Search NASA⌕ Search

SEARCH · Search NASA

Results for “Network Models”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,693 records · Page 94

Analysis-Specific Fast Simulation at the LHC with Deep Learning

Abstract We present a fast-simulation application based on a deep neural network, designed to create large analysis-specific datasets. Taking as an example the generation of W + jet events produced in $$\sqrt{s}=$$ s = 13 TeV proton–proton collisions, we train a neural network to model detector resolution effects as a transfer function acting on an analysis-specific set of relevant features, computed at generation level, i.e., in absence of detector effects. Based on this model, we propose a novel fast-simulation workflow that starts from a large amount of generator-level events to deliver large analysis-specific samples. The adoption of this approach would result in about an order-of-magnitude reduction in computing and storage requirements for the collision simulation workflow. This strategy could help the high energy physics community to face the computing challenges of the future High-Luminosity LHC.

Chen, C.↗

Convergence of Hyperbolic Neural Networks Under Riemannian Stochastic Gradient Descent

Abstract We prove, under mild conditions, the convergence of a Riemannian gradient descent method for a hyperbolic neural network regression model, both in batch gradient descent and stochastic gradient descent. We also discuss a Riemannian version of the Adam algorithm. We show numerical simulations of these algorithms on various benchmarks.

Whiting, Wes (ORCID:0000000247505060)↗

An in vivo imaging-based assay for detecting protein interactions over a wide range of binding affinities

Identifying and characterizing protein interactions are fundamental steps towards understanding and modeling biological networks. Methods that detect protein interactions in intact cells rather than buffered solutions are likely more relevant to natural systems since molecular crowding events in the cytosol can influence the diffusion and reactivity of individual proteins. One in vivo, imaging-based method relies on the co-localization of two proteins of interest fused to DivIVA, a cell division protein from Bacillus subtilis, and green fluorescent protein (GFP). We have modified this imaging-based assay to facilitate rapid cloning by constructing new vectors encoding N- and C-terminal DivIVA or GFP molecular tag fusions based on site-specific recombination technology. The sensitivity of the assay was defined using a well-characterized protein interaction system involving the eukaryotic nuclear import receptor subunit, Importin (Imp ) and variant nuclear localization signals (NLS) representing a range of binding affinities. These data demonstrate that the modified co-localization assay is sensitive enough to detect protein interactions with Kd values that span over four orders of magnitude (1nM to 15 M). Lastly, this assay was used to confirm numerous protein interactions identified from mass spectrometry-based analyses of affinity isolates as part of an interactome mapping project in Rhodopseudomonas palustris

59 BASIC BIOLOGICAL SCIENCES↗

Striving to translate shale physics across ten orders of magnitude: What have we learned?

Shales will play an important role in the successful transition of energy from fossil-based resources to renewables in the coming decades. Aside from being a significant source of low-carbon intensity fuels, like natural gas, they also serve as geologic seals of subsurface formations that may be used to isolate nuclear waste, sequester CO 2 , or store intermittent energy (e.g., solar hydrogen). Despite their importance, shales pose significant engineering and environmental challenges due to their nanoporous structure and extreme heterogeneity that spans at least ~10 orders of magnitude in spatial scale. Two challenges inhibit a system-level understanding: (1) the physics of fluid flow and phase behavior in shales are poorly understood due to the dominant molecular interactions between minerals and fluids under confinement, and (2) the apparent lack of scale separation that prevents a reliable (closed) description of the physics at any single scale of observation. In this review, we focus on the latter issue and discuss scale translation, which in its broadest sense is transforming data or simulations from one spatiotemporal scale to another. While effective scale translation is not exclusive to shales, but all geologic porous media, the need for it is especially acute in shales given their high degree of heterogeneity. Classical theories like homogenization, while indispensable, fail when scales are not separated. Other methods, like numerical upscaling, scale-translate in only one direction: small to large, but not the reverse, called downscaling. However, the confluence of advances in three areas are bringing challenging problems such as shales within reach: increased computational power and scalable algorithms; high-resolution imaging and multi-modal data acquisition; and machine learning to process massive amounts of data. While these advances equip geoscientists with a wide array of experimental and computational tools, no individual tool can probe the entire gamut of heterogeneity in shales. Their effective use, therefore, requires an ability to bridge between various data types obtained at different scales. The aim of this review is to present a coherent account of computational and experimental methods that may be used to achieve just that, i.e., to perform scale translation. We provide a broader definition of scale translation, one that transcends classical homogenization and upscaling methods, but is consistent with them and accommodates notions like downscaling and data translation. After a brief introduction to homogenization, we review hybrid methods, numerical upscaling and its recent extensions, multiscale computing, high-resolution imaging, and machine learning. We place particular emphasis on multiscale computing and propose an algorithmic framework to bridge between the pore (micro) and Darcy (macro) scales. Throughout the paper, we draw comparisons between the various methods and highlight their (often hidden) similarities, differences, benefits, and pitfalls. We finally conclude with two case studies on shales that exemplify some of the methods presented.

58 GEOSCIENCES↗

Influence of building heat distribution temperatures on the energy performance and sizing of 5th generation district heating and cooling networks

This paper investigates the energy performance and sizing criteria of 5th generation district heating and cooling (5GDHC) networks as a function of the heat distribution temperature in the building systems connected to the district network. An energy simulation model of a 5GDHC network was developed in Modelica for a case study located in Denmark. Calculations were carried out for four different building heating systems. Simulation results show that reducing the heat distribution temperatures from 70 °C to 23 °C leads to around 40% annual electric energy savings (from 10.4 kWh/m 2 to 6.2 kWh/m 2 ) for the operation of the heat pumps. Heat distribution temperatures of 23 °C cause higher water mass flow rates through the network, leading to annual electric energy consumption for the circulation pumps that are almost doubled (from 0.16 kWh/m 2 to 0.3 kWh/m 2 ) compared to the reference case at 70 °C. Furthermore, the paper discusses how the results obtained from the Danish case study can be generalized and applied to other cases using a simplified mathematical approach. In conclusion, it is found that about 1.5% of electric energy savings can be achieved for each temperature degree reduction in the heat distribution system.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Soil porous microstructure control over soil organic matter mobility: A multimethod workflow for understanding chemistry-dependent organic matter binding in soil

Soil organic matter (SOM) has attracted a great deal of interest; particularly for its potential to mitigate human derived CO 2 emissions. Studies have demonstrated that SOM plays a critical role in carbon storage and CO 2 sequestration. However, the sorption properties of SOM, which influence its transport in pore water and stabilization within the soil, remain poorly understood. This study develops a workflow to: (1) examine compound-specific advective and diffusive transport and desorption behaviors, (2) quantify desorption rates through stop-flow and continuous-flow column experiments, and (3) evaluate the impact of soil microporosity on SOM mobility using high-resolution imaging and extractions. Intact core column experiments were conducted on Uncultivated (Natural) and Cultivated soil samples, both were arid soils, collected in Washington State. X-ray computed tomography was employed to measure porosity and pore connectivity, while Fourier-transform ion cyclotron resonance mass spectrometry was used to analyze SOM composition. The findings revealed that cultivation increased total carbon and nitrogen levels due to irrigation and fertilization, enhancing carbon capture potential in arid soils. In contrast, the Natural soil, characterized by higher porosity and connectivity, contained more oxidized carbon. Pore network analysis indicated that soil compaction in the Cultivated soil may lead to longer diffusion pathways, significantly influencing SOM transport and stability.

Hydraulic Properties↗

Structural correlations in light irradiated Ge{sub 36}Se{sub 64} amorphous films-Radial distribution function study

The effect of laser irradiation on the electronic structure of amorphous Ge{sub 36}Se{sub 64} films has been detected by studying the variation of the bond length (r) and the coordination number C{sub N}. The total distribution function T(r) of the as deposited film is characterized by the first coordination sphere corresponding to the superposition of the correlation Ge-Se and Se-Se situated at 2.53 A. The average estimated C{sub N} is 2.519. The second peak ascribed to the correlation Se-Se lies at 3.85 A showing good agreement with other published data. After irradiation, the first peak of T(r) shows a considerable shift towards a small r and a reduction of C{sub N}. On the contrary, the second neighbor data shows a slight increase of r and a great increment of C{sub N} value (5.11 before irradiation against 6.59 after irradiation). Study of the variation of both r and C{sub N} values induced by subsequent annealing of the film is also given. The relative concentrations of the Ge-Ge, Ge-Se and Se-Se bonds, as well as, the number of the GeSe{sub 4} tetrahedral per atom are calculated using the continuous random network (CRN) and the chemically order continuous random network (COCRN) models. These calculations argue the presence of Ge{sub 2}(Se{sub 1/2}){sub 6} ethane like unit in addition to Ge(Se{sub 4}){sub 1/2} even with the COCRN model. The formation of dynamical bonds during irradiation of the film under study is suggested. Correlation to volume changes during illumination studied by tight binding molecular dynamics computer simulation has been also considered.

36 MATERIALS SCIENCE↗

Beyond interpolation: Physics-inspired gating transformers for extrapolating irradiation conditions to novel nuclear fuels

The qualification of advanced nuclear fuels relies on irradiation experiments in test reactors that emulate commercial conditions. Designing these tests requires accurate prediction of key irradiation quantities, particularly heat generation rate and burnup, yet obtaining them typically involves computationally expensive multi-step simulation workflows. We propose a physics-inspired gating transformer (PIGT) that integrates an inverse-square, distance-based attenuation into the encoder representation to bias attention toward physically relevant spatial relationships while retaining data-driven flexibility. Using MiniFuel irradiation data from the High Flux Isotope Reactor at Oak Ridge National Laboratory, we benchmark against ensemble methods, feedforward and recurrent networks, convolutional models, and standard transformers. While baseline models perform well under interpolation, they exhibit a pronounced generalization gap when evaluated on fuels not included in the training set. The proposed model consistently improves extrapolative accuracy and stability, yielding the strongest performance on unseen fuel configurations. These results indicate that a lightweight physics structure embedded within attention mechanisms can substantially improve robustness, enabling more reliable surrogate predictions to accelerate the design of nuclear fuel irradiation experiments.

Fuel qualification↗

Role of water states on water uptake and proton transport in Nafion using molecular simulations and bimodal network

In this paper, using molecular simulations and a bimodal-domain network, the role of water state on Nafion water uptake and water and proton transport is investigated. Although the smaller domains provide moderate transport pathways, their effectiveness remains low due to strong, resistive water molecules/domain surface interactions. Finally, the water occupancy of the larger domains yields bulk-like water, and causes the observed transition in the water uptake and significant increases in transport properties.

30 DIRECT ENERGY CONVERSION↗

Identification of serum biomarkers for occupational medicamentosa-like dermatitis induced by trichloroethylene using mass spectrometry

Occupational medicamentosa-like dermatitis induced by trichloroethylene (OMLDT) is an autoimmune disease and it has become a serious occupational health hazard. In the present study, we collected fasting blood samples from patients with OMLDT (n = 18) and healthy volunteers (n = 33) to explore serum peptidome patterns. Peptides in sera were purified using weak cation exchange magnetic beads (MB-WCX), and analyzed by matrix-assisted laser desorption ionization time-of-flight-mass spectrometry (MALDI-TOF-MS) and ClinProTools bioinformatics software. The intensities of thirty protein/peptide peaks were significantly different between the healthy control and OMLDT patients. A pattern of three peaks (m/z 2106.3, 2134.5, and 3263.67) was selected for supervised neural network (SNN) model building to separate the OMLDT patients from the healthy controls with a sensitivity of 95.5% and a specificity of 73.8%. Furthermore, two peptide peaks of m/z 4091.61 and 4281.69 were identified as fragments of ATP-binding cassette transporter family A member 12 (ABCA12), and cationic trypsinogen (PRRS1), respectively. Our findings not only show that specific proteomic fingerprints in the sera of OMLDT patients can be served as a differentiated tool of OMLDT patients with high sensitivity and high specificity, but also reveal the novel correlation between OMLDT with ABC transports and PRRS1, which will be of potential value for clinical and mechanistic studies of OMLDT. - Highlights: • Identify 30 differential protein/peptide peaks between OMLDT and healthy control • The test sensitivity and test specificity were 95.5% and 73.8%, respectively. • ABCA12 and PRSS1 were identified as potential biomarkers in OMLDT patients.

60 APPLIED LIFE SCIENCES↗

Within-Day Recharge of Plug-In Hybrid Electric Vehicles: Energy Impact of Public Charging Infrastructure

This paper examines the role of public charging infrastructure in increasing the share of driving on electricity that plug-in hybrid electric vehicles might exhibit, thus reducing their gasoline consumption. Vehicle activity data obtained from a global positioning system tracked household travel survey in Austin, Texas, is used to estimate gasoline and electricity consumptions of plug-in hybrid electric vehicles. Drivers within-day recharging behavior, constrained by travel activities and public charger availability, is modeled. It is found that public charging offers greater fuel savings for hybrid electric vehicles s equipped with smaller batteries, by encouraging within-day recharge, and providing an extensive public charging service is expected to reduce plug-in hybrid electric vehicles gasoline consumption by more than 30% and energy cost by 10%, compared to the scenario of home charging only.

Dong, Jing↗

AutonomieAI: An efficient and deployable vehicle energy consumption estimation toolkit

Here, this paper presents AutonomieAI, a novel toolkit designed for efficient energy estimation of vehicles across diverse trip scenarios, routes, and drive cycles, applicable to a broad range of vehicle powertrain technologies. It leverages state-of-the-art Machine Learning techniques to deliver real-time energy prediction of vehicles, enabling co-simulation with transportation level system tools and opening doors for large-scale optimization at city, network or national level. Benchmark results show that AutonomieAI achieves high accuracy, with an average percentage error below 2% for most powertrain types, and computational efficiency capable of processing over 10,000 trips per second. Applications of AutonomieAI have potential to offer the flexibility to assist in solving eco-routing problems, optimize for vehicle and powertrain selection, study charging decision behavior, and optimize for charging station placement. AutonomieAI is the result of large neural network based model architectures, trained on very large and unique high fidelity vehicle simulation data. It is lightweight, deployable, efficient and has accuracy comparable to specialized and complex physics based simulation softwares.

Autonomie↗

Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals

Graph networks are a new machine learning (ML) paradigm that supports both relational reasoning and combinatorial generalization. Here, we develop universal MatErials Graph Network (MEGNet) models for accurate property prediction in both molecules and crystals. We demonstrate that the MEGNet models outperform prior ML models such as the SchNet in 11 out of 13 properties of the QM9 molecule data set. Similarly, we show that MEGNet models trained on $\sim 60,000$ crystals in the Materials Project substantially outperform prior ML models in the prediction of the formation energies, band gaps and elastic moduli of crystals, achieving better than DFT accuracy over a much larger data set. Here, we present two new strategies to address data limitations common in materials science and chemistry. First, we demonstrate a physically-intuitive approach to unify four separate molecular MEGNet models for the internal energy at 0 K and room temperature, enthalpy and Gibbs free energy into a single free energy MEGNet model by incorporating the temperature, pressure and entropy as global state inputs. Second, we show that the learned element embeddings in MEGNet models encode periodic chemical trends and can be transfer-learned from a property model trained on a larger data set (formation energies) to improve property models with smaller amounts of data (band gaps and elastic moduli).

36 MATERIALS SCIENCE↗