Search NASA⌕ Search

SEARCH · Search NASA

Results for “Edge Computing”

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 1,657 records · Page 92

High-fidelity kinetic modeling of plasmas in Z-pinch and pulsed power configurations (18-ERD-048)

Plasma dynamics can be strongly influenced by fine-scale kinetic physics, which is not captured by standard fluid treatments, is challenging to simulate, and is poorly understood. The aim of this research was to investigate the fundamental fine-scale properties of collisionless plasmas in Z-pinch and pulsed power configurations by deploying novel noisefree high-accuracy computational algorithms. Significant advances were achieved through the development of robust numerical capabilities to exploit axisymmetry in kinetic simulations, the development of new methodologies to construct kinetic equilibria, and the theoretical and computational study of kinetic sheared-flow-driven transport in pulsed power systems. These new capabilities shed light on the anisotropies that can develop in Z-pinches and the transport processes that can lead to parasitic currents in pulsed power inertial confinement fusion experiments. The use of cutting-edge massively-parallel numerical algorithms to address outstanding scientific questions in plasma physics advances the mission of Lawrence Livermore National Laboratory in the areas of high energy density science and high performance computing. By bringing together kinetic theory, high-order numerical methods, high performance computing, and experimental applications, this research advances the state of the art in plasma physics modeling.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A Discussion of the Failure of a Quad Diode Module and Efforts to Assure the Flight Spares

In 2019 the International Space Station (ISS) experienced an on on-orbit failure that affected 1 of 28 Battery Charge Discharge Units (BCDUs). Telemetry pointed to a short circuit failure of a Power Rectifier Quad Diode Module. Astronauts removed the failed unit from service which was then returned to Earth for failure analysis. The failure analysis confirmed that the quad diode module had failed short circuit. The investigation identified silver dendrites had grown on the insulated, sloped edge of the mesa semiconductor die of the failed diode and also a 2nd diode in the same quad diode module. Voids between the diode’s protective encapsulating ring and the die provided space within which dendrites were able to form and cause catastrophic failure. The NASA Engineering & Safety Center (NESC) convened a EEE Parts Sub-Team to investigate root cause and to assist with risk assessment for all of the flight diode modules (4 distinct production lots) and the flight spares (from a 5th lot). Analysis of the original diode manufacturer’s read and record screening test data identified some lots which contained diodes with parametric instabilities that could signal conditions suited for silver dendrite formation. Computed Tomography (CT) X-ray screening inspection was performed on flight spare modules to identify diodes that do not exhibit voids between the edge of die and encapsulating ring as a mitigation against dendrite formation.

Jay Brusse↗

Modeling Boundary-Layer Transition in Subsonic Flow over a Swept Wing

Predicting the onset of boundary-layer transition is often more accurate using physics-based models that directly compute disturbance growth rather than phenomenological models often implemented into industrial CFD codes. The aim of this ongoing study is to calibrate linear, physics-based computations of transition in subsonic flows over swept wings against a large set of experimental data. Advancing the calibration of linear models of transition contributes to the CFD-Vision-2030 goal of automated boundary-layer transition prediction. This progress report uses the dual N-factor method to model transition over the swept NACA 64-2-015A wing. The flow conditions match selected test conditions from an extensive experimental dataset acquired from the NASA Ames 12-ft Pressure Tunnel. The OVERFLOW 2.4b flow solver is used to obtain laminar basic states based on an infinite-span assumption. Stability analyses are performed on 365 distinct configurations with linear stability theory (LST) and parabolized stability equations (PSE) from the Langley Stability and Transition Analysis Codes (LASTRAC), modeling the growth of Tollmien-Schlichting (TS) and stationary crossflow (SCF) disturbances. From a total of 67 data points for unswept, i.e., TS-dominant configurations, the critical N-factor based on PSE is found to be N_TS = 9. The SCF critical N-factor is found to be near 8 for the highly swept, SCF-dominant configurations. Dual N-factor curves for both LST and PSE computations demonstrate a high level of interaction between TS and SCF. It may be worthwhile to investigate an alternate metric to visualize maximal SCF amplification upstream of the transition location to account for the growth of SCF modes near the leading edge, which is not considered in the conventional applications of the dual N-factor criterion.

computational modeling↗

Dramatic Influence of an Anionic Donor on the Oxygen-Atom Transfer Reactivity of a Mn V -Oxo Complex

Addition of an anionic donor to an Mn V (O) porphyrinoid complex causes a dramatic increase in 2-electron oxygen-atom-transfer (OAT) chemistry. The 6-coordinate [Mn V (O)(TBP 8 Cz)(CN)] – was generated from addition of Bu 4 N + CN – to the 5-coordinate Mn V (O) precursor. The cyanide-ligated complex was characterized for the first time by Mn K-edge X-ray absorption spectroscopy (XAS) and gives Mn—O=1.53 Å, Mn—CN=2.21 Å. In combination with computational studies these distances were shown to correlate with a singlet ground state. Reaction of the CN– complex with thioethers results in OAT to give the corresponding sulfoxide and a 2e – -reduced Mn III (CN) – complex. Kinetic measurements reveal a dramatic rate enhancement for OAT of approximately 24 000-fold versus the same reaction for the parent 5-coordinate complex. An Eyring analysis gives ΔH ≠ = 14 kcal mol –1 , ΔS ≠ = –10 cal mol –1 K –1 . Computational studies fully support the structures, spin states, and relative reactivity of the 5- and 6-coordinate Mn V (O) complexes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

In Vivo Proton Beam Range Verification Using Spine MRI Changes

Purpose: In proton therapy, uncertainty in the location of the distal dose edge can lead to cautious treatment plans that reduce the dosimetric advantage of protons. After radiation exposure, vertebral bone marrow undergoes fatty replacement that is visible on magnetic resonance imaging (MRI). This presents an exciting opportunity to observe radiation dose distribution in vivo. We used quantitative spine MRI changes to precisely detect the distal dose edge in proton radiation patients. Methods and Materials: We registered follow-up T1-weighted MRI images to planning computed tomography scans from 10 patients who received proton spine irradiation. A radiation dose-MRI signal intensity curve was created using the lateral beam penumbra in the sacrum. This curve was then used to measure range errors in the lumbar spine. Results: In the lateral penumbra, there was an increase in signal intensity with higher dose throughout the full range of 0-37.5 Gy (RBE). In the distal fall-off region, the beam sometimes appeared to penetrate farther than planned. The mean overshoot in 10 patients was 1.9 mm (95% confidence interval, 0.8-3.1 mm), on the order of the uncertainties inherent to our range verification method. Conclusions: We have demonstrated in vivo proton range verification using posttreatment spine MRI changes. Our analysis suggests the presence of a systematic overshoot of a few millimeters in some proton spine treatments, but the range error does not exceed the uncertainty incorporated into the treatment planning margin. It may be possible to extend our technique to MRI sequences that show early bone marrow changes, enabling adaptive treatment modification.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Coupling 1D xRAGE simulations with machine learning for graded inner shell design optimization in double shell capsules

Advances in machine learning provide the ability to leverage data from expensive simulations of high-energy-density experiments to significantly cut down on computational time and costs associated with the search for optimal target designs. This study presents an application of cutting-edge Bayesian optimization methods to the one-dimensional (1D) design optimization of double shell graded layer targets for inertial confinement fusion experiments. This investigation attempts to reduce hydrodynamic instabilities while retaining high yields for future NIF experiments. Machine learning methods can use predictive physics simulations to identify graded layer designs from within the vast design space that demonstrate high predicted performance, including novel designs with high uncertainty in performance that may hold unexpected promise. By applying machine learning tools to the simulation design, we map the trade-off between 1D yield and instability, specifically isolating parameter ranges, which maintain high performance while showing significantly improved Rayleigh–Taylor stability over the point design. Furthermore, the groundwork laid in this study will be a useful design tool for future NIF experiments with graded layer targets.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Neoclassical transport in toroidal plasmas with nonaxisymmetric flux surfaces

The capability to treat nonaxisymmetric flux surface geometry has been added to the drift-kinetic code NEO. Geometric quantities (i.e. metric elements) are supplied by a recently-developed local 3D equilibrium solver, allowing neoclassical transport coefficients to be systematically computed while varying the 3D plasma shape in a simple and intuitive manner. Code verification is accomplished via detailed comparison with 3D Pfirsch–Schlüter theory. A discussion of the various collisionality regimes associated with 3D transport is given, with an emphasis on non-ambipolar particle flux, neoclassical toroidal viscosity, energy flux and bootstrap current. As a result, we compute the transport in the presence of ripple-type perturbations in a DIII-D-like H-mode edge plasma.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Self-Supervised and Interpretable Anomaly Detection Using Network Transformers

Machine learning and deep neural networks (DNNs) have been proposed as a tool to identify anomalies in computer network communications. However, due the obfuscated nature of off-the-shelf machine learning models, their output often does not provide enough information to isolate the source of the anomaly to take corrective measures. In this article, we introduce the network transformer (NeT), a DNN model for anomaly detection that incorporates the graph structure of the communication network in order to improve interpretability. Further, the presented approach has the following advantages: first, enhanced interpretability by incorporating the graph structure of computer networks; second, provides a hierarchical set of features that enables analysis at different levels of granularity; second, self-supervised training that does not require labeled data. The NeT model was evaluated on a set of anomalous scenarios executed in a real industrial control system. The presented approach successfully identified the anomalies, the devices affected, and the specific connections causing the anomalies, providing a data-driven hierarchical approach to analyze the behavior of a cyber network.

97 MATHEMATICS AND COMPUTING↗

INDDGO: Integrated Network Decomposition & Dynamic programming for Graph Optimization

It is well-known that dynamic programming algorithms can utilize tree decompositions to provide a way to solve some NP -hard problems on graphs where the complexity is polynomial in the number of nodes and edges in the graph, but exponential in the width of the underlying tree decomposition. However, there has been relatively little computational work done to determine the practical utility of such dynamic programming algorithms. We have developed software to construct tree decompositions using various heuristics and have created a fast, memory-efficient dynamic programming implementation for solving maximum weighted independent set. We describe our software and the algorithms we have implemented, focusing on memory saving techniques for the dynamic programming. We compare the running time and memory usage of our implementation with other techniques for solving maximum weighted independent set, including a commercial integer programming solver and a semi-definite programming solver. Our results indicate that it is possible to solve some instances where the underlying decomposition has width much larger than suggested by the literature. For certain types of problems, our dynamic programming code runs several times faster than these other methods.

Groër, Christopher S.↗

FY 2016 Fusion Energy Sciences (FES) Theory and Simulation Performance Target

Annual target: Predicting the magnitude and scaling of the divertor heat load width in magnetically confined burning plasmas is a high priority for the fusion program and ITER. One of the key unresolved physics issues is what sets the heat flux width at the entrance to the divertor region. Perform massively parallel simulations using 3D edge kinetic and fluid codes to determine the parameter dependence of the heat load width at the divertor entrance and compute the divertor plate heat flux applicable to moderate particle recycling conditions. Comparisons will be made with data from DIII-D, NSTX-U, and C-Mod.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A Comparison between Invariant and Equivariant Classical and Quantum Graph Neural Networks

Machine learning algorithms are heavily relied on to understand the vast amounts of data from high-energy particle collisions at the CERN Large Hadron Collider (LHC). The data from such collision events can naturally be represented with graph structures. Therefore, deep geometric methods, such as graph neural networks (GNNs), have been leveraged for various data analysis tasks in high-energy physics. One typical task is jet tagging, where jets are viewed as point clouds with distinct features and edge connections between their constituent particles. The increasing size and complexity of the LHC particle datasets, as well as the computational models used for their analysis, have greatly motivated the development of alternative fast and efficient computational paradigms such as quantum computation. In addition, to enhance the validity and robustness of deep networks, we can leverage the fundamental symmetries present in the data through the use of invariant inputs and equivariant layers. In this paper, we provide a fair and comprehensive comparison of classical graph neural networks (GNNs) and equivariant graph neural networks (EGNNs) and their quantum counterparts: quantum graph neural networks (QGNNs) and equivariant quantum graph neural networks (EQGNN). The four architectures were benchmarked on a binary classification task to classify the parton-level particle initiating the jet. Based on their area under the curve (AUC) scores, the quantum networks were found to outperform the classical networks. However, seeing the computational advantage of quantum networks in practice may have to wait for the further development of quantum technology and its associated application programming interfaces (APIs).

Forestano, Roy T. (ORCID:0000000203552076)↗

Pressure investigation of NASA leading edge vortex flaps on a 60 deg Delta wing

Pressure distributions on a 60 deg Delta Wing with NASA designed leading edge vortex flaps (LEVF) were found in order to provide more pressure data for LEVF and to help verify NASA computer codes used in designing these flaps. These flaps were intended to be optimized designs based on these computer codes. However, the pressure distributions show that the flaps wre not optimum for the size and deflection specified. A second drag-producing vortex forming over the wing indicated that the flap was too large for the specified deflection. Also, it became apparent that flap thickness has a possible effect on the reattachment location of the vortex. Research is continuing to determine proper flap size and deflection relationships that provide well-behaved flowfields and acceptable hinge-moment characteristics.

Marchman, J. F., III↗

The mechanics of delamination in fiber-reinforced composite materials. Part 2: Delamination behavior and fracture mechanics parameters

Based on theories of laminate anisotropic elasticity and interlaminar fracture, the complete solution structure associated with a composite delamination is determined. Fracture mechanics parameters characterizing the interlaminar crack behavior are defined from asymptotic stress solutions for delaminations with different crack-tip deformation configurations. A numerical method employing singular finite elements is developed to study delaminations in fiber composites with any arbitrary combinations of lamination, material, geometric, and crack variables. The special finite elements include the exact delamination stress singularity in its formulation. The method is shown to be computationally accurate and efficient, and operationally simple. To illustrate the basic nature of composite delamination, solutions are shown for edge-delaminated (0/-0/-0/0) and (+ or - 0/+ or - 0/90/90 deg) graphite-epoxy systems under uniform axial extenstion. Three-dimensional crack-tip stress intensity factors, associated energy release rates, and delamination crack-closure are determined for each individual case. The basic mechanics and mechanisms of composite delamination are studied, and fundamental characteristics unique to recently proposed tests for interlaminar fracture toughness of fiber composite laminates are examined.

Wang, S. S.↗

The mechanics of delamination in fiber-reinforced composite materials. II - The delamination behavior and fracture mechanics parameters

Based on theories of laminate anisotropic elasticity and interlaminar fracture, the complete solution structure associated with a composite delamination is determined. Fracture mechanics parameters characterizing the interlaminar crack behavior are defined from asymptotic stress solutions for delaminations with different crack-tip deformation configurations. A numerical method employing singular finite elements is developed to study delaminations in fiber composites with any arbitrary combinations of lamination, material, geometric, and crack variables. The special finite elements include the exact delamination stress singularity in its formulation. The method is shown to be computationally accurate and efficient, and operationally simple. To illustrate the basic nature of composite delamination, solutions are shown for edge-delaminated (0/-0/-0/0) and (+ or - 0/+ or - 0/90/90 deg) graphite-epoxy systems under uniform axial extension. Three-dimensional crack-tip stress intensity factors, associated energy release rates, and delamination crack-closure are determined for each individual case. The basic mechanics and mechanisms of composite delamination are studied, and fundamental characteristics unique to recently proposed tests for interlaminar fracture toughness of fiber composite laminates are examined. Previously announced in STAR as N84-13222

Wang, S. S.↗

Wing-Design And -Analysis Code

WINGDES2 computer program provides wing-design algorithm based on modified linear theory taking into account effects of attainable leading-edge thrust. Features improved numerical accuracy and additional capabilities. Provides analysis as well as design capability and applicable to both subsonic and supersonic flow. Replaces earlier wing-design code designated WINGDES (see LAR-13315). Written in FORTRAN V.

Darden, Christine M.↗