A parallel implicit adaptive-mesh-refinement scheme is proposed for the solution of the Navier-Stokes equations as applied to two-dimensional steady-state hypersonic laminar flows in conjunction with an equilibrium high-temperature equation of state. A finite-volume discretization is applied to the governing equations. Limited piecewise-linear solution reconstruction and Riemann solvers (Roe and HLLE, both modified for a general equation of state) are used to evaluate the inviscid fluxes. The gradients in the viscous fluxes are calculated using diamond-path reconstruction. The system of non-linear algebraic equations resulting from the finite-volume discretization are solved using an inexact Newton method with GMRES to solve the update step of the Newton method. GMRES is preconditioned with Schwarz preconditioning with local block-fill incomplete lower-upper factorization. Multigrid and pseudo-transient continuation are used for startup. Numerical results, including flows at Mach numbers of 7.0, are discussed and demonstrate the validity and efficiency of the scheme.

A parallel implicit adaptive-mesh-refinement scheme is proposed for the solution of the Navier-Stokes equations as applied to two-dimensional steady-state hypersonic laminar flows in conjunction with an equilibrium high-temperature equation of state. A finite-volume discretization is applied to the governing equations. Limited piecewise-linear solution reconstruction and Riemann solvers (Roe and HLLE, both modified for a general equation of state) are used to evaluate the inviscid fluxes. The gradients in the viscous fluxes are calculated using diamond-path reconstruction. The system of non-linear algebraic equations resulting from the finite-volume discretization are solved using an inexact Newton method with GMRES to solve the update step of the Newton method. GMRES is preconditioned with Schwarz preconditioning with local block-fill incomplete lower-upper factorization. Multigrid and pseudo-transient continuation are used for startup. Numerical results, including flows at Mach numbers of 7.0, are discussed and demonstrate the validity and efficiency of the scheme.

No description available.

Aerothermodynamics is an integral component in the design and implementation of hypersonic transport systems. Accurate estimates of the aerodynamic forces and heat transfer rates are critical in trajectory analysis and for payload weight considerations. The present work seeks to investigate the ability of an unstructured Cartesian grid framework in modeling hypersonic viscous flows. The effectiveness of modeling viscous phenomena in hypersonic flows using the immersed boundary ghost cell methodology of this solver is analyzed. The capacity of this framework to predict the surface physics in a hypersonic non-reacting environment is investigated. High velocity argon gas flows past a 2-D cylinder are simulated for a set of freestream conditions (Reynolds numbers), and impact of the grid cell sizes on the quality of the solution is evaluated. Additionally, the formulation is verified over a series of hypersonic Mach numbers for the flow past a hemisphere, and compared to experimental results and empirical estimates. Next, a test case that involves flow separation and the interaction between a hypersonic shock wave and a boundary layer, and a separation bubble is investigated using various adaptive mesh refinement strategies. The immersed boundary ghost cell approach is tested with two temperature clipping strategies, and their impact on the overall solution accuracy and smoothness of the surface property predictions are compared. Finally, species diffusion terms in the conservation equations, and collision cross-section based transport coefficients are installed, and hypersonic flows in thermochemical nonequilibrium environments are studied, and comparisons of the off-surface flow properties and the surface physics predictions are evaluated. First, a 2-D cylinder in a hypersonic reacting air flow is tested with an adiabatic wall boundary condition. Next, the same geometry is tested to evaluate the viscous chemistry prediction capability of the solver with an isothermal wall boundary condition, and to identify the strengths and weaknesses of the immersed boundary ghost cell methodology in computing convective heating rates in such an environment.

The thermo-chemical relaxation of nitrogen hypersonic flows behind strong shocks and in nozzle expansions is investigated with 1D flow simulations and detailed vibrational kinetics. This work aims at deriving from detailed vibrational models accurate reduced models easy to implement in multidimensional reentry flow codes. First, nonequilibrium couplings between vibrational excitation, dissociation and recombination reactions are considered. Vibrational kinetics is described using accurate vibrational state-to-state rate constant databases of the literature completed with the forced harmonic oscillator model. The key role of multiquanta vibration-translation processes on the relaxation of the vibrational distribution function and the dissociation/recombination processes is put forward behind shocks and in nozzles. The vibrational distributions, which deviate strongly from equilibrium for nozzle expansions, are driven by vibration-translation processes and dissociation/recombination processes. A macroscopic model using groups of vibrational levels is developed to derive consistently the chemical and vibrational energy source terms from the vibrational state-to-state database.This model successfully reproduces the thermal, chemical and vibrational distribution function dynamics predicted by the state-to-state model with one group of levels behind a shock wave, and with three groups of levels in nozzle expansions. In a second step, the detailed vibrational model is extended to ionized nitrogen flows, including in particular a detailed modeling of the resonant electronvibration processes. Behind shocks, these processes control the rate of ionization by feeding energy to the electrons, up until the time when the elastic electron-ion exchanges takes over. It is shown that the widely used assumption of equilibrium between the electron and vibration temperatures predicts a too fast relaxation behind shock waves. In nozzle expansions, it is shown that for low electron concentration, the electron temperature is driven by electronvibration processes. Moreover, it is found that electrons are strongly coupled to low vibrational levels, and that more levels are coupled when the electron temperature increases. Coupling of the flow field with radiation is performed using the tangent slab approximation, and it is shown that the population of a metastable and two higher electronic levels are strongly impacted. Finally, the macroscopic model is extended to ionized nitrogen flows and is successfully applied on shock waves with one group of levels and with three groups of levels in nozzle expansions. In particular, the proposed macroscopic model represents more accurately the electron-vibration coupling than the widely used Landau-Teller model.

Hypersonic vehicles generate shocks that can heat the air sufficiently to partially ionize the air and create an electrically conducting plasma that can be studied using the equations of single fluid magnetohydrodynamics (MHD). Introducing strong applied magnetic and electric fields into the flow could have beneficial effects such as reducing heat damage, providing a sort of MHD parachute, and generating electric power or thrust in the vehicle. The Low Diffusion E-CUSP (LDE) scheme with a fifth order WENO scheme has recently been developed by Zha et al. [1, 2]. The purpose of this work is to incorporate the low magnetic Reynolds number MHD model and the thermodynamics of high temperature air to the above CFD algorithm so that it can be used to simulate hypersonic flows with MHD effects. In this work we compare results treating air as chemically frozen, neglecting all high temperature real gas effects with results obtained treating the air as a real gas in thermodynamic equilibrium, whose thermodynamic properties are changed by the high temperature. The hypersonic flows at high altitudes considered in this study have low Reynolds numbers. The Reynolds numbers range from about 2000 to 5000 for Mach 6 flows and reach up to 1200000 for Mach 15 flows. Thus, the flows are treated as laminar for the former cases and as turbulent for the latter using the Baldwin-Lomax turbulence model.

La relaxation thermo-chimique d’écoulements hypersoniques d’azote derrière des chocs forts et pour des détentes en tuyères est étudiée en effectuant des simulations d’écoulements 1D basées sur une cinétique vibrationnelle détaillée. Ces modèles vibrationnels détaillés sont utilisés pour développer des modèles macroscopiques précis et peu coûteux en temps de calcul pour les codes multidimensionels d’écoulements de rentrée. On considère d’abord les couplages hors équilibre entre l’excitation vibrationnelle et les réactions de dissociation / recombinaison. La cinétique vibrationnelle est décrite en utilisant des bases de constantes de réaction vibrationnelles précises de la littérature, complétées par le modèle de l’oscillateur harmoniques forcé. Le rôle prépondérant des processus vibration-translation multiquanta sur la relaxation de la distribution vibrationnelle et les processus de dissociation / recombinaison est mis en évidence derrière les chocs et dans les tuyères. Les distributions vibrationnelles, qui dévient fortement de l’équilibre dans les détentes en tuyères, résultent des processus vibration-translation et de dissociation / recombinaison. Un modèle macroscopique utilisant des groupes de niveaux vibrationnels est développé pour calculer de manière consistante les termes sources de chimie et d’énergie vibrationnelle à partir de la base de constantes de réaction vibrationnelles. Ce modèle reproduit précisément les dynamiques des températures, de la chimie et des distributions vibrationnelles avec un groupe de niveaux derrière un choc et trois groupes de niveaux pour les détentes. Dans un second temps, le modèle détaillé est généralisé aux écoulements d’azote ionisé en adoptant en particulier un modèle détaillé des processus résonants électron-vibration. Derrière les chocs, ces processus contrôlent la dynamique d’ionisation en alimentant les électrons en énergie, jusqu’à ce que les échanges élastiques électron-ion prennent le relais. Il est montré que l’hypothèse couramment utilisée d’équilibre entre les températures des électrons et de vibration conduit à une relaxation trop rapide derrière les chocs. Dans les détentes en tuyère pour lesquelles la concentration en électrons est faible, la température des électrons est contrôlée par les processus électron-vibration. On observe que les électrons sont fortement couplés aux bas niveaux vibrationnels, et que le nombre de niveaux couplés augmente avec la température des électrons. Le couplage de l’écoulement avec le rayonnement, modélisé dans l’approximation des plans tangents, impacte fortement la population du second métastable et de deux états électroniques plus élevés de N. Finalement, le modèle macroscopique est généralisé à l’azote ionisé. Un bon accord avec le modèle détaillé est obtenu avec un groupe de niveaux derrière un choc et trois groupes de niveaux pour les détentes en tuyère. En particulier, le modèle macroscopique proposé décrit plus précisément les échanges électron-vibration que le modèle de Landau-Teller couramment utilisé. / The thermo-chemical relaxation of nitrogen hypersonic flows behind strong shocks and in nozzle expansions is investigated with 1D flow simulations and detailed vibrational kinetics. This work aims at deriving from detailed vibrational models accurate reduced models easy to implement in multidimensional reentry flow codes. First, nonequilibrium couplings between vibrational excitation, dissociation and recombination reactions are considered. Vibrational kinetics is described using accurate vibrational state-to-state rate constant databases of the literature completed with the forced harmonic oscillator model. The key role of multiquanta vibration-translation processes on the relaxation of the vibrational distribution function and the dissociation/recombination processes is put forward behind shocks and in nozzles. The vibrational distributions, which deviate strongly from equilibrium for nozzle expansions, are driven by vibration-translation processes and dissociation/recombination processes. A macroscopic model using groups of vibrational levels is developed to derive consistently the chemical and vibrational energy source terms from the vibrational state-to-state database.This model successfully reproduces the thermal, chemical and vibrational distribution function dynamics predicted by the state-to-state model with one group of levels behind a shock wave, and with three groups of levels in nozzle expansions. In a second step, the detailed vibrational model is extended to ionized nitrogen flows, including in particular a detailed modeling of the resonant electronvibration processes. Behind shocks, these processes control the rate of ionization by feeding energy to the electrons, up until the time when the elastic electron-ion exchanges takes over. It is shown that the widely used assumption of equilibrium between the electron and vibration temperatures predicts a too fast relaxation behind shock waves. In nozzle expansions, it is shown that for low electron concentration, the electron temperature is driven by electronvibration processes. Moreover, it is found that electrons are strongly coupled to low vibrational levels, and that more levels are coupled when the electron temperature increases. Coupling of the flow field with radiation is performed using the tangent slab approximation, and it is shown that the population of a metastable and two higher electronic levels are strongly impacted. Finally, the macroscopic model is extended to ionized nitrogen flows and is successfully applied on shock waves with one group of levels and with three groups of levels in nozzle expansions. In particular, the proposed macroscopic model represents more accurately the electron-vibration coupling than the widely used Landau-Teller model.

<div>High speed flow problems of practical interest require a solution of nonequilibrium aerothermochemistry to accurately predict important flow phenomena including surface heat transfer and stresses. As a majority of these flow problems are in the continuum regime, Computational Fluid Dynamics (CFD) is a useful tool for flow modeling. This work presents the development of a nonequilibrium add-on solver to ANSYS Fluent utilizing user-defined-functions to model salient aspects of nonequilibrium flow in air. The developed solver was verified for several benchmark nonequilibrium flow problems and compared with the available experimental data and other nonequilibrium flow simulations. <br></div><div><br></div><div>The rate of dissociation behind a strong shock in thermochemical nonequilibrium depends on the vibrational excitation of molecules. The Macheret-Fridman (MF) classical impulsive model provides analytical expressions for nonequilibrium dissociation rates. The original form of the model was limited to the dissociation of homonuclear molecules. In this work, a general form of the MF model has been derived and present macroscopic rates applicable for modeling dissociation in CFD. Additionally, some improvements to the prediction of mean energy removed in dissociation in the MF-CFD model has been proposed based on the comparisons with available QCT data. In general, the results from the MF-CFD model upon investigating numerous nonequilibrium flows are promising and the model shows a possibility of becoming the standard tool for investigating nonequilibrium flows in CFD.</div><div><br></div><div>The aerodynamic deorbit experiment (ADE) CubeSat has dragsail to accompany accelerated deorbiting of a CubeSat post-mission. A good estimation of the aerothermal load on a reentry CubeSat is paramount to ensure a predictable reentry. This study investigates the aerothermal load on an ADE CubeSat using the direct simulation Monte Carlo (DSMC) methods and Navier-Stokes-Fourier continuum based methods with slip boundary conditions. The aerothermal load on an ADE CubeSat at 90 km altitude from the DSMC and continuum methods were consistent with each other. The continuum breakdown at a higher altitude of 95 km resulted in a strong disagreement between the continuum and DSMC solutions. Overall, the continuum methods could offer a considerable computational cost saving to the DSMC methods in predicting aerothermal load on an ADE CubeSat at low altitudes.<br> </div>

No description available.

<p> Understanding boundary layer transition is fundamental to hypersonic vehicle design as the significant heating induced by the transition process informs the development of vehicle thermal protection systems. Carbon-based thermal protection systems have been shown to decrease thermal loads and delay transition by absorbing thermal energy during ablative mass transfer into the boundary layer. To better understand this process, a high-repetition rate measurement technique is needed to temporally resolve carbon species concentrations as they propagate through the boundary layer at frequencies where boundary layer instabilities occur. Carbon monoxide is a dominant product from the chemical reactions that take place during the ablation process and is the species of interest considered in this work. A proposed approach is applying carbon monoxide two-photon laser-induced fluorescence (CO TP-LIF) at 100 kHz+ during a simulated ablation experiment where CO is injected into the boundary layer of an axisymmetric slender-body cone model in the Boeing/AFOSR Mach 6 Quiet Tunnel (BAM6QT) at Purdue University. To develop this capability, a custom-built optical parametric oscillator (OPO) was designed and used in conjunction with a burst-mode laser system to achieve narrowband excitation of CO at 100 kHz. The OPO was designed, built, and characterized through measurements of cavity energy efficiency, spectral bandwidth, and beam quality. Initial demonstrations to show the OPO could successfully achieve CO LIF were conducted in a vacuum cell at 10 Hz. The feasibility of performing CO LIF at 100 kHz in the BAM6QT was then assessed on a bench-scale using a burst-mode laser, a high speed camera, and an imaging intensifier. CO number densities in a vacuum cell were related to those that would be expected within the boundary layer of a 3 degree half-angle cone in the BAM6QT, and a series of measurements were made at these representative conditions. Appreciable signal levels were attained for single dimensional focused line measurements demonstrating high potential for using this technique in the BAM6QT at 100 kHz. The potential for a two-dimensional planar measurement was also assessed with decent promise for success for planar laser sheets of small dimensions (2 mm tall or less). Additionally, an initial BAM6QT test entry was carried out to gain experience with experimental setup; lessons learned from this experience are examined and discussed. To date, CO TP-LIF has only been applied up to 1 kHz repetition rates. This work represents a 100 fold increase over the current CO LIF state of the art and the first reported measurements, bench scale or otherwise, of 100 kHz-rate CO LIF. This lays the foundation for future CO LIF experiments in the BAM6QT at kHz-MHz repetition rates. </p>