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.
A numerical solution precedure of a simplified unsteady transonic equation which is fast, reasonably accurate, and takes into account many of the effects of the steady flow field is described. The numeric solution of this equation is accurate and is accomplished on an IBM 360/65 computer. Arbitrary planform shape is accommodated and variable local Mach number effects from the steady flow are easily handled.
Two- and three-dimensional, viscous blunt body flows with planar impinging shocks are computed using an explicit, time-dependent, finite-difference method to solve the complete set of Navier-Stokes equations. The bow shock is treated as a discontinuity, while all interior shock layer detail such as shear layers, shock waves, jets and the wall boundary layer are automatically captured in the solution. Numerical results are presented for cases in which planar shock waves of different strengths and orientations are allowed to impinge on the flow field surroundings an infinite cylinder resulting in two- and three-dimensional shock interference patterns. The numerical results are compared with experiment.
Abstracts are presented on a method of numerical solution of the Navier-Stokes equation for the flow about arbitrary airfoils, using a numerically generated curvilinear coordinate system having a coordinate line coincident with the body contour. Results of continuing research are reported and include: application of the Navier-Stokes solution in the vorticity-stream function formulation to a number of single airfoils at Reynolds numbers up to 2000; programming of the Navier-Stokes solution for multiple airfoils in the primitive variable formulation; testing of the potential flow solution of multiple bodies; and development of a generalized coordinate system program.
A time-marching finite-difference method was used to solve the compressible Navier-Stokes equations for the three-dimensional wing-leading-edge shock impingement problem. The bow shock was treated as a discontinuity across which the exact shock jump conditions were applied. All interior shock layer detail such as shear layers, shock waves, jets, and the wall boundary layer were automatically captured in the solution. The impinging shock was introduced by discontinuously changing the freestream conditions across the intersection line at the bow shock. A special storage-saving procedure for sweeping through the finite-difference mesh was developed which reduces the required amount of computer storage by at least a factor of two without sacrificing the execution time. Numerical results are presented for infinite cylinder blunt body cases as well as the three-dimensional shock impingement case. The numerical results are compared with existing experimental and theoretical results.
A parametric numerical procedure permitting the rapid determination of the performance of a class of scramjet nozzle configurations is presented. The geometric complexity of these configurations ruled out attempts to employ conventional nozzle design procedures. The numerical program developed permitted the parametric variation of cowl length, turning angles on the cowl and vehicle undersurface and lateral expansion, and was subject to fixed constraints such as the vehicle length and nozzle exit height. The program required uniform initial conditions at the burner exit station and yielded the location of all predominant wave zones, accounting for lateral expansion effects. In addition, the program yielded the detailed pressure distribution on the cowl, vehicle undersurface and fences, if any, and calculated the nozzle thrust, lift and pitching moments.
The cubic spline collocation procedure for the numerical solution of partial differential equations was reformulated so that the accuracy of the second-derivative approximation is improved and parallels that previously obtained for lower derivative terms. The final result is a numerical procedure having overall third-order accuracy for a nonuniform mesh and overall fourth-order accuracy for a uniform mesh. Application of the technique was made to the Burger's equation, to the flow around a linear corner, to the potential flow over a circular cylinder, and to boundary layer problems. The results confirmed the higher-order accuracy of the spline method and suggest that accurate solutions for more practical flow problems can be obtained with relatively coarse nonuniform meshes.
The previously obtained second-order-accurate partial implicitization numerical technique used in the solution of fluid dynamic problems was modified with little complication to achieve fourth-order accuracy. The Von Neumann stability analysis demonstrated the unconditional linear stability of the technique. The order of the truncation error was deduced from the Taylor series expansions of the linearized difference equations and was verified by numerical solutions to Burger's equation. For comparison, results were also obtained for Burger's equation using a second-order-accurate partial-implicitization scheme, as well as the fourth-order scheme of Kreiss.
The principal disadvantage of the method of numerical averaging is that it provides only the average time history of the orbital elements and yields no information about the high-frequency or short-periodic variations that occur inside the averaging interval. This paper contains a description of a technique for recovering the short-periodic variations by minor modifications to the averaging process so as to permit the construction of a Fourier series for the osculating elements. The availability of this series permits the extension of the averaging technique to higher order and allows us to account for short-periodic coupling of the perturbations. Comparisons of the results with numerically integrated solutions are presented for three distinct orbit prediction problems.
Recent developments in the use of numerical methods for fluid flow simulation show an increasing tendency to use numerical operators that can, in various ways, be factored. Use of methods having this property often increases the accuracy and efficiency of computer codes. Tracing the factorization property provides a unity to the basic concepts involved in the development of cyclic reduction, predictor-corrector, splitting, fast Fourier transform, and pseudospectral methods.-
A cubic spline collocation procedure has recently been developed for the numerical solution of partial differential equations. In the present paper, this spline procedure is reformulated so that the accuracy of the second-derivative approximation is improved and parallels that previously obtained for lower derivative terms. The final result is a numerical procedure having overall third-order accuracy for a non-uniform mesh and overall fourth-order accuracy for a uniform mesh. Solutions using both spline procedures, as well as three-point finite difference methods, will be presented for several model problems.-
A procedure for numerical solution of the time-dependent, two-dimensional incompressible Navier-Stokes equations that can treat the unsteady laminar flow about bodies of arbitrary shape, such as two-dimensional airfoils, multiple airfoils, and submerged hydrofoils, as naturally as it can deal with the flow about simple bodies. The solution is based on a method of automatic numerical generation of a general curvilinear coordinate system with coordinate lines coincident with all boundaries of a general multiconnected region containing any number of arbitrarily shaped bodies. The curvilinear coordinates are generated as the solution of two elliptical partial differential equations with Dirichlet boundary conditions, one coordinate being specified to be constant on each of the boundaries, and a distribution of the other being specified along the boundaries. The solution compares excellently with the Blasius boundary layer solution for the flow past a semiinfinite flat plate.
A method is described for the numerical calculation of the hydrodynamic evolution of a self-gravitating configuration in two space dimensions with assumed axial symmetry. The calculation is formulated in cylindrical coordinates with respect to a moving Eulerian grid and is solved using explicit hydrodynamics combined with implicit radiative transfer. The physics included is appropriate for calculation of the collapse of a rotating protostellar cloud. The gravitational field is obtained by means of an alternating-direction iterative technique. Numerical tests to demonstrate the correctness of the method are presented for special cases.
A numerical procedure is proposed for analyzing Langmuir probe data in the presence of a two-group plasma containing both primary and Maxwellian electrons. The procedure is known as a least-squares differential-correction technique for determining the unknown coefficients of the governing equation. It is shown that for a given set of input data the results of the analysis are unique and independent of the initial estimate of electron temperature, that convergence is fastest when electron temperature is overestimated, that the results are sensitive to the region of the curve used as input and also to the voltage increment, and that plasma properties determined by the proposed numerical procedure are either consistent with those determined graphically or closer to expected values. With a suitable data acquisition system, the Fortran IV program worked out for this procedure could be used to provide real-time plasma diagnostic information for an operating ion thruster.
A method for numerical solution of the Navier-Stokes equations for the flow about arbitrary airfoils or other bodies is presented. This method utilizes a numerically generated curvilinear coordinate system having a coordinate line coincident with the body contour. Streamlines, velocity profiles, and pressure and force coefficients for several airfoils and an arbitrary rock are given. Potential flow solutions are also presented. The procedure capable of treating multiple-element airfoils, and potential flow results are presented.
Numerical codes developed for the calculation of three-dimensional nozzle exhaust flow fields associated with hypersonic airbreathing aircraft are described. Both codes employ reference plane grid networks with respect to three coordinate systems. Program CHAR3D is a characteristic code utilizing a new wave preserving network within the reference planes, while program BIGMAC is a finite difference code utilizing conservation variables and a one-sided difference algorithm. Secondary waves are numerically captured by both codes, while the underexpansion shock and plume boundary are treated discretely. The exhaust gas properties consist of hydrogen-air combustion product mixtures in local chemical equilibrium. Nozzle contours are treated by a newly developed geometry package based on dual cubic splines. Results are presented for simple configurations demonstrating two- and three-dimensional multiple wave interactions.
A two-dimensional numerical model was used to investigate the formation of marine advection fog. The model predicts the evolution of potential temperature, horizontal wind, water vapor content, and liquid water content in a vertical cross section of the atmosphere as determined by vertical turbulent transfer and horizontal advection, as well as radiative cooling and drop sedimentation. The model is designed to simulate the formation, development, or dissipation of advection fog in response to transfer of heat and moisture between the atmosphere and the surface as driven by advection over horizontal discontinuities in the surface temperature. Results from numerical simulations of advection fog formation are discussed with reference to observations of marine fog. A survey of candidate fog or cloud microphysics experiments which might be performed in the low gravity environment of a shuttle-type spacecraft in presented. Recommendations are given for relatively simple experiments which are relevent to fog modification problems.
A cubic spline collocation procedure was developed for the numerical solution of partial differential equations. This spline procedure is reformulated so that the accuracy of the second-derivative approximation is improved and parallels that previously obtained for lower derivative terms. The final result is a numerical procedure having overall third-order accuracy of a nonuniform mesh. Solutions using both spline procedures, as well as three-point finite difference methods, are presented for several model problems.
A numerical solution for chemically reacting supersonic gas-particle flows in rocket nozzles and exhaust plumes was described. The gas-particle flow solution is fully coupled in that the effects of particle drag and heat transfer between the gas and particle phases are treated. Gas and particles exchange momentum via the drag exerted on the gas by the particles. Energy is exchanged between the phases via heat transfer (convection and/or radiation). Thermochemistry calculations (chemical equilibrium, frozen or chemical kinetics) were shown to be uncoupled from the flow solution and, as such, can be solved separately. The solution to the set of governing equations is obtained by utilizing the method of characteristics. The equations cast in characteristic form are shown to be formally the same for ideal, frozen, chemical equilibrium and chemical non-equilibrium reacting gas mixtures. The particle distribution is represented in the numerical solution by a finite distribution of particle sizes.