Search NASA⌕ Search

SEARCH · Search NASA

Results for “gradient calculation”

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 37 records · Page 2

Computational aspects of unsteady flows

The calculation of unsteady flows and the development of numerical methods for solving unsteady boundary layer equations and their application to the flows around important configurations such as oscillating airfoils are presented. A brief review of recent work is provided with emphasis on the need for numerical methods which can overcome possible problems associated with flow reversal and separation. The zig-zag and characteristic box schemes are described in this context, and when embodied in a method which permits interaction between solutions of inviscid and viscous equations, the characteristic box scheme is shown to avoid the singularity associated with boundary layer equations and prescribed pressure gradient. Calculations were performed for a cylinder started impulsively from rest and oscillating airfoils. The results are presented and discussed. It is conlcuded that turbulence models based on an algebraic specification of eddy viscosity can be adequate, that location of translation is important to the calculation of the location of flow separation and, therefore, to the overall lift of an oscillating airfoil.

Cebeci, T.↗

Very Large Scale Optimization

The purpose of this research under the NASA Small Business Innovative Research program was to develop algorithms and associated software to solve very large nonlinear, constrained optimization tasks. Key issues included efficiency, reliability, memory, and gradient calculation requirements. This report describes the general optimization problem, ten candidate methods, and detailed evaluations of four candidates. The algorithm chosen for final development is a modern recreation of a 1960s external penalty function method that uses very limited computer memory and computational time. Although of lower efficiency, the new method can solve problems orders of magnitude larger than current methods. The resulting BIGDOT software has been demonstrated on problems with 50,000 variables and about 50,000 active constraints. For unconstrained optimization, it has solved a problem in excess of 135,000 variables. The method includes a technique for solving discrete variable problems that finds a "good" design, although a theoretical optimum cannot be guaranteed. It is very scalable in that the number of function and gradient evaluations does not change significantly with increased problem size. Test cases are provided to demonstrate the efficiency and reliability of the methods and software.

Vanderplaats, Garrett↗

Simulating Atmospheric Turbulence

Dimensionless time series for gusts and gradients calculated with model. Single-sided one-dimensional spectra obtained from Fourier analysis of simulated vertical gust. Each of four plots represents spectrum in one of four altitude bands ranging from 0 to 10 km. Model differs from most models; is nondimensional for greater generality and uses spectra with finite wave number limits. Developed to predict effects of atmospheric turbulence in horizontal and near-horizontal flight also applicable to commercial aviation.

Tatom, F. B.↗

A robust Feasible Directions algorithm for design synthesis

A nonlinear optimization algorithm is developed which combines the best features of the Method of Feasible Directions and the Generalized Reduced Gradient Method. This algorithm utilizes the direction-finding sub-problem from the Method of Feasible Directions to find a search direction which is equivalent to that of the Generalized Reduced Gradient Method, but does not require the addition of a large number of slack variables associated with inequality constraints. This method provides a core-efficient algorithm for the solution of optimization problems with a large number of inequality constraints. Further optimization efficiency is derived by introducing the concept of infrequent gradient calculations. In addition, it is found that the sensitivity of the optimum design to changes in the problem parameters can be obtained using this method without the need for second derivatives or Lagrange multipliers. A numerical example is given in order to demonstrate the efficiency of the algorithm and the sensitivity analysis.

Vanderplaats, G. N.↗

Optimal control concepts in design sensitivity analysis

A close link is established between open loop optimal control theory and optimal design by noting certain similarities in the gradient calculations. The resulting benefits include a unified approach, together with physical insights in design sensitivity analysis, and an efficient approach for simultaneous optimal control and design. Both matrix displacement and matrix force methods are considered, and results are presented for dynamic systems, structures, and elasticity problems.

Belegundu, Ashok D.↗

Application of a new K-tau model to near wall turbulent flows

A recently developed K-tau model for near wall turbulent flows is applied to a variety of test cases. The turbulent flows considered include the incompressible flat plate boundary layer with adverse pressure gradients, incompressible flow past a backward facing step, and the supersonic flat plate boundary layer at zero pressure gradient. Calculations are performed for this two-equation model using an anisotropic as well as isotropic eddy-viscosity. The model predictions are shown to compare quite favorably with experimental data.

Abid, R.↗

Theoretical characterization of the potential energy surface for NH + NO

The potential energy surface (PES) for NH + NO was characterized using complete active space self-consistent field (CASSCF) gradient calculations to determine the stationary point geometries and frequencies followed by CASSCF/internally contracted configuration interaction (CCI) calculations to refine the energetics. The present results are in qualitative accord with the BAC-MP4 calculations, but there are differences as large as 8 kcal/mol in the detailed energetics. Addition of NH to NO on a (2)A' surface, which correlated with N2 + OH or H + N2O products, involves barriers of 3.2 kcal/mol (trans) and 6.3 kcal/mol (cis). Experimental evidence for these barriers is found in earlier works. The (2)A' surface has no barrier to addition, but does not correlate with products. Surface crossings between the barrierless (2)A' surface and the (2)A' surface may be important. Production of N2 + OH products is predicted to occur via a planar saddle point of (2)A' symmetry. This is in accord with the preferential formation of II(A') lambda doublet levels of OH in earlier experiments. Addition of NH (1)delta to NO is found to occur on an excited state surface and is predicted to lead to N2O product as observed in earlier works.

Walch, Stephen P.↗

Theoretical characterization of the reaction CH3 +OH yields CH3OH yeilds products: The (1)CH2 + H2O, H2 + HCOH, and H2 + H2CO channels

The potential energy surface (PES) for the CH3OH system has been characterized for the (1)CH2 + H2O, H2 + HCOH, and H2 + H2CO product channels using complete-active-space self-consistent-field (CASSCF) gradient calculations to determine the stationary point geometries and frequencies followed by CASSCF/internally contracted configuration-interaction (CCI) calculations to refine the energetics. The (1)CH2 + H2O channel is found to have no barrier. The long range interaction is dominated by the dipole-dipole term, which orients the respective dipole moments parallel to each other but pointing in opposite directions. At shorter separations there is a dative bond structure in which a water lone pair donates into the empty a" orbital of CH2. Subsequent insertion of CH2 into an OH bond of water have barriers located at -5.2 kcal/mol and 1.7 kcal/mol, respectively, with respect to CH3 + OH. From comparison of the computed energetics of the reactants and products to known thermochemical data it is estimated that the computed PES is accurate to plus or minus 2 kcal/mol.

Walch, Stephen P.↗

Theoretical characterization of the potential energy surface for NH + NO

The potential energy surface for NH + NO was characterized using complete active space self-consistent field (CASSCF) gradient calculation to determine the stationary point geometries and frequencies followed by CASSCF/internally contracted configuration interaction calculations to refine the energetics. The present results are in qualitative accord with the BAC-MP4 calculations, but there are differences as large as 8 kcal/mol in the detailed energetics.

Walch, Stephen P.↗

Theoretical characterization of the reaction CH3 + OH yields CH3OH yields products - The (1)CH2 + H2O, H2 + HCOH, and H2 + H2CO channels

The potential energy surface (PES) for the CH3OH system has been characterized for the (1)CH2 + H2O, H2 + HCOH, and H2 + H2CO product channels using complete-active-space self-consistent-field (CASSCF) gradient calculations to determine the stationary point geometries and frequencies followed by CASSCF/internally contracted configuration-interaction (CCI) calculations to refine the energetics. The (1)CH2 + H2O channel is found to have no barrier. The long range interaction is dominated by the dipole-dipole term, which orients the respective dipole moments parallel to each other but pointing in opposite directions. At shorter separations there is a dative bond structure in which a water lone pair donates into the empty 'a' orbital of CH2. Subsequent insertion of CH2 into an OH bond of water have barriers located at -5.2 kcal/mol and 1.7 kcal/mol, respectively, with respect to CH3 + OH. From comparison of the computed energetics of the reactants and products to known thermochemical data it is estimated that the computed PES is accurate to plus or minus 2 kcal/mol.

Walch, Stephen P.↗

A structural fabric defined by topographic lineaments: Correlation with Tertiary deformation of Ellesmere and Axel Heiberg Islands, Canadian Arctic

Digital topographic contours from four 1:250000 scale maps have been gridded to produce a digital elevation model for part of Ellesmere and Axel Heiberg islands in the Canadian Arctic Islands. Gradient calculations were used to define both east and west dipping slopes defining a pattern of lineaments that have been compared with mapped geological structures. In ice-covered areas, where geological mapping was not possible, well-defined topographic lineaments have been identified and are correlated to extensions of major structural features. The northeast-southwest patterns of both topographic lineaments and mapped structures are strongly unimodal and support a single compressive event oriented at 67 deg west of north. This orientation is coincidental with the convergence direction calculated from the kinematic poles of rotation for Greenland relative to North America between 56 and 35 Ma. A minor secondary peak at 70 east of north is observed for thrust and normal fault solutions and is not directly related to the predicted convergence direction. Whether this represents a unique phase of deformation or is a subcomponent of a single event is not known. The agreement of structural components, lineament orientations, and convergence direction suggests an overwhelming over print of Eurekan deformation on any preexisting structural fabric. This study confirms, for the first time, an excellent compatibility between geological and geophysical constraints for the timing and geometry of the Eurekan orogeny.

Oakey, Gordon↗

Sensitivity Analysis for Coupled Aero-structural Systems

A novel method has been developed for calculating gradients of aerodynamic force and moment coefficients for an aeroelastic aircraft model. This method uses the Global Sensitivity Equations (GSE) to account for the aero-structural coupling, and a reduced-order modal analysis approach to condense the coupling bandwidth between the aerodynamic and structural models. Parallel computing is applied to reduce the computational expense of the numerous high fidelity aerodynamic analyses needed for the coupled aero-structural system. Good agreement is obtained between aerodynamic force and moment gradients computed with the GSE/modal analysis approach and the same quantities computed using brute-force, computationally expensive, finite difference approximations. A comparison between the computational expense of the GSE/modal analysis method and a pure finite difference approach is presented. These results show that the GSE/modal analysis approach is the more computationally efficient technique if sensitivity analysis is to be performed for two or more aircraft design parameters.

Giunta, Anthony A.↗

Parallelization of Program to Optimize Simulated Trajectories (POST3D)

This paper describes the parallelization of the Program to Optimize Simulated Trajectories (POST3D). POST3D uses a gradient-based optimization algorithm that reaches an optimum design point by moving from one design point to the next. The gradient calculations required to complete the optimization process, dominate the computational time and have been parallelized using a Single Program Multiple Data (SPMD) on a distributed memory NUMA (non-uniform memory access) architecture. The Origin2000 was used for the tests presented.

Hammond, Dana P.↗

Calculation of a realistic gradient as a function of radial distance from the sun

Numerical calculations of radial gradients of galactic cosmic ray protons have been performed for a model of the interplanetary cavity which includes processes of diffusion along the spiral interplanetary magnetic field lines, convection, and adiabatic energy change. The increasing spiral nature of the field lines with increasing distance from the sun results in the bulk of the modulation region being beyond the earth. The radial gradients for this model are found to be small near the earth and out to a few AU, increasing as one approaches the boundary.

Lezniak, J. A.↗