Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Function Estimation”

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 1,567 records · Page 87

Location and characteristics of the reconnection X-line deduced from low-altitude satellite and radar observations

We present an analysis of a cusp ion step observed between two poleward-moving events of enhanced ionospheric electron temperature. From the computed variation of the reconnection rate and the onset times of the associated ionospheric events, the distance between the satellite and the X-line can be estimated, but with a large uncertainty due to that in the determination of the low-energy cut-off of the ion velocity distribution function, f(E). Nevertheless, analysis of the time series f(t) shows the reconnection site to be on the dayside magnetopause, consistent with the pulsating cusp model, and the best estimate of the X-line location is 13 R(E) from the satellite. The ion precipitation is used to reconstruct the field-parallel part of the Cowley-D ion distribution function injected into the open low latitude boundary layer (LLBL) in the vicinity of the X-line. From this the Alfven speed, plasma density, magnetic field, parallel ion temperature, and flow velocity of the magnetosheath near the X-line can be derived.

Lockwood, M.↗

Causal correlation of foliar biochemical concentrations with AVIRIS spectra using forced entry linear regression

A major goal of airborne imaging spectrometry is to estimate the biochemical composition of vegetation canopies from reflectance spectra. Remotely-sensed estimates of foliar biochemical concentrations of forests would provide valuable indicators of ecosystem function at regional and eventually global scales. Empirical research has shown a relationship exists between the amount of radiation reflected from absorption features and the concentration of given biochemicals in leaves and canopies (Matson et al., 1994, Johnson et al., 1994). A technique commonly used to determine which wavelengths have the strongest correlation with the biochemical of interest is unguided (stepwise) multiple regression. Wavelengths are entered into a multivariate regression equation, in their order of importance, each contributing to the reduction of the variance in the measured biochemical concentration. A significant problem with the use of stepwise regression for determining the correlation between biochemical concentration and spectra is that of 'overfitting' as there are significantly more wavebands than biochemical measurements. This could result in the selection of wavebands which may be more accurately attributable to noise or canopy effects. In addition, there is a real problem of collinearity in that the individual biochemical concentrations may covary. A strong correlation between the reflectance at a given wavelength and the concentration of a biochemical of interest, therefore, may be due to the effect of another biochemical which is closely related. Furthermore, it is not always possible to account for potentially suitable waveband omissions in the stepwise selection procedure. This concern about the suitability of stepwise regression has been identified and acknowledged in a number of recent studies (Wessman et al., 1988, Curran, 1989, Curran et al., 1992, Peterson and Hubbard, 1992, Martine and Aber, 1994, Kupiec, 1994). These studies have pointed to the lack of a physical link between wavelengths chosen by stepwise regression and the biochemical of interest, and this in turn has cast doubts on the use of imaging spectrometry for the estimation of foliar biochemical concentrations at sites distant from the training sites. To investigate this problem, an analysis was conducted on the variation in canopy biochemical concentrations and reflectance spectra using forced entry linear regression.

Dawson, Terence P.↗

Comparative evaluation of workload estimation techniques in piloting tasks

Techniques to measure operator workload in a wide range of situations and tasks were examined. The sensitivity and intrusion of a wide variety of workload assessment techniques in simulated piloting tasks were investigated. Four different piloting tasks, psychomotor, perceptual, mediational, and communication aspects of piloting behavior were selected. Techniques to determine relative sensitivity and intrusion were applied. Sensitivity is the relative ability of a workload estimation technique to discriminate statistically significant differences in operator loading. High sensitivity requires discriminable changes in score means as a function of load level and low variation of the scores about the means. Intrusion is an undesirable change in the task for which workload is measured, resulting from the introduction of the workload estimation technique or apparatus.

Wierwille, W. W.↗

Rao-Blackwellization for Adaptive Gaussian Sum Nonlinear Model Propagation

When dealing with imperfect data and general models of dynamic systems, the best estimate is always sought in the presence of uncertainty or unknown parameters. In many cases, as the first attempt, the Extended Kalman filter (EKF) provides sufficient solutions to handling issues arising from nonlinear and non-Gaussian estimation problems. But these issues may lead unacceptable performance and even divergence. In order to accurately capture the nonlinearities of most real-world dynamic systems, advanced filtering methods have been created to reduce filter divergence while enhancing performance. Approaches, such as Gaussian sum filtering, grid based Bayesian methods and particle filters are well-known examples of advanced methods used to represent and recursively reproduce an approximation to the state probability density function (pdf). Some of these filtering methods were conceptually developed years before their widespread uses were realized. Advanced nonlinear filtering methods currently benefit from the computing advancements in computational speeds, memory, and parallel processing. Grid based methods, multiple-model approaches and Gaussian sum filtering are numerical solutions that take advantage of different state coordinates or multiple-model methods that reduced the amount of approximations used. Choosing an efficient grid is very difficult for multi-dimensional state spaces, and oftentimes expensive computations must be done at each point. For the original Gaussian sum filter, a weighted sum of Gaussian density functions approximates the pdf but suffers at the update step for the individual component weight selections. In order to improve upon the original Gaussian sum filter, Ref. [2] introduces a weight update approach at the filter propagation stage instead of the measurement update stage. This weight update is performed by minimizing the integral square difference between the true forecast pdf and its Gaussian sum approximation. By adaptively updating each component weight during the nonlinear propagation stage an approximation of the true pdf can be successfully reconstructed. Particle filtering (PF) methods have gained popularity recently for solving nonlinear estimation problems due to their straightforward approach and the processing capabilities mentioned above. The basic concept behind PF is to represent any pdf as a set of random samples. As the number of samples increases, they will theoretically converge to the exact, equivalent representation of the desired pdf. When the estimated qth moment is needed, the samples are used for its construction allowing further analysis of the pdf characteristics. However, filter performance deteriorates as the dimension of the state vector increases. To overcome this problem Ref. [5] applies a marginalization technique for PF methods, decreasing complexity of the system to one linear and another nonlinear state estimation problem. The marginalization theory was originally developed by Rao and Blackwell independently. According to Ref. [6] it improves any given estimator under every convex loss function. The improvement comes from calculating a conditional expected value, often involving integrating out a supportive statistic. In other words, Rao-Blackwellization allows for smaller but separate computations to be carried out while reaching the main objective of the estimator. In the case of improving an estimator's variance, any supporting statistic can be removed and its variance determined. Next, any other information that dependents on the supporting statistic is found along with its respective variance. A new approach is developed here by utilizing the strengths of the adaptive Gaussian sum propagation in Ref. [2] and a marginalization approach used for PF methods found in Ref. [7]. In the following sections a modified filtering approach is presented based on a special state-space model within nonlinear systems to reduce the dimensionality of the optimization problem in Ref. [2]. First, the adaptive Gaussian sum propagation is explained and then the new marginalized adaptive Gaussian sum propagation is derived. Finally, an example simulation is presented.

state estimation↗

Error Estimation of An Ensemble Statistical Seasonal Precipitation Prediction Model

This NASA Technical Memorandum describes an optimal ensemble canonical correlation forecasting model for seasonal precipitation. Each individual forecast is based on the canonical correlation analysis (CCA) in the spectral spaces whose bases are empirical orthogonal functions (EOF). The optimal weights in the ensemble forecasting crucially depend on the mean square error of each individual forecast. An estimate of the mean square error of a CCA prediction is made also using the spectral method. The error is decomposed onto EOFs of the predictand and decreases linearly according to the correlation between the predictor and predictand. Since new CCA scheme is derived for continuous fields of predictor and predictand, an area-factor is automatically included. Thus our model is an improvement of the spectral CCA scheme of Barnett and Preisendorfer. The improvements include (1) the use of area-factor, (2) the estimation of prediction error, and (3) the optimal ensemble of multiple forecasts. The new CCA model is applied to the seasonal forecasting of the United States (US) precipitation field. The predictor is the sea surface temperature (SST). The US Climate Prediction Center's reconstructed SST is used as the predictor's historical data. The US National Center for Environmental Prediction's optimally interpolated precipitation (1951-2000) is used as the predictand's historical data. Our forecast experiments show that the new ensemble canonical correlation scheme renders a reasonable forecasting skill. For example, when using September-October-November SST to predict the next season December-January-February precipitation, the spatial pattern correlation between the observed and predicted are positive in 46 years among the 50 years of experiments. The positive correlations are close to or greater than 0.4 in 29 years, which indicates excellent performance of the forecasting model. The forecasting skill can be further enhanced when several predictors are used.

Shen, Samuel S. P.↗

Flexural analysis of uplifted rift flanks on Venus

Knowledge of the thermal structure of a planet is vital to a thorough understanding of its general scheme of tectonics. Since no direct measurements of heat flow or thermal gradient are available for Venus, most estimates have been derived from theoretical considerations or by analog with the Earth. The flexural response of the lithosphere to applied loads is sensitive to regional thermal structure. Under the assumption that the yield strength as a function of depth can be specified, the temperature gradient can be inferred from the effective elastic plate thickness. Previous estimates of the effective elastic plate thickness of Venus range from 11-18 km for the foredeep north of Uorsar Rupes to 30-60 km for the annular troughs around several coronae. Thermal gradients inferred for these regions are 14-23 K km(exp -1) and 4-9 K km(exp -1) respectively. In this study, we apply the same techniques to investigate the uplifted flanks of an extensional rift. Hypotheses for the origin of uplifted rift flanks on Earth include lateral transport of heat from the center of the rift, vertical transport of heat by small-scale convection, differential thinning of the lithosphere, dynamical uplift, and isostatic response to mechanical uploading of the lithosphere. The 1st hypothesis is considered the dominant contributor to terrestrial rift flanks lacking evidence for volcanic activity, particularly for rift structures that are no longer active. In this study, we model the uplifted flanks of a venusian rift as the flexural response to a vertical end load.

Evans, Susan A.↗

On the correction for quantization effects in signal-to-noise ratio estimation

In sampled data digital telemetry systems the signal to noise ratio (SNR) is typically derived as a function of the moments of the digitized input stream (e.g., the receiver output). This analog to digital conversion process is itself an additional noise source known as the quantization noise. Thus a digitally measured SNR will only approximately represent the SNR of the analog input signal. A procedure (Sheppard's corrections) for correcting moments of any order for this quantization effect is presented.

Howard, L.↗

Tracing structural optima as a function of available resources by a homotopy method

Optimization problems are typically solved by starting with an initial estimate and proceeding iteratively to improve it until the optimum is found. The design points along the path from the initial estimate to the optimum are usually of no value. The present work proposes a strategy for tracing a path of optimum solutions parameterized by the amount of available resources. The paper specifically treats the optimum design of a structure to maximize its buckling load. Equations for the optimum path are obtained using Lagrange multipliers, and solved by a homotopy method. The solution path has several transitions from unimodal to bimodal solutions. The Lagrange multipliers and second-order optimality conditions are used to detect branching points and to switch to the optimum solution path. The procedure is applied to the design of a foundation which supports a column for maximum buckling load. Using the total available foundation stiffness as a homotopy parameter, a set of optimum foundation designs is obtained.

Haftka, Raphael T.↗

Magnetospheric Multiscale (MMS) Mission Attitude Ground System Design

This paper presents an overview of the attitude ground system (AGS) currently under development for the Magnetospheric Multiscale (MMS) mission. The primary responsibilities for the MMS AGS are definitive attitude determination, validation of the onboard attitude filter, and computation of certain parameters needed to improve maneuver performance. For these purposes, the ground support utilities include attitude and rate estimation for validation of the onboard estimates, sensor calibration, inertia tensor calibration, accelerometer bias estimation, center of mass estimation, and production of a definitive attitude history for use by the science teams. Much of the AGS functionality already exists in utilities used at NASA's Goddard Space Flight Center with support heritage from many other missions, but new utilities are being created specifically for the MMS mission, such as for the inertia tensor, accelerometer bias, and center of mass estimation. Algorithms and test results for all the major AGS subsystems are presented here.

Sedlak, Joseph E.↗

Static shape determination and control for large space structures. I - The flexible beam. II - A large space antenna

A method for determining and controlling the shape of large, continuous space structures by means of discrete or pointwise observations and control devices is presented. The general linear boundary value problem satisfied by a one-dimensional shape function is defined, and the existence of solutions is studied. The static shape control problems for one-dimensional systems with and without rigid body modes and the static shape estimation problem are presented and solved. Eigenfunction expansions are presented which provide approximations to the algorithm terms when the associated Green's function is not known. An integral operator approach is applied to the multidimensional static problem, and the results are illustrated with a finite element model of the disk of a large space antenna which assumes no rigid body modes. It is shown that the shape control algorithm must be modified for systems with rigid body modes.

Weeks, C. J.↗

The computation of C-C and N-N bond dissociation energies for singly, doubly, and triply bonded systems

The bond dissociation energies (De) of C2H2, C2H4, C2H6, N2, N2H2, and N2H4 are studied at various levels of correlation treatment. The convergence of De with respect to the one-particle basis is studied at the single-reference modified coupled-pair-functional (MCPF)level. At all levels of correlation treatment, the errors in the bond dissociation energies increase with the degree of multiple bond character. The multireference configuration-interaction (MRCI) De values, corrected for an estimate of higher excitations, are in excellent agreement with those determined using the size-extensive averaged-coupled-pair-functional (ACPF) method. The full-valence complete-active-space self-consistent-field (CASSCF)/MRCI calculations are reproduced very well by MRCI calculations based on a CASSCF calculation that includes in the active space only those electrons involved in the C-C or N-N bonds.

Langhoff, Stephen R.↗

Predictor-corrector procedures for thermal buckling analysis of multilayered composite plates

A study is made of two predictor-corrector procedures for the accurate determination of global, as well as detailed, thermal buckling response characteristics of composite plates. Both procedures use first-order shear deformation theory in the predictor phase, but differ in the elements of the computational model being adjusted in the corrector phase. The first procedure calculates a posteriori estimates of the composite correction factors and uses them to adjust the transverse shear stiffnesses of the plate. The second procedure calculates a posteriori the functional dependence of the displacement components on the thickness coordinate. The corrected quantities are then used in conjunction with three-dimensional equations to obtain better estimates for the different response quantities. Extensive numerical results are presented, showing the effects of variation in the geometric and lamination parameters for antisymmetrically laminated composite plates subjected to uniform temperature rise, on the accuracy of the thermal buckling response obtained by predictor-corrector procedures.

Noor, A. K.↗

Satellite Estimation of Spectral Surface UV Irradiance: Effect of Horizontally Homogeneous Clouds - 2

The local variability of UV irradiance at the Earth's surface is mostly caused by clouds in addition to the seasonal variability. Parametric representations of radiative transfer RT calculations are presented for the convenient solution of the transmission T of ultraviolet radiation through plane parallel clouds over a surface with reflectivity R(sub s). The calculations are intended for use with the Total Ozone Mapping Spectrometer (TOMS) measured radiances to obtain the calculated Lambert equivalent scene reflectivity R for scenes with and without clouds. The purpose is to extend the theoretical analysis of the estimation of UV irradiance from satellite data for a cloudy atmosphere. Results are presented for a range of cloud optical depths and solar zenith angles for the cases of clouds over a low reflectivity surface R(sub s) less than 0.1, over a snow or ice surface R(sub s) greater than 0.3, and for transmission through a non-conservative scattering cloud with single scattering albedo omega(sub 0) = 0.999. The key finding for conservative scattering is that the cloud-transmission function C(sub T), the ratio of cloudy-to clear-sky transmission, is roughly C(sub T) = 1 - R(sub c) with an error of less than 20% for nearly overhead sun and snow-free surfaces. For TOMS estimates of UV irradiance in the presence of both snow and clouds, independent information about snow albedo is needed for conservative cloud scattering. For non-conservative scattering with R(sub s) greater than 0.5 (snow) the satellite measured scene reflectance cannot be used to estimate surface irradiance. The cloud transmission function has been applied to the calculation of UV irradiance at the Earth's surface and compared with ground-based measurements.

Krothov, N.↗

Probabilities and statistics for backscatter estimates obtained by a scatterometer

Methods for the recovery of winds near the surface of the ocean from measurements of the normalized radar backscattering cross section must recognize and make use of the statistics (i.e., the sampling variability) of the backscatter measurements. Radar backscatter values from a scatterometer are random variables with expected values given by a model. A model relates backscatter to properties of the waves on the ocean, which are in turn generated by the winds in the atmospheric marine boundary layer. The effective wind speed and direction at a known height for a neutrally stratified atmosphere are the values to be recovered from the model. The probability density function for the backscatter values is a normal probability distribution with the notable feature that the variance is a known function of the expected value. The sources of signal variability, the effects of this variability on the wind speed estimation, and criteria for the acceptance or rejection of models are discussed. A modified maximum likelihood method for estimating wind vectors is described. Ways to make corrections for the kinds of errors found for the Seasat SASS model function are described, and applications to a new scatterometer are given.

Pierson, Willard J., Jr.↗

An Empirical Temperature Variance Source Model in Heated Jets

An acoustic analogy approach is implemented that models the sources of jet noise in heated jets. The equivalent sources of turbulent mixing noise are recognized as the differences between the fluctuating and Favre-averaged Reynolds stresses and enthalpy fluxes. While in a conventional acoustic analogy only Reynolds stress components are scrutinized for their noise generation properties, it is now accepted that a comprehensive source model should include the additional entropy source term. Following Goldstein s generalized acoustic analogy, the set of Euler equations are divided into two sets of equations that govern a non-radiating base flow plus its residual components. When the base flow is considered as a locally parallel mean flow, the residual equations may be rearranged to form an inhomogeneous third-order wave equation. A general solution is written subsequently using a Green s function method while all non-linear terms are treated as the equivalent sources of aerodynamic sound and are modeled accordingly. In a previous study, a specialized Reynolds-averaged Navier-Stokes (RANS) solver was implemented to compute the variance of thermal fluctuations that determine the enthalpy flux source strength. The main objective here is to present an empirical model capable of providing a reasonable estimate of the stagnation temperature variance in a jet. Such a model is parameterized as a function of the mean stagnation temperature gradient in the jet, and is evaluated using commonly available RANS solvers. The ensuing thermal source distribution is compared with measurements as well as computational result from a dedicated RANS solver that employs an enthalpy variance and dissipation rate model. Turbulent mixing noise predictions are presented for a wide range of jet temperature ratios from 1.0 to 3.20.

Khavaran, Abbas↗

Robust Inversion and Data Compression in Control Allocation

We present an off-line computational method for control allocation design. The control allocation function delta = F(z)tau = delta (sub 0) (z) mapping commanded body-frame torques to actuator commands is implicitly specified by trim condition delta (sub 0) (z) and by a robust pseudo-inverse problem double vertical line I - G(z) F(z) double vertical line less than epsilon (z) where G(z) is a system Jacobian evaluated at operating point z, z circumflex is an estimate of z, and epsilon (z) less than 1 is a specified error tolerance. The allocation function F(z) = sigma (sub i) psi (z) F (sub i) is computed using a heuristic technique for selecting wavelet basis functions psi and a constrained least-squares criterion for selecting the allocation matrices F (sub i). The method is applied to entry trajectory control allocation for a reusable launch vehicle (X-33).

Hodel, A. Scottedward↗

X-ray and gamma-ray emission of Sagittarius A* as a wind-accreting black hole

If, as many believe, Sgr A* is a massive black hole at the Galactic center, one should expect it to be a source of X-ray and gamma-ray activity, behaving basically as a scaled-down active galactic nucleus. An unavoidable source of accretion is the wind from IRS 16, a nearby group of hot, massive stars. Since the density and velocity of the accreting matter are known from observations, the accretion rate is basically a function of the putative black hole mass, M(sub h), only; this value represents a reliable lower limit to a real rate, given the other possible sources of accreting matter. Based on this and on the theories about shock acceleration in active galactic nuclei, we have estimated the expected production of relativistic particles and their hard radiation. These values turn out to be a function of M(sub h) as well. Comparing our results with available X-ray and gamma-ray observations which show Sgr A* to have a relatively low activity level, we conclude tentatively that the putative black hole in the Galactic center cannot have a mass greater than approximately 6 x 10(exp 3) solar mass. This conclusion is consistent with the upper limits to the black hole mass found by different methods earlier, although much more work is needed to make calculations of shock acceleration around black holes more reliable.

Mastichiadis, A.↗

Determination of design and operation parameters for upper atmospheric research instrumentation to yield optimum resolution with deconvolution, appendix 4

The power spectrum for a stationary random process can be defined with the Wiener-Khintchine Theorem, which says that the power spectrum and the auto correlation function are a Fourier transform pair. To implement this theorem for signals that are discrete and of finite length we can use the Blackman-Tukey method. Blackman and Tukey (1958) show that a function w(tau), called a lag window, can be applied to the auto correlation estimates to obtain power spectrum estimates that are statistically stable. The Fourier transform of w(r) is called a spectral window. Typical choices for spectral windows show a distinct trade-off between the main lobe width and side lobe strength. A new idea for designing windows by taking linear combinations of the standard windows to produce hybrid windows was introduced by Smith (1985). We implement Smith's idea to obtain spectral windows with narrow main lobes and smaller (compared with typical windows) near side lobes. One of the main contributions of this thesis is that we show that Smith's problem is equivalent to a Quadratic Programming (QP) problem with linear equality and inequality constraints. A computer program was written to produce hybrid windows by setting up and solving the QP problem. We also developed and solved two variations of the original problem. The two variations involved changing the inequality constraints in both cases from non negativity on the combination coefficients to non negativity on the hybrid lag window itself. For the second variation, the window functions used to construct the hybrid window were changed to a frequency-variable set of truncated cosinusoids. A series of tests was run with the three computer programs to investigate the behavior of the hybrid spectral and lag windows. Emphasis was put on obtaining spectral windows with both relatively narrow main lobes and the lowest possible (for these algorithms) near side lobes. Some success was achieved for this goal. A 10 dB peak side lobe reduction over the rectangular spectral window without significant main lobe broadening was achieved. Also, average side lobe levels of -117 dB were reached at a cost of doubling the main lobe width (at the -3 dB point).

Ioup, George E.↗