SEARCH · Engineering Papers

Results for “clustering statistics”

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 235 records · Page 13

All sources NASA NTRS DOE OSTI

NASA NTRS2018

K-Means Cluster Study for Radiofrequency Propagation Characterization

The objective of this study is to design a simple method for mining radio frequency (RF) propagation data. The study explored the characteristics of a large dataset of propagation experiments conducted over the span of years and using several ground stations around the world. Furthermore, this study developed simple predictive models that can be used for link characterization and overall propagation behavior description, without the need for physical measurements on-site. It is understood that such statistical learning has several drawbacks in terms of accuracy and precision. K-means clustering was used to characterize the data set in a way never explored before in an attempt to create useful tools that reduce cost, time and risk. K-means clustering was used to characterize the data set. Cosine distance was used as a method to determine the optimal number for clustering each feature. Dependence and independence analysis was performed to explore intra and inter-sensitivity between the presented features, with respect to each other and time. Several predicative models were generated and evaluated with respect to a test set to assess a measure of prediction accuracy and precision. A simple method for data analysis was developed and tested as the basis for further studies and future refinement to produce optimal performing models.

Cognitive↗

NASA NTRS1995

Deep ROSAT HRI observations of the Orion nebula region

We present results from three deep ROSAT high-resolution imager observations of the Orion Nebula star-forming region. The X-ray images contain over 1500 cataloged stars in a roughly 0.8 sq deg region centered on the Trapezium. In all, 389 distinct X-ray sources have been detected, at least two-thirds of which are associated with a single proper-motion cluster member. X-ray emission is detected from stars of all spectral types, from massive O- and B-type components of the Trapezium to the coolest, low-mass pre-main-sequence (PMS) stars. In this paper we focus primarily on X-ray emission from the late-type PMS stars. Of the approximately 100 late-type cluster members with measured spectral types, approximately three-fourths have been detected; we have derived X-ray luminosity upper limits for the remaining stars. We found coronal X-ray emission turns on around spectral type F6, with the upper envelope of activity increasing with deceasing effective temperature. When plotted in an X-ray luminosity versus bolometric luminosity diagram, late-type PMS stars lie below a 'saturation' line corresponding to L(sub x)/L(sub bol) approximately 10(exp -3). For approximately solar-mass PMS stars, we find a median X-ray luminosity approximately 1 x 10(exp 30) ergs/s. The late G, K, and M stars exhibit nearly a two order of magnitude spread in X-ray luminosity and in L(sub x)/L(sub bol) at a given effective temperature. Plots of X-ray activity versus v sin i rotational velocity and rotational period appear to show no clear dependence of activity on rotation. However, because only a small fraction of late-type PMS stars in the Orion Nebula have measured v sin i or P(sub rot) and because of uncertainties in L(sub x) and L(sub x)/L(sub bol), we believe the data are not conclusive on this point. Light curves of the detected X-ray sources have revealed at least 10 strong X-ray flares with characteristic rise times greater than or approximately equal to 1 hr and decay times ranging from approximately 2 to 12 hr. All the flares have X-ray energies in excess of 3 x 10(exp 35) ergs. Many of the X-ray sources associated with late-type cluster members have been observed in more than one ROSAT HRI exposure, allowing us to look for variability on the approximately 1 yr timescale between HRI observations. A statistical analysis of the resulting count rate ratios suggests that at least one-fourth of the late-type cluster stars are significantly variable on this timescale.

Gagne, Marc↗

NASA NTRS2003

Density Irregularities Inside the Plasmasphere: Cluster Observations

The electron density profiles derived from the EFW and WHISPER instruments onboard the four Cluster spacecraft reveal small-scale density irregularities inside the plasmasphere and at its outer boundary, the plasmapause. We review statistics of the plasmapause position and thickness, as well as statistics of these density structures. We focus on a particular plasmasphere crossing on 11 April 2002, with several density irregularities, as well as two plasma tails, observed by Cluster on the legs of the inbound and outbound passes. We derive the density gradient vectors from simultaneous density measurements by the four spacecraft. We determine also the normal velocity of the surface of these irregularities, assuming they are planar boundaries, from the time delays between density structures in the four individual density profiles. These new observations yield novel insights about (1) the dimensions of plasma irregularities across and parallel to magnetic field lines, (2) the dynamics of these small-scale structures, (3) their bulk velocities, and (4) their position and distribution as a function of the magnetic local time (MLT) and the geomagnetic conditions (as determined by Kp).

Darrouzet, F.↗

NASA NTRS1993

A statistical analysis of mesoscale rainfall as a random cascade

The paper considers the random cascade theory for spatial rainfall. Particular attention was given to the following four areas: (1) the relationship of the random cascade theory of rainfall to the simple scaling and the hierarchical cluster-point-process theories, (2) the mathematical foundations for some of the formalisms commonly applied in the develpment of statistical cascade theory, (3) the empirical evidence for a random cascade theory of rainfall, and (4) the way of using data for making estimates of parameters and for making statistical inference within this theoretical framework. An analysis of space-time rainfall data is presented. Cascade simulations are carried out to provide a comparison with methods of analysis that are applied to the rainfall data.

Gupta, Vijay K.↗

NASA NTRS1983

Cloud cover typing from environmental satellite imagery. Discriminating cloud structure with Fast Fourier Transforms (FFT)

The use of two dimensional Fast Fourier Transforms (FFTs) subjected to pattern recognition technology for the identification and classification of low altitude stratus cloud structure from Geostationary Operational Environmental Satellite (GOES) imagery was examined. The development of a scene independent pattern recognition methodology, unconstrained by conventional cloud morphological classifications was emphasized. A technique for extracting cloud shape, direction, and size attributes from GOES visual imagery was developed. These attributes were combined with two statistical attributes (cloud mean brightness, cloud standard deviation), and interrogated using unsupervised clustering amd maximum likelihood classification techniques. Results indicate that: (1) the key cloud discrimination attributes are mean brightness, direction, shape, and minimum size; (2) cloud structure can be differentiated at given pixel scales; (3) cloud type may be identifiable at coarser scales; (4) there are positive indications of scene independence which would permit development of a cloud signature bank; (5) edge enhancement of GOES imagery does not appreciably improve cloud classification over the use of raw data; and (6) the GOES imagery must be apodized before generation of FFTs.

Logan, T. L.↗

NASA NTRS1986

Incipient fault detection study for advanced spacecraft systems

A feasibility study to investigate the application of vibration monitoring to the rotating machinery of planned NASA advanced spacecraft components is described. Factors investigated include: (1) special problems associated with small, high RPM machines; (2) application across multiple component types; (3) microgravity; (4) multiple fault types; (5) eight different analysis techniques including signature analysis, high frequency demodulation, cepstrum, clustering, amplitude analysis, and pattern recognition are compared; and (6) small sample statistical analysis is used to compare performance by computation of probability of detection and false alarm for an ensemble of repeated baseline and faulted tests. Both detection and classification performance are quantified. Vibration monitoring is shown to be an effective means of detecting the most important problem types for small, high RPM fans and pumps typical of those planned for the advanced spacecraft. A preliminary monitoring system design and implementation plan is presented.

Milner, G. Martin↗

NASA NTRS1989

Land Analysis System (LAS)

Version 4.1 of LAS provides flexible framework for algorithm development and processing and analysis of image data. Over 500,000 lines of code enable image repair, clustering, classification, film processing, geometric registration, radiometric correction, and manipulation of image statistics.

Pease, P. B.↗

NASA NTRS1992

Compositional analysis of projectile residues on LDEF instrument AO187-1

Impact craters greater than 30 microns and associated projectile residues were analyzed by Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Analysis (EDXA). Objectives were to analyze a statistically significant number of projectiles to evaluate their chemical variability and possible clustering into discrete particle types. Bay A11 exposed six collector surfaces of anodized 1100-T4 (greater than 99 percent pure) aluminum sheets, 0.32 cm thick, yielding an exposed surface area of 1.1 sq. m. Four of the six panels have been retained at JSC, and were optically scanned, one (A11E00E) was prepared for SEM/EDX analysis. Bay A03 was occupied by high purity (99.99 percent) gold sheets, 0.5 mm thick, yielding an exposed surface area of 0.85 sq. m. Sample processing included the optical scanning (6X), labeling, and dislodging (by a punch-die device) of each individual impact greater than 75 microns for the aluminum and 30 microns for the gold. The 209 craters were dislodged form A11E00E, having crater diameters up to 3500 microns. Optical examination of the gold surfaces detected 238 craters, 198 of which were retained at JSC and analyzed via SEM/EDX. The analytical procedures included maximizing the geometric efficiency (take-off angles), using relatively long count times (500-1000 sec) and sufficiently high accelerating currents (25-30Kev). Despite diligent examination, a large number of craters did not exhibit measurable signals above background. Detectable resides were classified as either micrometeoritic or as man-made debris.

Bernhard, Ronald P.↗

NASA NTRS2004

Small-Scale Variability of Large Cloud Drops

Cloud droplet size distribution is one of the most fundamental subjects in cloud physics. Understanding of spatial distribution and small-scale fluctuations of cloud droplets is essential for both cloud physics and atmospheric radiation. For cloud physics, it relates to the coalescence growth of raindrops while for radiation, it has a strong impact on a cloud's radiative properties. Most of the existing cloud radiation and precipitation formation models assume that the mean number of drops with a given radius varies proportionally to volume. The analysis of microphysical data on liquid water drop sizes shows that, for sufficiently small volumes, the number is proportional to the drop size dependent power of the volume. For abundant small drops present, the exponent is 1 as assumed in the conventional approach. However, for rarer large drops, the exponents fall below unity. At small scales, therefore, the mean number of large drops decreases with volume at a slower rate than the conventional approach assumes, suggesting more large drops at these scales than conventional models account for; their impact is consequently underestimated. Size dependent models of spatial distribution of cloud drops that simulate the observed power laws show strong drop clustering, the more so the larger the drops. The degree of clustering is determined by the observed exponents. The strong clustering of large drops arises naturally from the observed power-law statistics. Current theories of photon-cloud interaction and warm rain formation will need radical revision in order to produce these statistics; their underlying equations are unable to yield the observed power law.

Marshak, Alexander↗

NASA NTRS2006

Comparison of Orbital Parameters for GEO Debris Predicted by LEGEND and Observed by MODEST: Can Sources of Orbital Debris be Identified?

Since 2002 the National Aeronautics and Space Administration (NASA) has carried out an optical survey of the debris environment in the geosynchronous Earth-orbit (GEO) region with the Michigan Orbital Debris Survey Telescope (MODEST) in Chile. The survey coverage has been similar for 4 of the 5 years allowing us to follow the orbital evolution of Correlated Targets (CTs), both controlled and un-controlled objects, and Un-Correlated Targets (UCTs). Under gravitational perturbations the distributions of uncontrolled objects, both CTs and UCTs, in GEO orbits will evolve in predictable patterns, particularly evident in the inclination and right ascension of the ascending node (RAAN) distributions. There are several clusters (others have used a "cloud" nomenclature) in observed distributions that show evolution from year to year in their inclination and ascending node elements. However, when MODEST is in survey mode (field-of-view approx.1.3deg) it provides only short 5-8 minute orbital arcs which can only be fit under the assumption of a circular orbit approximation (ACO) to determine the orbital parameters. These ACO elements are useful only in a statistical sense as dedicated observing runs would be required to obtain sufficient orbital coverage to determine a set of accurate orbital elements and then to follow their evolution. Identification of the source(s) for these "clusters of UCTs" would be advantageous to the overall definition of the GEO orbital debris environment. This paper will set out to determine if the ACO elements can be used to in a statistical sense to identify the source of the "clustering of UCTs" roughly centered on an inclination of 12deg and a RAAN of 345deg. The breakup of the Titan 3C-4 transtage on February 21, 1992 has been modeled using NASA s LEGEND (LEO-to-GEO Environment Debris) code to generate a GEO debris cloud. Breakup fragments are created based on the NASA Standard Breakup Model (including fragment size, area-to-mass (A/M), and delta-V distributions). Once fragments are created, they are propagated forward in time with a subroutine GEOPROP. Perturbations included in GEOPROP are those due to solar/lunar gravity, radiation pressure, and major geopotential terms. The question to be addressed: are the UCTs detected by MODEST in this inclination/RAAN region related to the Titan 3C-4 breakup? Discussion will include the observational biases in attempting to detect a specific, uncontrolled target during given observing session. These restrictions include: (1) the length of the observing session which is 8 hours or less at any given date or declination; (2) the assumption of ACO elements for detected object when the breakup model predicts debris with non-zero eccentricities; (3) the size and illumination or brightness of the debris predicted by the model and the telescope/sky limiting magnitude.

Barker, Edwin S.↗

NASA NTRS1986

Statistics of density maxima and the large-scale matter distribution

High peaks in Gaussian noise display enhanced clustering. The enhancement takes two forms: on large scales one obtains a linear amplification of the correlation function which is independent of scale. On smaller scales, but larger than the mass scale of the peaks themselves, a nonlinear (exponential) enhancement of the number density of high peaks in overdense regions arises. The large-scale correlations of Abell's rich clusters can be understood as a manifestation of this phenomenon. If the formation of bright galaxies favors the high overdensity peaks then the number of galaxies (per unit mass) in clusters and groups may be considerably enhanced. Consequences of these ideas for the density parameter and the large-scale matter distribution are discussed.

Kaiser, N.↗

NASA NTRS2013

Application of Statistically Derived CPAS Parachute Parameters

The Capsule Parachute Assembly System (CPAS) Analysis Team is responsible for determining parachute inflation parameters and dispersions that are ultimately used in verifying system requirements. A model memo is internally released semi-annually documenting parachute inflation and other key parameters reconstructed from flight test data. Dispersion probability distributions published in previous versions of the model memo were uniform because insufficient data were available for determination of statistical based distributions. Uniform distributions do not accurately represent the expected distributions since extreme parameter values are just as likely to occur as the nominal value. CPAS has taken incremental steps to move away from uniform distributions. Model Memo version 9 (MMv9) made the first use of non-uniform dispersions, but only for the reefing cutter timing, for which a large number of sample was available. In order to maximize the utility of the available flight test data, clusters of parachutes were reconstructed individually starting with Model Memo version 10. This allowed for statistical assessment for steady-state drag area (CDS) and parachute inflation parameters such as the canopy fill distance (n), profile shape exponent (expopen), over-inflation factor (C(sub k)), and ramp-down time (t(sub k)) distributions. Built-in MATLAB distributions were applied to the histograms, and parameters such as scale (sigma) and location (mu) were output. Engineering judgment was used to determine the "best fit" distribution based on the test data. Results include normal, log normal, and uniform (where available data remains insufficient) fits of nominal and failure (loss of parachute and skipped stage) cases for all CPAS parachutes. This paper discusses the uniform methodology that was previously used, the process and result of the statistical assessment, how the dispersions were incorporated into Monte Carlo analyses, and the application of the distributions in trajectory benchmark testing assessments with parachute inflation parameters, drag area, and reefing cutter timing used by CPAS.

Romero, Leah M.↗

NASA NTRS1989

Superclustering in the explosion scenario

A simple toy model is used to study the spatial distribution of rich clusters in a generic type of explosion scenario. The model, parameterized by the distribution of shell radii and the filling factor, places spherical shells at random and identifies each 'knot' as a cluster. The resulting cluster correlation function is close to a power law extending to the diameter of the largest spheres. Richer clusters form at the intersections of bigger shells and so have stronger correlations. Typical shell radii and filling factors are required to produce the observed number density of clusters. Models with a power-law radius distribution also reproduce the richness distribution of clusters in the Abell catalog. Supercluster multiplicity functions, void probabilities, number counts, topology statistics, and velocity correlations confirm the presence of strong superclustering and quantify the non-Gaussian nature of the model.

Weinberg, David H.↗

NASA NTRS1991

Collected Notes on the Workshop for Pattern Discovery in Large Databases

These collected notes are a record of material presented at the Workshop. The core data analysis is addressed that have traditionally required statistical or pattern recognition techniques. Some of the core tasks include classification, discrimination, clustering, supervised and unsupervised learning, discovery and diagnosis, i.e., general pattern discovery.

Buntine, Wray↗

NASA NTRS2018

Fire Induced Rock Spalls as Long-Term Traps for Ash

Severe fires accelerate rock weathering by spalling and exfoliation, creating abundant peels, flakes or spalls. In the following years, these spalls serve as physical traps which accommodate fine particles of dust, ash, organic matter, etc. We searched for traces of ash trapped under the spalls, after four major fires: 1989, 2005, 2010 and 2015 in Israel. Samples were collected beneath the spalls that formed on the rock outcrops, and in the immediate vicinity above and below them. Three laboratory analyses were performed: pH, EC and color. Five mineral/organic compounds were measured across the Mid-Infrared (MIR) spectral region at diffuse reflectance infrared Fourier transform mode: Hydroxylapatite (HAp), charcoal, organic carbon, montmorillonite and kaolinite. Several statistical analyses were performed: MANOVA, PCA and silhouette analysis on K-means clustering. The results show evidence of ash trapped under the spalls formed during the 2005 and 2010 fires, 6 to 11 years after the fires. Charcoal presence is evident, as well as increased amounts of HAp and organic carbon. In the exposed soil above or below the burned rock outcrop, these values are lower. Negligible amounts of ash were measured 27 years after the fire. In the 2015 burned outcrop, large amounts of charcoal were found above and below the outcrop, but not under the spalls. It seems that on the carbonate slopes of Israel and under Mediterranean climate, the time required for spalls to begin functioning as traps is longer than one rainy season, while ash traces are preserved in these traps for a period of two-three decades.

ash↗

NASA NTRSDate not supplied

The Experiment for Cryogenic Large-aperture Intensity Mapping (EXCLAIM)

Submillimeter and far-IR spectroscopy provides insight into galaxy evolution through measurement of atomic and molecular line emission. The EXperiment for Cryogenic Large-Aperture Intensity Mapping (EXCLAIM) is a cryogenic balloon-borne instrument designed to carry out intensity mapping to measure the cumulative redshifted line emission from carbon monoxide and singly-ionized carbon to probe star formation in windows from the present to z=3.5. During this time, the rate of star formation dropped dramatically, while dark matter continued to cluster. Intensity mapping permits a blind and complete survey of emitting gas through statistics of cumulative brightness fluctuations. EXCLAIM achieves high sensitivity using a cryogenic telescope coupled to six integrated spectrometers with spectral resolving power R=512 and employing microwave kinetic inductance detectors. Here we summarize the status of the mission.

Maryam Rahmani↗

NASA NTRSDate not supplied

The Experiment for Cryogenic Large-aperture Intensity Mapping (EXCLAIM)

Submillimeter and far-IR spectroscopy provides insight into galaxy evolution through atomic and molecular line emission. The EXperiment for Cryogenic Large-Aperture Intensity Mapping (EXCLAIM) is a cryogenic balloon-borne instrument designed to carry out an intensity mapping to measure the cumulative redshifted line emission from carbon monoxide and singly-ionized carbon to probe star formation in windows from the present to z=3.5. During this time, the rate of star formation dropped dramatically, while dark matter continued to cluster. Intensity mapping permits a blind and complete survey of emitting gas through statistics of cumulative brightness fluctuations. EXCLAIM achieves high sensitivity using a cryogenic telescope coupled to six integrated spectrometers with spectral resolving power R = 512 and employing kinetic inductance detectors. Here we summarize the status of the mission.

Maryam Rahmani↗

NASA NTRSDate not supplied

The Experiment for Cryogenic Large-aperture Intensity Mapping (EXCLAIM)

Maryam Rahmani↗