vast number of images collected by Earth-observing satellites. While powerful, these approaches require hand-labeling of training data, which is prohibitively time intensive. This study evaluates a recently developed self-supervised deep learning method that does not require human-labeled training data for processing images of clouds. We show that the trained algorithm performs competitively with algorithms trained on hand-labeled data for image classification tasks. We also discuss potential downstream uses and demonstrate some exciting features of the approach including application to multiple satellite instruments, resilience against changing image brightness, and its learned internal representations of cloud types. The self-supervised technique removes one of the major hurdles for applying deep learning to very large atmospheric datasets.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2015

CLOUD STRUCTURE OF THE NEAREST BROWN DWARFS. II. HIGH-AMPLITUDE VARIABILITY FOR LUHMAN 16 A AND B IN AND OUT OF THE 0.99 μm FeH FEATURE

The re-emergence of the 0.99 μm FeH feature in brown dwarfs of early- to mid-T spectral type has been suggested as evidence for cloud disruption where flux from deep, hot regions below the Fe cloud deck can emerge. The same mechanism could account for color changes at the L/T transition and photometric variability. We present the first observations of spectroscopic variability of brown dwarfs covering the 0.99 μm FeH feature. We observed the spatially resolved very nearby brown dwarf binary WISE J104915.57–531906.1 (Luhman 16AB), a late-L and early-T dwarf, with Hubble Space Telescope/WFC3 in the G102 grism at 0.8–1.15 μm. We find significant variability at all wavelengths for both brown dwarfs, with peak-to-valley amplitudes of 9.3% for Luhman 16B and 4.5% for Luhman 16A. This represents the first unambiguous detection of variability in Luhman 16A. We estimate a rotational period between 4.5 and 5.5 hr, very similar to Luhman 16B. Variability in both components complicates the interpretation of spatially unresolved observations. The probability for finding large amplitude variability in any two brown dwarfs is less than 10%. Our finding may suggest that a common but yet unknown feature of the binary is important for the occurrence of variability. For both objects, the amplitude is nearly constant at all wavelengths except in the deep K i feature below 0.84 μm. No variations are seen across the 0.99 μm FeH feature. The observations lend strong further support to cloud height variations rather than holes in the silicate clouds, but cannot fully rule out holes in the iron clouds. We re-evaluate the diagnostic potential of the FeH feature as a tracer of cloud patchiness.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Digital twins and deep learning segmentation of defects in monolayer MX 2 phases

Developing methods to understand and control defect formation in nanomaterials offers a promising route for materials discovery. Monolayer MX 2 phases represent a particularly compelling case for defect engineering of nanomaterials due to the large variability in their physical properties as different defects are introduced into their structure. However, effective identification and quantification of defects remain a challenge even as high-throughput scanning transmission electron microscopy methods improve. This study highlights the benefits of employing first principles calculations to produce digital twins for training deep learning segmentation models for defect identification in monolayer MX 2 phases. Around 600 defect structures were obtained using density functional theory calculations, with each monolayer MX 2 structure being subjected to multislice simulations for the purpose of generating the digital twins. Several deep learning segmentation architectures were trained on this dataset, and their performances evaluated under a variety of conditions such as recognizing defects in the presence of unidentified impurities, beam damage, grain boundaries, and with reduced image quality from low electron doses. Further, this digital twin approach allows benchmarking different deep learning architectures on a theory dataset, which enables the study of defect classification under a broad array of finely controlled conditions. It thus opens the door to resolving the underpinning physical reasons for model shortcomings and potentially chart paths forward for automated discovery of materials defect phases in experiments.

36 MATERIALS SCIENCE↗
NASA NTRSDate not supplied

The Impact of Stable Layers on Downdrafts in Tropical Deep Convection

Previous studies have demonstrated that one pathway by which dry air can interrupt tropical cyclone intensification is by inducing downdrafts that flood the surface inflow layer with low entropy air. While work has been performed examining the relationship between dry air and downdraft intensity, very little has been done to investigate the potential for features of the convective environment, such as stable layers, to inhibit the downward progression of downdrafts. This study aims to understand the role that stable layers have on inhibiting the downward progression of downdrafts and their associated low entropy air in the context of tropical deep convection. Specifically, this study asks two key questions: can a stable layer halt the downward progression of a downdraft or merely slow its velocity and can a stable layer stop the injection of low entropy air into the boundary layer? The present study aims to answer these questions using a combination of high-altitude airborne radar and dropsonde observations, both collected within four tropical cyclones (Gaston, Hermine, Karl, and Matthew) during the 2016 NOAA SHOUT field campaign, as well as idealized simulations of tropical deep convection. Preliminary results suggest that precipitation-free downdrafts can be blocked by a stable layer while precipitation-laden downdrafts are weakened, but not stopped, by the presence of a stable layer. This presentation will primarily focus on the precipitation-laden downdrafts that can be observed via airborne radar. Additionally, the presentation will conclude with a brief discussion of the implications of downdraft/stable-layer interactions for a tropical cyclone ingesting dry environmental air.

Charles N Helms↗
DOE OSTI2022

Moving the needle: Employing deep reinforcement learning to push the boundaries of coarse-grained vaccine models

Highly mutable infectious disease pathogens (hm-IDPs) such as HIV and influenza evolve faster than the human immune system can contain them, allowing them to circumvent traditional vaccination approaches and causing over one million deaths annually. Agent-based models can be used to simulate the complex interactions that occur between immune cells and hm-IDP-like proteins (antigens) during affinity maturation—the process by which antibodies evolve. Compared to existing experimental approaches, agent-based models offer a safe, low-cost, and rapid route to study the immune response to vaccines spanning a wide range of design variables. However, the highly stochastic nature of affinity maturation and vast sequence space of hm-IDPs render brute force searches intractable for exploring all pertinent vaccine design variables and the subset of immunization protocols encompassed therein. To address this challenge, we employed deep reinforcement learning to drive a recently developed agent-based model of affinity maturation to focus sampling on immunization protocols with greater potential to improve the chosen metrics of protection, namely the broadly neutralizing antibody (bnAb) titers or fraction of bnAbs produced. Using this approach, we were able to coarse-grain a wide range of vaccine design variables and explore the relevant design space. Our work offers new testable insights into how vaccines should be formulated to maximize protective immune responses to hm-IDPs and how they can be minimally tailored to account for major sources of heterogeneity in human immune responses and various socioeconomic factors. Our results indicate that the first 3 to 5 immunizations, depending on the metric of protection, should be specially tailored to achieve a robust protective immune response, but that beyond this point further immunizations require only subtle changes in formulation to sustain a durable bnAb response.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Using pollen in turbidites for vegetation reconstructions

Turbidites, deposited by sub-aqueous gravity flows, are common in sedimentary archives worldwide and present a unique challenge and opportunity when reconstructing past vegetation through pollen analysis. When sampling pollen from a sediment core for palaeovegetation records, it is common practice to target background sediments (i.e. pelagic sediment) and avoid sampling turbidites, as they are presumed to portray a misleading picture of past vegetation. This assumption stems from our limited understanding of pollen abundance and distribution through turbidites, meaning that palynologists overlook deposits that could potentially be used to reconstruct past vegetation and climate. We present pollen assemblage and sedimentological data from four recent (<150 years) deep marine turbidite deposits from the Hikurangi Subduction Margin, Aotearoa-New Zealand, with the aim of understanding the abundance and distribution of pollen in fine-grained turbidites. We find that pollen is diluted in the bases of turbidites, but despite this dilution, the proportions of different pollen taxa remain consistent through each turbidite. These results confirm that pollen can be sampled from turbidites for palaeovegetation reconstructions and that sampling the fine-grained upper parts of turbidites will provide the best pollen recovery.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2017

Falcon: Visual analysis of large, irregularly sampled, and multivariate time series data in additive manufacturing

Flexible visual analysis of long, high-resolution, and irregularly sampled time series data from multiple sensor streams is a challenge in several domains. In the field of additive manufacturing, this capability is critical for realizing the full potential of large-scale 3D printers. Here, we propose a visual analytics approach that helps additive manufacturing researchers acquire a deep understanding of patterns in log and imagery data collected by 3D printers. Our specific goals include discovering patterns related to defects and system performance issues, optimizing build configurations to avoid defects, and increasing production efficiency. We introduce Falcon, a new visual analytics system that allows users to interactively explore large, time-oriented data sets from multiple linked perspectives. Falcon provides overviews, detailed views, and unique segmented time series visualizations, all with adjustable scale options. To illustrate the effectiveness of Falcon at providing thorough and efficient knowledge discovery, we present a practical case study involving experts in additive manufacturing and data from a large-scale 3D printer. The techniques described are applicable to the analysis of any quantitative time series, though the focus of this paper is on additive manufacturing.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Adsorption of molecular iodine and alkyl iodides from spent-nuclear-fuel-reprocessing off-gas using reduced silver mordenite

Radioactive iodine is a hazardous byproduct of spent-nuclear-fuel reprocessing that must be removed from the off-gas stream before it can be discharged. Reduced silver mordenite (Ag 0 Z) is currently the baseline material for iodine capture in the U.S. Although the performance characteristics and capture mechanisms of I 2 and CH 3 I have been established, similar investigations into long-chain organic iodides have yet to be performed. In this study, thin beds of Ag 0 Z were loaded with I 2 , CH 3 I, C 4 H 9 I, and C 12 H 25 I (5 ppm to 50 ppm) carried in a dry air stream at 150°C. The maximum iodine capacity was 105 ± 5 mg I/g Ag 0 Z for all species. Saturated Ag 0 Z samples were characterized using scanning electron microscopy, X-ray fluorescence, X-ray photoelectron spectroscopy, diffuse reflectance UV–visible spectroscopy, pair distribution function analysis, and thermogravimetric analysis. Further, near-complete Ag utilization and similar physical/chemical properties were observed for all samples. Through a comparison with previous studies and an investigation of aged Ag 0 Z, we propose that iodine species react with Ag + at exchange sites, forming α-AgI within the mordenite channels, and with surface Ag 0 nanoparticles, yielding β-/γ-AgI. The available Ag sites in the interior (Ag + ) and exterior (Ag 0 ) of mordenite determine the adsorption capacity since α-AgI formation is limited by the total pore volume. Potential iodine uptake routes were summarized for aging and non-aging environments. A scalable predictive model was implemented for deep-bed iodine removal, and predictions were in good agreement with experimental data. Sensitivity analysis suggests that iodine uptake kinetics is governed by pore diffusion.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2019

Ureadepsipeptides as ClpP Activators

Acyldepsipeptides are a unique class of antibiotics that act via allosterically dysregulated activation of the bacterial caseinolytic protease (ClpP). The ability of ClpP activators to kill nongrowing bacteria represents a new opportunity to combat deep-seated biofilm infections. However, the acyldepsipeptide scaffold is subject to rapid metabolism. In this study, we explore alteration of the potentially metabolically reactive α,β unsaturated acyl chain. Through targeted synthesis, a new class of phenyl urea substituted depsipeptide ClpP activators with improved metabolic stability is described. The ureadepsipeptides are potent activators of Staphylococcus aureus ClpP and show activity against Gram-positive bacteria, including S. aureus biofilms. These studies demonstrate that a phenyl urea motif can successfully mimic the double bond, maintaining potency equivalent to acyldepsipeptides but with decreased metabolic liability. Although removal of the double bond from acyldepsipeptides generally has a significant negative impact on potency, structural studies revealed that the phenyl ureadepsipeptides can retain potency through the formation of a third hydrogen bond between the urea and the key Tyr63 residue in the ClpP activation domain. Ureadepsipeptides represent a new class of ClpP activators with improved drug-like properties, potent antibacterial activity, and the tractability to be further optimized.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2026

Learning neural representations for X-ray ptychography reconstruction with unknown probes

X-ray ptychography provides exceptional nanoscale resolution and is widely applied in materials science, biology, and nanotechnology. However, its full potential is constrained by the critical challenge of accurately reconstructing images when the illuminating probe is unknown. Conventional iterative methods and deep learning approaches are often suboptimal, particularly under the low-signal conditions inherent to low-dose and high-speed experiments. These limitations compromise reconstruction fidelity and restrict the broader adoption of the technique. In this work, we introduce the Ptychographic Implicit Neural Representation (PtyINR), a self-supervised framework that simultaneously addresses the object- and probe-recovery problem. By parameterizing both as continuous neural representations, PtyINR performs end-to-end reconstruction directly from raw diffraction patterns without requiring any pre-characterization of the probe. Extensive evaluations demonstrate that PtyINR achieves superior reconstruction quality on both simulated and experimental data, with remarkable robustness under challenging low-signal conditions. Furthermore, PtyINR offers a generalizable, physics-informed framework for addressing probe-dependent inverse problems, making it applicable to a wide range of computational microscopy problems.

36 MATERIALS SCIENCE↗
DOE OSTI2016

The effects of nanoparticles and organic additives with controlled dispersion on dielectric properties of polymers: Charge trapping and impact excitation

This work presents a comprehensive investigation into the effects of nanoparticles and organic additives on the dielectric properties of insulating polymers using reinforced silicone rubber as a model system. TiO{sub 2} and ZrO{sub 2} nanoparticles (d = 5 nm) were well dispersed into the polymer via a bimodal surface modification approach. Organic molecules with the potential of voltage stabilization were further grafted to the nanoparticle to ensure their dispersion. These extrinsic species were found to provide deep traps for charge carriers and exhibited effective charge trapping properties at a rather small concentration (∼10{sup 17} cm{sup −3}). The charge trapping is found to have the most significant effect on breakdown strength when the electrical stressing time is long enough that most charges are trapped in the deep states. To establish a quantitative correlation between the trap depth and the molecular properties, the electron affinity and ionization energy of each species were calculated by an ab initio method and were compared with the experimentally measured values. The correlation however remains elusive and is possibly complicated by the field effect and the electronic interactions between different species that are not considered in this computation. At high field, a super-linear increase of current density was observed for TiO{sub 2} filled composites and is likely caused by impact excitation due to the low excitation energy of TiO{sub 2} compared to ZrO{sub 2}. It is reasoned that the hot charge carriers with energies greater than the excitation energy of TiO{sub 2} may excite an electron-hole pair upon collision with the NP, which later will be dissociated and contribute to free charge carriers. This mechanism can enhance the energy dissipation and may account for the retarded electrical degradation and breakdown of TiO{sub 2} composites.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

pnnl/mol_dgnn

Molecular dynamics simulations often rely on parameterized atomistic potentials, which can be inaccurate compared to ab initio simulation methods. However, dynamic ab initio simulations come with high computational cost. Deep learning techniques offer a method to obtain both high accuracy and low cost predictions of dynamics. Our method gives a prediction of the structural dynamics of a molecule accounting for the prior dynamics, giving high-accuracy predictions at low computational cost

Ashby, Michael↗