SEARCH · Engineering Papers

Results for “random sampling”

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 487 records · Page 27

All sources NASA NTRS DOE OSTI

NASA NTRS2015

Star Tracker Performance Estimate with IMU

A software tool for estimating cross-boresight error of a star tracker combined with an inertial measurement unit (IMU) was developed to support trade studies for the Integrated Radio and Optical Communication project (iROC) at the National Aeronautics and Space Administration Glenn Research Center. Typical laser communication systems, such as the Lunar Laser Communication Demonstration (LLCD) and the Laser Communication Relay Demonstration (LCRD), use a beacon to locate ground stations. iROC is investigating the use of beaconless precision laser pointing to enable laser communication at Mars orbits and beyond. Precision attitude knowledge is essential to the iROC mission to enable high-speed steering of the optical link. The preliminary concept to achieve this precision attitude knowledge is to use star trackers combined with an IMU. The Star Tracker Accuracy (STAcc) software was developed to rapidly assess the capabilities of star tracker and IMU configurations. STAcc determines the overall cross-boresight error of a star tracker with an IMU given the characteristic parameters: quantum efficiency, aperture, apparent star magnitude, exposure time, field of view, photon spread, detector pixels, spacecraft slew rate, maximum stars used for quaternion estimation, and IMU angular random walk. This paper discusses the supporting theory used to construct STAcc, verification of the program and sample results.

Error Budget↗

NASA NTRSDate not supplied

Mapping Post-fire Conifer Regeneration using Snow-on Imagery

The 2000 Jasper Fire in the Black Hills of South Dakota was the largest wildfire to date in the region, burning over 83,000 acres of ponderosa pine forest. In collaboration with partners from the United States Forest Service (USFS) Black Hills Experimental Forest, USFS Rocky Mountain Research Station, and United States Geological Survey Geosciences and Environmental Change Science Center, we characterized post-fire forest regeneration within high severity burn patches. We accomplished this by implementing novel conifer detection techniques using a snow index mask to create a winter, snow-on image composite from Landsat 8 Operational Land Imager (OLI) and Sentinel-2 Multispectral Instrument (MSI) data. We utilized 2015 USFS stem maps of field-observed regeneration plots and ocularly sampled reforestation sites planted from 2001–2013. The field data and imagery were used to train a Random Forest (RF) model in Google Earth Engine. The RF model classified conifer regeneration density as low, medium, or high across the high-severity burn area with an overall accuracy of 81.3% for 2021. Approximately 45.9% of the high-severity burn area had low or no regeneration (0-40 trees per acre) 20 years post-fire. Given our partners' desire to find easily accessible low conifer regeneration zones, we identified 4,079 acres of priority planting sites that were within 1,500 feet of roads, had not been planted previously, and were larger than 50 acres. This method supports the use of snow-on imagery as a successful technique to identify conifer regeneration in a post-wildfire landscape.

Yeshey Seldon↗

NASA NTRS1997

Visual signal detection in structured backgrounds. II. Effects of contrast gain control, background variations, and white noise

Studies of visual detection of a signal superimposed on one of two identical backgrounds show performance degradation when the background has high contrast and is similar in spatial frequency and/or orientation to the signal. To account for this finding, models include a contrast gain control mechanism that pools activity across spatial frequency, orientation and space to inhibit (divisively) the response of the receptor sensitive to the signal. In tasks in which the observer has to detect a known signal added to one of M different backgrounds grounds due to added visual noise, the main sources of degradation are the stochastic noise in the image and the suboptimal visual processing. We investigate how these two sources of degradation (contrast gain control and variations in the background) interact in a task in which the signal is embedded in one of M locations in a complex spatially varying background (structured background). We use backgrounds extracted from patient digital medical images. To isolate effects of the fixed deterministic background (the contrast gain control) from the effects of the background variations, we conduct detection experiments with three different background conditions: (1) uniform background, (2) a repeated sample of structured background, and (3) different samples of structured background. Results show that human visual detection degrades from the uniform background condition to the repeated background condition and degrades even further in the different backgrounds condition. These results suggest that both the contrast gain control mechanism and the background random variations degrade human performance in detection of a signal in a complex, spatially varying background. A filter model and added white noise are used to generate estimates of sampling efficiencies, an equivalent internal noise, an equivalent contrast-gain-control-induced noise, and an equivalent noise due to the variations in the structured background.

NASA Discipline Space Human Factors↗

NASA NTRS1993

Improving the chi-squared approximation for bivariate normal tolerance regions

Let X be a two-dimensional random variable distributed according to N2(mu,Sigma) and let bar-X and S be the respective sample mean and covariance matrix calculated from N observations of X. Given a containment probability beta and a level of confidence gamma, we seek a number c, depending only on N, beta, and gamma such that the ellipsoid R = (x: (x - bar-X)'S(exp -1) (x - bar-X) less than or = c) is a tolerance region of content beta and level gamma; i.e., R has probability gamma of containing at least 100 beta percent of the distribution of X. Various approximations for c exist in the literature, but one of the simplest to compute -- a multiple of the ratio of certain chi-squared percentage points -- is badly biased for small N. For the bivariate normal case, most of the bias can be removed by simple adjustment using a factor A which depends on beta and gamma. This paper provides values of A for various beta and gamma so that the simple approximation for c can be made viable for any reasonable sample size. The methodology provides an illustrative example of how a combination of Monte-Carlo simulation and simple regression modelling can be used to improve an existing approximation.

Feiveson, Alan H.↗

NASA NTRS2006

Sensitivity Analysis Applied to Atomic Data Used for X-ray Spectrum Synthesis

A great deal of work has been devoted to the accumulation of accurate quantities describing atomic processes for use in analysis of astrophysical spectra. But in many situations of interest the interpretation of a quantity which is observed, such as a line flux, depends on the results of a modeling- or spectrum synthesis code. The results of such a code depends in turn on many atomic rates or cross sections, and the sensitivity of the observable quantity on the various rates and cross sections may be non-linear and if so cannot easily be derived analytically. In such cases the most practical approach to understanding the sensitivity of observables to atomic cross sections is to perform numerical experiments, by calculating models with various rates perturbed by random (but known) factors. In addition, it is useful to compare the results of such experiments with some sample observations, in order to focus attention on the rates which are of the greatest relevance to real observations. In this paper I will present some attempts to carry out this program, focussing on two sample datasets taken with the Chandra HETG. I will discuss the sensitivity of synthetic spectra to atomic data affecting ionization balance, temperature, and line opacity or emissivity, and discuss the implications for the ultimate goal of inferring astrophysical parameters.

Kallman, Tim↗

NASA NTRS2006

Comparison of Interstitial Fluid pH, PCO2, PO2 with Venous Blood Values During Repetitive Handgrip Exercise

We evaluated the use of a small, fiber optic sensor to measure pH, PCO2 and PO2 from forearm muscle interstitial fluid (IF) during handgrip dynamometry. PURPOSE: Compare pH, PCO2 and PO2 values obtained from venous blood with those from the IF of the flexor digitorum superficialis (FDS) during three levels of exercise intensity. METHODS: Six subjects (5M/1F), average age 29+/-5 yrs, participated in the study. A venous catheter was placed in the retrograde direction in the antecubital space and a fiber optic sensor (Paratrend, Diametrics Medical, Inc.) was placed through a 22 G catheter into the FDS muscle under ultrasound guidance. After a 45 min rest period, subjects performed three 5-min bouts of repetitive handgrip exercise (2s contraction/1 s relaxation) at attempted levels of 15%, 30% and 45% of maximal voluntary contraction. The order of the exercise bouts was random with the second and third bouts started after blood lactate had returned to baseline. Venous blood was sampled every minute during exercise and analyzed with an I-Stat CG-4+ cartridge, while IF fiber optic sensor measurements were obtained every 2 s. Change from pre-exercise baseline to end of exercise was computed for pH, PCO2 and PO2. Blood and IF values were compared with a paired t-test. RESULTS: Baseline values for pH, PCO2 and PO2 were 7.37+/-0.02, 46+/-4 mm Hg, and 36+/-6 mm Hg respectively in blood and 7.39+/-0.02, 44+/-6 mm Hg, and 35+/-14 mm Hg in IF. Average changes over all exercise levels are noted in the Table below. For each parameter the exercise-induced change was at least twice as great in IF as in blood. In blood and IF, pH and PCO2 increases were directly related to exercise intensity. Change in venous PO2 was unrelated to exercise intensity, while IF PO2 decreased with increases in exercise intensity. CONCLUSIONS: Measurement of IF pH, PCO2 and PO2 is more sensitive to exercise intensity than measurement of the same parameters in venous blood and provides continuous assessment during and after exercise.

Hagan, Ronald Donald↗

NASA NTRS2006

Sensitivity Analysis Applied to Atomic Data Used for X-ray Spectrum Synthesis

A great deal of work has been devoted to the accumulation of accurate quantities describing atomic processes for use in analysis of astrophysical spectra. But in many situations of interest the interpretation of a quantity which is observed, such as a line flux, depends on the results of a modeling- or spectrum synthesis code. The results of such a code depends in turn 011 many atomic rates or cross sections, and the sensitivity of the observable quantity on the various rates and cross sections may be non-linear and if so cannot easily be derived analytically. In such cases the most practical approach to understanding the sensitivity of observables to atomic cross sections is to perform numerical experiments, by calculating models with various rates perturbed by random (but known) factors. In addition, it is useful to compare the results of such experiments with some sample observations, in order to focus attention on the rates which are of the greatest relevance to real observations. In this paper I will present some attempts to carry out this program, focussing on two sample datasets taken with the Chandra HETG. I will discuss the sensitivity of synthetic spectra to atomic data affecting ionization balance, temperature, and line opacity or emissivity, and discuss the implications for the ultimate goal of inferring astrophysical parameters.

Kallman, Tim↗

NASA NTRS2008

Testing Photoionization Calculations Using Chandra X-ray Spectra

Kallman, Tim↗

NASA NTRS2019

An Ensemble of Bayesian Neural Networks for Exoplanetary Atmospheric Retrieval

Machine learning (ML) is now used in many areas of astrophysics, from detecting exoplanets in Kepler transit signals to removing telescope systematics. Recent work demonstrated the potential of using ML algorithms for atmospheric retrieval by implementing a random forest (RF) to perform retrievals in seconds that are consistent with the traditional, computationally expensive nested-sampling retrieval method. We expand upon their approach by presenting a new ML model, plan-net, based on an ensemble of Bayesian neural networks (BNNs) that yields more accurate inferences than the RF for the same data set of synthetic transmission spectra. We demonstrate that an ensemble provides greater accuracy and more robust uncertainties than a single model. In addition to being the first to use BNNs for atmospheric retrieval, we also introduce a new loss function for BNNs that learns correlations between the model outputs. Importantly, we show that designing ML models to explicitly incorporate domain-specific knowledge both improves performance and provides additional insight by inferring the covariance of the retrieved atmospheric parameters. We apply plan-net to the Hubble Space Telescope Wide Field Camera 3 transmission spectrum for WASP-12b and retrieve an isothermal temperature and water abundance consistent with the literature. We highlight that our method is flexible and can be expanded to higher resolution spectra and a larger number of atmospheric parameters.

Adam D. Cobb↗

NASA NTRS2021

Salvaging Data Records with Missing Data: Data Imputation using the Multivariate t Distribution

When doing multivariate data analysis, one commonobstacle is the presence of incomplete observations, i.e., observationsfor which one or more key fields are blank. Missing datais often countered by deleting entire observations that containmissing data. The negative effects of deleting entire observationsare multiple: deleting observations reduces sample size andcan also result in biased inferences even if data is missing atrandom. In addition, knowledge contained within incompleteobservations is knowledge lost when they are deleted– and theeffort spent collecting that knowledge is effort wasted. Data imputationmethods, or methods of statistically “filling-in” missingdata, can help combat small sample sizes by using the existinginformation in partially complete observations with the end goalof producing less biased and higher confidence inferences. Whena sample from a multivariate normal population is only partiallycomplete, and the missing data meets appropriate assumptions(missing at random), robust data imputation of the missing datacan be implemented with monotone data augmentation (MDA)using the multivariate t distribution.Missing data imputation is applied to data from the NASA InstrumentCost Model (NICM) using the MDA algorithm underthe assumption of having a multivariate t distribution with fixeddegrees of freedom. A sensitivity analysis to the degrees offreedom parameter is presented to demonstrate robustness ofthe multivariate t distribution when dealing with small samplesas compared to the multivariate normal distribution.

DiNicola, Michael↗

NASA NTRS2010

Process to Selectively Distinguish Viable from Non-Viable Bacterial Cells

The combination of ethidium monoazide (EMA) and post-fragmentation, randomly primed DNA amplification technologies will enhance the analytical capability to discern viable from non-viable bacterial cells in spacecraft-related samples. Intercalating agents have been widely used since the inception of molecular biology to stain and visualize nucleic acids. Only recently, intercalating agents such as EMA have been exploited to selectively distinguish viable from dead bacterial cells. Intercalating dyes can only penetrate the membranes of dead cells. Once through the membrane and actually inside the cell, they intercalate DNA and, upon photolysis with visible light, produce stable DNA monoadducts. Once the DNA is crosslinked, it becomes insoluble and unable to be fragmented for post-fragmentation, randomly primed DNA library formation. Viable organisms DNA remains unaffected by the intercalating agents, allowing for amplification via post-fragmentation, randomly primed technologies. This results in the ability to carry out downstream nucleic acid-based analyses on viable microbes to the exclusion of all non-viable cells.

LaDuc, Myron T.↗

NASA NTRS1983

Effect of leaf variables on visible, near-infrared and mid-infrared reflectance of excised leaves

Effects of an imposed (excised) leaf orientation, differing species and differing venation patterns on reflectance measurements in the LANDSAT-4 thematic mapper (TM) channels TM3 (0.63 to 0.69 microns), TM4 (0.76 to 0.90 microns), and TM5 (1.55 to 1.75 microns) were investigated. Orientation of leaves (random vs. systematic placement) was found to affect measurements in the TM4 channel, but not the TM3 and TM5 measurements. Venation caused no significant changes for any band. Azimuth of incident radiation was not a significant main effect, but in conjunction with changes in orientation, angle did have a significant effect on reflectance values in TM3, TM4 and TM5. Specific differences were highly significant (P f or = 0.006) in all but one borderline (P F or = 0.0222) case for TM5. For spectral examination of excised leaves, the sampling arrangement of the leaves should as closely approximate in situ positioning as possible (with respect to remote sensing instrumentation). This dictates a random rather than aligned arrangement.

Bell, R.↗

NASA NTRS1987

Probabilistic finite elements

In the Probabilistic Finite Element Method (PFEM), finite element methods have been efficiently combined with second-order perturbation techniques to provide an effective method for informing the designer of the range of response which is likely in a given problem. The designer must provide as input the statistical character of the input variables, such as yield strength, load magnitude, and Young's modulus, by specifying their mean values and their variances. The output then consists of the mean response and the variance in the response. Thus the designer is given a much broader picture of the predicted performance than with simply a single response curve. These methods are applicable to a wide class of problems, provided that the scale of randomness is not too large and the probabilistic density functions possess decaying tails. By incorporating the computational techniques we have developed in the past 3 years for efficiency, the probabilistic finite element methods are capable of handling large systems with many sources of uncertainties. Sample results for an elastic-plastic ten-bar structure and an elastic-plastic plane continuum with a circular hole subject to cyclic loadings with the yield stress on the random field are given.

Belytschko, Ted↗

NASA NTRS1989

Fatigue strength reduction model: RANDOM3 and RANDOM4 user manual. Appendix 2: Development of advanced methodologies for probabilistic constitutive relationships of material strength models

FORTRAN programs RANDOM3 and RANDOM4 are documented in the form of a user's manual. Both programs are based on fatigue strength reduction, using a probabilistic constitutive model. The programs predict the random lifetime of an engine component to reach a given fatigue strength. The theoretical backgrounds, input data instructions, and sample problems illustrating the use of the programs are included.

Boyce, Lola↗

NASA NTRS1993

The large-scale morphology of IRAS galaxies

At present, visual inspection is the only method for comparing the large-scale morphologies in the distribution of galaxies to those in model universes generated by N-body simulations. To remedy the situation, we have developed a set of three structure functions (S1, S2, S3) that quantify the degree of large-scale prolateness, oblateness, and sphericity/uniformity of a 3-D particle distribution and have applied them to a volume-limited (less than = 4000 km/s) sample of 699 IRAS galaxies with f sub 60 greater than 1.2 Jy. To determine the structure functions, we randomly select 500 galaxies as origins of spherical windows of radius R sub w, locate the centroid of the galaxies in the window (assuming all galaxies have equal mass) and then, compute the principal moments of inertia (I sub 1, I sub 2, I sub 3) about the centroid. Each S sub i is a function of (I sub 2)/(I sub 1) and (I sub 3)/I sub 1). S1, S2, and S3 tend to unity for highly prolate, oblate, and uniform distributions, respectively and tend to zero otherwise. The resulting 500 values of S sub i at each scale R sub w are used to construct a histogram.

Babul, Arif↗

NASA NTRS1958

A Method for the Calculation of the Lateral Response of Airplanes to Random Turbulence

In this method, the gust velocities are represented as rolling gusts, yawing gusts, and side gusts. Random distributions of gust velocities across the span are taken into account in defining the rolling and yawing gusts. Complex stability derivatives are used to account for the random distribution of side gusts along the fuselage and vertical tail and the lag effects in gust penetration. A sample calculation procedure is presented for obtaining the response of the airplane in each degree of freedom.

LOADS, GUST - WINGS↗

NASA NTRS1993

Finite-size effects on long-range correlations: implications for analyzing DNA sequences

We analyze the fluctuations in the correlation exponents obtained for noncoding DNA sequences. We find prominent sample-to-sample variations as well as variations within a single sample in the scaling exponent. To determine if these fluctuations may result from finite system size, we generate correlated random sequences of comparable length and study the fluctuations in this control system. We find that the DNA exponent fluctuations are consistent with those obtained from the control sequences having long-range power-law correlations. Finally, we compare our exponents for the DNA sequences with the exponents obtained from power-spectrum analysis and correlation-function techniques, and demonstrate that the original "DNA-walk" method is intrinsically more accurate due to reduced noise.

NASA Discipline Cardiopulmonary↗

NASA NTRS2007

Derivation and Application of a Global Albedo yielding an Optical Brightness To Physical Size Transformation Free of Systematic Errors

Orbital object data acquired via optical telescopes can play a crucial role in accurately defining the space environment. Radar systems probe the characteristics of small debris by measuring the reflected electromagnetic energy from an object of the same order of size as the wavelength of the radiation. This signal is affected by electrical conductivity of the bulk of the debris object, as well as its shape and orientation. Optical measurements use reflected solar radiation with wavelengths much smaller than the size of the objects. Just as with radar, the shape and orientation of an object are important, but we only need to consider the surface electrical properties of the debris material (i.e., the surface albedo), not the bulk electromagnetic properties. As a result, these two methods are complementary in that they measure somewhat independent physical properties to estimate the same thing, debris size. Short arc optical observations such as are typical of NASA's Liquid Mirror Telescope (LMT) give enough information to estimate an Assumed Circular Orbit (ACO) and an associated range. This information, combined with the apparent magnitude, can be used to estimate an "absolute" brightness (scaled to a fixed range and phase angle). This absolute magnitude is what is used to estimate debris size. However, the shape and surface albedo effects make the size estimates subject to systematic and random errors, such that it is impossible to ascertain the size of an individual object with any certainty. However, as has been shown with radar debris measurements, that does not preclude the ability to estimate the size distribution of a number of objects statistically. After systematic errors have been eliminated (range errors, phase function assumptions, photometry) there remains a random geometric albedo distribution that relates object size to absolute magnitude. Measurements by the LMT of a subset of tracked debris objects with sizes estimated from their radar cross sections indicate that the random variations in the albedo follow a log-normal distribution quite well. In addition, this distribution appears to be independent of object size over a considerable range in size. Note that this relation appears to hold for debris only, where the shapes and other properties are not primarily the result of human manufacture, but of random processes. With this information in hand, it now becomes possible to estimate the actual size distribution we are sampling from. We have identified two characteristics of the space debris population that make this process tractable and by extension have developed a methodology for performing the transformation.

Mulrooney, Dr. Mark K.↗