Search NASA⌕ Search

SEARCH · Search NASA

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Assessment of alternative radionuclides for use in a radioisotope thermoelectric generator

238 Pu is the current fuel source for MMRTGs. This is due to favorable properties, such as high decay heat and low gamma emissions. 238 Pu supplies are dwindling rapidly, necessitating the review of potential replacement isotopes for deep space missions. This work considers nine potential replacements, Ac-227, Am-241, Cf-252, Cm-244, Cs-137, Po-210, Sr-90, Th-228, and Th-170. We have assessed performance of alternatives to 238 Pu by modeling the specific decay-heat and gamma spectra of several potential radioisotopes in SCALE-ORIGEN, and comparing these with 238 Pu. An analysis of heat output, heat stability, safety, availability and required mass, baselined against 238 Pu demonstrates that of those radioisotopes studied, 241 Am has properties that make it most conducive for 238 Pu replacement.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
NASA NTRS2003

Neuroprotection trek--the next generation: neuromodulation I. Techniques--deep brain stimulation, vagus nerve stimulation, and transcranial magnetic stimulation

Neuromodulation denotes controlled electrical stimulation of the central or peripheral nervous system. The three forms of neuromodulation described in this paper-deep brain stimulation, vagus nerve stimulation, and transcranial magnetic stimulation-were chosen primarily for their demonstrated or potential clinical usefulness. Deep brain stimulation is a completely implanted technique for improving movement disorders, such as Parkinson's disease, by very focal electrical stimulation of the brain-a technique that employs well-established hardware (electrode and pulse generator/battery). Vagus nerve stimulation is similar to deep brain stimulation in being well-established (for the treatment of refractory epilepsy), completely implanted, and having hardware that can be considered standard at the present time. Vagus nerve stimulation differs from deep brain stimulation, however, in that afferent stimulation of the vagus nerve results in diffuse effects on many regions throughout the brain. Although use of deep brain stimulation for applications beyond movement disorders will no doubt involve placing the stimulating electrode(s) in regions other than the thalamus, subthalamus, or globus pallidus, the use of vagus nerve stimulation for applications beyond epilepsy-for example, depression and eating disorders-is unlikely to require altering the hardware significantly (although stimulation protocols may differ). Transcranial magnetic stimulation is an example of an external or non-implanted, intermittent (at least given the current state of the hardware) stimulation technique, the clinical value of which for neuromodulation and neuroprotection remains to be determined.

Review, Tutorial↗
DOE OSTI2022

A Length Adaptive Algorithm-Hardware Co-design of Transformer on FPGA Through Sparse Attention and Dynamic Pipelining

Transformers are considered one of the most important deep learning models since 2018, in part because it establishes state-of-the-art (SOTA) records and could potentially replace existing Deep Neural Networks (DNNs). Despite the remarkable triumphs, the prolonged turnaround time of Transformer models is a widely recognized roadblock. The variety of sequence lengths imposes additional computing overhead where inputs need to be zero-padded to the maximum sentence length in the batch to accommodate the parallel computing platforms. This paper targets the field-programmable gate array (FPGA) and proposes a coherent sequence length adaptive algorithm–hardware co-design for Transformer acceleration. Particularly, we develop a hardware-friendly sparse attention operator and a length-aware hardware resource scheduling algorithm. The proposed sparse attention operator brings the complexity of attention-based models down to linear complexity and alleviates the off-chip memory traffic. The proposed length-aware resource hardware scheduling algorithm dynamically allocates the hardware resources to fill up the pipeline slots and eliminates bubbles for NLP tasks. Experiments show that our design has very small accuracy loss and has 80.2 × and 2.6 × speedup compared to CPU and GPU implementation, and 4 × higher energy efficiency than state-of-the-art GPU accelerator optimized via CUBLAS GEMM.

Peng, Hongwu↗
DOE OSTI2022

Cooperative Research and Development Agreement between National Energy Technology Laboratory and University of Kentucky (Abstract)

The National Energy Technology Laboratory (NETL) and University of Kentucky (Participant) will collaborate in the development of “Deep Learning Potential-based Molecular Dynamics Simulation Investigation of Chemical Conversion-Absorption Coupling of CO 2 at Air-Reactive Deep Eutectic Solvent Interface.” This project is supported by DOE-Office of Science Graduate Student Research (SCGSR) program and managed by Oak Ridge Institute for Science and Education (ORISE). Deep eutectic solvents (DESs) are promising alternative sorbents for direct air capture (DAC) CO 2 due to their low vapor pressure, low corrosion, non-toxicity, relatively low cost, and biodegradable nature. Understanding the CO 2 reaction-transport mechanisms at the interface can help develop cost-effective DAC by DESs with high capacity and stability. The collaboration will investigate the absorption and chemical conversion of CO 2 at the air-DES interface to facilitate the development of novel DAC technology using reactive DESs. The anticipated results will make DAC more efficient, allow more sustainable use of our Nation’s fossil energy resources, and advance NETL’s ongoing CO 2 capture efforts.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1982

New CCIR papers on telecommunications for deep space research

Three draft reports submitted by JPL to the International Radio Consultative Committee (CCIR) for inclusion in their 1982 edition of reports and recommendations are presented. Potential interference between deep space telecommunications and the fixed satellite and broadcasting satellite services in harmonically related bands, protection criteria and sharing considerations relating to deep space research, and preferred frequencies and bandwidths for deep space research are considered.

Degroot, N. F.↗
DOE OSTI2023

Water dissociation at the water–rutile TiO 2 (110) interface from ab initio-based deep neural network simulations

The interaction of water with TiO 2 surfaces is of crucial importance in various scientific fields and applications, from photocatalysis for hydrogen production and the photooxidation of organic pollutants to self-cleaning surfaces and bio-medical devices. In particular, the equilibrium fraction of water dissociation at the TiO 2 –water interface has a critical role in the surface chemistry of TiO 2 , but is difficult to determine both experimentally and computationally. Among TiO 2 surfaces, rutile TiO 2 (110) is of special interest as the most abundant surface of TiO 2 ’s stable rutile phase. While surface-science studies have provided detailed information on the interaction of rutile TiO 2 (110) with gas-phase water, much less is known about the TiO 2 (110)–water interface, which is more relevant to many applications. In this work, we characterize the structure of the aqueous TiO 2 (110) interface using nanosecond timescale molecular dynamics simulations with ab initio-based deep neural network potentials that accurately describe water/TiO 2 (110) interactions over a wide range of water coverages. Simulations on TiO 2 (110) slab models of increasing thickness provide insight into the dynamic equilibrium between molecular and dissociated adsorbed water at the interface and allow us to obtain a reliable estimate of the equilibrium fraction of water dissociation. We find a dissociation fraction of 22 ± 6% with an associated average hydroxyl lifetime of 7.6 ± 1.8 ns. These quantities are both much larger than corresponding estimates for the aqueous anatase TiO 2 (101) interface, consistent with the higher water photooxidation activity that is observed for rutile relative to anatase.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1992

Ozone production potential following convective redistribution of biomass burning emissions

The effects of deep convection on the potential for forming ozone in the free troposphere have been simulated for regions where the trace gas composition is influenced by biomass burning. Cloud photochemical and dynamic simulations based on observations in the 1980 and 1985 Brazilian campaigns form the basis of a sensitivity study of the ozone production potential under differing conditions. It is seen that there is considerably more ozone formed in the middle and upper troposphere when convection has redistributed hydrocarbons, NO(x), and CO compared to the example of no convection.

Pickering, Kenneth E.↗
NASA NTRS2013

Does Aspartic Acid Racemization Constrain the Depth Limit of the Subsurface Biosphere?

Previous studies of the subsurface biosphere have deduced average cellular doubling times of hundreds to thousands of years based upon geochemical models. We have directly constrained the in situ average cellular protein turnover or doubling times for metabolically active micro-organisms based on cellular amino acid abundances, D/L values of cellular aspartic acid, and the in vivo aspartic acid racemization rate. Application of this method to planktonic microbial communities collected from deep fractures in South Africa yielded maximum cellular amino acid turnover times of approximately 89 years for 1 km depth and 27 C and 1-2 years for 3 km depth and 54 C. The latter turnover times are much shorter than previously estimated cellular turnover times based upon geochemical arguments. The aspartic acid racemization rate at higher temperatures yields cellular protein doubling times that are consistent with the survival times of hyperthermophilic strains and predicts that at temperatures of 85 C, cells must replace proteins every couple of days to maintain enzymatic activity. Such a high maintenance requirement may be the principal limit on the abundance of living micro-organisms in the deep, hot subsurface biosphere, as well as a potential limit on their activity. The measurement of the D/L of aspartic acid in biological samples is a potentially powerful tool for deep, fractured continental and oceanic crustal settings where geochemical models of carbon turnover times are poorly constrained. Experimental observations on the racemization rates of aspartic acid in living thermophiles and hyperthermophiles could test this hypothesis. The development of corrections for cell wall peptides and spores will be required, however, to improve the accuracy of these estimates for environmental samples.

cellular↗
DOE OSTI2024

Explaining Neural Spike Activity for Simulated Bio-plausible Network through Deep Sequence Learning

With significant improvements in large-scale simulations of brain models, there is a growing need to develop tools for rapid analysis and interpreting the simulation results. In this work, we explore the potential of sequential deep learning models to understand and explain the network dynamics among the neurons extracted from a large-scale neural simulation in STACS (Simulation Tool for Asynchronous Cortical Stream). Our method employs a representative neuroscience model that abstracts the cortical dynamics with a reservoir of randomly connected spiking neurons with a low stable spike firing rate throughout the simulation duration. We subsequently analyze the spike dynamics of the simulated spiking neural network through an autoencoder model and an attention-based mechanism.

Kulkarni, Shruti↗
DOE OSTI2023

Techno-Economic Analysis for a Potential Geothermal District Heating System in Tuttle, Oklahoma

Geothermal deep direct use (DDU) has potential across a wide swath of the United States but is underutilized due to challenging project economics associated with developing a deep geothermal resource for a large-scale and variable heat demand. The National Renewable Energy Laboratory (NREL) and University of Oklahoma evaluated the feasibility of a geothermal district heating (GDH) and cooling system in two schools and 250 houses by utilizing existing oil and gas (O&G) wells in Tuttle, Oklahoma. Heating and cooling demand in the two schools and a typical single-family house were modeled using EnergyPlus building energy simulation software. The modeling results indicated that annual heating demand in two schools and 250 houses is approximately 2.61 GWhth, and cooling demand in the two schools is approximately 2.65 GWhth. In this scope, the techno-economic analysis (TEA) was conducted using the GEOPHIRES tool combined with the TOUGH2 reservoir simulator. The reservoir performance, including geothermal heat production capacity, was modeled by the reservoir simulator TOUGH2. Then, levelized cost of heat (LCOH) was calculated using GEOPHIRES version 3.0, which includes new features such as hourly heat load optimization and peak performance evaluation. Geothermal reservoir temperature was estimated as 90.5 degrees C at a total depth of 3.3 km by the regional average temperature gradient of 22.8 degrees C/km and validated by cation geothermometer calculations. Four production scenarios with two different well configurations and two different heat load profiles have been developed for well flow rates ranging between 3.1 kg/s and 9.3 kg/s. The LCOH of the district heating and cooling system was calculated between $95 and $210/MWh ($28/MMBtu to $62/MMBtu) for four different production scenarios. Typical natural gas prices for residential customers in Oklahoma have ranged from 9 to 19 $/MMBtu over the past decade, which indicates a challenge for deployment of such a GDH system.

deep direct-use↗
DOE OSTI2023

Molecular dynamics simulation of Fe-Si alloys using a neural network machine learning potential

Interatomic potential development using machine learning (ML) approaches has attracted a lot of attention in recent years because these potentials can effectively describe the structural and dynamical properties of complex materials at the atomistic level. Here, in this work, we present the development of a neural network (NN) deep ML interatomic potential for Fe-Si alloys, and we demonstrate the effectiveness of the NN-ML potential in predicting the structures and energies of liquid and crystalline phases of Fe-Si alloys in comparison with the results from ab initio molecular dynamics simulations or experimental data. The developed NN-ML potential is also used to perform molecular dynamics simulations to study the structures of Fe-Si alloys with various compositions under rapid solidification conditions. The short-ranged orders in the rapidly solidified Fe-Si alloys are also analyzed by a cluster alignment method.

36 MATERIALS SCIENCE↗
DOE OSTI2020

GEOPHIRES Simulations for Deep Direct Use (DDU) Projects

This folder contains the GEOPHIRES codes and input files for running the base case scenarios for the six deep direct-use (DDU) projects. The six DDU projects took place during 2017-2020 and were funded by the U.S. Department of Energy Geothermal Technologies Office. They investigated the potential of geothermal deep direct-use at six locations across the country. The projects were conducted by Cornell University, West Virginia University (WVU), University of Illinois (U of IL), Sandia National Laboratory (SNL), Portland State University (PSU), and National Renewable Energy Laboratory (NREL). Four projects (Cornell, WVU, U of IL, SNL) investigated geothermal for direct heating of a local campus or community, the project by PSU considered seasonal subsurface storage of solar heating, and the NREL project investigated geothermal heating for turbine inlet cooling using absorption chillers. To allow comparison of techno-economic results across the six DDU projects, GEOPHIRES simulations were set up and conducted for each project. The GEOPHIRES code was modified for each project to simulate the local application and incorporate project-specific assumptions and results such as reservoir production temperature or financing conditions. The base case input file is included which simulates the base case conditions assumed by each project team. The levelized cost of heat (LCOH) is calculated and matches the base case LCOH reported by the project teams.

15 GEOTHERMAL ENERGY↗
NASA NTRS2012

Potential Applications of Modularity to Enable a Deep Space Habitation Capability for Future Human Exploration Beyond Low-Earth Orbit

Evaluating preliminary concepts of a Deep Space Habitat (DSH) enabling long duration crewed exploration of asteroids, the Moon, and Mars is a technically challenging problem. Sufficient habitat volumes and equipment, necessary to ensure crew health and functionality, increase propellant requirements and decrease launch flexibility to deliver multiple elements on a single launch vehicle; both of which increase overall mission cost. Applying modularity in the design of the habitat structures and subsystems can alleviate these difficulties by spreading the build-up of the overall habitation capability across several smaller parts. This allows for a more flexible habitation approach that accommodates various crew mission durations and levels of functionality. This paper provides a technical analysis of how various modular habitation approaches can impact the parametric design of a DSH with potential benefits in mass, packaging volume, and architectural flexibility. This includes a description of the desired long duration habitation capability, the definition of a baseline model for comparison, a small trade study to investigate alternatives, and commentary on potentially advantageous configurations to enable different levels of habitability. The approaches investigated include modular pressure vessel strategies, modular subsystems, and modular manufacturing approaches to habitat structure. The paper also comments upon the possibility of an integrated habitation strategy using modular components to create all short and long duration habitation elements required in the current exploration architectures.

Simon, Matthew A.↗
DOE OSTI2026

Patch-Based Convolutional Neural Networks for Multiple Microstructural Features Detection in FIB-SEM Micrographs of Irradiated Nuclear Fuel

Focused ion beam scanning electron microscopy (FIB-SEM) tomography has increasingly been utilized for acquiring three-dimensional (3D) microstructure features at the sub-micron scale in irradiated nuclear materials. This technique involves sequential ion beam slicing followed by electron beam imaging and compositional mapping using energy dispersive spectroscopy (EDS). Despite its growing use, several challenges persist. These include the time-intensive nature of data collection of EDS data, difficulties in distinguishing between various microstructures, and issues with image alignment. These challenges currently limit the broader application of FIB-SEM tomography in the field. To overcome these limitations, we propose using convolutional neural networks (CNNs) to automate microstructure identification in SEM images. Our study introduces a new framework for identifying microstructures in irradiated U-10Zr (wt. %) metallic fuel with limited annotated data. The framework includes the creation of a reliable annotated dataset with paired SEM and ground truth data from EDS maps, the applications of CNNs for microstructure identification, and the validation of model performance. Specifically, we employed the Segment Anything Model (SAM) to align SEM images with corresponding EDS maps and focused ion beam (FIB) tomography SEM data. We evaluate several models, including Patch-based U-Net, Attention U-Net, and Residual U-Net, finding that patch-based U-Net exhibits superior segmentation performance and consistency. This approach reduces reliance on EDS detectors and aids in accelerating nuclear material analysis process, highlighting the potential of advanced deep learning techniques to improve microstructural understanding in nuclear material. This is the first framework to integrate SAM and Patch-based CNN models for semantic segmentation of irradiated nuclear materials, with potential applicability to other tomography datasets.

36 - MATERIALS SCIENCE↗
DOE OSTI2021

Neural network potential for Zr–Rh system by machine learning

Zr–Rh metallic glass has enabled its many applications in vehicle parts, sports equipment and so on due to its outstanding performance in mechanical property, but the knowledge of the microstructure determining the superb mechanical property remains yet insufficient. Here, we develop a deep neural network potential of Zr–Rh system by using machine learning, which breaks the dilemma between the accuracy and efficiency in molecular dynamics simulations, and greatly improves the simulation scale in both space and time. Here, the results show that the structural features obtained from the neural network method are in good agreement with the cases in ab initio molecular dynamics simulations. Furthermore, we build a large model of 5400 atoms to explore the influences of simulated size and cooling rate on the melt-quenching process of Zr 77 Rh 23 . Our study lays a foundation for exploring the complex structures in amorphous Zr 77 Rh 23 , which is of great significance for the design and practical application.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2021

Deep Learning Segmentation of Complex Features in Atomic-Resolution Phase-Contrast Transmission Electron Microscopy Images

Phase-contrast transmission electron microscopy (TEM) is a powerful tool for imaging the local atomic structure of materials. TEM has been used heavily in studies of defect structures of two-dimensional materials such as monolayer graphene due to its high dose efficiency. However, phase-contrast imaging can produce complex nonlinear contrast, even for weakly scattering samples. It is, therefore, difficult to develop fully automated analysis routines for phase-contrast TEM studies using conventional image processing tools. For automated analysis of large sample regions of graphene, one of the key problems is segmentation between the structure of interest and unwanted structures such as surface contaminant layers. In this study, we compare the performance of a conventional Bragg filtering method with a deep learning routine based on the U-Net architecture. We show that the deep learning method is more general, simpler to apply in practice, and produces more accurate and robust results than the conventional algorithm. We provide easily adaptable source code for all results in this paper and discuss potential applications for deep learning in fully automated TEM image analysis.

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

Deep Learning for Pion Identification and Energy Calibration with the ATLAS Detector

Separating charged and neutral pions as well as calibrating the pion energy response is a core component of reconstruction in the ATLAS calorimeter. This note presents an investigation of deep learning techniques for these tasks, representing the signal in the ATLAS calorimeter layers as pixelated images. Deep learning approaches outperform the classification applied in the baseline local hadronic calibration and are able to improve the energy resolution for a wide range in particle momenta, especially for low energy pions. This work demonstrates the potential of deep-learning-based low-level hadronic calibrations to significantly improve the quality of particle reconstruction in the ATLAS calorimeter.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗