Search NASA⌕ Search

SEARCH · Search NASA

Results for “PROCESS”

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 685 records · Page 38

Enhanced Data Efficiency Using Deep Neural Networks and Gaussian Processes for Aerodynamic Design Optimization

Adjoint-based optimization methods are attractive for aerodynamic shape design primarily due to their computational costs being independent of the dimensionality of the input space and their ability to generate high-fidelity gradients that can then be used in a gradient-based optimizer. This makes them very well suited for high-fidelity simulation based aerodynamic shape optimization of highly parametrized geometries such as aircraft wings. However, the development of adjoint-based solvers involve careful mathematical treatment and their implementation require detailed software development. Furthermore, they can become prohibitively expensive when multiple optimization problems are being solved, each requiring multiple restarts to circumvent local optima. In this work, we propose a machine learning enabled, surrogate-based framework that replaces the expensive adjoint solver, without compromising on predicting predictive accuracy. Specifically, we first train a deep neural network (DNN) from training data generated from evaluating the high-fidelity simulation model on a model-agnostic design of experiments on the geometry shape parameters. The optimum shape may then be computed by using a gradient-based optimizer coupled with the trained DNN. Subsequently, we also perform a gradient-free Bayesian optimization, where the trained DNN is used as the prior mean. We observe that the latter framework (DNN-BO) improves upon the DNN-only based optimization strategy for the same computational cost. Overall, this framework predicts the true optimum with very high accuracy, while requiring far fewer high-fidelity function calls compared to the adjoint-based method. Furthermore, we show that multiple optimization problems can be solved with the same machine learning model with high accuracy, to amortize the offline costs associated with constructing our models. Our methodology finds applications in the early stages of aerospace design. (C) 2021 Published by Elsevier Masson SAS.

Renganathan, S. Ashwin↗

Processing effects on microstructure in Er and ErD 2 thin-films

Erbium metal thin-films have been deposited on molybdenum-on-silicon substrates and then converted to erbium dideuteride (ErD 2 ). Here in this work, we study the effects of deposition temperature (≈300 or 723 K) and deposition rate (1 or 20 nm/s) upon the initial Er metal microstructure and subsequent ErD 2 microstructure. We find that low deposition temperature and low deposition rate lead to small Er metal grain sizes, and high deposition temperature and deposition rate led to larger Er metal grain sizes, consistent with published models of metal thin-film growth. ErD 2 grain sizes are strongly influenced by the prior-metal grain size, with small metal grains leading to large ErD 2 grains. A novel sample preparation technique for electron backscatter diffraction of air-sensitive ErD 2 was developed, and allowed the quantitative measurement of ErD 2 grain size and crystallographic texture. Finer-grained ErD 2 showed a strong (1 1 1) fiber texture, whereas larger grained ErD 2 had only weak texture. Finally, we hypothesize that this inverse correlation may arise from improved hydrogen diffusion kinetics in the more defective fine-grained metal structure or due to improved nucleation in the textured large-grain Er.

36 MATERIALS SCIENCE↗

The impact of updated Zr neutron-capture cross sections and new asymptotic giant branch models on our understanding of the S process and the origin of stardust

We present model predictions for the Zr isotopic ratios produced by slow neutron captures in C-rich asymptotic giant branch (AGB) stars of masses 1.25-4 M {sub ☉} and metallicities Z = 0.01-0.03, and compare them to data from single meteoritic stardust silicon carbide (SiC) and high-density graphite grains that condensed in the outflows of these stars. We compare predictions produced using the Zr neutron-capture cross sections from Bao et al. and from n{sub T}OF experiments at CERN, and present a new evaluation for the neutron-capture cross section of the unstable isotope {sup 95}Zr, the branching point leading to the production of {sup 96}Zr. The new cross sections generally present an improved match with the observational data, except for the {sup 92}Zr/{sup 94}Zr ratios, which are on average still substantially higher than predicted. The {sup 96}Zr/{sup 94}Zr ratios can be explained using our range of initial stellar masses, with the most {sup 96}Zr-depleted grains originating from AGB stars of masses 1.8-3 M {sub ☉} and the others from either lower or higher masses. The {sup 90,} {sup 91}Zr/{sup 94}Zr variations measured in the grains are well reproduced by the range of stellar metallicities considered here, which is the same needed to cover the Si composition of the grains produced by the chemical evolution of the Galaxy. The {sup 92}Zr/{sup 94}Zr versus {sup 29}Si/{sup 28}Si positive correlation observed in the available data suggests that stellar metallicity rather than rotation plays the major role in covering the {sup 90,} {sup 91,} {sup 92}Zr/{sup 94}Zr spread.

79 ASTRONOMY AND ASTROPHYSICS↗

r -process nucleosynthesis from matter ejected in binary neutron star mergers [On r -process nucleosynthesis from matter ejected in binary neutron star mergers]

Here, when binary systems of neutron stars merge, a very small fraction of their rest mass is ejected, either dynamically or secularly. This material is neutron-rich and its nucleosynthesis provides the astrophysical site for the production of heavy elements in the Universe, together with a kilonova signal confirming neutron-star mergers as the origin of short gamma-ray bursts. We perform full general-relativistic simulations of binary neutron-star mergers employing three different nuclear-physics equations of state (EOSs), considering both equal- and unequal-mass configurations, and adopting a leakage scheme to account for neutrino radiative losses. Using a combination of techniques, we carry out an extensive and systematic study of the hydrodynamical, thermodynamical, and geometrical properties of the matter ejected dynamically, employing the WinNet nuclear-reaction network to recover the relative abundances of heavy elements produced by each configurations. Among the results obtained, three are particularly important. First, we find that, within the sample considered here, both the properties of the dynamical ejecta and the nucleosynthesis yields are robust against variations of the EOS and masses. Second, using a conservative but robust criterion for unbound matter, we find that the amount of ejected mass is ≲10 –3 M⊙, hence at least one order of magnitude smaller than what normally assumed in modelling kilonova signals. Finally, using a simplified and gray-opacity model we assess the observability of the infrared kilonova emission finding, that for all binaries the luminosity peaks around ~1/2 day in the H-band, reaching a maximum magnitude of –13, and decreasing rapidly after one day.

79 ASTRONOMY AND ASTROPHYSICS↗

Verification Of the Defense Waste Processing Facility's (DWPF) Process Digestion Methods for the Sludge Batch 8 Qualification Sample

This report contains the results and comparison of data generated from inductively coupled plasma – atomic emission spectroscopy (ICP-AES) analysis of Aqua Regia (AR), Sodium Peroxide/Sodium Hydroxide Fusion Dissolution (PF) and Cold Chem (CC) method digestions and Cold Vapor Atomic Absorption analysis of Hg digestions from the DWPF Hg digestion method of Sludge Batch 8 (SB8) Sludge Receipt and Adjustment Tank (SRAT) Receipt and SB8 SRAT Product samples. The SB8 SRAT Receipt and SB8 SRAT Product samples were prepared in the SRNL Shielded Cells, and the SRAT Receipt material is representative of the sludge that constitutes the SB8 Batch or qualification composition. This is the sludge in Tank 51 that is to be transferred into Tank 40, which will contain the heel of Sludge Batch 7b (SB7b), to form the SB8 Blend composition.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

TRACER Radar b1 Data Processing: Corrections, Calibrations, and Processing Report

The U.S. Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) user facility deployed the first ARM Mobile Facility (AMF1) to Houston, Texas for the Tracking Aerosol Convection Interactions Experiment (TRACER) field campaign. The TRACER campaign was conducted from October 1, 2021 to September 30, 2022, with an intensive operational period (IOP) from June 1 to September 30, 2022. To investigate the life cycles of convective cells in the polluted and humid urban environment of Houston, ARM deployed cloud and precipitation radars, including the C-band Scanning ARM Precipitation Radar (CSAPR2), the X/Ka-band Scanning ARM Cloud Radar (SACR), and the Ka-band ARM Zenith cloud Radar (KAZR), as shown in Figure 1. This report presents an analysis of radar data quality, hardware calibrations, and radar data corrections.

54 ENVIRONMENTAL SCIENCES↗