SEARCH · Search NASA

Results for “Vector Processing”

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.

SourceResearch connectionFrom yearThrough year

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

At least 451 records · Page 25

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Atomistic modeling of surface and grain boundary dislocation nucleation in FCC metals

Dislocation nucleation plays a critical role in the plastic deformation of crystalline materials. However, it is challenging to predict the active mode and associated rate of dislocation nucleation under typical experimental loading conditions through molecular dynamics simulation due to timescale limitations. In this report we use the free-end nudged elastic band method to determine the activation energies and activation volumes of dislocation nucleation in four typical face-centered cubic metals of Au, Al, Cu and Ni. We focus on the representative processes of surface and grain boundary dislocation nucleation. The atomistically determined activation volumes of these dislocation nucleation processes are larger than 10b3 (with b being the Burgers vector length) under typical experimental loading conditions. These results are compared with experimentally measured activation volumes in ultrafine-grained and nanocrystalline metals, thereby providing mechanistic insight into their rate-controlling deformation mechanisms.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Inferring Neutron Star Merger Ejecta Morphology with Kilonovae

In this study we incorporate a new grid of kilonova simulations produced by the Monte Carlo radiative transfer code SuperNu in an inference pipeline for astrophysical transients and evaluate their performance. These simulations contain four different two-component ejecta morphology classes. We analyze follow-up observational strategies by Vera Rubin Observatory in optical and James Webb Space Telescope (JWST) in mid-infrared (MIR). Our analysis suggests that, within these strategies, it is possible to discriminate between the four different morphologies only when late-time JWST observations in MIR are available. We conclude that follow-ups by the new Vera Rubin Observatory alone are not sufficient to determine ejecta morphology. Additionally, we make comparisons between surrogate models based on radiative transfer simulation grids by SuperNu and POSSIS, by analyzing the historic kilonova AT2017gfo that accompanied the gravitational wave event GW170817. We show that both SuperNu and POSSIS models provide similar fits to photometric observations but their qualitative interpretations differ.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2022

Climatic clustering and longitudinal analysis with impacts on food, bioenergy, and pandemics

Predicted growth in world population will put unparalleled stress on the need for sustainable energy and global food production, as well as increase the likelihood of future pandemics. In this work, we identify high-resolution environmental zones in the context of a changing climate and predict longitudinal processes relevant to these challenges. We do this using exhaustive vector comparison methods that measure the climatic similarity between all locations on earth at high geospatial resolution relative to global-scale analyses. The results are captured as networks, in which edges between geolocations are defined if their historical climate similarities exceed a threshold. We apply Markov clustering and our novel Correlation of Correlations method to the resulting climatic networks, which provides unprecedented agglomerative and longitudinal views of climatic relationships across the globe. The methods performed here resulted in the fastest (9.37x10 18 operations/sec) and one of the largest (168.7x10 21 operations) scientific computations ever performed, with more than 100 quadrillion edges considered for a single climatic network. Our climatic analysis reveals areas of the world experiencing rapid environmental changes, which can have important implications for global carbon fluxes and zoonotic spillover events. Correlation and network analyses of this kind are widely applicable across computational and predictive biology domains, including systems biology, ecology, carbon cycles, biogeochemistry, and zoonosis research.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2019

Adversarial Training for Privacy-Preserving Deep Learning Model Distribution

Collaboration among cancer registries is essential to develop accurate, robust, and generalizable deep learning models for automated information extraction from cancer pathology reports. Sharing data presents a serious privacy issue, especially in biomedical research and healthcare delivery domains. Distributing pretrained deep learning (DL) models has been proposed to avoid critical data sharing. However, there is growing recognition that collaboration among clinical institutes through DL model distribution exposes new security and privacy vulnerabilities. These vulnerabilities increase in natural language processing (NLP) applications, in which the dataset vocabulary with word vector representations needs to be associated with the other model parameters. In this paper, we propose a novel privacy-preserving DL model distribution across cancer registries for information extraction from cancer pathology reports with privacy and confidentiality considerations. The proposed approach exploits the adversarial training framework to distinguish private features from shared features among different datasets. It only shares registry-invariant model parameters, without sharing raw data nor registry-specific model parameters among cancer registries. Thus, it protects both the data and the trained model simultaneously. We compare our proposed approach to single-registry models, and a model trained on centrally hosted data from different cancer registries. The results show that the proposed approach significantly outperforms the single-registry models and achieves statistically indistinguishable micro and macro F1-score as compared to the centralized model.

Alawad, Mohammed↗

DOE OSTI2022

Azimuthal correlations in diffractive scattering at the Electron-Ion Collider

We calculate azimuthal correlations between the exclusively produced vector meson and the scattered electron in Deep Inelastic Scattering processes at the future Electron-Ion Collider (EIC). We identify ``kinematical'' and ``intrinsic'' contributions to these correlations, and show that the correlations are sensitive to the non-trivial correlations in the gluon distribution of the target. Realistic predictions at the EIC kinematics are provided using two different approaches to describe the dipole-proton interaction at relatively small x.

Dumitru, Adrian↗

DOE OSTI2024

LLNL FESP Theory Highlights: October 2024

I. Novikau, I. Y. Dodin, E. A. Startsev, I. Joseph, Quantum algorithms for simulating dissipative linear and nonlinear dynamics of plasmas. Invited talk at the 66th Annual Meeting of the APS Division of Plasma Physics, Atlanta, Georgia. Novikau I., Dodin I.Y., Startsev E.A., Encoding of linear kinetic plasma problems in quantum circuits via data compression, Journal of Plasma Physics. 2024;90(4):805900401, doi:10.1017/S0022377824000795. We propose an algorithm for encoding linear kinetic plasma problems in quantum circuits. The focus is on modelling electrostatic linear waves in a one-dimensional Maxwellian electron plasma. The waves are described by the linearized Vlasov–Ampère system with a spatially localized external current that drives plasma oscillations. This system is formulated as a boundary-value problem and cast in the form of a linear vector equation to be solved by using the quantum signal processing algorithm. The latter requires encoding of a matrix in a quantum circuit as a sub-block of a unitary matrix. We propose how to encode in a circuit in a compressed form and discuss how the resulting circuit scales with the problem size and the desired precision.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2020

A Deep Neural Network for Accurate and Robust Prediction of the Glass Transition Temperature of Polyhydroxyalkanoate Homo- and Copolymers

The purpose of this study was to develop a data-driven machine learning model to predict the performance properties of polyhydroxyalkanoates (PHAs), a group of biosourced polyesters featuring excellent performance, to guide future design and synthesis experiments. A deep neural network (DNN) machine learning model was built for predicting the glass transition temperature, Tg, of PHA homo- and copolymers. Molecular fingerprints were used to capture the structural and atomic information of PHA monomers. The other input variables included the molecular weight, the polydispersity index, and the percentage of each monomer in the homo- and copolymers. The results indicate that the DNN model achieves high accuracy in estimation of the glass transition temperature of PHAs. In addition, the symmetry of the DNN model is ensured by incorporating symmetry data in the training process. The DNN model achieved better performance than the support vector machine (SVD), a nonlinear ML model and least absolute shrinkage and selection operator (LASSO), a sparse linear regression model. The relative importance of factors affecting the DNN model prediction were analyzed. Sensitivity of the DNN model, including strategies to deal with missing data, were also investigated. Compared with commonly used machine learning models incorporating quantitative structure–property (QSPR) relationships, it does not require an explicit descriptor selection step but shows a comparable performance. The machine learning model framework can be readily extended to predict other properties.

quantitative structure–property relationship (QSPR↗

NASA NTRS1980

Navigation accuracy analysis for the Halley flyby phase of a dual comet mission using ion drive

A dual comet (Halley Flyby/Tempel 2 Rendezvous) mission, making use of the solar electric propulsion system, is under consideration for a 1985 launch. This paper presents navigation accuracy analysis results for the Halley flyby phase of this mission. Orbit determination and guidance accuracies are presented for the baseline navigation strategy, along with the results of a number of sensitivity studies involving parameters such as data frequencies, data accuracies, ion drive thrust vector errors, comet ephemeris uncertainties, time lags associated with data processing and command sequence generation, probe release time, and navigation coast arc duration.

Wood, L. J.↗

NASA NTRS1980

Navigation accuracy analyses for two comet rendezvous missions using ion drive

Two cometary missions, making use of the solar electric propulsion system, have recently been considered for launches in the mid to late 1980's. This paper presents navigation accuracy analysis results for the rendezvous portions of these missions, the target bodies being Comet Tempel 2 and Comet Tuttle-Giacobini-Kresak. Orbit determination and guidance accuracies are presented for the baseline navigation strategies, along with the results of a number of sensitivity studies involving parameters such as data frequencies, data accuracies, ion drive thrust vector errors, comet emphemeris uncertainties, time lags associated with data processing and command sequence generation, and certain guidance law parameters. The accuracies obtained are, in some respects, significantly better than the results of previous solar electric propulsion comet rendezvous studies.

Wood, L. J.↗

NASA NTRS1981

Study of gravity and magnetic anomalies using MAGSAT data

The results of modeling satellite-elevation magnetic and gravity data using the constraints imposed by near surface data and seismic evidence shows that the magnetic minimum can be accounted for by either an intracrustal lithologic variation or by an upwarp of the Curie point isotherm. The long wavelength anomalies of the NOO's-vector magnetic survey of the conterminous U.S. were contoured and processed by various frequency filters to enhance particular characteristics. A preliminary inversion of the data was completed and the anomaly field calculated at 450 km from the equivalent magnet sources to compare with the POGO satellite data. Considerable progress was made in studing the satellite magnetic data of South America and adjacent marine areas. Preliminary versions of the 1 deg free-air gravity anomaly map (20 m gal contour interval) and the high cut (lambda approximately 8 deg) filtered anomaly maps are included.

Braile, L. W.↗

NASA NTRS1983

Calculation of viscous supersonic flows over finned bodies

The parabolized Navier-Stokes (PNS) equations are used to calculate the viscous, supersonic flow fields about a six-finned projectile and a generic four-finned missile at angles of attack. Since current computer speeds and storage preclude a fully three-dimensional calculation using the unsteady, Reynolds-averaged, Navier-Stokes equations, the applicability of the PNS equations to the above flow fields is of considerable interest. Two important aspects of the calculation are grid generation and the type of smoothing used to prevent nonphysical solutions. This paper includes a description of the grid-generation process. Results in the form of density contours and velocity vector plots are presented for the two configurations. The applicability of the PNS equations to the complicated flow fields considered is successfully demonstrated.

Rai, M. M.↗

NASA NTRS1985

Onboard rendezvous navigation for the Space Shuttle

The onboard rendezvous navigation software for Shuttle flight 41-C are described. Particular attention is given to the inputs, models, and outputs of the software. The performance of the rendezvous navigation system is compared to predicted performance profiles in connection with the relative vehicle geometry, as well as the location of the navigation sensor tracking arcs. The methods used to process navigational sensor measurements in order to update the state vector are also summarized. A table listing the sources of maneuver targeting errors is provided.

Wylie, A. D.↗

NASA NTRS1987

Parallel processing in finite element structural analysis

A brief review is made of the fundamental concepts and basic issues of parallel processing. Discussion focuses on parallel numerical algorithms, performance evaluation of machines and algorithms, and parallelism in finite element computations. A computational strategy is proposed for maximizing the degree of parallelism at different levels of the finite element analysis process including: 1) formulation level (through the use of mixed finite element models); 2) analysis level (through additive decomposition of the different arrays in the governing equations into the contributions to a symmetrized response plus correction terms); 3) numerical algorithm level (through the use of operator splitting techniques and application of iterative processes); and 4) implementation level (through the effective combination of vectorization, multitasking and microtasking, whenever available).

Noor, Ahmed K.↗

NASA NTRS1991

Software manual for operating particle displacement tracking data acquisition and reduction system

The software manual is presented. The necessary steps required to record, analyze, and reduce Particle Image Velocimetry (PIV) data using the Particle Displacement Tracking (PDT) technique are described. The new PDT system is an all electronic technique employing a CCD video camera and a large memory buffer frame-grabber board to record low velocity (less than or equal to 20 cm/s) flows. Using a simple encoding scheme, a time sequence of single exposure images are time coded into a single image and then processed to track particle displacements and determine 2-D velocity vectors. All the PDT data acquisition, analysis, and data reduction software is written to run on an 80386 PC.

Wernet, Mark P.↗

NASA NTRS1994

Studies on image compression and image reconstruction

During this six month period our works concentrated on three, somewhat different areas. We looked at and developed a number of error concealment schemes for use in a variety of video coding environments. This work is described in an accompanying (draft) Masters thesis. In the thesis we describe application of this techniques to the MPEG video coding scheme. We felt that the unique frame ordering approach used in the MPEG scheme would be a challenge to any error concealment/error recovery technique. We continued with our work in the vector quantization area. We have also developed a new type of vector quantizer, which we call a scan predictive vector quantization. The scan predictive VQ was tested on data processed at Goddard to approximate Landsat 7 HRMSI resolution and compared favorably with existing VQ techniques. A paper describing this work is included. The third area is concerned more with reconstruction than compression. While there is a variety of efficient lossless image compression schemes, they all have a common property that they use past data to encode future data. This is done either via taking differences, context modeling, or by building dictionaries. When encoding large images, this common property becomes a common flaw. When the user wishes to decode just a portion of the image, the requirement that the past history be available forces the decoding of a significantly larger portion of the image than desired by the user. Even with intelligent partitioning of the image dataset, the number of pixels decoded may be four times the number of pixels requested. We have developed an adaptive scanning strategy which can be used with any lossless compression scheme and which lowers the additional number of pixels to be decoded to about 7 percent of the number of pixels requested! A paper describing these results is included.

Sayood, Khalid↗

NASA NTRS1992

An implementation of the QMR method based on coupled two-term recurrences

The authors have proposed a new Krylov subspace iteration, the quasi-minimal residual algorithm (QMR), for solving non-Hermitian linear systems. In the original implementation of the QMR method, the Lanczos process with look-ahead is used to generate basis vectors for the underlying Krylov subspaces. In the Lanczos algorithm, these basis vectors are computed by means of three-term recurrences. It has been observed that, in finite precision arithmetic, vector iterations based on three-term recursions are usually less robust than mathematically equivalent coupled two-term vector recurrences. This paper presents a look-ahead algorithm that constructs the Lanczos basis vectors by means of coupled two-term recursions. Implementation details are given, and the look-ahead strategy is described. A new implementation of the QMR method, based on this coupled two-term algorithm, is described. A simplified version of the QMR algorithm without look-ahead is also presented, and the special case of QMR for complex symmetric linear systems is considered. Results of numerical experiments comparing the original and the new implementations of the QMR method are reported.

Freund, Roland W.↗

NASA NTRS1995

Experimental confirmation of a PDE-based approach to design of feedback controls

Issues regarding the experimental implementation of partial differential equation based controllers are discussed in this work. While the motivating application involves the reduction of vibration levels for a circular plate through excitation of surface-mounted piezoceramic patches, the general techniques described here will extend to a variety of applications. The initial step is the development of a PDE model which accurately captures the physics of the underlying process. This model is then discretized to yield a vector-valued initial value problem. Optimal control theory is used to determine continuous-time voltages to the patches, and the approximations needed to facilitate discrete time implementation are addressed. Finally, experimental results demonstrating the control of both transient and steady state vibrations through these techniques are presented.

Banks, H. T.↗

NASA NTRS1998

Numerical Modeling of Solidification in Space With MEPHISTO-4

A pre-flight analysis of the directional solidification of BiSn with MEPHISTO-4 is presented. Simplified Bridgman growth under microgravity conditions is simulated using a two dimensional finite element model. This numerical model is a single domain, pseudo-steady state model, and includes the effects of both thermal and solutal convection. The results show that for all orientations of the applied steady state gravity vector, of magnitude 1 micro-g, the directional solidification process remains diffusion controlled. The maximum convective velocity was found to be 4.424 x 10(exp -5) cm/s for the horizontal Bridgman growth configuration. This value is an order of magnitude lower than the growth velocity. The maximum and minimum values or solute concentration in the liquid at the crystal-melt interface were 13.867 at.% and 13.722 at.%, respectively. This gives a radial segregation value of xi = 1.046% at the interface. A secondary objective of this work was to compare the results obtained to those that consider thermal convection only (no solutal convection). It was found that the convective flow patterns in simulations which included solutal convection were significantly different from those which ignored solutal convection. The level of radial segregation predicted by the current simulations is an order of magnitude lower than that found in simulations which ignore solutal convection. The final aim was to investigate the effect of g-jitter on the crystal growth process. A simulation was performed to calculate the system response to a 1 second, 100 micro-g gravity impulse acting normal to the direction of growth. This pulse is consistent with that induced by Orbiter thruster firings. The results obtained indicate that such a gravity pulse causes an increase in the level of radial solute segregation at the interface from the steady state values. The maximum value of solute concentration in the liquid was found to be 13.888 at.%, the minimum value calculated was 13.706 at.%, yielding a radial segregation value of xi = 1.31% at the interface. These values occurred 126 seconds after the pulse terminated. Thus it is anticipated that the process will remain diffusion controlled even when subjected to such g-jitter.

Simpson, James E.↗