SEARCH · Engineering Papers
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.
The linearized equation of motion for the NASA/IEEE SCOLE configuration are developed. The derivation is based on the method of Lagrange and the equations are assembled into matrix second order form.
The spatial coherence method for obtaining fractional cloud cover from satellite imagery is extended to the case of multilayered cloud systems. Examples are presented in which simultaneous observations at 3.7 microns and 11 microns are used to solve a system of linear equations for the nonoverlapped fractional cover contributed by each of two layers. The retrieval relies on the assumption that the clouds reside in distinct, well-defined layers and are optically thick at the wavelengths of observation. Simultaneous observations at 3.7 microns and 11 microns of the separate layers indicate that the assumptions are generally valid. Owing to the reflection of solar radiation at 3.7 microns by low-level water clouds, the method is limited to nighttime observations.
Previous work in identification theory has been concerned with the general first order time derivative form. Linear mechanical systems, a large and important class, naturally have a second order form. This paper utilizes this additional structural information for the purpose of identification. A realization is obtained from input-output data, and then knowledge of the system input, output, and inertia matrices is used to determine a set of linear equations whereby we identify the remaining unknown system matrices. Necessary and sufficient conditions on the number, type and placement of sensors and actuators are given which guarantee identificability, and less stringent conditions are given which guarantee generic identifiability. Both a priori identifiability and a posteriori identifiability are considered, i.e., identifiability being insured prior to obtaining data, and identifiability being assured with a given data set.
A new analytic approach is applied to the problem of star formation fronts in differentially rotating disks recently studied by Cowie and Rybicki. A quasi-linear equation for the inclination angle of the front is derived, and its characteristic equations are solved. The stability of these fronts found numerically in the previous study is revealed to be a direct consequence of the structure of the front characteristics. The time-dependent solutions found here are potentially useful in describing the evolution of local wave fronts shearing in a galactic velocity field, as observed in spirals lacking 'grand-design' structure.
This paper presents a new syndrome decoding algorithm for the (n, k) convolutional codes (CC) which differs completely from an earlier syndrome decoding algorithm of Schalkwijk and Vinck. The new algorithm is based on the general solution of the syndrome equation, a linear Diophantine equation for the error polynomial vector E(D). The set of Diophantine solutions is a coset of the CC. In this error coset a recursive, Viterbi-like algorithm is developed to find the minimum weight error vector (circumflex)E(D). An example, illustrating the new decoding algorithm, is given for the binary nonsystemmatic (3, 1)CC. Previously announced in STAR as N83-34964
Theoretical predictions of thermospheric gravity wave oscillations are compared with observed neutral temperatures and velocities. The data were taken in February 1983 using a Fabry-Perot interferometer located on Greenland, close to impulse heat sources in the auroral oval. The phenomenon was modeled in terms of linearized equations of motion of the atmosphere on a slowly rotating sphere. Legendre polynomials were used as eigenfunctions and the transfer function amplitude surface was characterized by maxima in the wavenumber frequency plane. Good agreement for predicted and observed velocities and temperatures was attained in the 250-300 km altitude. The amplitude of the vertical velocity, however, was not accurately predicted, nor was the temperature variability. The vertical velocity did exhibit maxima and minima in response to corresponding temperature changes.
Much of the computational effort of the finite element process involves the solution of a system of linear equations. The coefficient matrix of this system, known as the global stiffness matrix, is symmetric, positive definite, and generally sparse. An important technique for reducing the time required to solve this system is substructuring or matrix partitioning. Substructuring is based on the idea of dividing a structure into pieces, each of which can then be analyzed relatively indepenently. As a result of this division, each point in the finite element discretization is either interior to a substructure or on a boundary between substructures. Contributions to the global stiffness matrix from connections between boundary points from the K(bb) matrix are reported. The triangularization of a general K(bb) matrix on a parallel machine is specifically discussed.
A method of determining spatial derivatives of flow quantities behind an expansion fan as a function of the curvature of the streamline behind the fan is developed. Taylor series expansions of flow quantities within the fan are used and boundary conditions satisfied to the first and second order so that the curvature of the characteristics in the fan may be determined. A system of linear equations for the spatial derivatives is then developed. An application of the method to shock coalescence including asymmetric effects is described.
The univariate and multivariate relationships of provocative measures used to produce motion sickness symptoms were described. Normative subjects were used to develop and cross-validate sets of linear equations that optimally predict motion sickness in parabolic flights. The possibility of reducing the number of measurements required for prediction was assessed. After describing the variables verbally and statistically for 159 subjects, a factor analysis of 27 variables was completed to improve understanding of the relationships between variables and to reduce the number of measures for prediction purposes. The results of this analysis show that none of variables are significantly related to the responses to parabolic flights. A set of variables was selected to predict responses to KC-135 flights. A series of discriminant analyses were completed. Results indicate that low, moderate, or severe susceptibility could be correctly predicted 64 percent and 53 percent of the time on original and cross-validation samples, respectively. Both the factor analysis and the discriminant analysis provided no basis for reducing the number of tests.
Many applications using LANDSAT images face a dilemma: the user needs a certain scene (for example, a flooded region), but that particular image may present interference or noise in form of horizontal stripes. During automatic analysis, this interference or noise may cause false readings of the region of interest. In order to minimize this interference or noise, many solutions are used, for instane, that of using the average (simple or weighted) values of the neighboring vertical points. In the case of high interference (more than one adjacent line lost) the method of averages may not suit the desired purpose. The solution proposed is to use a spline-like algorithm (weighted splines). This type of interpolation is simple to be computer implemented, fast, uses only four points in each interval, and eliminates the necessity of solving a linear equation system. In the normal mode of operation, the first and second derivatives of the solution function are continuous and determined by data points, as in cubic splines. It is possible, however, to impose the values of the first derivatives, in order to account for shapr boundaries, without increasing the computational effort. Some examples using the proposed method are also shown.
Synopses are given for NASA supported work in computer science at the University of Virginia. Some areas of research include: error seeding as a testing method; knowledge representation for engineering design; analysis of faults in a multi-version software experiment; implementation of a parallel programming environment; two computer graphics systems for visualization of pressure distribution and convective density particles; task decomposition for multiple robot arms; vectorized incomplete conjugate gradient; and iterative methods for solving linear equations on the Flex/32.
The integral-operator method is extended to line and continuum transfer calculations for static and expanding spherical radiating atmospheres. A relationship is defined between radiative transfer and the source function at particular optical depths. Linearized equations are developed for the mean intensity, monochromatic scattering, line transfer, and line transfer with absorption. Numerical solutions are obtained for each equation, as well as for background absorption and scattering, partial redistribution when the line source is frequency dependent, and line transfer in connection with the expansion velocity. Finally, a solution is found for spherical expansion in concert with partial redistribution and Doppler line broadening.
The cosmic ray source composition is shown to be related to the observed cosmic ray abundances by a set of linear equations which are valid for either ad hoc pathlength distributions or quantitative solutions to Galactic propagation models. These relations can be used to propagate abundances from the source or to the source, and the method described here allows a simple propagation of abundance uncertainties. A simultaneous analysis of B/C and (21-25)Fe indicates a need for truncation of the pathlength distribution.
The minimal residual (MR) method for the numerical solution of transonic potential flows is closely related to the conjugate gradient method, which has found widespread use in the solution of large sparse, symmetric, and positive-definite linear equations. The primary advantage of the MR method is its applicability to both symmetric and nonsymmetric matrices.
An integrated theory of cooling and resistive reconnection processes in flares and filaments of a sheared magnetic structure is proposed. Thermal and tearing phenomena in the structure were coupled by means of a temperature-dependent Coulomb instability, and the relevant linearized equations were solved numerically over a range of parameters. It is shown that an increase in the growth rate of instabilities in the structure was associated with a significant amount of magnetic reconnection, particularly at shorter wavelengths. Long wavelength reconnection was abetted by an increase in Coulomb resistivity due to the dominance of cooling at the X-point. Some implications of the above results for current models of solar coronal reconnection processes are discussed.
Estimates of wall effects in two-dimensional wind tunnel obtained using upwash measurements on two contours near test model. Method derived from combination of prior techniques that correct for wall effects. Improved method limited to flows described by linear equations. Method accurately predicted wall-induced velocities along centerline of theoretical wind tunnel and confirmed that wall adjustments substantially reduced wall interference.
Various graduate research activities in the field of computer science are reported. Among the topics discussed are: (1) failure probabilities in multi-version software; (2) Gaussian Elimination on parallel computers; (3) three dimensional Poisson solvers on parallel/vector computers; (4) automated task decomposition for multiple robot arms; (5) multi-color incomplete cholesky conjugate gradient methods on the Cyber 205; and (6) parallel implementation of iterative methods for solving linear equations.
Work performed during Summer 1984 (from June to Sept. 30) in the area of porous media for use in low temperature applications is discussed. Recent applications are in the area of vapor - liquid phase separation, pumping based on the fountain effect and related subsystems. Areas of potential applications of the latter are outlined in supplementary work. Experimental data have been developed. The linear equations of the two-fluid model are inspected critically in the light of forced convection evidence reported recently. It is emphasized that the Darcy permeability is a unique throughput quantity in the porous media application areas whose use will permit meaningful comparisons of data not only in one lab but also within a group of labs doing porous plug studies.