SEARCH · Search NASA
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.
Explore the source record for details and available documents.
Sequential vector estimation of matrix of second partial derivatives, suggesting alternative least squares method
Explore the source record for details and available documents.
Wake/blade-row interaction noise produced by the Annular Cascade Facility at Purdue University has been modeled using the LINFLO analysis. Actuator displacements needed for complete cancellation of the propagating acoustic response modes have been determined, along with the associated actuator power requirements. As an alternative, weighted least squares minimization of the total far-field sound power using individual actuators has also been examined. Attempts were made to translate the two-dimensional aerodynamic results into three-dimensional actuator requirements. The results lie near the limit of present actuator technology. In order to investigate the concept of noise control at the source for active rotor/stator noise control at the source, various techniques for embedding miniature actuators into vanes were examined. Numerous miniature speaker arrangements were tested and analyzed to determine their suitability as actuators for a demonstration test in the Annular Cascade Facility at Purdue. The best candidates demonstrated marginal performance. An alternative concept to using vane mounted speakers as control actuators was developed and tested. The concept uses compression drivers which are mounted externally to the stator vanes. Each compression driver is connected via a tube to an air cavity in the stator vane, from which the driver signal radiates into the working section of the experimental rig. The actual locations and dimensions of the actuators were used as input parameters for a LINFLO computational analysis of the actuator displacements required for complete cancellation of tones in the Purdue experimental rig. The actuators were designed and an arrangement determined which is compatible with the Purdue experimental rig and instrumentation. Experimental tests indicate that the actuators are capable of producing equivalent displacements greater than the requirements predicted by the LINFLO analysis. The acoustic output of the actuators was also found to be unaffected by the presence of air flow representative of the Purdue experimental rig. A test of the active noise control at the source concept for rotor/stator active noise control was demonstrated. This 2-D test demonstrated conclusively the simultaneous reduction of two acoustic modes. Reductions of over 10 dB were obtained over a wide operating range.
Research activities conducted under the auspices of National Aeronautics and Space Administration Cooperative Agreement NCC 9-9 are discussed. During this contract period research efforts are concentrated in two primary areas. The first are is an investigation of the use of measurement error models as alternatives to least squares regression estimators of crop production or timber biomass. The secondary primary area of investigation is on the estimation of the mixing proportion of two-component mixture models. This report lists publications, technical reports, submitted manuscripts, and oral presentation generated by these research efforts. Possible areas of future research are mentioned.
Principal Component Thermography applies Singular Value Decomposition (SVD) to post-process data that are derived from active thermographic inspections. SVD provides useful compression of the data and allows for better understanding of substructure and indications of potential damage. In the standard approach, SVD is applied to a certain reshaping of a three-dimensional data stack into a two-dimensional array. This work applies the CANDECOMP-PARAFAC (CP) tensor rank decomposition directly to the three-dimensional data to avoid the initial reshaping step in order to begin to develop an inspection method that can more accurately detect defects in non-homogeneous and anisotropic materials. Tests against simulated data that compare the CP decomposition method with traditional Principal Component Thermography based on SVD are described. Finally, the method of Proper Generalized Decomposition (PGD) is used to derive the CP decomposition, and its performance against other algorithms is also discussed.
Earth-based spacecraft tracking data have historically been processed with classical least squares filtering techniques both for navigation purposes and for physical constant determination. The small, stochastic nongravitational forces acting on the spacecraft are described to motivate the use of sequential estimation as an alternative to the least squares fitting procedures. The stochastic forces are investigated both in terms of their effect on the tracking data and their influence on estimation accuracy. A flexible sequential filter design which leaves the existing trajectory, variational equations, data observable and partial computations undisturbed is described. A detailed filter design is presented that meets the precision demands and flexibility requirements of deep space navigation and of scientific problems.
The equations of rotational motion for a spacecraft equipped with external jets and internal reaction wheels are shown to be feedback-equivalent to those of a linear system in attitude parameter space. Reorientation maneuvers are thereby formulated as linear optimal control problems with least mean square acceleration in attitude parameter space, solved in closed form and implementable either with internal or external torque commands, the choice depending on power and throttling requirements. For prior detumbling, an alternative solution with least mean square torque by angular momentum feedback is also given, that is implementable with gimbaled pairs of thrusters at constant throttle. Such a detumbling maneuver may then be followed by an acceleration-commanded rest-to-rest maneuver by means of the reaction wheels.
The first-order system least-squares methodology represents an alternative to standard mixed finite element methods. Among its advantages is the fact that the finite element spaces approximating the pressure and flux variables are not restricted by the inf-sup condition and that the least-squares functional itself serves as an appropriate error measure. This paper studies the first-order system least-squares approach for scalar second-order elliptic boundary value problems with discontinuous coefficients. Ellipticity of an appropriately scaled least-squares bilinear form of the size of the jumps in the coefficients leading to adequate finite element approximation results. The occurrence of singularities at interface corners and cross-points is discussed. and a weighted least-squares functional is introduced to handle such cases. Numerical experiments are presented for two test problems to illustrate the performance of this approach.
In this paper, we introduce implicit gradient methods as alternatives to conventional least-squares gradient methods for second- and third-order nodal-gradient cell-centered finite-volume discretizations, where solutions are stored at cells but gradients are stored at nodes. Because of the unique configuration of solutions and gradients, implicit gradient systems developed for the node-centered edge-based discretization method can be directly applied once the numerical solutions are interpolated from cells to nodes with sufficient accuracy. The resulting defect-correction solver can be loosely coupled with a flow-equation solver, and at convergence, solutions and gradients that satisfy the corresponding residual equations are obtained. Each iteration is relatively cheap compared with least-squares methods involving hundreds of neighbors. Numerical results are presented for accuracy verification studies and some simple but realistic flow problems.
Usually the theoretical analysis of the Navier-Stokes equations is conducted via the Galerkin method which leads to difficult saddle-point problems. This paper demonstrates that the least-squares method is a useful alternative tool for the theoretical study of partial differential equations since it leads to minimization problems which can often be treated by an elementary technique. The principal part of the Navier-Stokes equations in the first-order velocity-pressure-vorticity formulation consists of two div-curl systems, so the three-dimensional div-curl system is thoroughly studied at first. By introducing a dummy variable and by using the least-squares method, this paper shows that the div-curl system is properly determined and elliptic, and has a unique solution. The same technique then is employed to prove that the Stokes equations are properly determined and elliptic, and that four boundary conditions on a fixed boundary are required for three-dimensional problems. This paper also shows that under four combinations of non-standard boundary conditions the solution of the Stokes equations is unique. This paper emphasizes the application of the least-squares method and the div-curl method to derive a high-order version of differential equations and additional boundary conditions. In this paper, an elementary method (integration by parts) is used to prove Friedrichs' inequalities related to the div and curl operators which play an essential role in the analysis.
Crustal magnetic anomaly signals over satellite orbits were simulated to investigate numerical averaging as an anomaly estimator. Averaging as an anomaly estimator involves significant problems concerning spatial and amplitude smoothing of the satellite magnetic observations. The results of simulations suggest that the error of numerical averaging constitutes a small and relatively minor component of the total error-budget of higher orbital anomaly estimates, whereas for lower orbital estimates numerical averaging error increases substantially. As an alternative to numerical averaging, least-squares collocation was investigated and observed to produce substantially more accurate anomaly estimates, particularly as the orbital elevation of prediction was decreased towards the crustal sources. In contrast to averaging, collocation is a significantly more resource-intensive procedure to apply because of the practical, but surmountable problems related to establishing and inverting the covariance matrix for accurate anomaly prediction. However, collocation may be much more effectively used to exploit the anomaly details contained in the lower orbital satellite magnetic data for geologic analysis.
Vector potential and related methods, for the simulation of both inviscid and viscous flows over aerodynamic configurations, are briefly reviewed. The advantages and disadvantages of several formulations are discussed and alternate strategies are recommended. Scalar potential, modified potential, alternate formulations of Euler equations, least-squares formulation, variational principles, iterative techniques and related methods, and viscous flow simulation are discussed.
The errors of numerically averaging satellite magnetic anomaly data for geologic analysis are investigated using orbital anomaly simulations of crustal magnetic sources by Gauss-Legendre quadrature integration. These simulations suggest that numerical averaging errors constitute small and relatively minor contributions to the total error-budget of higher orbital estimates (approx. 400 km), whereas for lower orbital estimates the error of averaging may increase substantially. Least-squares collocation is also investigated as an alternative to numerical averaging and found to produce substantially more accurate anomaly estimates as the elevation of prediction is decreased towards the crustal sources.
Calibration algorithms for a synthetic aperture microwave radiometer are presented. The calibration is geared to Earth remote sensing applications, and is demonstrated on an airborne prototype of thinned array imager. Two approaches to the system calibration are presented; the first utilizes commonly available reference brightness temperature scenes, such as open water, and the second utilizes data collected on the antenna range. Both algorithms yield spatial response information which is cast in matrix form and inverted to obtain the image reconstruction formula. Experimental results are examined, and errors in some reconstructed images are linked to the present prototype antenna design. Algorithms for improving the synthesized antenna pattern side lobe performance are also presented. In one solution, the pattern efficiency is optimized by minimizing the pattern outside a defined beam. In another solution, the patterns are matched to a desired model pattern by the method of least squared errors. Both techniques offer an attractive alternative to aperture weighting.
In this paper new real-time approaches for three-axis magnetometer sensor calibration are derived. These approaches rely on a conversion of the magnetometer-body and geomagnetic-reference vectors into an attitude independent observation by using scalar checking. The goal of the full calibration problem involves the determination of the magnetometer bias vector, scale factors and non-orthogonality corrections. Although the actual solution to this full calibration problem involves the minimization of a quartic loss function, the problem can be converted into a quadratic loss function by a centering approximation. This leads to a simple batch linear least squares solution. In this paper we develop alternative real-time algorithms based on both the extended Kalman filter and Unscented filter. With these real-time algorithms, a full magnetometer calibration can now be performed on-orbit during typical spacecraft mission-mode operations. Simulation results indicate that both algorithms provide accurate integer resolution in real time, but the Unscented filter is more robust to large initial condition errors than the extended Kalman filter. The algorithms are also tested using actual data from the Transition Region and Coronal Explorer (TRACE).
Remote sensing uses of principal component analysis (PCA) of multispectral images include band selection and optimal color selection for display of information content. PCA has also been used for quantitative determination of mineral types and abundances given end member spectra. The preliminary results of the investigation of target transformation PCA (TTPCA) in the fourier domain to both identify end member spectra in an unknown spectrum, and to then calculate the relative concentrations of these selected end members are presented. Identification of endmember spectra in an unknown sample has previously been performed through bandmatching, expert systems, and binary classifiers. Both bandmatching and expert system techniques require the analyst to select bands or combinations of bands unique to each endmember. Thermal infrared mineral spectra have broad spectral features which vary subtly with composition. This makes identification of unique features difficult. Alternatively, whole spectra can be used in the classification process, in which case there is not need for an expert to identify unique spectra. Use of binary classifiers on whole spectra to identify endmember components has met with some success. These techniques can be used, along with a least squares fit approach on the endmembers identified, to derive compositional information. An alternative to the approach outlined above usese target transformation in conjunction with PCA to both identify and quantify the composition of unknown spectra. Preprocessing of the library and unknown spectra into the fourier domain, and using only a specific number of the components, allows for significant data volume reduction while maintaining a linear relationship in a Beer's Law sense. The approach taken here is to iteratively calculate concentrations, reducing the number of endmember components until only non-negative concentrations remain.
Stigmatic spectra between 4160 and 4385 A of the central regions of M31 and M32 as well as the K0 III star 51 Ori are presented which were obtained using a SEC-vidicon integrating television tube and the coude spectrograph of the Hale telescope. Line-of-sight velocity dispersions of approximately 130 and 55 km/s are determined for the nuclei of M31 and M32, respectively, by directly comparing their spectra with Gaussian-broadened spectra of 51 Ori. The analytical techniques are outlined, and alternative methods for estimating velocity dispersions are evaluated, including least-squares fitting of intensities and division of Fourier transforms. It is noted that the broadened spectrum of 51 Ori fits the nucleus of M32 and the bulge of M31 much better than the nucleus of M31 in the region between 4180 and 4370 A.