Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “network shape”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Source shape estimation for neutron imaging systems using convolutional neural networks

Neutron imaging systems are important diagnostic tools for characterizing the physics of inertial confinement fusion reactions at the National Ignition Facility (NIF). In particular, neutron images give diagnostic information on the size, symmetry, and shape of the fusion hot spot and surrounding cold fuel. Images are formed via collection of neutron flux from the source using a system of aperture arrays and scintillator-based detectors. Currently, reconstruction of fusion source geometry from the collected neutron images is accomplished by solving a computationally intensive maximum likelihood estimation problem via expectation maximization. In contrast, it is often useful to have simple representations of the overall source geometry that can be computed quickly. In this work, we develop convolutional neural networks (CNNs) to reconstruct the outer contours of simple source geometries. We compare the performance of the CNN for penumbral and pinhole data and provide experimental demonstrations of our methods on both non-noisy and noisy data.

Machine learning, neutron imaging, source reconstr↗
DOE OSTI2022

Water Network Shape-Dependence of Local Interactions with the Microhydrated –NO 2 – and –CO 2 – Anionic Head Groups by Cold Ion Vibrational Spectroscopy

We report the structural evolutions of water networks and solvatochromic response of the CH 3 NO 2 – radical anion in the OH and CH stretching regions by analysis of the vibrational spectra displayed by cryogenically cooled CH 3 NO 2 – ·(H 2 O) n=1–6 clusters. The OH stretching bands evolve with a surprisingly large discontinuity at n = 6, which features the emergence of an intense, strongly red-shifted band along with a weaker feature that appears in the region assigned to a free OH fundamental. Very similar behavior is displayed by the perdeuterated carboxylate clusters, RCO 2 – ·(H 2 O) n=5–7 (R = CD 3 CD 2 ), indicating that this behavior is a general feature in the microhydration of the triatomic anionic domain and not associated with CH oscillators. Electronic structure calculations trace this behavior to the formation of a “book” isomer of the water hexamer that adopts a configuration in which one of the water molecules resides in an acceptor–acceptor–donor (AAD) (A = acceptor, D = donor) H-bonding site. Excitation of the bound OH in the AAD site explores the local network topology best suited to stabilize an incipient –XO 2 H–OH–(H 2 O) 2 intracluster proton-transfer reaction. These systems thus provide particularly clear examples where the network shape controls the potential energy landscape that governs water network-mediated, intracluster proton transfer. The CH stretching bands of the CH 3 NO 2 – ·(H 2 O) n=1–6 clusters also exhibit strong solvatochromic shifts, but in this case, they smoothly blue-shift with increasing hydration with no discontinuity at n = 6. Furthermore, this behavior is analyzed in the context of the solute-ion polarizability response and partial charge transfer to the water networks.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

BUTTER - Empirical Deep Learning Dataset

The BUTTER Empirical Deep Learning Dataset represents an empirical study of the deep learning phenomena on dense fully connected networks, scanning across thirteen datasets, eight network shapes, fourteen depths, twenty-three network sizes (number of trainable parameters), four learning rates, six minibatch sizes, four levels of label noise, and fourteen levels of L1 and L2 regularization each. Multiple repetitions (typically 30, sometimes 10) of each combination of hyperparameters were preformed, and statistics including training and test loss (using a 80% / 20% shuffled train-test split) are recorded at the end of each training epoch. In total, this dataset covers 178 thousand distinct hyperparameter settings ("experiments"), 3.55 million individual training runs (an average of 20 repetitions of each experiments), and a total of 13.3 billion training epochs (three thousand epochs were covered by most runs). Accumulating this dataset consumed 5,448.4 CPU core-years, 17.8 GPU-years, and 111.2 node-years.

Array↗
NASA NTRS2003

The Temporal Configuration of Airline Networks

The deregulation of US aviation in 1978 resulted in the reconfiguration of airline networks into hub-and-spoke systems, spatially concentrated around a small number of central airports or 'hubs' through which an airline operates a number of daily waves of flights. A hub-and-spoke network requires a concentration of traffic in both space and time. In contrast to the U.S. airlines, European airlines had entered the phase of spatial network concentration long before deregulation. Bilateral negotiation of traffic fights between governments forced European airlines to focus their networks spatially on small number of 'national' airports. In general, these star-shaped networks were not coordinated in time. Transfer opportunities at central airports were mostly created 'by accident'. With the deregulation of the EU air transport market from 1988 on, a second phase of airline network concentration started. European airlines concentrated their networks in time by adopting or intensifying wave-system structures in their flight schedules. Temporal concentration may increase the competitive position of the network in a deregulated market because of certain cost and demand advantages.

Burghouwt, Guillaume↗
DOE OSTI2022

BUTTER-E - Energy Consumption Data for the BUTTER Empirical Deep Learning Dataset

The BUTTER-E - Energy Consumption Data for the BUTTER Empirical Deep Learning Dataset adds node-level energy consumption data from watt-meters to the primary sweep of the BUTTER - Empirical Deep Learning Dataset. This dataset contains energy consumption and performance data from 63,527 individual experimental runs spanning 30,582 distinct configurations: 13 datasets, 20 sizes (number of trainable parameters), 8 network "shapes", and 14 depths on both CPU and GPU hardware collected using node-level watt-meters. This dataset reveals the complex relationship between dataset size, network structure, and energy use, and highlights the impact of cache effects. BUTTER-E is intended to be joined with the BUTTER dataset (see "BUTTER - Empirical Deep Learning Dataset on OEDI" resource below) which characterizes the performance of 483k distinct fully connected neural networks but does not include energy measurements.

Array↗
DOE OSTI2024

CryoSegNet: accurate cryo-EM protein particle picking by integrating the foundational AI image segmentation model and attention-gated U-Net

Picking protein particles in cryo-electron microscopy (cryo-EM) micrographs is a crucial step in the cryo-EM-based structure determination. However, existing methods trained on a limited amount of cryo-EM data still cannot accurately pick protein particles from noisy cryo-EM images. The general foundational artificial intelligence–based image segmentation model such as Meta’s Segment Anything Model (SAM) cannot segment protein particles well because their training data do not include cryo-EM images. Here, we present a novel approach (CryoSegNet) of integrating an attention-gated U-shape network (U-Net) specially designed and trained for cryo-EM particle picking and the SAM. The U-Net is first trained on a large cryo-EM image dataset and then used to generate input from original cryo-EM images for SAM to make particle pickings. CryoSegNet shows both high precision and recall in segmenting protein particles from cryo-EM micrographs, irrespective of protein type, shape and size. On several independent datasets of various protein types, CryoSegNet outperforms two top machine learning particle pickers crYOLO and Topaz as well as SAM itself. The average resolution of density maps reconstructed from the particles picked by CryoSegNet is 3.33 Å, 7% better than 3.58 Å of Topaz and 14% better than 3.87 Å of crYOLO. It is publicly available at https://github.com/jianlin-cheng/CryoSegNet

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

A Nickel-Decorated Carbon Flower/Sulfur Cathode for Lean-Electrolyte Lithium–Sulfur Batteries

Lithium–sulfur (Li–S) batteries involve a reversible conversion reaction between sulfur and lithium sulfide (Li 2 S) via a series of soluble lithium polysulfide intermediates (LiPSs), enabling a high theoretical specific capacity of 1675 mAh g –1 . However, this process exhibits large polarization and low sulfur utilization and suffers critical capacity fade. The primary approach to tackle the problem has so far been to infiltrate sulfur into nanostructured carbon. Furthermore, most studies using porous carbon as host materials have tested with high electrolyte to sulfur ratios (E/S) (generally > 15 µL mg –1 ) that compromise the cell-level energy density. Here, a flower-shaped porous carbon structure with nickel nanoparticles that can address the problems discussed above is designed. First, the 3D flower-shaped carbon structure enables short ionic transport lengths. Second, the small pore diameters <10 nm and high specific surface areas > 3300 m 2 g –1 with sufficient pore volume are ideal for charging performance for low E/S ratios. Finally, Ni nanoparticles are employed onto the flower-shaped network to improve the reaction kinetics. Collectively, it is successfully demonstrated that the batteries with a high mass loading of 5 mg cm –2 and a 5 µL mg –1 E/S ratio can retain cycle retention of 70% after 150 cycles.

25 ENERGY STORAGE↗
NASA NTRS2005

ALI (Autonomous Lunar Investigator): Revolutionary Approach to Exploring the Moon with Addressable Reconfigurable Technology

Addressable Reconfigurable Technology (ART) based structures: Mission Concepts based on Addressable Reconfigurable Technology (ART), originally studied for future ANTS (Autonomous Nanotechnology Swarm) Space Architectures, are now being developed as rovers for nearer term use in lunar and planetary surface exploration. The architecture is based on the reconfigurable tetrahedron as a building block. Tetrahedra are combined to form space-filling networks, shaped for the required function. Basic structural components are highly modular, addressable arrays of robust nodes (tetrahedral apices) from which highly reconfigurable struts (tetrahedral edges), acting as supports or tethers, are efficiently reversibly deployed/stowed, transforming and reshaping the structures as required.

Clark, P. E.↗
DOE OSTI2024

3D Printing ofThermally Responsive Shape Memory LiquidCrystalline Epoxy Networks

Abstract A two-component liquid crystalline epoxy network (LCEN) with shape memory behavior was developed and evaluated as a candidate material for 3D printing. The cure kinetics of the uncured material and the shape memory properties of the cured LCEN were investigated by using parallel plate rheology and dynamic mechanical analysis, respectively. A commercially available fumed silica additive was introduced to the neat, uncured material to improve the rheological properties for 3D printing. The addition of fumed silica was found to increase the yield stress, shear-thinning behavior, and toughness of the uncured epoxy ink. Polarized light microscopy, differential scanning calorimetry, and wide-angle X-ray scattering measurements between the neat and additive-modified LCEN suggested a reduction in liquid crystalline alignment in the modified LCEN, owing to interactions between crystalline domains and fumed silica, which in turn influenced the mechanical behavior. Overall, the additive was found to be successful in preserving the shape memory properties of LCEN while improving its printability.

Chemistry↗
DOE OSTI2021

Shape Persistent, Highly Conductive Ionogels from Ionic Liquids Reinforced with Cellulose Nanocrystal Network

Abstract Shape‐persistent, conductive ionogels where both mechanical strength and ionic conductivity are enhanced are developed using multiphase materials composed of cellulose nanocrystals and hyperbranched polymeric ionic liquids (PILs) as a mechanically strong supporting network matrix for ionic liquids with an interrupted ion‐conducting pathway. The integration of needlelike nanocrystals and PIL promotes the formation of multiple hydrogen bonding and electrostatic ionic interaction capacitance, resulting in the formation of interconnected networks capable of confining a high amount of ionic liquid (≈95 wt%) without losing its self‐sustained shape. The resulting nanoporous and robust ionogels possess outstanding mechanical strength with a high compressive elastic modulus (≈5.6 MPa), comparable to that of tough, rubbery materials. Surprisingly, these rigid materials preserve the high ionic conductivity of original ionic liquids (≈7.8 mS cm −1 ), which are distributed within and supported by the nanocrystal network‐like rigid frame. On the one hand, such stable materials possess superior ionic conductivities in comparison to traditional solid electrolytes; on the other hand, the high compression resistance and shape‐persistence allow for easy handling in comparison to traditional fluidic electrolytes. The synergistic enhancement in ion transport and solid‐like mechanical properties afforded by these ionogel materials make them intriguing candidates for sustainable electrodeless energy storage and harvesting matrices.

Lee, Hansol↗
DOE OSTI2022

Coarse-grained molecular dynamics integrated with convolutional neural network for comparing shapes of temperature sensitive bottlebrushes

Quantification of shape changes in nature-inspired soft material architectures of stimuli-sensitive polymers is critical for controlling their properties but is challenging due to their softness and flexibility. Here, we have computationally designed uniquely shaped bottlebrushes of a thermosensitive polymer, poly(N-isopropylacrylamide) (PNIPAM), by controlling the length of side chains along the backbone. Coarse-grained molecular dynamics simulations of solvated bottlebrushes were performed below and above the lower critical solution temperature of PNIPAM. Conventional analyses (free volume, asphericity, etc.) show that lengths of side chains and their immediate environments dictate the compactness and bending in these architectures. We further developed 100 unique convolutional neural network models that captured molecular-level features and generated a statistically significant quantification of the similarity between different shapes. Thus, our study provides insights into the shapes of complex architectures as well as a general method to analyze them. The shapes presented here may inspire the synthesis of new bottlebrushes.

36 MATERIALS SCIENCE↗
DOE OSTI2024

The dynamics and stratigraphy of distributary channel networks (Final Technical Report)

Sedimentary reservoirs are frequently composed of the deposits of distributary channel networks. These networks are associated with coarse sediments (and high permeability) within the channel network, and relatively fine sediments (and low permeability) outside the channel network. The shape of the network is complex, and a product of the environmental boundary conditions under which it grows (Figure 1). These systems control heterogeneity at small and large scales within sedimentary basins, and understanding this heterogeneity and the links between scales represents the important scientific frontier addressed by this project.

58 GEOSCIENCES↗
NASA NTRS1988

A Method for Producing a Shaped Contour Radiation Pattern Using a Single Shaped Reflector and a Single Feed

Eliminating the corporate feed network in shaped contour beam antennas will reduce the expense, weight, and RF loss of the antenna system. One way of producing a shaped contour beam without using a feed network is to use a single shaped reflector with a single feed element. For a prescribed contour beam and feed, an optimization method for designing the reflector shape is given. As a design example, a shaped reflector is designed to produce a continental U.S. coverage (CONUS) beam. The RF performance of the shaped reflector is then verified by physical optics.

Cherrette, A. R.↗
NASA NTRS1989

A method for producing a shaped contour radiation pattern using a single shaped reflector and a single feed

Eliminating the corporate feed network in shaped contour beam antennas will reduce the expense, weight, and RF loss of the antenna system. One way of producing a shaped contour beam without using a feed network is to use a single shaped reflector with a single feed element. For a prescribed contour beam and feed, an optimization method for designing the reflector shape is given. As a design example, a shaped reflector is designed to produce a continental U.S. coverage (CONUS) beam. The RF performance of the shaped reflector is then verified by physical optics.

Cherrette, Alan R.↗
NASA NTRS2000

Efficient Simulation of Wing Modal Response: Application of 2nd Order Shape Sensitivities and Neural Networks

At the preliminary design stage of a wing structure, an efficient simulation, one needing little computation but yielding adequately accurate results for various response quantities, is essential in the search of optimal design in a vast design space. In the present paper, methods of using sensitivities up to 2nd order, and direct application of neural networks are explored. The example problem is how to decide the natural frequencies of a wing given the shape variables of the structure. It is shown that when sensitivities cannot be obtained analytically, the finite difference approach is usually more reliable than a semi-analytical approach provided an appropriate step size is used. The use of second order sensitivities is proved of being able to yield much better results than the case where only the first order sensitivities are used. When neural networks are trained to relate the wing natural frequencies to the shape variables, a negligible computation effort is needed to accurately determine the natural frequencies of a new design.

Kapania, Rakesh K.↗