Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Conditional Normalizing Flow”

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 343 records · Page 19
All sourcesNASA NTRSDOE OSTI
NASA NTRS1982

Determination of boundary shape of cooled porous region

The shape of a cooled porous wall section is found that will provide a uniform surface temperature, as dictated by material limitations, when the surface is subjected to spatially nonuniform heating. In the analysis, local temperatures and pressures in the porous material are expressed in terms of a potential function. From the imposed thermal conditions, this potential function is governed by the dual constraints of both its value and its normal derivative being specified along the heated surface. The unknown shape of this surface is obtained by meeting these dual conditions. The analytical method uses a generalized conformal mapping procedure that includes a curved boundary. The coolant flow can be compressible or incompressible, and its viscosity can depend on temperature.

Snyder, A.↗
NASA NTRS1988

Investigation of the validity of Reynolds averaged turbulence models at the frequencies that occur in turbomachinery

Turbulent flows subjected to various kinds of unsteady disturbances were simulated using a large-eddy-simulation computer code for flow in a channel. The disturbances were: a normal velocity expressed as a traveling wave on one wall of the channel; staggered blowing and suction distributions on the opposite walls of the channel; and oscillations of the mean flow through the channel. The wall boundary conditions were designed to simulate the effects of wakes of a stator stage passing through a rotor channel in a turbine. The oscillating flow simulated the effects of a pressure pulse moving over the rotor blade boundary layer. The objective of the simulations was to provide better understanding of the effects of time-dependent disturbances on the turbulence of a boundary layer and of the underlying physical phenomena regarding the basic interaction between the turbulence and external disturbances of the type found in turbomachinery. Results showed that turbulence is sensitive to certain ranges of frequencies of disturbances. However, no direct connection was found between the frequency of imposed disturbances and characteristic burst frequency of turbulence. New insight into the nature of turbulence at high frequencies was found. The viscous phenomena near solid walls was found to be the dominant influence for high frequency perturbations. At high frequencies, the turbulence was found to be undisturbed, remaining the same as for the steady mean flow. A transition range exists between the high frequency range and the low, or quasi-steady, range in which the turbulence is not predictable by either quasi-steady models or the steady flow model. The limiting lowest frequency for use of the steady flow turbulence model is that for which the viscous Stokes layer based on the blade passing frequency is thicker than the laminar sublayer.

Kuhn, Gary D.↗
NASA NTRS1994

Studies of two phase flow

The development of instrumentation for the support of research in two-phase flow in simulated microgravity conditions was performed. The funds were expended in the development of a technique for characterizing the motion and size distribution of small liquid droplets dispersed in a flowing gas. Phenomena like this occur in both microgravity and normal earth gravity situations inside of conduits that are carrying liquid-vapor mixtures at high flow rates. Some effort to develop a conductance probe for the measurement of liquid film thickness was also expended.

Witte, Larry C.↗
NASA NTRS2008

Estimation of the Unsteady Aerodynamic Load on Space Shuttle External Tank Protuberances from a Component Wind Tunnel Test

At the wake of the Columbia (STS-107) accident it was decided to remove the Protuberance Aerodynamic Load (PAL) Ramp that was originally intended to protect various protuberances outside of the Space Shuttle External Tank from high buffet load induced by cross-flows at transonic speed. In order to establish the buffet load without the PAL ramp, a wind tunnel test was conducted where segments of the protuberances were instrumented with dynamic pressure transducers; and power-spectra of sectional lift and drag forces at various span-wise locations between two adjacent support brackets were measured under different cross flow angles, Mach number and other conditions. Additionally, frequency-dependent spatial correlations between the sectional forces were also established. The sectional forces were then adjusted by the correlation length to establish span-averaged spectra of normal and lateral forces that can be suitably "added" to various other unsteady forces encountered by the protuberance. This paper describes the methodology used for calculating the correlation-adjusted power spectrum of the buffet load. A second part of the paper describes wind-tunnel results on the difference in the buffet load on the protuberances with and without the PAL ramp. In general when the ramp height is the same as that of the protuberance height, such as that found on the liquid Oxygen part of the tank, the ramp is found to cause significant reduction of the unsteady aerodynamic load. However, on the liquid Hydrogen part of the tank, where the Oxygen feed-line is far larger in diameter than the height of the PAL ramp, little protection is found to be available to all but the Cable Tray.

Panda, J.↗
NASA NTRS2008

Estimation of the Unsteady Aerodynamic Load on Space Shuttle External Tank Protuberances from a Component Wind Tunnel Test

At the wake of the Columbia (STS-107) accident it was decided to remove the Protuberance Aerodynamic Load (PAL) Ramp that was originally intended to protect various protuberances outside of the Space Shuttle External Tank from high buffet load induced by cross-flows at transonic speed. In order to establish the buffet load without the PAL ramp, a wind tunnel test was conducted where segments of the protuberances were instrumented with dynamic pressure transducers; and power-spectra of sectional lift and drag forces at various span-wise locations between two adjacent support brackets were measured under different cross flow angles, Mach number and other conditions. Additionally, frequency-dependent spatial correlations between the sectional forces were also established. The sectional forces were then adjusted by the correlation length to establish span-averaged spectra of normal and lateral forces that can be suitably "added" to various other unsteady forces encountered by the protuberance. This paper describes the methodology used for calculating the correlation-adjusted power spectrum of the buffet load. A second part of the paper describes wind-tunnel results on the difference in the buffet load on the protuberances with and without the PAL ramp. In general when the ramp height is the same as that of the protuberance height, such as that found on the liquid Oxygen part of the tank, the ramp is found to cause significant reduction of the unsteady aerodynamic load. However, on the liquid Hydrogen part of the tank, where the Oxygen feed-line is far larger in diameter than the height of the PAL ramp, little protection is found to be available to all but the Cable Tray.

Panda, Jayatana↗
NASA NTRS2008

Survivability of Psychrobacter cryohalolentis K5 Under Simulated Martian Surface Conditions

Spacecraft launched to Mars can retain viable terrestrial microorganisms on board that may survive the interplanetary transit. Such biota might compromise the search for life beyond Earth if capable of propagating on Mars. The current study explored the survivability of Psychrobacter cryohalolentis K5, a psychrotolerant microorganism obtained from a Siberian permafrost cryopeg, under simulated martian surface conditions of high ultraviolet irradiation, high desiccation, low temperature, and low atmospheric pressure. First, a desiccation experiment compared the survival of P. cryohalolentis cells embedded, or not embedded, within a medium/salt matrix (MSM) maintained at 25 degrees C for 24 hr within a laminar flow hood. Results indicate that the presence of the MSM enhanced survival of the bacterial cells by 1 to 3 orders of magnitude. Second, tests were conducted in a Mars Simulation Chamber to determine the UV tolerance of the microorganism. No viable vegetative cells of P. cryohalolentis were detected after 8 hr of exposure to Mars-normal conditions of 4.55 W/m(2) UVC irradiation (200-280 nm), -12.5 degrees C, 7.1 mbar, and a Mars gas mix composed of CO2 (95.3%), N2 (2.7%), Ar (1.6%), O2 (0.2%), and H(2)O (0.03%). Third, an experiment was conducted within the Mars chamber in which total atmospheric opacities were simulated at tau = 0.1 (dust-free CO2 atmosphere at 7.1 mbar), 0.5 (normal clear sky with 0.4 = dust opacity and 0.1 = CO2-only opacity), and 3.5 (global dust storm) to determine the survivability of P. cryohalolentis to partially shielded UVC radiation. The survivability of the bacterium increased with the level of UVC attenuation, though population levels still declined several orders of magnitude compared to UVC-absent controls over an 8 hr exposure period.

Smith, David J.↗
NASA NTRS1988

Computational and theoretical analysis of free surface flow in a thin liquid film under zero and normal gravity

The results of a numerical computation and theoretical analysis are presented for the flow of a thin liquid film in the presence and absence of a gravitational body force. Five different flow systems were used. Also presented are the governing equations and boundary conditions for the situation of a thin liquid emanating from a pressure vessel; traveling along a horizontal plate with a constant initial height and uniform initial velocity; and traveling radially along a horizontal disk with a constant initial height and uniform initial velocity.

Faghri, Amir↗
NASA NTRS1982

Tangential blowing for control of strong normal shock - Boundary layer interactions on inlet ramps

The use of tangential blowing from a row of holes in an aft facing step is found to provide good control of the ramp boundary layer, normal shock interaction on a fixed geometry inlet over a wide range of inlet mass flow ratios. Ramp Mach numbers of 1.36 and 1.96 are investigated. The blowing geometry is found to have a significant effect on system performance at the highest Mach number. The use of high-temperature air in the blowing system, however, has only a slight effect on performance. The required blowing rates are significantly high for the most severe test conditions. In addition, the required blowing coefficient is found to be proportional to the normal shock pressure rise.

Schwendemann, M. F.↗
NASA NTRS1983

Wind-tunnel investigation of aerodynamic loading on a 0.237-scale model of a remotely piloted research vehicle with a thick, high-aspect-ratio supercritical wing

Wind-tunnel measurements were made of the wing-surface static-pressure distributions on a 0.237 scale model of a remotely piloted research vehicle equipped with a thick, high-aspect-ratio supercritical wing. Data are presented for two model configurations (with and without a ventral pod) at Mach numbers from 0.70 to 0.92 at angles of attack from -4 deg to 8 deg. Large variations of wing-surface local pressure distributions were developed; however, the characteristic supercritical-wing pressure distribution occurred near the design condition of 0.80 Mach number and 2 deg angle of attack. The significant variations of the local pressure distributions indicated pronounced shock-wave movements that were highly sensitive to angle of attack and Mach number. The effect of the vertical pod varied with test conditions; however at the higher Mach numbers, the effects on wing flow characteristics were significant at semispan stations as far outboard as 0.815. There were large variations of the wing loading in the range of test conditions, both model configurations exhibited a well-defined peak value of normal-force coefficient at the cruise angle of attack (2 deg) and Mach number (0.80).

Byrdsong, T. A.↗
NASA NTRS1991

Does a threshold stress for creep exist in HfC-dispersed NiAl?

Recently it was proposed (Jha et al., 1989; Whittenberger et al., 1990) on the basis of constant velocity testing at 1300 K that dispersion strengthened NiAl composites containing about 4 wt pct HfC possess threshold stresses for creep. Further, 1300 K compression testing has been conducted on NiAl+4HfC, and diametrically opposite behavior has been found: for constant load creep tests a normal power law behavior was observed. However, additional constant velocity testing still indicates that the flow stress is essentially independent of strain rate below 10 exp -6/s. Examination of NiAl+4.3HfC specimens deformed under constant velocity conditions revealed that the original hot extruded small grain structure could be converted to large, elongated grains during testing. Such a transformation appears to be responsible for the apparent threshold stress behavior in HfC dispersed NiAl.

Whittenberger, J. D.↗
NASA NTRS1952

The Langley Annular Transonic Tunnel

Report describes the development of the Langley annular transonic tunnel, a facility in which test Mach numbers from 0.6 to slightly over 1.0 are achieved by rotating the test model in an annular passage between two concentric cylinders. Data obtained for two-dimensional airfoil models in the Langley annular transonic tunnel at subsonic and sonic speeds are shown to be in reasonable agreement with experimental data from other sources and with theory when comparisons are made for nonlifting conditions or for equal normal-force coefficients rather than for equal angles of attack. The trends of pressure distributions obtained from measurements in the Langley annular transonic tunnel are consistent with distributions calculated for Prandtl-Meyer flow.

Habel, Louis W↗
NASA NTRS1993

Aerodynamic characteristics and pressure distributions for an executive-jet baseline airfoil section

A wind tunnel test of an executive-jet baseline airfoil model was conducted in the adaptive-wall test section of the NASA Langley 0.3-Meter Transonic Cryogenic Tunnel. The primary goal of the test was to measure airfoil aerodynamic characteristics over a wide range of flow conditions that encompass two design points. The two design Mach numbers were 0.654 and 0.735 with corresponding Reynolds numbers of 4.5 x 10(exp 6) and 8.9 x 10(exp 6) based on chord, respectively, and normal-force coefficients of 0.98 and 0.51, respectively. The tests were conducted over a Mach number range from 0.250 to 0.780 and a chord Reynolds number range from 3 x 10(exp 6) to 18 x 10(exp 6). The angle of attack was varied from -2 deg to a maximum below 10 deg with one exception in which the maximum was 14 deg for a Mach number of 0.250 at a chord Reynolds number of 4.5 x 10(exp 6). Boundary-layer transition was fixed at 5 percent of chord on both the upper and lower surfaces of the model for most of the test. The adaptive-wall test section had flexible top and bottom walls and rigid sidewalls. Wall interference was minimized by the movement of the adaptive walls, and the airfoil aerodynamic characteristics were corrected for any residual top and bottom wall interference.

Allison, Dennis O.↗
NASA NTRS1995

Delta wing vortex manipulation using pulsed and steady blowing during ramp pitching

The effectiveness of steady and pulsed blowing as a method of controlling delta wing vortices during ramp pitching has been investigated in flow visualization experiments conducted in a water tunnel. The recessed angled spanwise blowing technique was utilized for vortex manipulation. This technique was implemented on a beveled 60 delta wing using a pair of blowing ports located beneath the vortex core at 40% chord. The flow was injected primarily in the spanwise direction but was also composed of a component normal to the wing surface. The location of vortex burst was measured as a function of blowing intensity and pulsing frequency under static conditions, and the optimum blowing case was applied at three different wing pitching rates. Experimental results have shown that, when the burst location is upstream of the blowing port, pulsed blowing delays vortex breakdown in static and dynamic cases. Dynamic tests verified the existence of a hysteresis effect and demonstrated the improvements offered by pulsed blowing over both steady blowing and no-blowing scenarios. The application of blowing, at the optimum pulsing frequency, made the vortex breakdown location comparable in static and ramp pitch-up conditions.

Moreira, J.↗
NASA NTRS1999

Estimation of cardiac reserve by peak power: validation and initial application of a simplified index

OBJECTIVES: To validate a simplified estimate of peak power (SPP) against true (invasively measured) peak instantaneous power (TPP), to assess the feasibility of measuring SPP during exercise and to correlate this with functional capacity. DESIGN: Development of a simplified method of measurement and observational study. SETTING: Tertiary referral centre for cardiothoracic disease. SUBJECTS: For validation of SPP with TPP, seven normal dogs and four dogs with dilated cardiomyopathy were studied. To assess feasibility and clinical significance in humans, 40 subjects were studied (26 patients; 14 normal controls). METHODS: In the animal validation study, TPP was derived from ascending aortic pressure and flow probe, and from Doppler measurements of flow. SPP, calculated using the different flow measures, was compared with peak instantaneous power under different loading conditions. For the assessment in humans, SPP was measured at rest and during maximum exercise. Peak aortic flow was measured with transthoracic continuous wave Doppler, and systolic and diastolic blood pressures were derived from brachial sphygmomanometry. The difference between exercise and rest simplified peak power (Delta SPP) was compared with maximum oxygen uptake (VO(2)max), measured from expired gas analysis. RESULTS: SPP estimates using peak flow measures correlated well with true peak instantaneous power (r = 0.89 to 0.97), despite marked changes in systemic pressure and flow induced by manipulation of loading conditions. In the human study, VO(2)max correlated with Delta SPP (r = 0.78) better than Delta ejection fraction (r = 0.18) and Delta rate-pressure product (r = 0.59). CONCLUSIONS: The simple product of mean arterial pressure and peak aortic flow (simplified peak power, SPP) correlates with peak instantaneous power over a range of loading conditions in dogs. In humans, it can be estimated during exercise echocardiography, and correlates with maximum oxygen uptake better than ejection fraction or rate-pressure product.

Non-NASA Center↗
NASA NTRS1999

Fundamental Study of Material Flow in Friction Stir Welds

The presented research project consists of two major parts. First, the material flow in solid-state, friction stir, butt-welds as been investigated using a marker insert technique. Changes in material flow due to welding parameter as well as tool geometry variations have been examined for different materials. The method provides a semi-quantitative, three-dimensional view of the material transport in the welded zone. Second, a FSW process model has been developed. The fully coupled model is based on fluid mechanics; the solid-state material transport during welding is treated as a laminar, viscous flow of a non-Newtonian fluid past a rotating circular cylinder. The heat necessary for the material softening is generated by deformation of the material. As a first step, a two-dimensional model, which contains only the pin of the FSW tool, has been created to test the suitability of the modeling approach and to perform parametric studies of the boundary conditions. The material flow visualization experiments agree very well with the predicted flow field. Accordingly, material within the pin diameter is transported only in the rotation direction around the pin. Due to the simplifying assumptions inherent in the 2-D model, other experimental data such as forces on the pin, torque, and weld energy cannot be directly used for validation. However, the 2-D model predicts the same trends as shown in the experiments. The model also predicts a deviation from the "normal" material flow at certain combinations of welding parameters, suggesting a possible mechanism for the occurrence of some typical FSW defects. The next step has been the development of a three-dimensional process model. The simplified FSW tool has been designed as a flat shoulder rotating on the top of the workpiece and a rotating, cylindrical pin, which extends throughout the total height of the flow domain. The thermal boundary conditions at the tool and at the contact area to the backing plate have been varied to fit experimental data such as temperature profiles, torque and tool forces. General aspects of the experimentally visualized material flow pattern are confirmed by the 3-D model.

Reynolds, Anthony P.↗
NASA NTRS1974

Thermal proton flow in the plasmasphere - The morning sector

Vertical profiles of electron density obtained in the vicinity of the plasmapause using the Alouette-2 topside sounder have been analyzed to assess the presence of H(+) flow in the topside ionosphere. The observations in the midnight sector show clearly the presence of the plasmapause - i.e., there is a sharp boundary separating the poleward regions of polar wind H(+) flow and the more gentle conditions of the plasmasphere where light ions are present in abundance. In contrast, in the sunlit morning sector upward H(+) flow is deduced to be present to invariant latitudes as low as 48 deg (L = 2.2) in the regions normally known to be well inside the plasmasphere. The upward H(+) flux is so large 300,000,00 ions per sq cm per sec that the plasmapause cannot be seen in the latitudinal electron density contours of the topside ionosphere.

Banks, P. M.↗
NASA NTRS1974

A finite-element analysis for steady and oscillatory subsonic flow around complex configurations

The problem of potential subsonic flow around complex configurations is considered. The solution is given of an integral equation relating the values of the potential on the surface of the body to the values of the normal derivative, which is known from the boundary conditions. The surface of the body is divided into small (hyperboloidal quadrilateral) surface elements, which are described in terms of the Cartesian components of the four corner points. The values of the potential (and its normal derivative) within each element is assumed to be constant and equal to its value at the centroid of the element. The coefficients of the equation are given by source and doublet integrals over the surface elements. Closed form evaluations of the integrals are presented. The results obtained with the above formulation are compared with existing analytical and experimental results.

Chen, L. T.↗
NASA NTRS1976

Liquid jet impingement normal to a disk in zero gravity

An experimental and analytical investigation was conducted to determine the free surface shapes of circular liquid jets impinging normal to sharp-edged disks under both normal and zero gravity conditions. An order of magnitude analysis was conducted indicating regions where viscous forces were not significant when computing free surface shapes. The demarcation between the viscous and inviscid region was found to depend upon the flow Reynolds number and the ratio between the jet and disk radius.

Labus, T. L.↗