Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “direct solver”

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 451 records · Page 25
All sourcesNASA NTRSDOE OSTI
NASA NTRS2008

Efficient Unstructured Grid Adaptation Methods for Sonic Boom Prediction

This paper examines the use of two grid adaptation methods to improve the accuracy of the near-to-mid field pressure signature prediction of supersonic aircraft computed using the USM3D unstructured grid flow solver. The first method (ADV) is an interactive adaptation process that uses grid movement rather than enrichment to more accurately resolve the expansion and compression waves. The second method (SSGRID) uses an a priori adaptation approach to stretch and shear the original unstructured grid to align the grid with the pressure waves and reduce the cell count required to achieve an accurate signature prediction at a given distance from the vehicle. Both methods initially create negative volume cells that are repaired in a module in the ADV code. While both approaches provide significant improvements in the near field signature (< 3 body lengths) relative to a baseline grid without increasing the number of grid points, only the SSGRID approach allows the details of the signature to be accurately computed at mid-field distances (3-10 body lengths) for direct use with mid-field-to-ground boom propagation codes.

Campbell, Richard L.↗
NASA NTRS2015

Aeroacoustic Simulations of a Nose Landing Gear Using FUN3D on Pointwise Unstructured Grids

Numerical simulations have been performed for a partially-dressed, cavity-closed (PDCC) nose landing gear configuration that was tested in the University of Florida's open-jet acoustic facility known as the UFAFF. The unstructured-grid flow solver FUN3D is used to compute the unsteady flow field for this configuration. Mixed-element grids generated using the Pointwise(TradeMark) grid generation software are used for these simulations. Particular care is taken to ensure quality cells and proper resolution in critical areas of interest in an effort to minimize errors introduced by numerical artifacts. A hybrid Reynolds-averaged Navier-Stokes/large eddy simulation (RANS/LES) turbulence model is used for these simulations. Solutions are also presented for a wall function model coupled to the standard turbulence model. Time-averaged and instantaneous solutions obtained on these Pointwise grids are compared with the measured data and previous numerical solutions. The resulting CFD solutions are used as input to a Ffowcs Williams-Hawkings noise propagation code to compute the farfield noise levels in the flyover and sideline directions. The computed noise levels compare well with previous CFD solutions and experimental data.

Vatsa, Veer N.↗
NASA NTRS1991

The improved robustness of multigrid elliptic solvers based on multiple semicoarsened grids

Multigrid convergence rates degenerate on problems with stretched grids or anisotropic operators, unless one uses line or plane relaxation. For 3-D problems, only plane relaxation suffices, in general. While line and plane relaxation algorithms are efficient on sequential machines, they are quite awkward and inefficient on parallel machines. A new multigrid algorithm is presented based on the use of multiple coarse grids, that eliminates the need for line or plane relaxation in anisotropic problems. This algorithm was developed and the standard multigrid theory was extended to establish rapid convergence for this class of algorithms. The new algorithm uses only point relaxation, allowing easy and efficient parallel implementation, yet achieves robustness and convergence rates comparable to line and plane relaxation multigrid algorithms. The algorithm described is a variant of Mulder's multigrid algorithm for hyperbolic problems. The latter uses multiple coarse grids to achieve robustness, but is unsuitable for elliptic problems, since its V-cycle convergence rate goes to one as the number of levels increases. The new algorithm combines the contributions from the multiple coarse grid via a local switch, based on the strength of the discrete operator in each coordinate direction.

Naik, Naomi H.↗
NASA NTRS2015

A Radiation Solver for the National Combustion Code

A methodology is given that converts an existing finite volume radiative transfer method that requires input of local absorption coefficients to one that can treat a mixture of combustion gases and compute the coefficients on the fly from the local mixture properties. The Full-spectrum k-distribution method is used to transform the radiative transfer equation (RTE) to an alternate wave number variable, g . The coefficients in the transformed equation are calculated at discrete temperatures and participating species mole fractions that span the values of the problem for each value of g. These results are stored in a table and interpolation is used to find the coefficients at every cell in the field. Finally, the transformed RTE is solved for each g and Gaussian quadrature is used to find the radiant heat flux throughout the field. The present implementation is in an existing cartesian/cylindrical grid radiative transfer code and the local mixture properties are given by a solution of the National Combustion Code (NCC) on the same grid. Based on this work the intention is to apply this method to an existing unstructured grid radiation code which can then be coupled directly to NCC.

National Combustion Code↗
NASA NTRS2001

Investigation of Conjugate Heat Transfer in Turbine Blades and Vanes

We report on work carried out to develop a 3-D coupled Finite Volume/BEM-based temperature forward/flux back (TFFB) coupling algorithm to solve the conjugate heat transfer (CHT) which arises naturally in analysis of systems exposed to a convective environment. Here, heat conduction within a structure is coupled to heat transfer to the external fluid which is convecting heat into or out of the solid structure. There are two basic approaches to solving coupled fluid structural systems. The first is a direct coupling where the solution of the different fields is solved simultaneously in one large set of equations. The second approach is a loose coupling strategy where each set of field equations is solved to provide boundary conditions for the other. The equations are solved in turn until an iterated convergence criterion is met at the fluid-solid interface. The loose coupling strategy is particularly attractive when coupling auxiliary field equations to computational fluid dynamics codes. We adopt the latter method in which the BEM is used to solve heat conduction inside a structure which is exposed to a convective field which in turn is resolved by solving the NASA Glenn compressible Navier-Stokes finite volume code Glenn-HT. The BEM code features constant and bi-linear discontinuous elements and an ILU-preconditioned GMRES iterative solver for the resulting non-symmetric algebraic set arising in the conduction solution. Interface of flux and temperature is enforced at the solid/fluid interface, and a radial-basis function scheme is used to interpolated information between the CFD and BEM surface grids. Additionally, relaxation is implemented in passing the fluxes from the conduction solution to the fluid solution. Results from a simple test example are reported.

Kassab, A. J.↗
NASA NTRS2016

An Overview of Spray Modeling With OpenNCC and its Application to Emissions Predictions of a LDI Combustor at High Pressure

The open national combustion code (Open- NCC) is developed with the aim of advancing the current multi-dimensional computational tools used in the design of advanced technology combustors. In this paper we provide an overview of the spray module, LSPRAY-V, developed as a part of this effort. The spray solver is mainly designed to predict the flow, thermal, and transport properties of a rapidly evaporating multi-component liquid spray. The modeling approach is applicable over a wide-range of evaporating conditions (normal, superheat, and supercritical). The modeling approach is based on several well-established atomization, vaporization, and wall/droplet impingement models. It facilitates large-scale combustor computations through the use of massively parallel computers with the ability to perform the computations on either structured & unstructured grids. The spray module has a multi-liquid and multi-injector capability, and can be used in the calculation of both steady and unsteady computations. We conclude the paper by providing the results for a reacting spray generated by a single injector element with 600 axially swept swirler vanes. It is a configuration based on the next-generation lean-direct injection (LDI) combustor concept. The results include comparisons for both combustor exit temperature and EINOX at three different fuel/air ratios.

spray modeling↗
NASA NTRS1987

High performance Solid Rocket Motor (SRM) submerged nozzle/combustion cavity flowfield assessment

Two and three dimensional internal flowfield solutions for critical points in the Space Shuttle solid rocket booster burn time were developed using the Lockheed Huntsville GIM/PAID Navier-Stokes solvers. These perfect gas, viscous solutions for the high performance motor characterize the flow in the aft segment and nozzle of the booster. Two dimensional axisymmetric solutions were developed at t = 20 and t = 85 sec motor burn times. The t = 85 sec solution indicates that the aft segment forward inhibitor stub produces vortices with are shed and convected downwards. A three dimensional 3.5 deg gimbaled nozzle flowfield solution was developed for the aft segment and nozzle at t = 9 sec motor burn time. This perfect gas, viscous analysis, provided a steady state solution for the core region and the flow through the nozzle, but indicated that unsteady flow exists in the region under the nozzle nose and near the flexible boot and nozzle/case joint. The flow in the nozzle/case joint region is characterized by low magnitude pressure waves which travel in the circumferential direction. From the two and three dimensional flowfield calculations presented it can be concluded that there is no evidence from these results that steady state gas dynamics is the primary mechanism resulting in the nozzle pocketing erosion experienced on SRM nozzles 8A or 17B. The steady state flowfield results indicate pocketing erosion is not directly initiated by a steady state gas dynamics phenomenon.

Freeman, J. A.↗
NASA NTRS2012

Large-Eddy / Reynolds-Averaged Navier-Stokes Simulations of a Dual-Mode Scramjet Combustor

Numerical simulations of reacting and non-reacting flows within a scramjet combustor configuration experimentally mapped at the University of Virginia s Scramjet Combustion Facility (operating with Configuration A ) are described in this paper. Reynolds-Averaged Navier-Stokes (RANS) and hybrid Large Eddy Simulation / Reynolds-Averaged Navier-Stokes (LES / RANS) methods are utilized, with the intent of comparing essentially blind predictions with results from non-intrusive flow-field measurement methods including coherent anti-Stokes Raman spectroscopy (CARS), hydroxyl radical planar laser-induced fluorescence (OH-PLIF), stereoscopic particle image velocimetry (SPIV), wavelength modulation spectroscopy (WMS), and focusing Schlieren. NC State's REACTMB solver was used both for RANS and LES / RANS, along with a 9-species, 19- reaction H2-air kinetics mechanism by Jachimowski. Inviscid fluxes were evaluated using Edwards LDFSS flux-splitting scheme, and the Menter BSL turbulence model was utilized in both full-domain RANS simulations and as the unsteady RANS portion of the LES / RANS closure. Simulations were executed and compared with experiment at two equivalence ratios, PHI = 0.17 and PHI = 0.34. Results show that the PHI = 0.17 flame is hotter near the injector while the PHI = 0.34 flame is displaced further downstream in the combustor, though it is still anchored to the injector. Reactant mixing was predicted to be much better at the lower equivalence ratio. The LES / RANS model appears to predict lower overall heat release compared to RANS (at least for PHI = 0.17), and its capability to capture the direct effects of larger turbulent eddies leads to much better predictions of reactant mixing and combustion in the flame stabilization region downstream of the fuel injector. Numerical results from the LES/RANS model also show very good agreement with OH-PLIF and SPIV measurements. An un-damped long-wave oscillation of the pre-combustion shock train, which caused convergence problems in some RANS simulations, was also captured in LES / RANS simulations, which were able to accommodate its effects accurately.

Fulton, Jesse A.↗
NASA NTRSDate not supplied

Multiphase Simulations of the SLS Launch Environment

NASA’s Space Launch System (SLS), which will send astronauts back to the Moon in the next few years, is powered by four RS-25 engines and two RSRMV solid rocket boosters (SRBs). During launch the SLS propulsion system generates intense acoustics and other powerful waves, such as ignition overpressure (IOP) which, if unmitigated, have the potential to damage the vehicle and possibly cause loss of mission or crew. To protect the vehicle from these powerful waves, the SLS launch pad design includes an ignition overpressure/sound suppression (IOP/SS) system which sprays 270,000 gallons per minute of water very close to the SRB and RS-25 nozzles. The SRB and RS-25 engine plumes, and the proximity of the IOP/SS water, create a complex multiphase (gas and liquid) environment during the SLS ignition sequence. The interplay among these systems creates challenges related to water spray into/onto engine nozzles, potential debris transport, and additional transient loads due to strong plume-water interactions - all of which the SLS vehicle must be able to withstand. Prior to the Artemis I launch, the SLS multiphase liftoff environment was largely unknown due to differences from the Space Shuttle and other programs. Some data was available from tests of individual systems, but no integrated testing or analysis was available. Even post-launch analysis of Artemis I cannot provide a full understanding of the complex physics involved due to limited (or obstructed) camera views and instrumentation. Computational fluid dynamics (CFD) is being used to investigate the details of the multiphase environment which could not be measured, help comprehend the data gathered from the launch, and ultimately identify phenomena that are a concern for future flights. Project Details Engineers at NASA’s Marshall Space Flight Center (MSFC) have executed simulations using the Loci/STREAM-Volume of Fluid (VoF) multiphase CFD solver to understand this environment. Initial efforts successfully validated the CFD solver on various tests, giving confidence to simulate the SLS multiphase liftoff environment prior to the Artemis I launch. The CFD simulation of the SLS ignition sequence was conducted in three phases. First the IOP/SS water system was simulated for approximately 6 seconds to reach a quasi-steady state. Next, the RS-25 engine plumes were activated and held at full power for 1 second. Lastly, the SRB booster was activated and the simulation was carried out until just prior to vehicle motion. This simulation process mimics the conditions that exist at launch. Results and Impact The SLS ignition sequence simulation results provide deep understanding of the underlying physics occuring during launch. Observations from the simulation include reduction of water splashing into/onto the engine nozzles, change in angling of the dense water sheets, and the origin of the powerful ignition overpressure (IOP) wave. These observations directly inform the SLS program on subjects including plume-water induced side loads, debris transport, and the acoustic launch environment. Additionally, with post launch comparison of CFD observations to flight data, these tools can be applied to launch vehicles and environments other than SLS with confidence. Why HPC Matters The SLS ignition sequence CFD simulations are conducted on meshes up to hundreds of millions of cells on thousands of processors for weeks at a time. These simulations generate terabytes of data that must also be stored and archived for future use on HPC systems. Simply put, the CFD simulations would not be possible without NASA HPC resources. What’s Next Comparisons between the Artemis I flight data and the CFD simulations will be continued to both improve confidence in the CFD results and provide deeper understanding into the SLS multiphase launch environment. This will be used to provide insight for decision making for the first manned SLS flight, Artemis II. Future simulations will target new configurations of the SLS IOP/SS water required to support the more powerful variants of the SLS vehicle, such as Block 1B. Additionally, this capability provides NASA the ability to investigate launch environments for vehicles other than SLS to support other missions.

Travis Rivord↗
NASA NTRSDate not supplied

3D Material Response of the MSL Heatshield Using NuSil-Coated PICA

The Mars Science Laboratory (MSL) was protected during its atmospheric entry by an instrumented heatshield that used NASA's Phenolic Impregnated Carbon Ablator (PICA) material [1]. PICA is a lightweight carbon fiber/polymeric resin material that offers outstanding performance for protecting probes during planetary entry. Data from the Mars Entry Descent and Landing Instrument (MEDLI) suite on MSL offers unique in-flight validation data for models of material response and atmospheric entry. MEDLI recorded, among other things, time-resolved in-depth temperature data of PICA using thermocouple sensors assembled in the MEDLI Integrated Sensor Plugs (MISP) [2]. A space-grade silicone-based coating commercially known as NuSil CV-1144-0 [3] was applied to the entire MSL heatshield, including the MEDLI plugs, to mitigate the spread of dust from PICA. Modeling the thermal response of PICA-NuSil (PICA-N) system is still an open challenge. Ground testing of PICA-N models exhibited surface temperature jumps of the order of 150 K due to oxide scale formation and sub-sequent NuSil burn-off. It is therefore critical to include a validated model for the material response of the coating in engineering codes. A test campaign has been conducted at the NASA’s Langley HyMETS [4] facility to screen the response of PICA-N and gather detailed data on its behavior [5]. A first model of PICA-N thermal response has been developed using the Hy-METS experiments [6]. The objective of this work is to analyze the material response of the latest PICA-N model compared to the engineering model used to simulate the entry of MSL. The environment and material response around the MSL aeroshell during Mars atmospheric entry is simulated using a collection of tools. The Direct Simulation Monte Carlo SPARTA code [7] is used in the rarefied regime, the Data Parallel Line Relaxation (DPLR) code [8] is used in the continuum regime and radiative heating conditions are provided by the Nonequilibrium air radiation (NEQAIR) code [9] to estimate the environmental conditions. The thermal response inside the material is computed using the Porous material Analysis Toolbox based on Open-FOAM (PATO) [10,11,12]. Thermodynamic and chemistry properties are estimated using the Mutation++ library [13]. The approach implemented in PATO as a first cut PICA-N thermal response model is outlined in Figure 1. While the recession is less than the coating thickness, the Surface mass and energy balance Boundary Condition (SBC) uses the NuSil B’ tables. Once the recession removes the coating, the usual PICA B’ tables are used for the SBC. The B’ tables are computed using an equilibrium solver implemented in Mutation++, given the temperature, pressure, blowing rate, composition of the pyrolysis and environment gases, and the condensed species at the surface. Preliminary results of the 3D material response of the MSL heat-shield at the peak heating (80 sec after Entry Interface) are shown in Figure 2. Current NASA’s mission to Mars, Mars 2020, used the spare heatshield of MSL for thermal protection during entry, descent, and landing. In preparation for Mars 2020 post-flight analysis, the PATO high-fidelity material response capability was benchmarked against flight data from MEDLI. This effort represents an important milestone toward the development of validated predictive capabilities for designing thermal protection systems for planetary probes. This bench-marking is awaiting the final release of the MEDLI-2 data.

Aerospace↗
NASA NTRS2018

Tetrahedral-Mesh Simulations of Shock-Turbulence Interaction

Despite decades of development of unstructured mesh methods, direct numerical simulations (DNS) of turbulent flows are still predominantly performed on structured or unstructured hexahedral meshes with high-order finite-difference methods, weighted essentially nonoscillatory (WENO) schemes, or hybrid schemes formed by their combinations. Tetrahedral meshes offer easy mesh generation and adaptation around complex geometries and the potential of an orientation-free grid that would benefit the isotropic nature of small-scale dissipation, as well as the solution accuracy of intermediate scales. To advance the state of the art of unstructured-mesh simulation capabilities for shock/turbulence interaction, DNS using pure tetrahedral meshes are carried out with the space-time conservation element, solution element (CESE) method in this research. By its design, the CESE method is constructed based on a non-dissipative scheme and is a genuinely multidimensional numerical framework that is free from the use of an approximate Riemann-solver. The numerical framework also provides the ability to add numerical dissipation (the nondissipative scheme acts as the reference state like that of the reversible state in thermodynamics) when needed (with justification from mathematics/physics). The above-mentioned features along with the CESE method's consistent shock-capturing approach and strong enforcement of flux conservation in spacetime offers a novel method to accurately simulate turbulent flows and their interaction with shocks using tetrahedral meshes. Two canonical problems, namely, isotropic turbulence interaction with a normal shock and a Mach 2.9 turbulent boundary layer flow over a 24deg compression corner are investigated in this study. Computational results show reasonably good agreement with experimental data and results from structured-mesh, high-order simulations available in the literature. Successful validation of these canonical problems demonstrated here paves the way for future high-fidelity supersonic flow simulations involving complex-geometries.

Venkatachari, Balaji Shankar↗
NASA NTRS2016

Simulation of Sweep-Jet Flow Control, Single Jet and Full Vertical Tail

This work is a simulation technology demonstrator, of sweep jets used to suppress boundary layer separation and increase maximum achievable load coefficients. A sweep jet is a discrete Coanda jet that oscillates in the plane parallel to an aerodynamic surface. It injects mass and momentum in the approximate stream wise direction. It also generate turbulent eddies at the oscillation frequency, which are typically large relative to boundary layer turbulence, and which augmenting mixing across the boundary layer to attack flow separation. Simulations of a fluidic oscillator, the sweep jet emerging from the oscillator, and the suppression of boundary layer separation by an array of sweep jets are performed. Simulation results are compared to data from a dedicated CFD validation experiment of a single oscillator and its sweep jet, and from a study of a full-scale Boeing 757 vertical tail augmented with an array of sweep jets.2, 20 A critical step in the work is the development of realistic time-dependent sweep-jet in flow boundary conditions, derived from the results of the single-oscillator simulations, which create the sweep jets in the full-tail simulations. Simulations were performed using the Over flow CFD solver, with high-order spatial discretization and a range of turbulence modeling. Good results were obtained for all flows simulated, when suitable turbulence modeling was used.

flow control↗
NASA NTRS2016

Simulation of Sweep-Jet Flow Control, Single Jet and Full Vertical Tail

This work is a simulation technology demonstrator, of sweep jet flow control used to suppress boundary layer separation and increase the maximum achievable load coefficients. A sweep jet is a discrete Coanda jet that oscillates in the plane parallel to an aerodynamic surface. It injects mass and momentum in the approximate streamwise direction. It also generates turbulent eddies at the oscillation frequency, which are typically large relative to the scales of boundary layer turbulence, and which augment mixing across the boundary layer to attack flow separation. Simulations of a fluidic oscillator, the sweep jet emerging from a nozzle downstream of the oscillator, and an array of sweep jets which suppresses boundary layer separation are performed. Simulation results are compared to data from a dedicated validation experiment of a single oscillator and its sweep jet, and from a wind tunnel test of a full-scale Boeing 757 vertical tail augmented with an array of sweep jets. A critical step in the work is the development of realistic time-dependent sweep jet inflow boundary conditions, derived from the results of the single-oscillator simulations, which create the sweep jets in the full-tail simulations. Simulations were performed using the computational fluid dynamics (CFD) solver Overow, with high-order spatial discretization and a range of turbulence modeling. Good results were obtained for all flows simulated, when suitable turbulence modeling was used.

flow control↗
NASA NTRS1995

Computational Modeling of the Ames 11-Ft Transonic Wind Tunnel in Conjunction with IofNEWT

Technical advances in Computational Fluid Dynamics have now made it possible to simulate complex three-dimensional internal flows about models of various size placed in a Transonic Wind Tunnel. TWT wall interference effects have been a source of error in predicting flight data from actual wind tunnel measured data. An advantage of such internal CFD calculations is to directly compare numerical results with the actual tunnel data for code assessment and tunnel flow analysis. A CFD capability has recently been devised for flow analysis of the NASA/Ames 11-Ft TWT facility. The primary objectives of this work are to provide a CFD tool to study the NASA/Ames 11-Ft TWT flow characteristics, to understand the slotted wall interference effects, and to validate CFD codes. A secondary objective is to integrate the internal flowfield calculations with the Pressure Sensitive Paint data, a surface pressure distribution capability in Ames' production wind tunnels. The effort has been part of the Ames IofNEWT, Integration of Numerical and Experimental Wind Tunnels project, which is aimed at providing further analytical tools for industrial application. We used the NASA/Ames OVERFLOW code to solve the thin-layer Navier-Stokes equations. Viscosity effects near the model are captured by Baldwin-Lomax or Baldwin-Barth turbulence models. The solver was modified to model the flow behavior in the vicinity of the tunnel longitudinal slotted walls. A suitable porous type wall boundary condition was coded to account for the cross-flow through the test section. Viscous flow equations were solved in generalized coordinates with a three-factor implicit central difference scheme in conjunction with the Chimera grid procedure. The internal flow field about the model and the tunnel walls were descretized by the Chimera overset grid system. This approach allows the application of efficient grid generation codes about individual components of the configuration; separate minor grids were developed to resolve the model and overset onto a main grid which discretizes the interior of the tunnel test section. Individual grid components axe not required to have mesh boundaries joined in any special way to each other or to the main tunnel grid. Programs have been developed to rotate the model about the tunnel pivot point and rotation axis, similar to that of the tunnel turntable mechanism for adjusting the pitch of the physical model in the test section.

Djomehri, M. Jahed↗
NASA NTRS1994

A Navier-Strokes Chimera Code on the Connection Machine CM-5: Design and Performance

We have implemented a three-dimensional compressible Navier-Stokes code on the Connection Machine CM-5. The code is set up for implicit time-stepping on single or multiple structured grids. For multiple grids and geometrically complex problems, we follow the 'chimera' approach, where flow data on one zone is interpolated onto another in the region of overlap. We will describe our design philosophy and give some timing results for the current code. A parallel machine like the CM-5 is well-suited for finite-difference methods on structured grids. The regular pattern of connections of a structured mesh maps well onto the architecture of the machine. So the first design choice, finite differences on a structured mesh, is natural. We use centered differences in space, with added artificial dissipation terms. When numerically solving the Navier-Stokes equations, there are liable to be some mesh cells near a solid body that are small in at least one direction. This mesh cell geometry can impose a very severe CFL (Courant-Friedrichs-Lewy) condition on the time step for explicit time-stepping methods. Thus, though explicit time-stepping is well-suited to the architecture of the machine, we have adopted implicit time-stepping. We have further taken the approximate factorization approach. This creates the need to solve large banded linear systems and creates the first possible barrier to an efficient algorithm. To overcome this first possible barrier we have considered two options. The first is just to solve the banded linear systems with data spread over the whole machine, using whatever fast method is available. This option is adequate for solving scalar tridiagonal systems, but for scalar pentadiagonal or block tridiagonal systems it is somewhat slower than desired. The second option is to 'transpose' the flow and geometry variables as part of the time-stepping process: Start with x-lines of data in-processor. Form explicit terms in x, then transpose so y-lines of data are in-processor. Form explicit terms in y, then transpose so z-lines are in processor. Form explicit terms in z, then solve linear systems in the z-direction. Transpose to the y-direction, then solve linear systems in the y-direction. Finally transpose to the x direction and solve linear systems in the x-direction. This strategy avoids inter-processor communication when differencing and solving linear systems, but requires a large amount of communication when doing the transposes. The transpose method is more efficient than the non-transpose strategy when dealing with scalar pentadiagonal or block tridiagonal systems. For handling geometrically complex problems the chimera strategy was adopted. For multiple zone cases we compute on each zone sequentially (using the whole parallel machine), then send the chimera interpolation data to a distributed data structure (array) laid out over the whole machine. This information transfer implies an irregular communication pattern, and is the second possible barrier to an efficient algorithm. We have implemented these ideas on the CM-5 using CMF (Connection Machine Fortran), a data parallel language which combines elements of Fortran 90 and certain extensions, and which bears a strong similarity to High Performance Fortran. We make use of the Connection Machine Scientific Software Library (CMSSL) for the linear solver and array transpose operations.

Jespersen, Dennis C.↗
NASA NTRS1998

Development of Numerical Extended Hydrodynamics for Transition-Regime Non-Equilibrium Flows Encountered in Semiconductor Manufacturing Processes

Six months of funding was received for the proposed three year research program (funding for the period from March 1, 1997 to August 31, 1997). Although the official starting date for the project was March 1, 1997, no funding for the project was received until July 1997. In the funded research period, considerable progress was made on Phase I of the proposed research program. The initial research efforts concentrated on applying the 10-, 20-, and 35-moment Gaussian-based closures to a series of standard two-dimensional non-reacting single species test flow problems, such as the flat plate, couette, channel, and rearward facing step flows, and to some other two-dimensional flows having geometries similar to those encountered in chemical-vapor deposition (CVD) reactors. Eigensystem analyses for these systems for the case of two spatial dimensions was carried out and efficient formulations of approximate Riemann solvers have been formulated using these eigenstructures. Formulations to include rotational non-equilibrium effects into the moment closure models for the treatment of polyatomic gases were explored, as the original formulations of the closure models were developed strictly for gases composed of monatomic molecules. The development of a software library and computer code for solving relaxing hyperbolic systems in two spatial dimensions of the type arising from the closure models was also initiated. The software makes use of high-resolution upwind finite-volumes schemes, multi-stage point implicit time stepping, and automatic adaptive mesh refinement (AMR) to solve the governing conservation equations for the moment closures. The initial phase of the code development was completed and a numerical investigation of the solutions of the 10-moment closure model for the simple two-dimensional test cases mentioned above was initiated. Predictions of the 10-moment model were compared to available theoretical solutions and the results of direct-simulation Monte Carlo (DSMC) calculations. The first results of this study were presented at a meeting last year.

Groth, Clinton P. T.↗
NASA NTRS2017

Reconfigurable Model Execution in the OpenMDAO Framework

NASA's OpenMDAO framework facilitates constructing complex models and computing their derivatives for multidisciplinary design optimization. Decomposing a model into components that follow a prescribed interface enables OpenMDAO to assemble multidisciplinary derivatives from the component derivatives using what amounts to the adjoint method, direct method, chain rule, global sensitivity equations, or any combination thereof, using the MAUD architecture. OpenMDAO also handles the distribution of processors among the disciplines by hierarchically grouping the components, and it automates the data transfer between components that are on different processors. These features have made OpenMDAO useful for applications in aircraft design, satellite design, wind turbine design, and aircraft engine design, among others. This paper presents new algorithms for OpenMDAO that enable reconfigurable model execution. This concept refers to dynamically changing, during execution, one or more of: the variable sizes, solution algorithm, parallel load balancing, or set of variables-i.e., adding and removing components, perhaps to switch to a higher-fidelity sub-model. Any component can reconfigure at any point, even when running in parallel with other components, and the reconfiguration algorithm presented here performs the synchronized updates to all other components that are affected. A reconfigurable software framework for multidisciplinary design optimization enables new adaptive solvers, adaptive parallelization, and new applications such as gradient-based optimization with overset flow solvers and adaptive mesh refinement. Benchmarking results demonstrate the time savings for reconfiguration compared to setting up the model again from scratch, which can be significant in large-scale problems. Additionally, the new reconfigurability feature is applied to a mission profile optimization problem for commercial aircraft where both the parametrization of the mission profile and the time discretization are adaptively refined, resulting in computational savings of roughly 10% and the elimination of oscillations in the optimized altitude profile.

Hwang, John T.↗
NASA NTRS1996

Aerothermodynamic Analysis of Commercial Experiment Transporter (COMET) Reentry Capsule

An aerothermodynamic analysis of the Commercial Experiment Transporter (COMET) reentry capsule has been performed using the laminar thin-layer Navier-Stokes solver Langley Aerothermodynamic Upwind Relaxation Algorithm. Flowfield solutions were obtained at Mach numbers 1.5, 2, 5, 10, 15, 20, 25, and 27.5. Axisymmetric and 5, 10, and 20 degree angles of attack were considered across the Mach-number range, with the Mach 25 conditions taken to 90 degrees angle of attack and the Mach 27.5 cases taken to 60 degrees angle of attack. Detailed surface heat-transfer rates were computed at Mach 20 and 25, revealing that heating rates on the heat-shield shoulder ,can exceed the stagnation-point heating by 230 percent. Finite-rate chemistry solutions were performed above Mach 10, otherwise perfect gas computations were made. Drag, lift, and pitching moment coefficients are computed and details of a wake flow are presented. The effect of including the wake in the solution domain was investigated and base pressure corrections to forebody drag coefficients were numerically determined for the lower Mach numbers. Pitching moment comparisons are made with direct simulation Monte Carlo results in the more rarefied flow at the highest Mach numbers, showing agreement within two-percent. Thin-layer Navier-Stokes computations of the axial force are found to be 15 percent higher across the speed range than the empirical/Newtonian based results used during the initial trajectory analyses.

Wood, William A.↗