Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data clustering”

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 991 records · Page 55
All sourcesNASA NTRSDOE OSTI
NASA NTRS2006

Value, Cost, and Sharing: Open Issues in Constrained Clustering

Clustering is an important tool for data mining, since it can identify major patterns or trends without any supervision (labeled data). Over the past five years, semi-supervised (constrained) clustering methods have become very popular. These methods began with incorporating pairwise constraints and have developed into more general methods that can learn appropriate distance metrics. However, several important open questions have arisen about which constraints are most useful, how they can be actively acquired, and when and how they should be propagated to neighboring points. This position paper describes these open questions and suggests future directions for constrained clustering research.

constraints↗
DOE OSTI2015

Galaxy cluster pressure profiles, as determined by Sunyaev-Zeldovich effect observations with MUSTANG and Bolocam. I. Joint analysis technique

We present a technique to constrain galaxy cluster pressure profiles by jointly fitting Sunyaev-Zeldovich effect (SZE) data obtained with MUSTANG and Bolocam for the clusters Abell 1835 and MACS0647. Bolocam and MUSTANG probe different angular scales and are thus highly complementary. We find that the addition of the high-resolution MUSTANG data can improve constraints on pressure profile parameters relative to those derived solely from Bolocam. In Abell 1835 and MACS0647, we find gNFW inner slopes of γ=0.36{sub −0.21}{sup +0.33} and γ=0.38{sub −0.25}{sup +0.20}, respectively, when α and β are constrained to 0.86 and 4.67, respectively. The fitted SZE pressure profiles are in good agreement with X-ray derived pressure profiles.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

MC{sup 2}: Multiwavelength and Dynamical Analysis of the Merging Galaxy Cluster ZwCl 0008.8+5215: An Older and Less Massive Bullet Cluster

We present and analyze a rich data set including Subaru/SuprimeCam, HST/Advanced Camera for Surveys and Wide Field Camera 3, Keck/DEIMOS, Chandra/ACIS-I, and JVLA/C and D array for the merging cluster of galaxies ZwCl 0008.8+5215. With a joint Subaru+HST weak gravitational lensing analysis, we identify two dominant subclusters and estimate the masses to be M{sub 200}=5.7{sub −1.8}{sup +2.8}×10{sup 14} M{sub ⊙} and 1.2{sub −0.6}{sup +1.4}×10{sup 14} M{sub ⊙}. We estimate the projected separation between the two subclusters to be 924{sub −206}{sup +243} kpc. We perform a clustering analysis of spectroscopically confirmed cluster member galaxies and estimate the line-of-sight velocity difference between the two subclusters to be 92±164 km s{sup −1}. We further motivate, discuss, and analyze the merger scenario through an analysis of the 42 ks of Chandra/ACIS-I and JVLA/C and D array polarization data. The X-ray surface brightness profile reveals a merging gas-core reminiscent of the Bullet Cluster. The global X-ray luminosity in the 0.5–7.0 keV band is 1.7±0.1×10{sup 44} erg s{sup −1} and the global X-ray temperature is 4.90 ± 0.13 keV. The radio relics are polarized up to 40%, and along with the masses, velocities, and positions of the two subclusters, we input these quantities into a Monte Carlo dynamical analysis and estimate the merger velocity at pericenter to be 1800{sub −300}{sup +400} km s{sup −1}. This is a lower-mass version of the Bullet Cluster and therefore may prove useful in testing alternative models of dark matter (DM). We do not find significant offsets between DM and galaxies, but the uncertainties are large with the current lensing data. Furthermore, in the east, the BCG is offset from other luminous cluster galaxies, which poses a puzzle for defining DM–galaxy offsets.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2014

A rare encounter with very massive stars in NGC 3125-A1

Super star cluster A1 in the nearby starburst galaxy NGC 3125 is characterized by broad He II λ1640 emission (FWHM ∼ 1200 km s{sup –1}) of unprecedented strength (equivalent width, EW = 7.1 ± 0.4 Å). Previous attempts to characterize the massive star content in NGC 3125-A1 were hampered by the low resolution of the UV spectrum and the lack of co-spatial panchromatic data. We obtained far-UV to near-IR spectroscopy of the two principal emitting regions in the galaxy with the Space Telescope Imaging Spectrograph and the Cosmic Origins Spectrograph on board the Hubble Space Telescope. We use these data to study three clusters in the galaxy, A1, B1, and B2. We derive cluster ages of 3-4 Myr, intrinsic reddenings of E(B – V) = 0.13, 0.15, and 0.13, and cluster masses of 1.7 × 10{sup 5}, 1.4 × 10{sup 5}, and 1.1 × 10{sup 5} M {sub ☉}, respectively. A1 and B2 show O V λ1371 absorption from massive stars, which is rarely seen in star-forming galaxies, and have Wolf-Rayet (WR) to O star ratios of N(WN5-6)/N(O) = 0.23 and 0.10, respectively. The high N(WN5-6)/N(O) ratio of A1 cannot be reproduced by models that use a normal initial mass function (IMF) and generic WR star line luminosities. We rule out that the extraordinary He II λ1640 emission and O V λ1371 absorption of A1 are due to an extremely flat upper IMF exponent, and suggest that they originate in the winds of very massive (>120 M {sub ☉}) stars. In order to reproduce the properties of peculiar clusters such as A1, the present grid of stellar evolution tracks implemented in Starburst99 needs to be extended to masses >120 M {sub ☉}.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

DICER: Data Intensive Computing Environment and Runtime for Evaluating Unprecedented Scale of Geospatial-Temporal Human Mobility Data

With the significant increase in sources and volume of human mobility data through commercial data vendors as well as microsimulation of cities, the scale of geospatial-temporal data to analyze and assess for mobility characterization has grown to the level of Big Data. There are mobility related commercial organizations deploying scalable computing, but often the system architecture, workflow, and intermediate processing components are not fully disclosed in relevant scope. Current research literature has a notable lack of studies demonstrating architectures and workflows for human mobility analytics that are implemented on a TeraByte scale of geospatial-temporal data. In this context, this paper presents a hyperscale-level system solution named DICER (Data Intensive Computing Environment and Runtime) for processing and analytics of geospatial-temporal data at big data scale. Although the cluster computing architecture of DICER with Apache Spark job running on Kubernetes cluster is not new, there are innovations in the workflow, hierarchical processing logic, and a wide range of intermediate preprocessing and mobility metrics calculation. We have performed case studies to validate the effectiveness of DICER system solution by performing detailed analytics and assessment of human mobility microsimulation output at three different scopes and scale, including a usecase with 16.97 TeraByte and 259.2 Billion rows of data. In addition, we have presented another case study of utilizing DICER to perform the same mobility processing and comparative analytics on large-scale commercially available geospatial-temporal data. All these case studies validate the efficiency and usefulness of DICER in computing population mobility characteristics from geospatial-temporal trajectory data at an unprecedented scale (not only just data volume, but also combination of: number of user entities, temporal frequency, spatial resolution, data duration).

De, Debraj↗
DOE OSTI2016

SU-D-204-01: A Methodology Based On Machine Learning and Quantum Clustering to Predict Lung SBRT Dosimetric Endpoints From Patient Specific Anatomic Features

Purpose: To develop a data-mining methodology based on quantum clustering and machine learning to predict expected dosimetric endpoints for lung SBRT applications based on patient-specific anatomic features. Methods: Ninety-three patients who received lung SBRT at our clinic from 2011–2013 were retrospectively identified. Planning information was acquired for each patient, from which various features were extracted using in-house semi-automatic software. Anatomic features included tumor-to-OAR distances, tumor location, total-lung-volume, GTV and ITV. Dosimetric endpoints were adopted from RTOG-0195 recommendations, and consisted of various OAR-specific partial-volume doses and maximum point-doses. First, PCA analysis and unsupervised quantum-clustering was used to explore the feature-space to identify potentially strong classifiers. Secondly, a multi-class logistic regression algorithm was developed and trained to predict dose-volume endpoints based on patient-specific anatomic features. Classes were defined by discretizing the dose-volume data, and the feature-space was zero-mean normalized. Fitting parameters were determined by minimizing a regularized cost function, and optimization was performed via gradient descent. As a pilot study, the model was tested on two esophageal dosimetric planning endpoints (maximum point-dose, dose-to-5cc), and its generalizability was evaluated with leave-one-out cross-validation. Results: Quantum-Clustering demonstrated a strong separation of feature-space at 15Gy across the first-and-second Principle Components of the data when the dosimetric endpoints were retrospectively identified. Maximum point dose prediction to the esophagus demonstrated a cross-validation accuracy of 87%, and the maximum dose to 5cc demonstrated a respective value of 79%. The largest optimized weighting factor was placed on GTV-to-esophagus distance (a factor of 10 greater than the second largest weighting factor), indicating an intuitively strong correlation between this feature and both endpoints. Conclusion: This pilot study shows that it is feasible to predict dose-volume endpoints based on patient-specific anatomic features. The developed methodology can potentially help to identify patients at risk for higher OAR doses, thus improving the efficiency of treatment planning. R01-184173.

60 APPLIED LIFE SCIENCES↗
NASA NTRS2005

Impacts of the Variability of Ice Types on the Decline of the Arctic Perennial Sea Ice Cover

The observed rapid decline in the Arctic perennial ice cover is one of the most remarkable signal of change in the Arctic region. Updated data now show an even higher rate of decline of 9.8% per decade than the previous report of 8.9% per decade mainly because of abnormally low values in the last 4 years. To gain insights into this decline, the variability of the second year ice, which is the relatively thin component of the perennial ice cover, and other ice types is studied. The perennial ice cover in the 1990s was observed to be highly variable which might have led to higher production of second year ice and may in part explain the observed ice thinning during the period and triggered further decline. The passive microwave signature of second year ice is also studied and results show that while the signature is different from that of the older multiyear ice, it is surprisingly more similar to that of first year ice. This in part explains why previous estimates of the area of multiyear ice during the winter period are considerably lower than the area of the perennial ice cover during the preceding summer. Four distinct clusters representing radiometrically different types have been identified using multi-channel cluster analysis of passive microwave data. Data from two of these clusters, postulated to come from second year and older multiyear ice regions are also shown to have average thicknesses of 2.4 and 4.1 m, respectively, indicating that the passive microwave data may contain some ice thickness information that can be utilized for mass balance studies. The yearly anomaly maps indicate high gains of first year ice cover in the Arctic during the last decade which means higher production of second year ice and fraction of this type in the declining perennial ice cover. While not the only cause, the rapid decline in the perennial ice cover is in part caused by the increasing fractional component of the thinner second year ice cover that is very vulnerable to total melt due to warming in the Arctic, especially in spring.

Comiso, Josefino C.↗
DOE OSTI2021

Irradiation-enhanced diffusion and diffusion-limited creep in U 3 Si 2

U 3 Si 2 is an advanced fuel candidate due to its relatively high fissile density and attractive thermal properties. Compared to standard UO 2 fuel, there are significant data gaps for the thermophysical and thermomechanical properties of U 3 Si 2 . Point defect concentrations and mobilities under irradiation govern a number of important fuel performance properties, such as creep and fission gas release. In this work, we utilized density functional theory (DFT) data to inform a cluster dynamics framework to predict point defect concentrations in U 3 Si 2 under irradiation. Molecular dynamics (MD) simulations were used to examine the contribution of atomic mixing during ballistic cascades to diffusion, as well as the diffusivity of U and Si at grain boundaries. These atomic scale models for diffusivity were then used to inform a creep model based on bulk (Nabarro-Herring) and grain boundary (Coble) diffusional creep, and climb-limited dislocation creep. The model compares well against available experimental data and has been implemented in the BISON fuel performance code. Finally, a demonstration case using simple power profiles has been carried out, showing that negligible creep occurs due to the low temperatures experienced by U 3 Si 2 in-reactor, a consequence of its high thermal conductivity.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2026

VoroClust: Scalable Clustering for Remote Sensing

Although supervised machine learning provides a powerful framework for image classification and segmentation, it requires comprehensive consistent datasets, which are not available for many remote-sensing applications. Remote-sensing datasets are expensive to collect, and each is acquired under different environmental conditions or with significant variations in system operating parameters. Unsupervised clustering algorithms analyze the structure of each dataset independently, rather than drawing on similarities with existing “training” examples, and are thus well suited for practical remote-sensing applications. We introduce VoroClust, a fast density-based unsupervised clustering algorithm applicable to high-resolution and high-dimensional data. VoroClust runs as fast as distance-based clustering methods, while capturing complex regional geometries at least as well as current-density-based methods. It uses a data-centered sphere cover to reduce computational demands, while still capturing data topology. It then propagates clusters outward from local peaks in density. We show that VoroClust provides fast state-of-the-art clustering for both high-resolution polarimetric synthetic aperture radar and high-dimensional hyperspectral imaging datasets.

42 ENGINEERING↗
NASA NTRS1991

The relation between FIR and H I emission in galaxies

This project explored the relationship between the global far-infrared and neutral hydrogen (H I) emission from galaxies, based on data from the Infrared Astronomy Satellite (IRAS) and published radio data. 100 and 60 micron IRAS fluxes were used to establish a temperature corrected measure of the cold dust emission, and H I fluxes were drawn from the literature with the greatest possible consistency. The degree of correlation between the FIR and H I fluxes was found to be better than in previous studies, comparable to the correlation previously found between FIR and CO fluxes. The improvement was obtained largely by (1) separating 'stripped' from 'unstripped' galaxies, and (2) using compatible sources of H I data. Stripping occurs in clusters of galaxies and is probably caused by ram-pressure effects as a galaxy travels through the intergalactic medium. Our results suggest that stripped galaxies have had their outer-disk gas removed (approximately 80% of their total H I) while retaining most of their 100-micron-emitting dust. This strongly shifts the ratio of their 100-micron-to-H I fluxes. The second problem, arising from diverse sources of data, arises because differing telescopes and observational techniques give rise to substantial disagreement in the measured H I flux, and this degrades the correlation of the FIR and H I fluxes.

Schneider, Stephen E.↗
DOE OSTI2018

Time resolved surface photovoltage measurements using a big data capture approach to KPFM

Optoelectronic behavior in materials such as organic/inorganic hybrid perovskites depend on a complex interplay between fast (electronic) and slower (ionic) processes. These processes are thought to be influenced by structural inhomogeneities (e.g. interfaces and grain boundaries) bringing forward the necessity for development of techniques capable of correlating nanostructure and photo-transport behavior. While Kelvin probe force microscopy (KPFM) is ideally suited to map surface potentials on relevant length scales, it lacks sufficient temporal resolution to extract the meaningful system dynamics. Here, we develop a time resolved surface photovoltage (SPV) measurement based on full information capture of the photodetector stream during open loop KPFM operation. G-Mode, or G-KPFM allows quantification of SPV with microsecond temporal and nanoscale spatial resolution. Using this technique, we observe concurrent spatial and fast temporal variations in the SPV generated across a methylammonium lead bromide (MAPbBr3) thin film, a possible indicator relating microstructure with heterogenous photo-transport behavior. Here, we further demonstrate the advantage of adopting big data analytics including unsupervised clustering methods to quickly discern spatial variability in the information rich SPV dataset. Beyond G-KPFM, such clustering methods will be useful for interpretation of the multidimensional datasets arising from the growing number of time resolved KPFM approaches now available.

36 MATERIALS SCIENCE↗
NASA NTRS2020

Assisted Authoring of Model-Based Systems Engineering Documents

In systems engineering practices system design and analysis have historically been performed using a document-centric approach where stakeholders produce a number of documents that represent their views on a system under development. Given the ad-hoc, disparate, and informal nature of natural language documents, these views become quickly inconsistent. Rigor in engineering work is also lost in the transition from model based engineering design and analysis to engineering documents. Once the documents are delivered, the engineering portion of the work is disconnected. In the Open Model Based Engineering Environment (OpenMBEE), Cross-References (aka transclusions) synthesize data for engineering when the information is not simply hyperlinked, but referenced in place in a document, upgrading a document-based process with model-based engineering technology. Those Cross-References are nowadays partially created manually, putting a burden on the engineer who is authoring the document.This paper presents an approach which can assist the engineer by providing machine-generated suggestions for Cross-References using language processing, graph analysis, and clustering technologies on model data managed by the OpenMBEE infrastructure.

Gomes, Ivan↗
DOE OSTI2015

The Atacama Cosmology Telescope: The LABOCA/ACT survey of clusters at all Redshifts

We present a multi-wavelength analysis of 11 Sunyaev–Zel’dovich effect (SZE)-selected galaxy clusters (10 with new data) from the Atacama Cosmology Telescope (ACT) southern survey. We have obtained new imaging from the Large APEX Bolometer Camera (345 GHz; LABOCA) on the Atacama Pathfinder EXperiment (APEX) telescope, the Australia Telescope Compact Array (2.1 GHz; ATCA), and the Spectral and Photometric Imaging Receiver (250, 350, and 500 μm; SPIRE) on the Herschel Space Observatory.{sup 24} Spatially resolved 345 GHz SZE increments with integrated signal-to-noise ratio (S/N)>5 are found in six clusters. We compute 2.1 GHz number counts as a function of cluster-centric radius and find significant enhancements in the counts of bright sources at projected radii θ<θ{sub 2500c}. By extrapolating in frequency, we predict that the combined signals from 2.1 GHz-selected radio sources and 345 GHz-selected submillimeter galaxies (SMGs) contaminate the 148 GHz SZE decrement signal by ∼5% and the 345 GHz SZE increment by ∼18%. After removing radio source and SMG emission from the SZE signals, we use ACT, LABOCA, and (in some cases) new Herschel SPIRE imaging to place constraints on the clusters’ peculiar velocities. The sample’s average peculiar velocity relative to the cosmic microwave background is 〈v{sub p}〉=153±383 km s{sup −1}.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

Imputation for multisource data with comparison and assessment techniques

Missing data are prevalent issue in analyses involving data collection. The problem of missing data is exacerbated for multisource analysis, where data from multiple sensors are combined to arrive at a single conclusion. In this scenario, it is more likely to occur and can lead to discarding a large amount of data collected; however, the information from observed sensors can be leveraged to estimate those values not observed. We propose two methods for imputation of multisource data, both of which take advantage of potential correlation between data from different sensors, through ridge regression and a state-space model. These methods, as well as the common median imputation, are applied to data collected from a variety of sensors monitoring an experimental facility. Performance of imputation methods is compared with the mean absolute deviation; however, rather than using this metric to solely rank themethods,we also propose an approach to identify significant differences. Imputation techniqueswill also be assessed by their ability to produce appropriate confidence intervals, through coverage and length, around the imputed values. Finally, performance of imputed datasets is compared with a marginalized dataset through a weighted k-means clustering. In general, we found that imputation through a dynamic linearmodel tended to be the most accurate and to produce the most precise confidence intervals, and that imputing the missing values and down weighting them with respect to observed values in the analysis led to the most accurate performance.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1978

Photometric studies of composite stellar systems. I - CO and JHK observations of E and S0 galaxies

Multiaperture infrared photometric observations of the central regions of 51 early-type galaxies and of the integrated light of five globular clusters are presented. These data are compared with selected optical observations and with various model predictions. The main results of the work are: (1) the observed parameters for the brighter galaxies, particularly the CO index and the V-K color, agree with the predictions of stellar synthesis models characterized by giant-dominated populations with mass/visual luminosity ratios less than 10; (2) the galaxian broad-band colors tend to redden with increasing luminosity and decreasing aperture size; (3) for the globular clusters, there is evidence that the integrated colors become redder with increasing metallicity; and (4) in bright galaxies the relative changes of U-V, V-J, and J-K as functions of radius may differ from the relative changes as functions of luminosity at a fixed radius.

Frogel, J. A.↗
DOE OSTI2014

Anthracene clusters and the interstellar infrared emission features

The unidentified infrared bands are ubiquitous in the interstellar medium and typically attributed to emission from neutral and ionized polycyclic aromatic hydrocarbons (or PAHs). The contribution of neutral PAH clusters to these bands has been impossible to determine due to a paucity of infrared spectral data. Here we investigated neutral clusters of the three-ring PAH anthracene using FTIR absorption spectroscopy of anthracene matrix-isolated at varying concentrations in solid argon. In order to determine likely cluster structures of the embedded molecules, we also calculated theoretical absorption spectra for the anthracene monomer through hexamer using density functional theory with a dispersion correction (DFT-D). The DFT-D calculations have been calibrated for the anthracene dimer using the second-order Møller-Plesset approach. Because there is some support for the hypothesis that three or four-ring PAHs are present in the Red Rectangle nebula, we discuss the application of our results to this nebula in particular as well as to the interstellar infrared emission in general.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1991

Monte Carlo Shower Counter Studies

Activities and accomplishments related to the Monte Carlo shower counter studies are summarized. A tape of the VMS version of the GEANT software was obtained and installed on the central computer at Gallaudet University. Due to difficulties encountered in updating this VMS version, a decision was made to switch to the UNIX version of the package. This version was installed and used to generate the set of data files currently accessed by various analysis programs. The GEANT software was used to write files of data for positron and proton showers. Showers were simulated for a detector consisting of 50 alternating layers of lead and scintillator. Each file consisted of 1000 events at each of the following energies: 0.1, 0.5, 2.0, 10, 44, and 200 GeV. Data analysis activities related to clustering, chi square, and likelihood analyses are summarized. Source code for the GEANT user subprograms and data analysis programs are provided along with example data plots.

Snyder, H. David↗
DOE OSTI2019

Experimental Demonstration of Ferroelectric Spiking Neurons for Unsupervised Clustering

© 2018 IEEE. We report the first experimental demonstration of ferroelectric field-effect transistor (FEFET) based spiking neurons. A unique feature of the ferroelectric (FE) neuron demonstrated herein is the availability of both excitatory and inhibitory input connections in the compact 1T-1FEFET structure, which is also reported for the first time for any neuron implementations. Such dual neuron functionality is a key requirement for bio-mimetic neural networks and represents a breakthrough for implementation of the third generation spiking neural networks (SNNs) - also reported herein for unsupervised learning and clustering on real world data for the first time. The key to our demonstration is the careful design of two important device level features: (1) abrupt hysteretic transitions of the FEFET with no stable states therein, and (2) the dynamic tunability of the FEFET hysteresis by bias conditions which allows for the inhibition functionality. Experimentally calibrated, multi-domain Preisach based FEFET models were used to accurately simulate the FE neurons and project their performance at scaled nodes. We also implement an SNN for unsupervised clustering and benchmark the network performance across analog CMOS and emerging technologies and observe (1) unification of excitatory and inhibitory neural connections, (2) STDP based learning, (3) lowest reported power (3.6nW) during classification, and (4) a classification accuracy of 93%.

Wang, Z↗