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.
Characteristics of NACA submerged duct entries and wing leading-edge inlets designed for a 1/4 scale flow model of a fighter-type airplane powered by a jet engine in the fuselage are presented.
It is hypothesized that the spatial structure of a turbulent diffusion flame plays an important role in determining the emissions of radiative energy, soot and NO, from a combustor. This structure, manifested in the two point statistics, is influenced by buoyancy. Radiation, soot and NOx emissions are the cumulative result of processes that occur throughout a flame. For example, radiation fluxes along a line of sight can be found from summing up the contributions from sources in individual pockets of hot soot that emit, and from sinks in cold soot that absorb. Soot and NOx are both the results of slow chemistry and are not equilibrium products. The time that is available for production and burnout is crucial in determining the eventual emissions of these pollutants. Turbulence models generally rely on a single point closure of the appropriate time averaged equations. Hence, spatial information is lost and needs to be modeled using solution variables such as turbulence kinetic energy and dissipation rate, often with the assumption of isotropy. However, buoyancy can affect the physical structure of turbulent flames and can change the spatial extent of soot bearing regions. Theoretical comparisons with models are best done in the limit of infinite Froude number because the inclusion of buoyancy in flow models introduces significant uncertainties. Hence, LII measurements of soot, measurements of radiation fluxes from soot, Particle Imaging Velocimetry (PIV) of the flow field and measurements of post flame NOX will be carried out on the NASA Lewis 2.2 sec drop tower and eventually on the parabolic flight aircraft. The drop rig will be a modified version of a unit that has been successfully used at Lewis in the past.
The role of experiment in the development of Computational Fluid Dynamics (CFD) for external aerodynamic flow-field prediction is discussed. Requirements for code verification from two sources pacing the development of CFD are described: (1) the need for developing adequate flow modeling, and (2) the need for establishing confidence in the use of CFD for predicting complex flows. The type of data needed to meet these needs differs in detail and scope and leads to definite wind tunnel requirements. Recent examples of testing to assess and develop turbulence models, and to verify code development are used to establish future requirements of wind tunnel testing for CFD verification.
We perform wall-modeled large eddy simulations (WMLES) of turbulent flow over a Gaussian-shaped bump geometry, to assess its performance in the acceleration and separation regions of the flow. An unstructured finite-volume solver is used along with an equilibrium wall model. The Mach number of the oncoming flow selected for the simulation is 0.176, and two Reynolds numbers (Re) are simulated that are about 10000 and 36000 based on boundary layer thickness upstream of the bump. Spanwise periodic results for the lower Re case are compared with the available DNS data, while the full three-dimensional simulation results for the higher Re case are compared to available experimental data. Sensitivity of WMLES results are assessed for a number of factors including grid resolution, wall model exchange location, type of wall model, and unstructured grid topology. The WMLES results for the lower Re case agree well with available DNS data in terms of the wall pressure variation, velocity and turbulent stress profile comparisons. The skin-friction predictions show a reduced tendency to separate. This appears to be accentuated for the higher Re case, for the medium resolution grid used in this study, where WMLES does not predict any flow separation in the mean, which is in contrast to the large separation bubble observed in the experiments.
A two-pass explicit scheme is developed in order to exploit some of the capabilities of finite difference modeling (FDM) for finite element modeling (FEM), which offers the opportunity to account for any type of geometry in fluid flow modeling. Features of the first-order upwind and the Lax-Wendroff high precision explicit finite difference algorithms are reviewed. A flux limiter is developed for FEM to serve as an analog for the single limiter function which has been defined for the various FDMs. It is shown that an antidiffusive limiter must be introduced into the weighting function which normally multiplies the time-derivative term in the variational equation. The two-pass scheme which results is demonstrated to be the equivalent of FDMs with five-point support. However, the present scheme is valid only for one-dimensional calculations and linear shape functions for shock tube flow phenomena. Further work is required for its use with nonlinear hyperbolic systems.
With the growing size and complexity of turbulent flow models, data compression approaches are of the utmost importance to analyze, visualize, or restart the simulations. Recently, in-situ autoencoder-based compression approaches have been proposed and shown to be effective at producing reduced representations of turbulent flow data. However, these approaches focus solely on training the model using point-wise sample reconstruction losses that do not take advantage of the physical properties of turbulent flows. In this paper, we show that training autoencoders with additional physics-informed regularizations, e.g., enforcing incompressibility and preserving enstrophy, improves the compression model in three ways: (i) the compressed data better conform to known physics for homogeneous isotropic turbulence without negatively impacting point-wise reconstruction quality, (ii) inspection of the gradients of the trained model uncovers changes to the learned compression mapping that can facilitate the use of explainability techniques, and (iii) as a performance byproduct, training losses are shown to converge up to 12x faster than the baseline model.
Small variations in gravity anomalies and topographic elevation observed in areas that have undergone highly variable amounts of upper crustal thinning can be satisfactorily explained by ductile flow of lower crustal material under the proper conditions. The boundary between the unextended Colorado Plateau and a strongly extended domain in the Basin and Range Province in the Lake Mead (Nevada) region is examined. Finite element modeling of Newtonian flow and power law creep shows that flow over the length scale of the eastern Basin and Range (500) km or more) corresponding to upper crustal extension by a factor of 1.4-3 over 10 million years requires effective viscosities less than 10 to the 18th - 10 to the 20th Pa s for ductile channels 10-25 km thick. Modeling suggests that these effective viscosities may be sustained by lower crustal material deforming at laboratory-derived power law creep rates. The longer-scale flow may require elevated crustal temperatures (more than 700 C), depending on the composition and material properties assumed. Under the boundary conditions assumed in this study the linear viscous flow models yield a satisfactory approximation to deformation by power law creep. This work suggests that flow in the lower crust may be a viable mechanism for producing small variations in total crustal thickness between strongly extended and less extended regions, and thereby explaining the relative uniformity in gravity and topography between such regions.
Experimental simulation to establish accuracy of nonequilibrium flow model with system simulating charring during ablation
Geologic hydrogen (GeoH 2 ) is emerging as a viable clean energy source. It is largely produced through serpentinization, a geological process in which ultramafic rocks react with water under suitable temperature and pressure. Here, this review synthesizes the current understanding of H 2 generation by serpentinization, with an emphasis on reaction mechanisms, kinetics, and thermodynamics, as well as on modeling flow and transport of reacting fluids in geological formations. We describe the role of mineral assemblages, such as olivine and pyroxene, fluid-rock interactions, and catalytic surfaces, in influencing GeoH 2 yield and reaction rates. By integrating models of reaction kinetics, subsurface reactive flow and transport, and the serpentinization process, and by accounting for the thermodynamic state of the system, this review aims to guide future GeoH 2 research and to evaluate the potential of natural hydrogen as a sustainable clean energy source.
The objective of the current research is the development of an improved k-epsilon two-equation compressible turbulence model for turbulent boundary layer flows experiencing strong viscous-inviscid interactions. The development of an improved model is important in the design of hypersonic vehicles such as the National Aerospace Plane (NASP) and the High Speed Civil Transport (HSCT). Improvements have been made to the low Reynolds number functions in the eddy viscosity and dissipation of solenoidal dissipation of the k-epsilon turbulence mode. These corrections offer easily applicable modifications that may be utilized for more complex geometries. The low Reynolds number corrections are functions of the turbulent Reynolds number and are therefore independent of the coordinate system. The proposed model offers advantages over some current models which are based upon the physical distance from the wall, that modify the constants of the standard model, or that make more corrections than are necessary to the governing equations. The code has been developed to solve the Favre averaged, boundary layer equations for mass, momentum, energy, turbulence kinetic energy, and dissipation of solenoidal dissipation using Keller's box scheme and the Newton spatial marching method. The code has been validated by removing the turbulent terms and comparing the solution with the Blasius solution, and by comparing the turbulent solution with an existing k-epsilon model code using wall function boundary conditions. Excellent agreement is seen between the computed solution and the Blasius solution, and between the two codes. The model has been tested for both subsonic and supersonic flat-plate turbulent boundary layer flow by comparing the computed skin friction with the Van Driest II theory and the experimental data of Weighardt; by comparing the transformed velocity profile with the data of Weighardt, and the Law of the Wall and the Law of the Wake; and by comparing the computed results of an adverse pressure gradient with the experimental data of Fernando and Smits. Good agreement is obtained with the experimental correlations for all flow conditions.
The performance of two conceptually-simple uncertainty quantification techniques are tested against the rigorous nested-loop sampling technique of Roy and Oberkampf (Comput Methods Appl Mech Eng, 200: 2131–2144, 2011) (herein called full-sampling) using two very small-scale DEM-based models of particulate flow (one gas-solid flow and one granular flow). The first simplified forward uncertainty propagation technique, reduced-sampling, uses a sensitivity analysis to eliminate uncertain inputs that have little impact on the model output prior to nested-loop sampling. The second technique, boundary-sampling, uses a sensitivity analysis to inform the selection of two bounding cases for each key model output. In conclusion, the uncertainties in the model outputs obtained via the reduced- and boundary-sampling methods agree well with those from full-sampling for both the gas-solid and granular flow models while yielding computational savings of 65–75% (reduced sampling) and 94–97% (boundary sampling).
Based on the potential flow model, computations were carried out for various three-dimensional inlet models. Some of these calculated results are presented in the forms of surface static pressure, flow angularity, surface flow pattern, and inlet flow field. Comparisons with experimental data are also made.
The 4.8 micron flux from the Galilean satellite Io was shown to fluctuate with an amplitude of approximately 10 percent on time scales of every 28 seconds, 40 minutes, night-to-night, and perhaps year-to-year. Such behavior was found to be the result of random fluctuations for which the mean square fluctuation varies inversely with frequency for constant bandwidth measurement. The theory developed for thermionic emission from barium oxide cathodes in vacuum tubes might be applicable to this situation. If so, the fluctuations in the flux from Io's volcanoes may be caused by diffusion of hot convective cells onto the surface of Io. Long term fluctuations may furnish a means by which the Io volcanism can shut down and conserve energy. Thus the discrepancy that exists between measurements of the current heat flow from Io and calculations of tidal dissipation may be resolved. Tests for rapid flickering at 10 microns showed no fluctuations greater than one percent. This agrees with the prediction of the flow model theory in which the 10 microns volcanic thermal emission arises from cooling of old flows.
Water tunnel tests have been conducted to study the flow associated with fin buffet for twin-fin fighter aircraft using 1/48th scale F/A-18 models. Flow visualization made use of colored dyes to determine vortex patterns, and surface hot-film anemometry was used to study the turbulent energy and the frequencies present in the flow. Configurations tested include the full aircraft, aircraft without fins, aircraft without leading-edge-extensions, and aircraft without wings. Test Reynolds number ranged from 4300 to 12,800, with corresponding Mach numbers less than 10 to the -6th. The flow studies show that the LEX vortices burst just forward of the fins at about 25-deg angle-of-attack. Hot film anemometer signals show that fin surface turbulence increases with angle of attack, and that dominant frequencies appear in the flow when bursting occurs. The dominant frequencies correspond to a Strouhal number of about 0.7 for all speeds tested, and for all angles of attack for which vortex bursting was present.
The X-ray properties of clusters of galaxies and elliptical galaxies are reviewed, with an emphasis on cooling flows and their effects on the structure and energetics of radio sources. It is found that the radial profiles of the hot (10 to the 8th K) diffuse (n = about 0.001/cu cm) gas in clusters do not require heating of the intracluster medium after the formation of the cluster. Consideration is given to theoretical formulas for cooling rates, the possible reasons for the observed correlation between cooling flows and the presence of a dominant central galaxy, cooling-flow models of elliptical galaxies, and the significant shear in the velocity field of a cooling flow.
Turbulence models for the pressure-strain term of the Reynolds-stress equations in the vicinity of a moving wall are evaluated for a high Reynolds number flow using decaying grid turbulence as a model problem. The data of Thomas and Hancock are used as a base for evaluating the different turbulence models. In particular, the Rotta model for return-to-isotropy is evaluated both in its inclusion into the Reynolds-stress equation model and in comparison to a nonlinear model advanced by Sarkar and Speziale. Further, models for the wall correction to the transfer term advanced by Launder et al., Shir, and Shih and Lumley are compared. Initial data using the decaying grid turbulence experiment as a base suggests that the coefficients proposed for these models are high perhaps by as much as an order of magnitude. The Shih and Lumley model which satisfies realizability constraints, in particular, seems to hold promise in adequately modeling the Reynolds stress components of this flow. Extensions of this work are to include testing the homogeneous transfer model by Shih and Lumley and the testing of the wall transfer models using their proposed coefficients and the coefficients chosen from this work in a flow with mean shear component.
There are gloveboxes that frequently exceed the 30% relative humidity level required for operations. These exceedances result in alternative operations needing to be utilized. This report documents a computational fluid dynamics (CFD) analysis performed to characterize airflow and humidity transport within a representative glovebox under nominal, low-flow conditions. The analysis focuses on establishing the flow regime, identifying transport-limited regions, and evaluating the implications of laminar flow on humidity removal. A single-phase, laminar flow model with a species transport model was used to represent air and water vapor mixing. Results demonstrate that, at the nominal flow rates, airflow is laminar with weak mixing, which will result in long residence times, tens of minutes long or more, in low-velocity regions. These conclusions suggests that humidity reduction is transport-limited. This analysis provides qualitative insight into airflow structure and humidity transport trends. In addition to characterization of glovebox airflow and humidity transport, results are intended to inform experimental validation and future transient modeling efforts.
A coupled zonal method has been developed for modeling flows about aircraft with highly integrated advanced nozzle installations in which the exhaust plume interacts with the aircraft flow field to enhance performance. A cost effective flow analysis is achieved by dividing the flow domain into zones and coupling the zonal solutions. PAN AIR is used to model the flow in the inviscid zone which surrounds the aircraft. The plume zone flow field is solved with a PNS code. The flow in the optional internal nozzle zone is modeled using a full Navier-Stokes code. An automated iterative procedure couples the inviscid and plume zones by passing updated boundary conditions between the zones. The coupled analysis provides converged solutions for the overall flow field. The method is demonstrated for a highly integrated nonaxisymmetric vectored nozzle installed on a swept-wing aircraft.