Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “least-squares regression”

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 19 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS1975

Consolidation of fatigue and fatigue-crack-propagation data for design use

Analytical methods developed for consolidation of fatigue and fatigue-crack-propagation data for use in design of metallic aerospace structural components are evaluated. A comprehensive file of data on 2024 and 7075 aluminums, Ti-6Al-4V alloy, and 300M steel was established by obtaining information from both published literature and reports furnished by aerospace companies. Analyses are restricted to information obtained from constant-amplitude load or strain cycling of specimens in air at room temperature. Both fatigue and fatigue-crack-propagation data are analyzed on a statistical basis using a least-squares regression approach. For fatigue, an equivalent strain parameter is used to account for mean stress or stress ratio effects and is treated as the independent variable; cyclic fatigue life is considered to be the dependent variable. An effective stress-intensity factor is used to account for the effect of load ratio on fatigue-crack-propagation and treated as the independent variable. In this latter case, crack-growth rate is considered to be the dependent variable. A two term power function is used to relate equivalent strain to fatigue life, and an arc-hyperbolic-tangent function is used to relate effective stress intensity to crack-growth rate.

Rice, R. C.↗
NASA NTRS1986

Rarefied aerodynamics and upper atmosphere density from multiple orbiter flight measurements

Flight data from six flights of the experimental High Resolution Accelerometer Package (HIRAP), a micro-g accelerometer system, have been analyzed to produce rarefied flow aerodynamic coefficients of the Orbiter and freestream density in the altitude range from 60 km to 160 km. The direct measurements of the lift to drag ratio (L/D) are used to obtain individual normalized body-axis force coefficients in a least-squares regression scheme. An aerodynamics model based on an exponential function of Knudsen number is used for the individual coefficients. The calculated L/D from the flight determined coefficients agree with the multiple flight data measurements to within about 5 percent. Simultaneously, an upper altitude density variation is also obtained from the data. Density altitude profiles exhibit a wave feature on all flights. The wave has an amplitude of about + or - 25 percent relative to the 1962 Standard atmosphere model.

Buck, G. M.↗
NASA NTRS1992

On the distribution of pitch angles in external galactic spirals NGC 1232 and NGC 5457

A numerical method, originally developed to analyze the morphology of global and local structure in prototype galaxies, is modified for analyzing observed disk-shape galaxies. Two digitized spiral galaxies NGC 1232 and NGC 5457 with varying degrees of contrast between arm and interarm regions are analyzed. A synergism of partitioning methods and a geometric mean least-squares regression algorithm serves to isolate local arm segments, spurs, feathers, and secondary features and to measure their pitch angles and lengths. The global arms are actually highly disjointed, with arm segments frequently revealing pitch angles between 30 and 50 deg, certainly greater than those of the parent arms. Prominent spurs tend to exhibit a much greater pitch angle. The automated mathematical algorithm is shown to have negligible numerical biasing and could be applied to any number of spiral galaxies manifesting flocculent structure, either prototype or observed, and could possibly be used as a tool for classification of multiple-armed-type galaxies.

Russell, William S.↗
NASA NTRS1990

Error analysis for reducing noisy wide-gap concentric cylinder rheometric data for nonlinear fluids - Theory and applications

This work discusses the propagation of errors for the recovery of the shear rate from wide-gap concentric cylinder viscometric measurements of non-Newtonian fluids. A least-square regression of stress on angular velocity data to a system of arbitrary functions is used to propagate the errors for the series solution to the viscometric flow developed by Krieger and Elrod (1953) and Pawlowski (1953) ('power-law' approximation) and for the first term of the series developed by Krieger (1968). A numerical experiment shows that, for measurements affected by significant errors, the first term of the Krieger-Elrod-Pawlowski series ('infinite radius' approximation) and the power-law approximation may recover the shear rate with equal accuracy as the full Krieger-Elrod-Pawlowski solution. An experiment on a clay slurry indicates that the clay has a larger yield stress at rest than during shearing, and that, for the range of shear rates investigated, a four-parameter constitutive equation approximates reasonably well its rheology. The error analysis presented is useful for studying the rheology of fluids such as particle suspensions, slurries, foams, and magma.

Borgia, Andrea↗
NASA NTRS1990

Constrained optimization using design of experiment surfaces

An algorithm for solving constrained optimization problems is presented. First, design of experiment techniques are used to survey the design space. After evaluating the objective and constraint functions, as specified by Taguchi orthogonal arrays, analytical models of these functions are generated using a least-squares regression analysis. Next, a nonlinear programming package is used to optimize the analytical model. Based on the optimization information, the design space is reduced so as to close in around the minimum, and the entire procedure is repeated until convergence. An important feature of the algorithm is that function gradients are not required; therefore, for problems in which gradients would have to be estimated using finite-differences the number of function evaluations required for the optimization is significantly reduced, when compared with traditional nonlinear programming techniques. In addition, there is no requirement that the gradients must be smooth and continuous.

Bolt, Marvin Vance↗
NASA NTRS1992

Linear regression in astronomy. II

A wide variety of least-squares linear regression procedures used in observational astronomy, particularly investigations of the cosmic distance scale, are presented and discussed. The classes of linear models considered are (1) unweighted regression lines, with bootstrap and jackknife resampling; (2) regression solutions when measurement error, in one or both variables, dominates the scatter; (3) methods to apply a calibration line to new data; (4) truncated regression models, which apply to flux-limited data sets; and (5) censored regression models, which apply when nondetections are present. For the calibration problem we develop two new procedures: a formula for the intercept offset between two parallel data sets, which propagates slope errors from one regression to the other; and a generalization of the Working-Hotelling confidence bands to nonstandard least-squares lines. They can provide improved error analysis for Faber-Jackson, Tully-Fisher, and similar cosmic distance scale relations.

Feigelson, Eric D.↗
NASA NTRS1990

Linear regression in astronomy. I

Five methods for obtaining linear regression fits to bivariate data with unknown or insignificant measurement errors are discussed: ordinary least-squares (OLS) regression of Y on X, OLS regression of X on Y, the bisector of the two OLS lines, orthogonal regression, and 'reduced major-axis' regression. These methods have been used by various researchers in observational astronomy, most importantly in cosmic distance scale applications. Formulas for calculating the slope and intercept coefficients and their uncertainties are given for all the methods, including a new general form of the OLS variance estimates. The accuracy of the formulas was confirmed using numerical simulations. The applicability of the procedures is discussed with respect to their mathematical properties, the nature of the astronomical data under consideration, and the scientific purpose of the regression. It is found that, for problems needing symmetrical treatment of the variables, the OLS bisector performs significantly better than orthogonal or reduced major-axis regression.

Isobe, Takashi↗
NASA NTRS1975

Comparison of some biased estimation methods (including ordinary subset regression) in the linear model

Ridge, Marquardt's generalized inverse, shrunken, and principal components estimators are discussed in terms of the objectives of point estimation of parameters, estimation of the predictive regression function, and hypothesis testing. It is found that as the normal equations approach singularity, more consideration must be given to estimable functions of the parameters as opposed to estimation of the full parameter vector; that biased estimators all introduce constraints on the parameter space; that adoption of mean squared error as a criterion of goodness should be independent of the degree of singularity; and that ordinary least-squares subset regression is the best overall method.

Sidik, S. M.↗
NASA NTRS1982

Evaluation of SEASAT-A SMMR derived wind speed measurements

Wind speeds derived from versions of the least-squares and regression algorithms developed after the JASIN Workshop were evaluated. The accuracy of scanning multichannel microwave radiometer (SMMR) wind retrievals was determined in terms of the intrinsic accuracy of a baseline surface truth data set in favorable conditions. Effects which degrade the wind retrievals or introduce biases were identified and assessed. The performance of the SMMR in storms was ascertained with particular emphasis on the effects of rain.

Source record↗
NASA NTRS1989

Probabilistic constitutive relationships for material strength degradation models

In the present probabilistic methodology for the strength of aerospace propulsion system structural components subjected to such environmentally-induced primitive variables as loading stresses, high temperature, chemical corrosion, and radiation, time is encompassed as an interacting element, allowing the projection of creep and fatigue effects. A probabilistic constitutive equation is postulated to account for the degradation of strength due to these primitive variables which may be calibrated by an appropriately curve-fitted least-squares multiple regression of experimental data. The resulting probabilistic constitutive equation is embodied in the PROMISS code for aerospace propulsion component random strength determination.

Boyce, L.↗
NASA NTRS1969

Probabilistic-graphical and phenomenological analysis of combined bending-torsion fatigue reliability data

Fatigue data generated by three combined bending-torsion fatigue reliability research machines at The University of Arizona are probabilistic-graphically and phenomenologically analyzed. Distributions that are applicable to fatique life and static strength data are discussed. Phenomenological justifications for the use of these distributions are presented. It is found that the normal distribution represents the cycles-to-failure data at the highest stress levels best. The lognormal distribution appears to fit the cycles-to-failure data at the lower stress levels best and quite well at all stress levels including the highest. A regression analysis and least-squares goodness-of-fit test was performed for normal and lognormal plots. In most cases, the correlation coefficient gave a better fit to the data using the normal distribution, but the difference between the two was so slight that positive discrimination could not be made.

Kececioglu, D.↗
NASA NTRS1993

Analysis of the distribution of pitch angles in model galactic disks - Numerical methods and algorithms

An automated mathematical method capable of successfully isolating the many different features in prototype and observed spiral galaxies and of accurately measuring the pitch angles and lengths of these individual features is developed. The method is applied to analyze the evolution of specific features in a prototype galaxy exhibiting flocculent spiral structure. The mathematical-computational method was separated into two components. Initially, the galaxy was partitioned into dense regions constituting features using two different methods. The results obtained using these two partitioning algorithms were very similar, from which it is inferred that no numerical biasing was evident and that capturing of the features was consistent. Standard least-squares methods underestimated the true slope of the cloud distribution and were incapable of approximating an orientation of 45 deg. The problems were overcome by introducing a superior fit least-squares method, developed with the intention of calculating true orientation rather than a regression line.

Russell, William S.↗
NASA NTRS1986

Rain cell size statistics derived from radar observations at Wallops Island, Virginia

An investigation of two-dimensional rain cell size statistics has analyzed regression relations relating radar-determined rain rates to disdrometer data. This has yielded least-squares fits of radar reflectivity factors, and the application of a contouring program has generated 22,000 contours in which each isopleth belongs to predefined rain-rate intervals. An abundance of total and cell contours were observed belonging to all rain-rate categories. Both the computed number distributions and the conditional cumulative distributions as a function of contour diameter were found to be represented with good approximation by given exponential functions.

Goldhirsh, J.↗
NASA NTRS1985

Identification of pilot dynamics from in-flight tracking data

Data from a representative flight task involving an F-14 'pursuer' aircraft tracking a T-38 'target' aircraft in a 3G wind-up turn and in level flight are processed using a least squares identification technique in an attempt to identify pilot/vehicle dynamics. Comparative identification results are provided by a Fourier coefficient method which requires a carefully designed and implemented input consisting of a sum of sinusoids. The least-squares results compare favorably with those obtained by the Fourier technique. An example of crossover frequency regression is discussed in the light of the conditions of one of the flight configurations.

Hess, R. A.↗
NASA NTRS1991

A case study demonstration of the soil temperature extrema recovery rates after precipitation cooling at 10-cm soil depth

Since the invention of maximum and minimum thermometers in the 18th century, diurnal temperature extrema have been taken for air worldwide. At some stations, these extrema temperatures were collected at various soil depths also, and the behavior of these temperatures at a 10-cm depth at the Tifton Experimental Station in Georgia is presented. After a precipitation cooling event, the diurnal temperature maxima drop to a minimum value and then start a recovery to higher values (similar to thermal inertia). This recovery represents a measure of response to heating as a function of soil moisture and soil property. Eight different curves were fitted to a wide variety of data sets for different stations and years, and both power and exponential curves were fitted to a wide variety of data sets for different stations and years. Both power and exponential curve fits were consistently found to be statistically accurate least-square fit representations of the raw data recovery values. The predictive procedures used here were multivariate regression analyses, which are applicable to soils at a variety of depths besides the 10-cm depth presented.

Welker, Jean Edward↗
NASA NTRS1973

Critical configurations (determinantal loci) for range and range difference satellite networks

The observational modes of Geometric Satellite Geodesy are discussed. The geometrical analysis of the problem yielded a regression model for the adjustment of the observations along with a suitable and convenient metric for the least-squares criterion. The determinantal loci (critical configurations) for range networks are analyzed. An attempt is made to apply elements of the theory of variants for this purpose. The use of continuously measured range differences for loci determination is proposed.

Tsimis, E.↗
NASA NTRS1981

Microwave scatterometer measurements of oceanic wind vector

Results from Seasat-A scatterometer (SASS) remote sensing of oceanic wind vectors are reviewed, together with the theory of radar backscattering. The SASS was designed to measure the surface wind stress and neutral stability wind vector at 19.5 m altitude. Requirements included windspeeds from 4-24 m/sec to within 2 m/sec or 10%, a 1000 km swath, directions from 0-360 deg to within 20 deg, a 50 km resolution cell, and cross-track and along-track spacing between resolution cells of 100 km. The method was based on Bragg scattering of microwaves from cm-length capillary ocean waves. The strength of the backscatter is proportional to the capillary wave amplitude, which is in equilibrium with the ocean surface wind speed. The measurement grouping, least-squares estimator, and dB deviation model function table for the SASS geophysical algorithm are discussed. Comparisons of SASS, GOASEX, and ground truth wind data are made, and a 0.92 correlation between SASS and ground truth data is determined from regression analysis.

Jones, W. L.↗
NASA NTRS1982

Least-squares sequential parameter and state estimation for large space structures

This paper presents the formulation of simultaneous state and parameter estimation problems for flexible structures in terms of least-squares minimization problems. The approach combines an on-line order determination algorithm, with least-squares algorithms for finding estimates of modal approximation functions, modal amplitudes, and modal parameters. The approach combines previous results on separable nonlinear least squares estimation with a regression analysis formulation of the state estimation problem. The technique makes use of sequential Householder transformations. This allows for sequential accumulation of matrices required during the identification process. The technique is used to identify the modal prameters of a flexible beam.

Thau, F. E.↗