Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “approximate domain boundaries”

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 235 records · Page 13
All sourcesNASA NTRSDOE OSTI
NASA NTRS2006

Using Multi-scale Modeling System to Study the Interactions between Clouds, Precipitation, Aerosols, Radiation and Land Surface

Numerical cloud models, which are based the non-hydrostatic equations of motion, have been extensively applied to cloud-scale and mesoscale processes during the past four decades. Because cloud-scale dynamics are treated explicitly, uncertainties stemming from convection that have to be parameterized in (hydrostatic) large-scale models are obviated, or at least mitigated, in cloud models. Global models will use the non-hydrostatic framework when their horizontal resolution becomes about 10 km, the theoretical limit for the hydrostatic approximation. This juncture will be reached one to two decades from now. Over the past generation, voluminous datasets on atmospheric convection have been accumulated from radar, instrumented aircraft, satellites, and rawinsonde measurements in field campaigns, enabling the detailed evaluation of models. Improved numerical methods have resulted in more accurate and efficient dynamical cores in models. Improvements have been made in the parameterizations of microphysical processes, radiation, boundary-layer effects, and turbulence; however, microphysical parameterizations remain a major source of uncertainty in all classes of atmospheric models. In recent years, exponentially increasing computer power has extended cloud-resolving-model integrations from hours to months, the number of computational grid points from less than a thousand to close to ten million. Three-dimensional models are now more prevalent. Much attention is devoted to precipitating cloud systems where the crucial 1-km scales are resolved in horizontal domains as large as 10,000 km in two-dimensions, and 1,000 x 1,000 sq km in three-dimensions. Cloud models now provide statistical information useful for developing more realistic physically based parameterizations for climate models and numerical weather prediction models. It is also expected that NWP and mesoscale model can be run in grid size similar to cloud resolving model through nesting technique. A review of developments, improvements and applications of cloud models (GCE and WRF) at Goddard will be presented in this talk. In particular, a new approach to using multi-scale modeling system to study the interactions between clouds, precipitation, aerosols and land will be presented.

Tao, Wei-Kuo↗
NASA NTRS2018

Direct Numerical Simulation of Acoustic Disturbances in the Rectangular Test Section of a Hypersonic Wind Tunnel

Direct numerical simulations (DNS) of the full-scale rectangular nozzle of a hypersonic wind tunnel are conducted to study the acoustic freestream fluctuations radiating from turbulent boundary layers (TBLs) along the nozzle walls. The nozzle geometry and the flow conditions of the DNS match those of the NASA 20-Inch Mach 6 Tunnel, and the DNS has been completed for a domain without spanwise sidewall boundary conditions. The turbulent boundary layer parameters based on the DNS compare well with those derived from Reynolds Averaged Navier-Stokes (RANS) calculations as well as with the predictions based on Pate’s correlation. A similarly good comparison is observed for both the Mach number distribution and the Reynold’s stresses obtained from the DNS and RANS calculations, respectively. Various characteristics of the acoustic pressure fluctuations within the inviscid core of the nozzle flow are compared with those associated with a single flat plate at a similar freestream Mach number. The frequency spectrum and bulk propagation speeds match well between the nozzle and the flat plate, but the rms pressure fluctuation is higher for the nozzle configuration, likely due to the combined effect of acoustic radiation from the top and bottom walls. Spatial contours of the two-point correlation coefficient display elliptical tails with approximately equal but opposite angles corresponding to the preferred directionality of acoustic structures radiated from both walls. Future work will focus on DNS of the full nozzle configuration, including the effects of the nozzle side walls.

Deegan, Cole P.↗
DOE OSTI2023

Thermomechanical Stress and Creep-Fatigue Analysis of a High-Temperature Prototype Receiver for Heating Particles

This work presents a three-dimensional (3D) thermomechanical model of a prototype-scale enclosed light trapping solar receiver for heating particles. Results of the thermoelastic model are used to estimate receiver lifetime under maximum flux conditions. A computational fluid dynamics (CFD) model is first developed to predict the temperature fields in a multi-panel assembly under steady operating conditions. Solar flux distributions on the receiver are obtained from the software package SolTrace and applied to the 3D thermal model. The subsequent particle heating is captured through a simplified 1D energy balance. Panel reradiation is considered through a surface-to-surface radiation model and natural convection loss to the surrounding air is captured in a representative fluid domain surrounding the receiver. The resulting temperature fields from the CFD analysis are used as inputs for a thermoelastic mechanical model with representative boundary conditions. With the resultant temperature and stress fields, a creep-fatigue damage and lifetime analysis is performed using the linear damage accumulation (LDA) theory. The Manson-Coffin formula and Larson Miller correlation are used to calculate the fatigue and creep, respectively. A maximum damage (corresponding to a 30-year service life) is defined for design assessment. The model was first developed and verified in detail by comparing with published results in the literature (temperature and stress profiles and distributions, and creep/fatigue damage fractions) for tubular solar receivers with supercritical carbon dioxide as the working fluid. It was then implemented to model a planar-cavity receiver with various design parameters. Specifically, three different design geometries are considered, and the results show that a maximum temperature of approximately 1200 K could be reached for each design with the given incident solar flux, with the main difference being the distribution of these temperatures. Preliminary resulting stresses for the small-scale prototype without design optimization vary from 20 MPa to 250 MPa for each design, with the maximum stresses occurring on the front face and concave geometry on the side of the panel. In future work, the developed methodology shown here will be applied to analyze a full-scale (50-150 MWth) receiver.

concentrated solar power↗
NASA NTRS2019

Stabilization of a Swept-Wing Boundary Layer by Discrete Roughness Elements at High Reynolds Numbers

Direct numerical simulations (DNS) are performed to study potential stabilizing ef- fect of spanwise periodic discrete roughness elements (DREs) on cross ow instabilities in a spatially developing three-dimensional boundary layer over an in nite-swept natural- laminar- ow wing at a freestream Mach number of 0:75 and a chord Reynolds number of approximately 25 million. In the DNS, both the spanwise periodic DREs and distributed roughness in the leading-edge region are implemented to simulate a typical experimen- tal scenario in which multiple steady cross ow modes including the most unstable mode (i.e., the \target" mode) emerge because of the presence of naturally distributed surface roughness in the leading edge region and spanwise periodic control cylinders of subcritical wavelength are used to force small-wavelength disturbances (i.e., the control mode) for damping the target mode. The DNS results show that the e ectiveness of DRE control is sensitive to roughness diameter, height, and chordwise placement. For the DRE parame- ters considered in this study, the stabilizing e ect on the target mode is small within the computational domain that ended at about 35% of the chord.

Nicholson, Gary↗
NASA NTRS1993

Massively parallel computation of RCS with finite elements

One of the promising combinations of finite element approaches for scattering problems uses Whitney edge elements, spherical vector wave-absorbing boundary conditions, and bi-conjugate gradient solution for the frequency-domain near field. Each of these approaches may be criticized. Low-order elements require high mesh density, but also result in fast, reliable iterative convergence. Spherical wave-absorbing boundary conditions require additional space to be meshed beyond the most minimal near-space region, but result in fully sparse, symmetric matrices which keep storage and solution times low. Iterative solution is somewhat unpredictable and unfriendly to multiple right-hand sides, yet we find it to be uniformly fast on large problems to date, given the other two approaches. Implementation of these approaches on a distributed memory, message passing machine yields huge dividends, as full scalability to the largest machines appears assured and iterative solution times are well-behaved for large problems. We present times and solutions for computed RCS for a conducting cube and composite permeability/conducting sphere on the Intel ipsc860 with up to 16 processors solving over 200,000 unknowns. We estimate problems of approximately 10 million unknowns, encompassing 1000 cubic wavelengths, may be attempted on a currently available 512 processor machine, but would be exceedingly tedious to prepare. The most severe bottlenecks are due to the slow rate of mesh generation on non-parallel machines and the large transfer time from such a machine to the parallel processor. One solution, in progress, is to create and then distribute a coarse mesh among the processors, followed by systematic refinement within each processor. Elimination of redundant node definitions at the mesh-partition surfaces, snap-to-surface post processing of the resulting mesh for good modelling of curved surfaces, and load-balancing redistribution of new elements after the refinement are auxiliary steps expected to result in a robust low i/o system for very large finite element problems.

Parker, Jay↗
NASA NTRS2007

Simulation of a Periodic Jet in a Crossflow with a RANS Solver Using an Unstructured Grid

A second-order unstructured-grid code, developed and used primarily for steady aerodynamic simulations, is applied to the synthetic jet in a cross flow. The code, FUN3D, is a vertex-centered finite-volume method originally developed by Anderson[1, 2], and is currently supported by members of the Fast Adaptive Aerospace Tools team at NASA Langley. Used primarily for design[3] and analysis[4] of steady aerodynamic configurations, FUN3D incorporates a discrete adjoint capability, and supports parallel computations using MPI. A detailed description of the FUN3D code can be found in the references given above. The code is under continuous development and contains a variety of flux splitting algorithms for the inviscid terms, two methods for computing gradients, several turbulence models, and several solution methodologies; all in varying states of development. Only the most robust and reliable components, based on experiences with steady aerodynamic simulations, were employed in this work. As applied in this work, FUN3D solves the Reynolds averaged Navier-Stokes equations using the one equation turbulence model of Spalart and Allmaras[5]. The spatial discretization is formed on unstructured meshes using a vertex-centered approach. The inviscid terms are evaluated by a flux-difference splitting formulation using least-squares reconstruction and Roe-type approximate Riemann fluxes. Green-Gauss gradient evaluations are used for viscous and turbulence modeling terms. The discrete spatial operator is combined with a backward time operator which is then solved iteratively using point or line Gauss-Seidel and local time stepping in a pseudo time. For steady flows, the physical time step is set to infinity and the pseudo time step is ramped up with the iteration count. A second-order backward in time operator is used for time accurate flows with 20 to 50 steps in the pseudo time applied at each physical time step. For this effort, FUN3D was modified to support spatially varying boundary and initial conditions, and unsteady boundary conditions. Also, a specialized in/out flow boundary condition was implemented to model the action of the diaphragm. This boundary condition is described below in more detail. The grids were generated using the internally developed codes GridEX[6] for meshing the surfaces and inviscid regions of the domain, and for CAD access; and MesherX[7] for meshing the viscous regions. Grid spacing in on the surfaces and in the inviscid regions are indirectly controlled by specifying sources. The viscous layers are generated using an advancing layer technique. MeshersX allows the user to control the spatial variation of the first step off the surface, growth rates, and the termination criterion by providing small problem dependent subroutines.

Atkins, H. L.↗
DOE OSTI2022

Generalized quasiharmonic approximation via space group irreducible derivatives

The quasiharmonic approximation (QHA) is the simplest nontrivial approximation for interacting phonons under constant pressure, bringing the effects of anharmonicity into temperature-dependent observables. Nonetheless, the QHA is often implemented with additional approximations due to the complexity of computing phonons under arbitrary strains, and the generalized QHA, which employs constant stress boundary conditions, has not been completely developed. In this work we formulate the generalized QHA, providing a practical algorithm for computing the strain state and other observables as a function of temperature and true stress. We circumvent the complexity of computing phonons under arbitrary strains by employing irreducible second-order displacement derivatives of the Born-Oppenheimer potential and their strain dependence, which are efficiently and precisely computed using the lone irreducible derivative approach. We formulate two complementary strain parametrizations: a discretized strain grid interpolation and a Taylor series expansion in symmetrized strain. We illustrate our approach by evaluating the temperature and pressure dependence of select elastic constants and the thermal expansion in thoria (ThO 2 ) using density functional theory with three exchange-correlation functionals. The QHA results are compared to our measurements of the elastic constant tensor using time-domain Brillouin scattering and inelastic neutron scattering. Our irreducible derivative approach simplifies the implementation of the generalized QHA, which will facilitate reproducible, data-driven applications.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2022

Measuring sub-surface spatially varying thermal conductivity of silicon implanted with krypton

The thermal properties of semiconductors following exposure to ion irradiation are of great interest for the cooling of electronic devices; however, gradients in composition and structure due to irradiation often make the measurement difficult. Furthermore, the nature of spatial variations in thermal resistances due to spatially varying ion irradiation damage is not well understood. In this work, we develop an advancement in the analysis of time-domain thermoreflectance to account for spatially varying thermal conductivity in a material resulting from a spatial distribution of defects. We then use this method to measure the near-surface (≲1 μm) thermal conductivity of silicon wafers irradiated with Kr + ions, which has an approximate Gaussian distribution centered 260 nm into the sample. Our numerical analysis presented here allows for the spatial gradient of thermal conductivity to be extracted via what is fundamentally a volumetric measurement technique. We validate our findings via transmission electron microscopy, which is able to confirm the spatial variation of the sub-surface silicon structure, and provide additional insight into the local structure resulting from the effects of ion bombardment. Thermal measurements found the ion stopping region to have a nearly 50x reduction in thermal conductivity as compared to pristine silicon, while TEM showed the region was not fully amorphized. Our results suggest this drastic reduction in silicon thermal conductivity is primarily driven by structural defects in crystalline regions along with boundary scattering between amorphous and crystalline regions, with a negligible contribution being due to implanted krypton ions themselves.

36 MATERIALS SCIENCE↗
NASA NTRS2017

Understanding Arctic Surface Temperature Differences in Reanalyses

Reanalyses in the Arctic are widely used for model evaluation and for understanding contemporary climate change. Nevertheless, differences among reanalyses in fundamental meteorological variables including surface air temperature are large. For example, the 1980-2009 mean surface air temperature for the north polar cap (70ÂdegN-90ÂdegN) among global reanalyses span a range of 2.4 K, which approximates the average warming trend from these reanalyses over the 30-year period of 2.1 K. Understanding these differences requires evaluation over the three principal surface domains of the Arctic: glaciated land, the unglaciated terrestrial surface, and sea ice/ocean. An examination is conducted of contemporary global reanalyses of the ECMWF Interim project, NASA MERRA, MERRA-2, JRA-55, and NOAA CFSR using available in situ data and assessments of the surface energy budget. Overly-simplistic representations of the Greenland Ice Sheet surface are found to be associated with local warm air temperature biases in winter. A review of progress made in the development of the MERRA-2 land-ice representation is presented. Large uncertainty is also found in temperatures over the Arctic tundra and boreal forest zone. But a key focus of temperature differences for northern high latitudes is the Arctic Ocean. Near-surface air temperature differences over the Arctic Ocean are found to be related to discrepancies in sea ice and sea surface temperature boundary data, which are severely compromised in current reanalyses. Issues with the modeled representation of sea ice cover are an additional factor in reanalysis temperature trends. Differences in the representation of the surface energy budget among the various reanalyses are also reviewed.

Surface Temperatur↗
DOE OSTI2022

Modeling Sequences of Earthquakes and Aseismic Slip (SEAS) in Elasto-Plastic Fault Zones With a Hybrid Finite Element Spectral Boundary Integral Scheme

We present a coupled finite element spectral boundary integral framework for modeling sequences of earthquakes and aseismic slip on a 2-D planar rate-and-state fault with off-fault visco-plastic response in the plane strain approximation. The model resolves both slow aseismic deformation and inertia effects during rapid slip. As an application, we perform two sets of simulations with different choices of cohesion to explore the co-evolution of fault slip, bulk plasticity and local stress fields. The first set implements a relatively large value of the cohesion parameter, which results in limiting inelastic strain accumulation to dynamic rupture phases. The second set implements a smaller cohesion, allowing for plastic strain to accumulate in both seismic and aseismic phases. For the first model, our results indicate that the extent and distribution of plastic strain depend on the angle of maximum compressive principal stress. At larger angles, inelastic strain accumulates on the extensional side of a dynamically propagating rupture. At smaller angles, the extent of plasticity is limited to the compressional side of the domain. At smaller cohesion values, off-fault plasticity may occur during aseismic slip, which alters the nucleation characteristics and earthquake sequence pattern. Furthermore, our results at lower cohesion values indicate that plastic strain accumulation may occur in both the extensional and compressional sides of the off-fault bulk even at higher angles of maximum compression. In conclusion, this produces damage patterns that deviate from the traditional off-fault fan-like distribution observed in dynamic rupture simulations and emphasizes the significance of long-term deformation in interpreting observations.

58 GEOSCIENCES↗
NASA NTRS2003

Synthetic Jets in Cross-flow: Round Jet - Part 1

Results of an experimental investigation on synthetic jets from round orifices with and without cross-flow are presented. Jet Reynolds number up to 46,000 with a fully turbulent approach boundary layer, and Stokes number up to 400. are covered. The threshold of stroke length for synthetic jet formation. in the absence of the cross-flow, is found to be Lo /D approximately 0.5. Above Lo /D is approximately 10, the profiles of normalized centerline mean velocity appear to become invariant. It is reasoned that the latter threshold may be related to the phenomenon of saturation of impulsively generated vortices. In the presence of the cross-flow, the penetration height of a synthetic jet is found to depend on the momentum- flux ratio . When this ratio is defined in terms of the maximum jet velocity and the cross-flow velocity. not only all data collapse but also the jet trajectory is predicted well by correlation equation available for steady jets-in-cross-flow. Distributions of mean velocity, streamwise vorticity as well as turbulence intensity for a synthetic jet in cross-flow are found to be similar to those of a steady jet-in-cross-flow. A pair of counter-rotating streamwise vortices, corresponding to the bound vortex pair of the steady case, is clearly observed. Mean velocity distribution exhibits a dome of low momentum fluid pulled up from the boundary layer, and the entire domain is characterized by high turbulence.

Zaman, K. B. M. Q.↗
NASA NTRS1986

Thermodynamic evaluation of transonic compressor rotors using the finite volume approach

A method was developed which calculates two-dimensional, transonic, viscous flow in ducts. The finite volume, time marching formulation is used to obtain steady flow solutions of the Reynolds-averaged form of the Navier Stokes equations. The entire calculation is performed in the physical domain. The method is currently limited to the calculation of attached flows. The features of the current method can be summarized as follows. Control volumes are chosen so that smoothing of flow properties, typically required for stability, is now needed. Different time steps are used in the different governing equations to improve the convergence speed of the viscous calculations. A new pressure interpolation scheme is introduced which improves the shock capturing ability of the method. A multi-volume method for pressure changes in the boundary layer allows calculations which use very long and thin control volumes. A special discretization technique is also used to stabilize these calculations. A special formulation of the energy equation is used to provide improved transient behavior of solutions which use the full energy equation. The method is then compared with a wide variety of test cases. The freestream Mach numbers range from 0.075 to 2.8 in the calculations. Transonic viscous flow in a converging diverging nozzle is calculated with the method; the Mach number upstream of the shock is approximately 1.25. The agreement between the calculated and measured shock strength and total pressure losses is good. Essentially incompressible turbulent boundary layer flow in a adverse pressure gradient is calculated and the computed distribution of mean velocity and shear stress are in good agreement with the measurements. At the other end of the Mach number range, a flat plate turbulent boundary layer with a freestream Mach number of 2.8 is calculated using the full energy equation; the computed total temperature distribution and recovery factor agree well with the measurements when a variable Prandtl number is used through the boundary layer.

Nicholson, S.↗
NASA NTRS2003

Effects of Wing Leading Edge Penetration with Venting and Exhaust Flow from Wheel Well at Mach 24 in Flight

A baseline solution for CFD Point 1 (Mach 24) in the STS-107 accident investigation was modified to include effects of: (1) holes through the leading edge into a vented cavity; and (2) a scarfed, conical nozzle directed toward the centerline of the vehicle from the forward, inboard corner of the landing gear door. The simulations were generated relatively quickly and early in the investigation because simplifications were made to the leading edge cavity geometry and an existing utility to merge scarfed nozzle grid domains with structured baseline external domains was implemented. These simplifications in the breach simulations enabled: (1) a very quick grid generation procedure; and (2) high fidelity corroboration of jet physics with internal surface impingements ensuing from a breach through the leading edge, fully coupled to the external shock layer flow at flight conditions. These simulations provided early evidence that the flow through a two-inch diameter (or larger) breach enters the cavity with significant retention of external flow directionality. A normal jet directed into the cavity was not an appropriate model for these conditions at CFD Point 1 (Mach 24). The breach diameters were of the same order or larger than the local, external boundary-layer thickness. High impingement heating and pressures on the downstream lip of the breach were computed. It is likely that hole shape would evolve as a slot cut in the direction of the external streamlines. In the case of the six-inch diameter breach the boundary layer is fully ingested. The intent of externally directed jet simulations in the second scenario was to approximately model aerodynamic effects of a relatively large internal wing pressure, fueled by combusting aluminum, which deforms the corner of the landing gear door and directs a jet across the windside surface. These jet interactions, in and of themselves, were not sufficiently large to explain observed aerodynamic behavior.

Gnoffo, Peter A.↗
NASA NTRS2007

High-Precision Nd Isotopes in Picrites from Hawaii and Iceland - No Evidence for an Early-Formed Enriched Reservoir

New high-precision Nd isotope measurements have shown that the present-day Nd-142/Nd-144 for average chondrites is approximately 20 ppm lower than that for the terrestrial standard and all recent mantle-derived samples measured to date. One explanation for these differences is that an enriched missing reservoir with lower Nd-142/Nd-144 resides in the mantle to balance the bulk Earth to chondritic. Data from Archean Greenland rocks show anomalies in Nd-142/Nd-144 of up to 40 ppm higher than the proposed average for chondrites. This difference between the Archean Greenland rocks and present-day mantle-derived samples has been interpreted to result from remixing of an early-formed enriched reservoir into the convecting mantle during the Archean. If so, the implication from this is that remixing of the enriched reservoir largely shut down some time in the past such that it must at present reside in a region in the mantle that infrequently participates dynamically and leading to volcanism at the surface. Several studies have suggested that the missing reservoir may be present just above the coremantle boundary (CMB). Depending on the size of this reservoir it could potentially make up all of D or exist as distinct domains within it. If such a reservoir does exist, then mantle-derived samples from plume systems that are thought to come from the CMB may be the best opportunity to identify this component using high-precision Nd isotope measurements. To test this, picrites from Hawaii with coupled enrichments in Os-186-Os-187 that has been proposed to be a signature of core-mantle interaction were measured on the JSC Triton for high-precision Nd-142/Nd-144. For comparison, picrites from Hawaii and Iceland that do not show coupled enrichments in Os-186-Os-187 were measured. We have established an external precision for Nd-142/Nd-144 of 3.6 ppm (2 sigma) during the analytical campaign. The Iceland (n=5) and Hawaiian data (n=9) have Nd-142 ranging from -0.01 plus or minus 0.03 to +0.03 plus or minus 0.03 (2 sigma) and there is no resolvable difference between samples with Os isotope enrichments versus those without. The average epsilon Nd-142 of +0.011 plus or minus 0.034 (2 sigma ) for all of the samples (n=14) is not resolvable from the Ames and La Jolla standards with epsilon Nd-142 of +0.000 plus or minus 0.036 (n=16, 2 sigma). These data confirm recent measurements on lavas for the absence of an ancient enriched Nd-142 isotopic signature in plume systems likely derived from D and indicates that such a reservoir, if existing, must now reside in areas of the deep mantle that plumes do not sample.

Brandon, A. D.↗
NASA NTRS1994

The Divnoe meteorite: Petrology, chemistry, oxygen isotopes and origin

The Divnoe meteorite is an olivine-rich primitive achondrite with subchondritic chemistry and mineralogy. It has a granoblastic, coarse-grained, olivine groundmass (CGL: coarse-grained lithology) with relatively large pyroxene-plagioclase poiklitic patches (PP) and small fine-grained domains of an opaque-rich lithology (ORL). Both PP and ORL are inhomogeneously distributed and display reaction boundaries with the groundmass. Major silicates, olivine Fa(20-28) and orthopyroxyene Fs(20-28 Wo(0.5-2.5), display systematic differences in composition between CGL and ORL as well as a complicated pattern of variations within CGL. Accessory plagioclase has low K content and displays regular igneous zoning with core compositions An(40-45) and rims An(32-37). The bulk chemical composition of Divnoe is similar to that of olivine-rich primitive achondrites, except for a depletion of incompatible elements and minor enrichment of refractory siderophiles. Oxygen isotope compositions for whole-rock and separated minerals from Divnoe fall in a narrow range, with mean delta O-18 = +4.91, delta O-17 = +2.24, and Delta O-17 = -0.26 +/- 0.11. The isotopic composition is not within the range of any previously recognized group but is very close to that of the brachinites. To understand the origin of Divnoe lithologies, partial melting and crystallization were modelled using starting compositions equal to that of Divnoe and some chondritic meteorites. It was found that the Divnoe composition could be derived from a chondritic source region by approximately 20 wt% partial melting at Ta approximately 1300 C and log(fO2) = IW-1.8, followed by approximtely 60 wt% crystallization of the partial melt formed, and removal of the still-liquid portion of the partial melt. Removal of the last partial melt resulted in depletion of the Divnoe plagioclase in Na and K. In this scenario, CGL represents the residue of partial melting, and PP is a portion of the partial melt that crystallized in situ. The ORL was formed during the final stages of partial melting by reaction between gaseous sulfur and residual olivine in the source region. A prominent feature of Divnoe is fine micron-scale chemical variations within olivine grains, related to lamellar structures the olivines display. The origin of these structures is not known.

Petaev, M. I.↗
NASA NTRS1994

Radiation boundary condition and anisotropy correction for finite difference solutions of the Helmholtz equation

In this paper finite-difference solutions of the Helmholtz equation in an open domain are considered. By using a second-order central difference scheme and the Bayliss-Turkel radiation boundary condition, reasonably accurate solutions can be obtained when the number of grid points per acoustic wavelength used is large. However, when a smaller number of grid points per wavelength is used excessive reflections occur which tend to overwhelm the computed solutions. Excessive reflections are due to the incompability between the governing finite difference equation and the Bayliss-Turkel radiation boundary condition. The Bayliss-Turkel radiation boundary condition was developed from the asymptotic solution of the partial differential equation. To obtain compatibility, the radiation boundary condition should be constructed from the asymptotic solution of the finite difference equation instead. Examples are provided using the improved radiation boundary condition based on the asymptotic solution of the governing finite difference equation. The computed results are free of reflections even when only five grid points per wavelength are used. The improved radiation boundary condition has also been tested for problems with complex acoustic sources and sources embedded in a uniform mean flow. The present method of developing a radiation boundary condition is also applicable to higher order finite difference schemes. In all these cases no reflected waves could be detected. The use of finite difference approximation inevita bly introduces anisotropy into the governing field equation. The effect of anisotropy is to distort the directional distribution of the amplitude and phase of the computed solution. It can be quite large when the number of grid points per wavelength used in the computation is small. A way to correct this effect is proposed. The correction factor developed from the asymptotic solutions is source independent and, hence, can be determined once and for all. The effectiveness of the correction factor in providing improvements to the computed solution is demonstrated in this paper.

Tam, Christopher K. W.↗
DOE OSTI2025

Direct numerical simulations for hybrid rocket boundary layers: Performance modeling and scaling

This paper presents a comprehensive performance and scaling analysis of direct numerical simulations for reacting boundary layers, focusing on slab burner configurations. Using a PETSc-based finite volume CFD framework, the study evaluates the scalability and computational cost of flow, chemistry, and radiation evaluations across 2D and 3D simulations. Polymethyl methacrylate (PMMA) is the fuel with pure O 2 as the oxidizer, modeled using a detailed chemical kinetics mechanism with 113 species and 660 reactions. A ray-tracing-based radiation solver, designed for distributed memory applications, is implemented to model radiation heat transfer. Parallel scalability is analyzed for the coupled flow, chemistry, and radiation heat transfer processes. Weak and strong scaling studies are conducted on up to 15,000 computational ranks, revealing robust performance when flow cells exceed 200 per rank. Chemistry evaluations dominate the computational cost in large 3D simulations, accounting for approximately 40% of the total runtime, while flow processes contribute around 35%, and radiation solver contributions remain below 10% due to reduced evaluation frequencies. GPU accelerated chemistry evaluation, implemented with Zero-RK, demonstrates significant promise, achieving up to a 4x speedup for workloads exceeding 30,000 cells per GPU. However, diminishing returns are observed for smaller workloads due to CPU-GPU communication overhead. This study identifies key challenges, including memory bottlenecks and the effects of domain partitioning on flow scalability, while highlighting the potential of GPU-accelerated chemistry to reduce computational costs. In conclusion, these findings provide realizable run configurations for 2D, 3D, and GPU-accelerated cases, offering insights for optimizing reactive flow solvers.

CFD Scalability↗
DOE OSTI2022

A comprehensive and fair comparison of two neural operators (with practical extensions) based on $\mathrm{FAIR}$ data

Neural operators can learn nonlinear mappings between function spaces and offer a new simulation paradigm for real-time prediction of complex dynamics for realistic diverse applications as well as for system identification in science and engineering. Herein, we investigate the performance of two neural operators, which have shown promising results so far, and we develop new practical extensions that will make them more accurate and robust and importantly more suitable for industrial-complexity applications. The first neural operator, DeepONet, was published in 2019 (Lu et al., 2019), and its original architecture was based on the universal approximation theorem of Chen & Chen (1995). The second one, named Fourier Neural Operator or FNO, was published in 2020, and it is based on parameterizing the integral kernel in the Fourier space. DeepONet is represented by a summation of products of neural networks (NNs), corresponding to the branch NN for the input function and the trunk NN for the output function; both NNs are general architectures, e.g., the branch NN can be replaced with a CNN or a ResNet. According to Kovachki et al. (2021), FNO in its continuous form can be viewed conceptually as a DeepONet with a specific architecture of the branch NN and a trunk NN represented by a trigonometric basis. In order to compare FNO with DeepONet computationally for realistic setups, we develop several extensions of FNO that can deal with complex geometric domains as well as mappings where the input and output function spaces are of different dimensions. We also develop an extended DeepONet with special features that provide inductive bias and accelerate training, and we present a faster implementation of DeepONet with cost comparable to the computational cost of FNO, which is based on the Fast Fourier Transform. Here we consider 16 different benchmarks to demonstrate the relative performance of the two neural operators, including instability wave analysis in hypersonic boundary layers, prediction of the vorticity field of a flapping airfoil, porous media simulations in complex-geometry domains, etc. We follow the guiding principles of FAIR (Findability, Accessibility, Interoperability, and Reusability) for scientific data management and stewardship. The performance of DeepONet and FNO is comparable for relatively simple settings, but for complex geometries the performance of FNO deteriorates greatly. We also compare theoretically the two neural operators and obtain similar error estimates for DeepONet and FNO under the same regularity assumptions.

42 ENGINEERING↗