Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “gradient methods”

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 433 records · Page 24
All sourcesNASA NTRSDOE OSTI
NASA NTRS1990

Research on diabatic initialization

The objective of this research project is to contribute to improvement in the synoptic analyses in the tropics for numerical weather prediction and climate research. In addition to a prediction model, four dimensional data assimilation systems have two principal components. One is objective analysis and the other is initialization. Various methods of objective analysis are designed primarily to analyze the mass and rotational wind fields. Methods of initialization are developed to obtain the irrotational wind and its associated vertical velocity field which are balanced with the mass field and free from meteorological noise. There are essentially three approaches to the problem of initialization: quasi-geostrophic theory, bounded derivative method and nonlinear normal mode method. In the midlatitudes, these approaches generally produce satisfactory results even without diabatic effects for large-scale motions. In the tropics, the situation is quite different from that in the midlatitudes. Because of a small magnitude of the Coriolis parameter and a weak horizontal temperature gradient in the tropics, any method of initialization must incorporate diabatic effects. In fact, it can be said that understanding the problem of diabatic initialization is the key to improving the analysis and weather forecasting in the tropics.

Kasahara, Akira↗
NASA NTRS1973

Function space quasi-Newton algorithms for optimal control problems with bounded controls and singular arcs

Two existing function space algorithms, Davidon and projected gradient, are modified so that they may handle directly control variable inequality constraints. A third quasi-Newton type algorithm developed by C. G. Broyden is extended to optimal control problems. The Broyden algorithm is further modified so that it also may handle directly control variable inequality constraints. These methods along with a pure gradient and two conjugate gradient algorithms are simulated on three relatively simple yet representative bounded control problems, two of which have singular subarcs. Overall the Broyden algorithm was found to be superior. The most notable result of the study was the clear superiority of the Broyden and Davidon algorithms in producing a sharp interior control subarc.

Edge, E. R.↗
NASA NTRS1980

A proposed method for wind velocity measurement from space

An investigation was made of the feasibility of making wind velocity measurements from space by monitoring the apparent change in the refractive index of the atmosphere induced by motion of the air. The physical principle is the same as that resulting in the phase changes measured in the Fizeau experiment. It is proposed that this phase change could be measured using a three cornered arrangement of satellite borne source and reflectors, around which two laser beams propagate in opposite directions. It is shown that even though the velocity of the satellites is much larger than the wind velocity, factors such as change in satellite position and Doppler shifts can be taken into account in a reasonable manner and the Fizeau phase measured. This phase measurement yields an average wind velocity along the ray path through the atmosphere. The method requires neither high accuracy for satellite position or velocity, nor precise knowledge of the refractive index or its gradient in the atmosphere. However, the method intrinsically yields wind velocity integrated along the ray path; hence to obtain higher spatial resolution, inversion techniques are required.

Censor, D.↗
NASA NTRS2002

On the Formulation of Weakly Singular Displacement/Traction Integral Equations; and Their Solution by the MLPG Method

In this paper, a very simple method is used to derive the weakly singular traction boundary integral equation based on the integral relationships for displacement gradients. The concept of the MLPG method is employed to solve the integral equations, especially those arising in solid mechanics. A moving Least Squares (MLS) interpolation is selected to approximate the trial functions in this paper. Five boundary integral Solution methods are introduced: direct solution method; displacement boundary-value problem; traction boundary-value problem; mixed boundary-value problem; and boundary variational principle. Based on the local weak form of the BIE, four different nodal-based local test functions are selected, leading to four different MLPG methods for each BIE solution method. These methods combine the advantages of the MLPG method and the boundary element method.

Atluri, Satya N.↗
NASA NTRS1973

Unsteady turbulent boundary layers in two-dimensional, incompressible flow.

The paper presents the results of some calculations aimed at studying the characteristics of turbulent boundary layers in unsteady flow. The calculations were performed using the method of Singleton and Nash, which is based on the turbulent kinetic energy equation. Emphasis is placed on the approach to zero wall stress, and the results show the extent to which onset of this condition can be delayed by the effects of time dependence. The delay is not associated merely with alleviation of the pressure gradients, and quasi-steady methods cannot readily be used to calculate this type of flow.

Nash, J. F.↗
NASA NTRS1992

Iterative solution of linear systems

Recent advances in the field of iterative methods for solving large linear systems are reviewed. The main focus is on developments in the area of conjugate gradient-type algorithms and Krylov subspace methods for nonHermitian matrices.

Freund, Roland W.↗
NASA NTRS1996

Computation of Kinetics for the Hydrogen/Oxygen System Using the Thermodynamic Method

A new method for predicting chemical rate constants using thermodynamics has been applied to the hydrogen/oxygen system. This method is based on using the gradient of the Gibbs free energy and a single proportionality constant D to determine the kinetic rate constants. Using this method the rate constants for any gas phase reaction can be computed from thermodynamic properties. A modified reaction set for the H/O system is determined. A11 of the third body efficiencies M are taken to be unity. Good agreement was obtained between the thermodynamic method and the experimental shock tube data. In addition, the hydrogen bromide experimental data presented in previous work is recomputed with M's of unity.

Marek, C. John↗
NASA NTRS1975

Airship stresses due to vertical velocity gradients and atmospheric turbulence

Munk's potential flow method is used to calculate the resultant moment experienced by an ellipsoidal airship. This method is first used to calculate the moment arising from basic maneuvers considered by early designers, and then expended to calculate the moment arising from vertical velocity gradients and atmospheric turbulence. This resultant moment must be neutralized by the transverse force of the fins. The results show that vertical velocity gradients at a height of 6000 feet in thunderstorms produce a resultant moment approximately three to four times greater than the moment produced in still air by realistic values of pitch angle or steady turning. Realistic values of atmospheric turbulence produce a moment which is significantly less than the moment produced by maneuvers in still air.

Sheldon, D.↗
NASA NTRS1983

Vertical gradients of sunspot magnetic fields

The results of a Solar Maximum Mission (SMM) guest investigation to determine the vertical gradients of sunspot magnetic fields for the first time from coordinated observations of photospheric and transition-region fields are described. Descriptions are given of both the photospheric vector field of a sunspot, derived from observations using the NASA Marshall Space Flight Center vector magnetograph, and of the line-of-sight component in the transition region, obtained from the SMM Ultraviolet Spectrometer and Polarimeter instrument. On the basis of these data, vertical gradients of the line-of-sight magnetic field component are calculated using three methods. It is found that the vertical gradient of Bz is lower than values from previous studies and that the transition-region field occurs at a height of approximately 4000-6000 km above the photosphere.

Hagyard, M. J.↗
NASA NTRS2010

Peak-Seeking Control Using Gradient and Hessian Estimates

A peak-seeking control method is presented which utilizes a linear time-varying Kalman filter. Performance function coordinate and magnitude measurements are used by the Kalman filter to estimate the gradient and Hessian of the performance function. The gradient and Hessian are used to command the system toward a local extremum. The method is naturally applied to multiple-input multiple-output systems. Applications of this technique to a single-input single-output example and a two-input one-output example are presented.

Ryan, John J.↗
NASA NTRS1983

Computational methods for aerodynamic design using numerical optimization

Five methods to increase the computational efficiency of aerodynamic design using numerical optimization, by reducing the computer time required to perform gradient calculations, are examined. The most promising method consists of drastically reducing the size of the computational domain on which aerodynamic calculations are made during gradient calculations. Since a gradient calculation requires the solution of the flow about an airfoil whose geometry was slightly perturbed from a base airfoil, the flow about the base airfoil is used to determine boundary conditions on the reduced computational domain. This method worked well in subcritical flow.

Peeters, M. F.↗
NASA NTRS1996

Aerodynamic shape optimization using control theory

Aerodynamic shape design has long persisted as a difficult scientific challenge due its highly nonlinear flow physics and daunting geometric complexity. However, with the emergence of Computational Fluid Dynamics (CFD) it has become possible to make accurate predictions of flows which are not dominated by viscous effects. It is thus worthwhile to explore the extension of CFD methods for flow analysis to the treatment of aerodynamic shape design. Two new aerodynamic shape design methods are developed which combine existing CFD technology, optimal control theory, and numerical optimization techniques. Flow analysis methods for the potential flow equation and the Euler equations form the basis of the two respective design methods. In each case, optimal control theory is used to derive the adjoint differential equations, the solution of which provides the necessary gradient information to a numerical optimization method much more efficiently then by conventional finite differencing. Each technique uses a quasi-Newton numerical optimization algorithm to drive an aerodynamic objective function toward a minimum. An analytic grid perturbation method is developed to modify body fitted meshes to accommodate shape changes during the design process. Both Hicks-Henne perturbation functions and B-spline control points are explored as suitable design variables. The new methods prove to be computationally efficient and robust, and can be used for practical airfoil design including geometric and aerodynamic constraints. Objective functions are chosen to allow both inverse design to a target pressure distribution and wave drag minimization. Several design cases are presented for each method illustrating its practicality and efficiency. These include non-lifting and lifting airfoils operating at both subsonic and transonic conditions.

Reuther, James↗
NASA NTRS1991

Optimal design of solidification processes

An optimal design algorithm is presented for the analysis of general solidification processes, and is demonstrated for the growth of GaAs crystals in a Bridgman furnace. The system is optimal in the sense that the prespecified temperature distribution in the solidifying materials is obtained to maximize product quality. The optimization uses traditional numerical programming techniques which require the evaluation of cost and constraint functions and their sensitivities. The finite element method is incorporated to analyze the crystal solidification problem, evaluate the cost and constraint functions, and compute the sensitivities. These techniques are demonstrated in the crystal growth application by determining an optimal furnace wall temperature distribution to obtain the desired temperature profile in the crystal, and hence to maximize the crystal's quality. Several numerical optimization algorithms are studied to determine the proper convergence criteria, effective 1-D search strategies, appropriate forms of the cost and constraint functions, etc. In particular, we incorporate the conjugate gradient and quasi-Newton methods for unconstrained problems. The efficiency and effectiveness of each algorithm is presented in the example problem.

Dantzig, Jonathan A.↗
NASA NTRS1995

SEEK: A FORTRAN optimization program using a feasible directions gradient search

This report describes the use of computer program 'SEEK' which works in conjunction with two user-written subroutines and an input data file to perform an optimization procedure on a user's problem. The optimization method uses a modified feasible directions gradient technique. SEEK is written in ANSI standard Fortran 77, has an object size of about 46K bytes, and can be used on a personal computer running DOS. This report describes the use of the program and discusses the optimizing method. The program use is illustrated with four example problems: a bushing design, a helical coil spring design, a gear mesh design, and a two-parameter Weibull life-reliability curve fit.

Savage, M.↗
NASA NTRS1998

Single-Transducer, Ultrasonic Imaging Method for High-Temperature Structural Materials Eliminates the Effect of Thickness Variation in the Image

NASA Lewis Research Center's Life Prediction Branch, in partnership with Sonix, Inc., and Cleveland State University, recently advanced the development of, refined, and commercialized an advanced nondestructive evaluation (NDE) inspection method entitled the Single Transducer Thickness-Independent Ultrasonic Imaging Method. Selected by R&D Magazine as one of the 100 most technologically significant new products of 1996, the method uses a single transducer to eliminate the superimposing effects of thickness variation in the ultrasonic images of materials. As a result, any variation seen in the image is due solely to microstructural variation. This nondestructive method precisely and accurately characterizes material gradients (pore fraction, density, or chemical) that affect the uniformity of a material's physical performance (mechanical, thermal, or electrical). Advantages of the method over conventional ultrasonic imaging include (1) elimination of machining costs (for precision thickness control) during the quality control stages of material processing and development and (2) elimination of labor costs and subjectivity involved in further image processing and image interpretation. At NASA Lewis, the method has been used primarily for accurate inspections of high temperature structural materials including monolithic ceramics, metal matrix composites, and polymer matrix composites. Data were published this year for platelike samples, and current research is focusing on applying the method to tubular components. The initial publicity regarding the development of the method generated 150 requests for further information from a wide variety of institutions and individuals including the Federal Bureau of Investigation (FBI), Lockheed Martin Corporation, Rockwell International, Hewlett Packard Company, and Procter & Gamble Company. In addition, NASA has been solicited by the 3M Company and Allison Abrasives to use this method to inspect composite materials that are manufactured by these companies.

Roth, Don J.↗
NASA NTRS1999

SAGE: The Self-Adaptive Grid Code

The multi-dimensional self-adaptive grid code, SAGE, is an important tool in the field of computational fluid dynamics (CFD). It provides an efficient method to improve the accuracy of flow solutions while simultaneously reducing computer processing time. Briefly, SAGE enhances an initial computational grid by redistributing the mesh points into more appropriate locations. The movement of these points is driven by an equal-error-distribution algorithm that utilizes the relationship between high flow gradients and excessive solution errors. The method also provides a balance between clustering points in the high gradient regions and maintaining the smoothness and continuity of the adapted grid, The latest version, Version 3, includes the ability to change the boundaries of a given grid to more efficiently enclose flow structures and provides alternative redistribution algorithms.

Davies, Carol B.↗
NASA NTRS2016

Hyperbolic Method for Dispersive PDEs: Same High-Order of Accuracy for Solution, Gradient, and Hessian

In this paper, we introduce a new hyperbolic first-order system for general dispersive partial differential equations (PDEs). We then extend the proposed system to general advection-diffusion-dispersion PDEs. We apply the fourth-order RD scheme of Ref. 1 to the proposed hyperbolic system, and solve time-dependent dispersive equations, including the classical two-soliton KdV and a dispersive shock case. We demonstrate that the predicted results, including the gradient and Hessian (second derivative), are in a very good agreement with the exact solutions. We then show that the RD scheme applied to the proposed system accurately captures dispersive shocks without numerical oscillations. We also verify that the solution, gradient and Hessian are predicted with equal order of accuracy.

Mazaheri, Alireza↗
NASA NTRS1994

Geometrical evidence for dark matter: X-ray constraints on the mass of the elliptical galaxy NGC 720

We describe (1) a new test for dark matter and alternate theories of gravitation based on the relative geometries of the X-ray and optical surface brightness distributions and an assumed form for the potential, of the optical light, (2) a technique to measure the shapes of the total gravitating matter and dark matter of an ellipsoidal system which is insensitive to the precise value of the temperature of the gas and to modest temperature gradients, and (3) a new method to determine the ratio of dark mass to stellar mass that is dependent on the functional forms for the visible star, gas and dark matter distributions, but independent of the distance to the galaxy or the gas temperature. We apply these techniques to X-ray data from the ROSAT Position Sensitive Proportional Counter (PSPC) of the optically flattened elliptical galaxy NGC 720; the optical isophotes have ellipticity epsilon approximately 0.40 extending out to approximately 120 sec. The X-ray isophotes are significantly elongated, epsilon = 0.20-0.30 for semimajor axis a approximately 100 sec. The major axes of the optical and X-ray isophotes are misaligned by approximately 30 deg +/- 15 deg. Spectral analysis of the X-ray data reveals no evidence of temperature gradients or anisotropies and demonstrates that a single-temperature plasma (T approximately 0.6 keV) having subsolar heavy element abundances and a two-temperature model having solar abundances describe the spectrum equally well. Considering only the relative geometries of the X-ray and optical surface brightness distributions and an assumed functional form for the potential of the optical light, we conclude that matter distributed like the optical light cannot produce the observed ellipticities of the X-ray isophotes, independent of the gas pressure, the gas temperature, and the value of the stellar mass; this comparison assumes a state of quasi-hydrostatic equilibrium so that the three-dimensional surfaces of the gas emissivity trace the three-dimensional isopotential surfaces -- we discuss the viability of this assumption for NGC 720. Milgrom's Modification of Newtonian Dynamics (MOND) cannot dispel this manifestation of dark matter. Hence, geometrical considerations require, without mention of pressure or temperature, the presence of an extended, massive dark matter halo in NGC 720. Employing essentially the technique of Buote & Canizares (1992; Buote 1992) we use the shape of the X-ray surface brightness to constrain the shape of the total gravitating matter. The total matter is modeled as either an oblate or prolate spheriod of constant shape and orientation having either a Ferrers (rho approximately r(exp -n)) or Hernquist density. Assuming the X-ray gas is in hydrostatic equilibrium, we construct a model X-ray gas distribution for various temperature profiles. We determine the ellipticity of the total gravitating matter to be epsilon approximately 0.50-0.70. Using the single-temperature model we estimate a total mass approximately (0.41-1.4) x 10(exp 12) h(sub 80) solar mass interior to the ellipsoid of semimajor axis 43.6 h(sub 80) kpc. Ferrers densities as steep as r(exp -3) do not fit the data, but the r(exp -2) and Hernquist models yield excellent fits. We estimate the mass distributions of the stars and the gas and fit the dark matter directly. For a given gas equation of state and functional forms for the visible stars, gas, and dark matter, these models yield a distance-independent and temperature-independent measurement of the ratio of dark mass to stellar mass M(sub DM)/M(sub stars). We estimate a minimum M(sub DM)/M(sub stars) greater than or equal to 4 which corresponds to a total mass slightly greater than that derived from the single-temperature models for distance D = 20h(sub 80) Mpc.

Buote, David A.↗