SEARCH · Search NASA

Results for “generalized gradient approximations”

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 217 records · Page 12

All sources NASA NTRS DOE OSTI

NASA NTRS1980

Structure and circulation of the Venus atmosphere

The Pioneer Venus data relevant to the dynamics and thermodynamics of the atmosphere is summarized and interpreted. On the day side there is a thermosphere in which temperatures increase with height to an exospheric temperature of about 300 K. On the night side there is a cryosphere in which temperatures decrease with height to an exospheric temperature of about 100 K. The atmosphere is stratified stably from the highest altitudes down to about 28 km except for a layer in the clouds between about 50 and 55 km which is nearly adiabatic. Horizontal thermal contrasts are approximately 1 to 2% in the deep atmosphere and 100% in the upper atmosphere. The temperatures generally decrease with latitude at and below the clouds on constant pressure surfaces. Above the clouds there is a reversed zonally averaged latitudinal temperature gradient. The dominant circulation of the atmosphere above the lowest one or two scale heights is a zonal retrograde motion with 100 m/s winds at 60 km altitude. There is also a superrotation at altitudes of 150 km and above.

Schubert, G.↗

NASA NTRS1996

Multigrid, Fractional-Step Computation Of Flow

Speed of computer code solving three-dimensional Navier-Stokes equations of flow of incompressible fluid by fractional-step method increased significantly by use of multigrid procedures. In method, equations solved on general nonorthogonal curvilinear coordinate grid, using volume fluxes. At each time step, computations performed in two fractional steps. In first step, equations of conservation of momentum solved by use of gradient of pressure from previous time step via explicit approximate-factorization method, yielding approximate flow field that does not satisfy equation of conservation of mass. In second step, discrete Poisson-like equation with Neumann-type boundary conditions, formed by combining equations of conservation of momentum and mass, solved iteratively.

Kwak, Dochan↗

DOE OSTI2021

Bayesian sparse learning with preconditioned stochastic gradient MCMC and its applications

Deep neural networks have been successfully employed in an extensive variety of research areas, including solving partial differential equations. Despite its significant success, there are some challenges in effectively training DNN, such as avoiding overfitting in over-parameterized DNNs and accelerating the optimization in DNNs with pathological curvature. Here, we propose a Bayesian type sparse deep learning algorithm. The algorithm utilizes a set of spike-and-slab priors for the parameters in the deep neural network. The hierarchical Bayesian mixture will be trained using an adaptive empirical method. That is, one will alternatively sample from the posterior using preconditioned stochastic gradient Langevin Dynamics (PSGLD), and optimize the latent variables via stochastic approximation. The sparsity of the network is achieved while optimizing the hyperparameters with adaptive searching and penalizing. A popular SG-MCMC approach is Stochastic gradient Langevin dynamics (SGLD). However, considering the complex geometry in the model parameter space in nonconvex learning, updating parameters using a universal step size in each component as in SGLD may cause slow mixing. To address this issue, we apply a computationally manageable preconditioner in the updating rule, which provides a step-size parameter to adapt to local geometric properties. Moreover, by smoothly optimizing the hyperparameter in the preconditioning matrix, our proposed algorithm ensures a decreasing bias, which is introduced by ignoring the correction term in the preconditioned SGLD. According to the existing theoretical framework, we show that the proposed algorithm can asymptotically converge to the correct distribution with a controllable bias under mild conditions. Numerical tests are performed on both synthetic regression problems and learning solutions of elliptic PDE, which demonstrate the accuracy and efficiency of the present work.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

NASA NTRS2002

Application of the generalized vertical coordinate ocean model for better representing satellite data

It is found that two adaptive parametric functions can be introduced into the basic ocean equations for utilizing the optimal or hybrid features of commonly used z-level, terrain- following, isopycnal, and pressure coordinates in numerical ocean models. The two parametric functions are formulated by combining three techniques: the arbitrary vertical coordinate system of Kasahara (1 974), the Jacobian pressure gradient formulation of Song (1 998), and a newly developed metric factor that permits both compressible (non-Boussinesq) and incompressible (Boussinesq) approximations. Based on the new formulation, an adaptive modeling strategy is proposed and a staggered finite volume method is designed to ensure conservation of important physical properties and numerical accuracy. Implementation of the combined techniques to SCRUM (Song and Haidvogel1994) shows that the adaptive modeling strategy can be applied to any existing ocean model without incurring computational expense or altering the original numerical schemes. Such a generalized coordinate model is expected to benefit diverse ocean modelers for easily choosing optimal vertical structures and sharing modeling resources based on a common model platform. Several representing oceanographic problems with different scales and characteristics, such as coastal canyons, basin-scale circulation, and global ocean circulation, are used to demonstrate the model's capability for multiple applications. New results show that the model is capable of simultaneously resolving both Boussinesq and non-Boussinesq, and both small- and large-scale processes well. This talk will focus on its applications of multiple satellite sensing data in eddy-resolving simulations of Asian Marginal Sea and Kurosio. Attention will be given to how Topex/Poseidon SSH, TRMM SST; and GRACE ocean bottom pressure can be correctly represented in a non- Boussinesq model.

numerical ocean model generalized coordinate syste↗

NASA NTRS2007

The Local Stellar Velocity Field via Vector Spherical Harmonics

We analyze the local field of stellar tangential velocities for a sample of 42,339 nonbinary Hipparcos stars with accurate parallaxes, using a vector spherical harmonic formalism. We derive simple relations between the parameters of the classical linear model (Ogorodnikov-Milne) of the local systemic field and low-degree terms of the general vector harmonic decomposition. Taking advantage of these relationships, we determine the solar velocity with respect to the local stars of (V(sub X), V(sub Y), V(sub Z)) (10.5, 18.5, 7.3) +/- 0.1 km s(exp -1) not corrected for the asymmetric drift with respect to the local standard of rest. If only stars more distant than 100 pc are considered, the peculiar solar motion is (V(sub X), V(sub Y), V(sub Z)) (9.9, 15.6, 6.9) +/- 0.2 km s(exp -1). The adverse effects of harmonic leakage, which occurs between the reflex solar motion represented by the three electric vector harmonics in the velocity space and higher degree harmonics in the proper-motion space, are eliminated in our analysis by direct subtraction of the reflex solar velocity in its tangential components for each star. The Oort parameters determined by a straightforward least-squares adjustment in vector spherical harmonics are A=14.0 +/- 1.4, B=13.1 +/- 1.2, K=1.1 +/- 1.8, and C=2.9 +/- 1.4 km s(exp -1) kpc(exp -1). The physical meaning and the implications of these parameters are discussed in the framework of a general linear model of the velocity field. We find a few statistically significant higher degree harmonic terms that do not correspond to any parameters in the classical linear model. One of them, a third-degree electric harmonic, is tentatively explained as the response to a negative linear gradient of rotation velocity with distance from the Galactic plane, which we estimate at approximately -20 km s(exp -1) kpc(exp -1). A similar vertical gradient of rotation velocity has been detected for more distant stars representing the thick disk (z greater than 1 kpc), but here we surmise its existence in the thin disk at z less than 200 pc. The most unexpected and unexplained term within the Ogorodnikov-Milne model is the first-degree magnetic harmonic, representing a rigid rotation of the stellar field about the axis -Y pointing opposite to the direction of rotation. This harmonic comes out with a statistically robust coefficient of 6.2 +/- 0.9 km s(exp -1) kpc(exp -1) and is also present in the velocity field of more distant stars. The ensuing upward vertical motion of stars in the general direction of the Galactic center and the downward motion in the anticenter direction are opposite to the vector field expected from the stationary Galactic warp model.

Galaxy↗

NASA NTRS1950

Approximate Turbulent Boundary-Layer Development in Plane Compressible Flow Along Thermally Insulated Surfaces with Application to Supersonic-Tunnel Contour Correction

Numerical solutions of the differential equation obtained from the momentum theorem for the development of a turbulent boundary layer along a thermally insulated surface in two-dimensional and in radial shock-free flow are presented in tabular form for a range of Mach numbers from 0.100 to 10. The solution can be used in a step-wise procedure with any given distribution of favorable pressure gradients and for zero pressure gradients. Solutions are also given for use with moderate adverse pressure gradients. The mean velocity in the boundary layer is approximated by a power-law profile. In view of the stepwise integration methods to be used, the exponent designated the profile shape can be varied along the surface between the integral fraction limits 1/5 and 1/11 through interpolation. Agreement obtained between theoretical and experimental boundary-layer development in a supersonic nozzle at a nominal Mach number of 2 indicates the general validity of the approximations used in the analysis - in particular, the method of extrapolating low-speed skin-friction relations to high Mach number flows. The extrapolation method used assumes that the skin-friction coefficient depend primarily on Reynolds number, provided that the density and the kinematic viscosity are evaluated at surface conditions.

BOUNDARY LAYER - WIND TUNNEL WALLS↗

NASA NTRS1988

In situ observations of ClO in the Antarctic: Evidence for chlorine catalyzed destruction of ozone

Results from a series of 12 ER-2 aircraft flights into the Antarctic polar vortex are summarized. These in situ data define the spatial and temporal distribution of ClO as the aircraft flew at an altitude of approx. 18 km from Punta Arenas (54 deg S latitude) to the base of the Palmer Peninsula (72 deg S latitude), executed a rapid descent to approx. 13 km, turned north and climbed bach to approximately 18 km, returning to Punta Arenas. A general pattern in the ClO distribution is reported: mixing ratios of approximately 10 ppt are found at altitude in the vicinity of 55 deg S increasing to 50 ppt at 60 degrees S. In the vicinity of 65 deg S latitude a steep gradient in the ClO mixing ratio is observed. At a fixed potential temperature, the ClO mixing ratio through this sharp transition increases by an order of magnitude within a very few degrees of latitude, thus defining the edge of the chemical containment vessel. From the edge of that containment vessel to the southern extension of the flights, 72 deg S, a dome of slowly increasing ClO best describes the distribution. Conclusion are drawn from the data.

Anderson, J. G.↗

NASA NTRS1989

Global properties of pseudospectral methods

The present application of polynomial interpolation methods to function-approximation and numerical solutions for hyperbolic and elliptic PDEs allows the explicit construction of the derivative matrix for a general sequence of collocation points. An evaluation of the effect of several factors on the performance of these methods indicates an inability to interpret global methods in terms of local ones; the accuracy of the approximation will differ when the function's large gradients occur near the center of the region or near the boundary, irrespective of the boundary vicinity's collocation-point density.

Solomonoff, A.↗

NASA NTRS1985

Solar modulation and interplanetary gradients of the galactic electron flux, 1977 - 1984

The flux of electrons with energy from approximately 10 to 180 MeV measured ith the electron telescope on the Voyager 1 and 2 spacecraft in the heliocentric radial range 1 to 22 AU between 1977 and 1984 is reported. Jovian electrons were clearly observable between 1978 and 1983 (radial range 2 to 12 AU) at energies below approximately 50 MeV. Above approximately 50 MeV the electron intensity exhibited temporal variations generally related to the 11 year modulation of protons 75 MeV. The overall magnitude of the electron intensity changes between the maximum intensity observed in 1977 and the minimum intensity in 1981 was a factor approximately 2, also comparable to that observed for 75 MeV protons. By early 1985 the electron intensity has apparently recovered to the level observed in 1977 whereas the proton intensity was still about 20% lower. A detailed interpretation of these electron variations in all energy channels depends on an accurate subtraction of background induced by energetic protons of a few 100 Mev. This subtraction is facilitated by calibration results at several energies. Further results on these temporal variations and limits on possible interplanetary gradients will be reported.

Christon, S. P.↗

DOE OSTI2020

SympNets: Intrinsic structure-preserving symplectic networks for identifying Hamiltonian systems

Here, we propose new symplectic networks (SympNets) for identifying Hamiltonian systems from data based on a composition of linear, activation and gradient modules. In particular, we define two classes of SympNets: the LA-SympNets composed of linear and activation modules, and the G-SympNets composed of gradient modules. Correspondingly, we prove two new universal approximation theorems that demonstrate that SympNets can approximate arbitrary symplectic maps based on appropriate activation functions. We then perform several experiments including the pendulum, double pendulum and three-body problems to investigate the expressivity and the generalization ability of SympNets. The simulation results show that even very small size SympNets can generalize well, and are able to handle both separable and non-separable Hamiltonian systems with data points resulting from short or long time steps. In all the test cases, SympNets outperform the baseline models, and are much faster in training and prediction. We also develop an extended version of SympNets to learn the dynamics from irregularly sampled data. This extended version of SympNets can be thought of as a universal model representing the solution to an arbitrary Hamiltonian system.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

NASA NTRS2008

Irregularities at Sub-Auroral, Middle, and Low Latitudes in the Topside Ionosphere Observed During Geomagnetic Storms with the DEMETER and DMSP Satellites

Detailed observations of the plasma structure and irregularities that characterize the topside ionosphere at sub-auroral, middle, and low-latitudes are presented that were gathered with probes on the DEMETER and DMSP satellites during geomagnetic storms. Data from successive orbits reveal how the density structure and irregularities evolve with changes in the Dst. The observations reveal that precisely during the main phase of severe geomagnetic storms, increased ambient plasma densities and broad regions of irregularities are observed at 700 km, initially at storm commencement near the magnetic equator and then extending to mid- and sub-auroral latitudes within the approximately 8 hour period corresponding to the negative Dst excursions. Furthermore, intense, broadband electric and magnetic field irregularities are often observed at sub-auroral latitudes and are typically associated with the trough region and its poleward plasma density gradient. The observations provide a general framework showing how low, mid, and sub-auroral latitude plasma density structuring and associated irregularities respond to geomagnetic storms.

Pfaff, R.↗

NASA NTRS1988

Domain decomposition methods for the parallel computation of reacting flows

Domain decomposition is a natural route to parallel computing for partial differential equation solvers. Subdomains of which the original domain of definition is comprised are assigned to independent processors at the price of periodic coordination between processors to compute global parameters and maintain the requisite degree of continuity of the solution at the subdomain interfaces. In the domain-decomposed solution of steady multidimensional systems of PDEs by finite difference methods using a pseudo-transient version of Newton iteration, the only portion of the computation which generally stands in the way of efficient parallelization is the solution of the large, sparse linear systems arising at each Newton step. For some Jacobian matrices drawn from an actual two-dimensional reacting flow problem, comparisons are made between relaxation-based linear solvers and also preconditioned iterative methods of Conjugate Gradient and Chebyshev type, focusing attention on both iteration count and global inner product count. The generalized minimum residual method with block-ILU preconditioning is judged the best serial method among those considered, and parallel numerical experiments on the Encore Multimax demonstrate for it approximately 10-fold speedup on 16 processors.

Keyes, David E.↗

NASA NTRS1999

Ice Thicknesses and Driving Stresses of Greenland Ice Sheet

Surface elevations from satellite radar altimetry (Geosat, Seasat, and ERS-1) and bedrock topography from airborne radar sounding (Simon Ekholm's Danish compilation) are combined to derive maps of the driving stresses in the Greenland ice sheet. The stress vector, tau = rho g h sin(alpha), is calculated using surface slope vectors, alpha, from surface elevations and ice thicknesses, h, from the difference between surface and basal elevations. Since the 5-km scale of the surface slope is only about 2 times the ice thickness, the stress maps show spatial variations indicative of longitudinal stress gradients associated with topographic undulations. Values of alpha generally vary from near zero at the ice divides to maxima values around 120 kpa, returning to near zero in a narrow band at the edges. The distribution of alpha's peaks at 60 kpa with an approximate sigma of +/- 20 kpa. Areas of very low alpha near the origin of the northeast ice stream may indicate small sub-glacial lakes. The profile of alpha, down the ice stream from near the ice divide, increases to a maximum of about 120 kpa near the margin, which is characteristic of East Antarctic outlet glaciers and in contrast to West Antarctic ice streams where alpha has maximum values 400 to 500 km inland from the grounding lines. Overall distributions of alpha values are compared with those for the Antarctic ice sheet and the Mars Northern ice cap.

Zwally, H. J.↗

NASA NTRS2006

Thermal Management Techniques for Oil-Free Turbomachinery Systems

Tests were performed to evaluate three different methods of utilizing air to provide thermal management control for compliant journal foil air bearings. The effectiveness of the methods was based on bearing bulk temperature and axial thermal gradient reductions during air delivery. The first method utilized direct impingement of air on the inner surface of a hollow test journal during operation. The second, less indirect method achieved heat removal by blowing air inside the test journal to simulate air flowing axially through a hollow, rotating shaft. The third method emulated the most common approach to removing heat by forcing air axially through the bearing s support structure. Internal bearing temperatures were measured with three, type K thermocouples embedded in the bearing that measured general internal temperatures and axial thermal gradients. Testing was performed in a 1 atm, 260 C ambient environment with the bearing operating at 60 krpm and supporting a load of 222 N. Air volumetric flows of 0.06, 0.11, and 0.17 cubic meters per minute at approximately 150 to 200 C were used. The tests indicate that all three methods provide thermal management but at different levels of effectiveness. Axial cooling of the bearing support structure had a greater effect on bulk temperature for each air flow and demonstrated that the thermal gradients could be influenced by the directionality of the air flow. Direct air impingement on the journal's inside surface provided uniform reductions in both bulk temperature and thermal gradients. Similar to the direct method, indirect journal cooling had a uniform cooling effect on both bulk temperatures and thermal gradients but was the least effective of the three methods.

Radil, Kevin↗

NASA NTRS2013

The Face that Launched a Thousand Slips

Helene, (approximately 17.6 kilometers mean radius) is an L4 Trojan co-orbital of Saturn's moon Dione. Its hemisphere features an unusual morphology consisting of broad depressions and a generally smooth surface patterned with streaks and grooves. The streaks appear to be oriented down-gradient, as are the grooves. This pattern suggests intensive mass-wasting as a dominant process on the leading hemisphere. Kilometer-scale impact craters are very sparse on the leading hemisphere other than the degraded kilometer-scale basins defining the overall satellite shape, and many small craters have a diffuse appearance suggesting ongoing mass wasting. Thus mass wasting must dominate surface-modifying processes at present. In fact, the mass wasting appears to have been sufficient in magnitude to narrow the divides between adjacent basins to narrow septa, similar, but in lower relief, to the honeycomb pattern of Hyperion. The prominent groves occur primarily near topographic divides and appear have cut into a broad, slightly lower albedo surface largely conforming to the present topography but elevated a few meters above the smooth surfaces undergoing mass wasting flow. Low ridges and albedo markings on the surface suggest surface flow of materials traveling up to several kilometers. Diffusive mass wasting produces smooth surfaces - such a pattern characterizes most of the low-lying surfaces. The grooves, however, imply that the transport process is advective at those locations where they occur, that is, erosion tends to concentrate along linear pathways separated by divides. In fact, in many places grooves have a fairly regular spacing of 125-160 meters, defining a characteristic erosional scale. Several questions are prompted by the unusual morphology of Helene: 1) What is the nature of the surface materials? 2) Are the transport processes gradual or catastrophic motion from one or a few events? 3) What mechanisms drive mass wasting and groove development? 4) Have the formative processes been active in the recent past? 5) Finally, is the surface accreting or eroding? The smooth character of the leading edge hemisphere of Helene and the dominance of mass wasting suggest that the surface is composed of fine-grained debris, probably dominated by dust-size to small gravel particles. The Lagrangian points of saturnian satellites are locations where planetesimals might have accreted to form co-orbital satellites such as and they may capture ejecta from their master moon. Published models suggest that Helene is a site of net accretion, but we find no extant explanation for the dominance of fine grain sizes for the surface (and probable subsurface) composition of Helene and the other Lagrangian satellites. Observation of the mass wasting tracks on Helene suggests the presence of well-defined streams of debris with low bordering levees that may be depositional features or remnants of the dissected higher surface. Some flows in grazing illumination appear to have a convex cross-section. This mass-flow morphology might be consistent with the occurrence of large-scale flow events, but which might have occurred through rapid emplacement or slow glacier-like creep. On the other hand, small craters appear to have been "softened" by creep-like processes, indicating ongoing mass wasting.

honeycomb pattern of Hyperion↗

NASA NTRS1976

A triangular element based on generalized potential energy concepts

Stiffness equations are formulated for a doubly-curved triangular thin shell finite element. The strain energy component of the potential energy is first expressed in terms of displacements and displacement gradients with the aid of consistent deep shell strain-displacement equations. The element in-plane and normal displacement fields are approximated by complete cubic polynomials. These functions do not satisfy the interelement displacement admissibility conditions. Satisfaction is forced by the imposition of constraint conditions on the interelement boundaries; the constraints represent the modification of the potential energy. Some numerical results for a pinched cylinder, a cylindrical sphere, and a pinched sphere are examined.

Thomas, G. R.↗

NASA NTRS2001

A {1,2}-Order Plate Theory Accounting for Three-Dimensional Thermoelastic Deformations in Thick Composite and Sandwich Laminates

A {1,2}-order theory for laminated composite and sandwich plates is extended to include thermoelastic effects. The theory incorporates all three-dimensional strains and stresses. Mixed-field assumptions are introduced which include linear in-plane displacements, parabolic transverse displacement and shear strains, and a cubic distribution of the transverse normal stress. Least squares strain compatibility conditions and exact traction boundary conditions are enforced to yield higher polynomial degree distributions for the transverse shear strains and transverse normal stress through the plate thickness. The principle of virtual work is used to derive a 10th-order system of equilibrium equations and associated Poisson boundary conditions. The predictive capability of the theory is demonstrated using a closed-form analytic solution for a simply-supported rectangular plate subjected to a linearly varying temperature field across the thickness. Several thin and moderately thick laminated composite and sandwich plates are analyzed. Numerical comparisons are made with corresponding solutions of the first-order shear deformation theory and three-dimensional elasticity theory. These results, which closely approximate the three-dimensional elasticity solutions, demonstrate that through - the - thickness deformations even in relatively thin and, especially in thick. composite and sandwich laminates can be significant under severe thermal gradients. The {1,2}-order kinematic assumptions insure an overall accurate theory that is in general superior and, in some cases, equivalent to the first-order theory.

Tessler, A.↗

NASA NTRS1995

A General Method for Solving Systems of Non-Linear Equations

The method of steepest descent is modified so that accelerated convergence is achieved near a root. It is assumed that the function of interest can be approximated near a root by a quadratic form. An eigenvector of the quadratic form is found by evaluating the function and its gradient at an arbitrary point and another suitably selected point. The terminal point of the eigenvector is chosen to lie on the line segment joining the two points. The terminal point found lies on an axis of the quadratic form. The selection of a suitable step size at this point leads directly to the root in the direction of steepest descent in a single step. Newton's root finding method not infrequently diverges if the starting point is far from the root. However, the current method in these regions merely reverts to the method of steepest descent with an adaptive step size. The current method's performance should match that of the Levenberg-Marquardt root finding method since they both share the ability to converge from a starting point far from the root and both exhibit quadratic convergence near a root. The Levenberg-Marquardt method requires storage for coefficients of linear equations. The current method which does not require the solution of linear equations requires more time for additional function and gradient evaluations. The classic trade off of time for space separates the two methods.

Nachtsheim, Philip R.↗