SEARCH · Search NASA
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.
Short duration simulation of rocket exhaust products, flow duration through nozzle and effect on recirculation flow fields
Laminar to turbulent flow transition concepts studied in relation to oscillatory flows found in circulatory system
Neural operators are promising surrogates for dynamical systems but when trained with standard L 2 losses they tend to oversmooth fine-scale turbulent structures. Here, we show that combining operator learning with generative modeling overcomes this limitation. We consider three practical turbulent-flow challenges where conventional neural operators fail: spatio-temporal super-resolution, forecasting, and sparse flow reconstruction. For Schlieren jet super-resolution, an adversarially trained neural operator (adv-NO) reduces the energy-spectrum error by 15 × while preserving sharp gradients at neural operator-like inference cost. For 3D homogeneous isotropic turbulence, adv-NO trained on only 160 timesteps from a single trajectory forecasts accurately for five eddy-turnover times and offers 114 × wall-clock speed-up at inference than the baseline diffusion-based forecasters, enabling near-real-time rollouts. For reconstructing cylinder wake flows from highly sparse Particle Tracking Velocimetry-like inputs, a conditional generative model infers full 3D velocity and pressure fields with correct phase alignment and statistics. These advances enable accurate reconstruction and forecasting at low compute cost, bringing near-real-time analysis and control within reach in experimental and computational fluid mechanics.
Methane and natural gas flow through critical flow nozzles, calculating real gas effects on mass flow rate
Comparisons of the results of testing a single-stage axial-flow compressor with a solid-wall casing and with grooved casings are presented. The depth, location, and number of circumferential grooves in the casing over the casing over the rotor tip were varied. The near-stall weight flow was lower than that with the solid-wall casing for all but one grooved configuration indicating an improvement in the stall margin. The greatest reduction in the near-stall weight flow was noted for the configuration with five grooves located over the blade midchord region.
Flow visualization studies were conducted to evaluate techniques for injecting simulated-fuel and simulated-propellant gases into a spherical cavity for application to open-cycle gaseous-core nuclear rockets. Preliminary studies were conducted with six simulated-fuel injectors and eight simulated-propellant injection configurations. Additional tests were conducted with the best configuration to determine the effect of weight-flow ratio, gas density ratio, injector location, and flow distribution on the simulated-fuel containment characteristics.
Composite photographs of the surface flow and shadowgraphs of the shock wave pattern are presented that depict the hypersonic flow field of a typical delta wing space shuttle orbiter. Results from a wind tunnel test in air are given in side, oblique, and projected plan views for angles of attack from 0 to 60 deg. The tests were conducted at a Mach number of 7.4 and for Reynolds numbers based on body length of 6,000,000 and 9,000,000. The interrelationship is shown for the intersecting bow and wing leading edge waves with the surface flow for angles of attack for which: (1) the leading edge wave is attached, and (2) the leading edge wave is detached.
Wall static pressure measurements and performance parameters are presented for axisymmetric supersonic nozzles with relatively steep convergent sections and comparatively small radius-of-curvature throats. The nozzle walls were essentially adiabatic. These results are compared with those obtained in other nozzles tested previously to appraise the influence of contraction shape on performance. Both the flow coefficient and the thrust were less than the corresponding values for one-dimensional, isentropic, plane flow for both the axial and radial inflow nozzles considered, but the specific impulse, the most important performance parameter, was found to be relatively unchanged. The thrust decrement for the axial inflow nozzles was established primarily by the shape of the contraction section, and could be estimated reasonably well from a conical sink flow consideration. The radial inflow nozzle has a potential advantage from a cooling point of view if used in a rocket engine.
A wind tunnel study of the basic nature of plume induced flow separation is reported with emphasis on the unsteady aspects of the flow. Testing was conducted in a 6 inch by 6 inch blow-down supersonic wind tunnel. A cone-cylinder model with a pluming jet was used as the test model. Tests were conducted with a systematic variation in Mach number and plume pressure. Results of the tests are presented in the form of root-mean-squared surface pressure levels, power spectral densities, photographs of the flow field from which shock angles and separation lengths were taken, and time-averaged surface pressure profiles.
A second-order theory for supersonic flow past slender bodies is presented. Through the introduction of characteristic coordinates as independent variables and the expansion procedure proposed by Lin and Oswatitsch, a uniformly valid solution is obtained for the whole flow field in the axisymmetric case and for far field in the general three-dimensional case. For distances far from the body the theory is an extension of Whitham's first-order solution and for the domain close to the body it is a modification of Van Dyke's second-order solution in the axisymmetric case. From the theory useful formulas relating flow deflections to the Whitham F-function are derived, which permits one to determine the sonic boom strength from wind tunnel measurements fairly close to the body.
Development of a method of estimating deviation angles by analytical procedures was begun. Solutions for inviscid, irrotational flow in the blade-to-blade plane were obtained with a finite-difference calculation method. Deviation angles for a plane cascade with a rounded trailing edge were estimated by using the inviscid-flow solutions and three trailing-edge hypotheses. The estimated deviation angles were compared with existing experimental data over a range of incidence angles at inlet flow angles of 30 deg and 60 deg. The results indicate that deviation angles can be estimated accurately (within 1 deg) by using one of the three trailing-edge hypotheses, but only when pressure losses are low. A new trailing-edge hypotheses is presented which is suitable (for the cascade considered) for both low- and high-loss operating points.
Measurements have been made of the mean flow properties and turbulent fluctuations in adiabatic turbulent boundary layer flows subjected to distributed adverse pressure gradients. In the freestream region upstream of the adverse pressure gradient the Mach number was 3.86, the unit Reynolds number 5.3 million per foot. The boundary layer developed on the wall of an axisymmetric nozzle and straight test section. In order to avoid the effects of streamwise surface curvature the adverse pressure gradients at the test section wall were induced by contoured centerbodies mounted on the wind tunnel centerline. The flow under study simulated that which might be found in an axially symmetric engine inlet of a supersonic aircraft.
Development of a more satisfactory method of estimating deviation angles was attempted using analytical procedures. Solutions for inviscid, irrotational flow in the blade-to-blade plane were obtained with a finite-difference calculation method. Deviation angles for three cascades with rounded trailing edges were estimated using the inviscid flow solutions and four existing trailing edge hypotheses. The estimated deviation angles were compared with experimental values for a range of incidence angle and inlet flow angle. None of the hypotheses were satisfactory at all incidence angles for all the cascades considered. An alternative hypothesis is presented which is suitable for one of the cascade profiles over a range of incidence and blade setting angle.
A new method of calculating transonic flows based on a 'rotated' difference scheme is described. It is suitable for the calculation of both two- and three-dimensional flows without restriction on the speed at infinity and is well adapted to computer use. The Murman procedure is modified to eliminate any assumptions about the direction of flow when constructing the difference scheme. The proper directional property is obtained by rotating the difference scheme to conform with the local stream direction. In the hyperbolic region retarded difference formulas are used for all contributions to the streamwise second derivative, producing a correctly oriented positive artificial viscosity. In the absence of a simple implicit scheme in the hyperbolic and elliptic regions, the concept of iterations as steps in artificial time is introduced. Computer testing of this procedure provides numerical confirmation of the existence and uniqueness of weak solutions of the potential equation when a suitable entropy inequality is enforced.
A method of calculating stage parameters and flow distribution of axial turbines is described. The governing equations apply to space between the blade rows and are based on the assumption of rotationally symmetrical, compressible, adiabatic flow conditions. Results are presented for stage design and flow analysis calculations. Theoretical results from the calculation system are compared with experimental data from low pressure steam turbine tests.
A combined experimental and computational research program is described for testing and guiding turbulence modeling within regions of separation induced by shock waves incident in turbulent boundary layers. Specifically, studies are made of the separated flow the rear portion of an 18%-thick circular-arc airfoil at zero angle of attack in high Reynolds number supercritical flow. The measurements include distributions of surface static pressure and local skin friction. The instruments employed include highfrequency response pressure cells and a large array of surface hot-wire skin-friction gages. Computations at the experimental flow conditions are made using time-dependent solutions of ensemble-averaged Navier-Stokes equations, plus additional equations for the turbulence modeling.
Noise spectra were measured in three dimensions for several surfaces immersed in turbulent flow from a jet and over a range of flow conditions. The data are free field and were corrected to remove the small contributions of jet noise, atmospheric attenuation and feedback tones. These broadband data were compared with the results of available theories which are only strictly applicable to simple geometries over a limited range of conditions. The available theories proved to be accurate over the range of flow, chord length, thickness, angle of attack, and surface geometries defined by the experiments. These results apply to the noise generated by fixed surfaces in engine passages, the lifting surfaces of aircraft and also to fan noise.
An approximate numerical method has been developed for computing the entire transonic flow field over an airfoil, including the boundary layer, the shock-wave/boundary-layer interaction, and the shock-induced-separation. The inviscid flow, the turbulent boundary layer in the weak interaction zone, and the strong interaction elliptic region are solved separately. The inviscid flow is represented by the solution of the transonic small-disturbance equation, while the boundary layer is solved by the generalized Galerkin method, and the strong interaction region (including the wake) is computed from the time-averaged Navier-Stokes equations. A second-order interaction is computed iteratively using the inviscid pressure distribution as a boundary condition for the boundary layer and Navier-Stokes codes, and the viscous displacement thickness to modify the airfoil geometry for the inviscid computation. The computed pressure distributions and shock-wave profiles are compared with experimental data. Agreement is very good except that the pressure recovery on the trailing edge is greater than the experimental data.