Le but de cette thèse est de faire des simulations prédictives à partir de modèles de turbulence de type RANS (Reynolds-Averaged Navier-Stokes). Ces simulations font l'objet d'un traitement systématique du modèle, de son incertitude et de leur propagation par le biais d'un modèle de calcul prédictif aux incertitudes quantifiées. Pour faire cela, nous utilisons le cadre robuste de la statistique Bayesienne.La première étape vers ce but a été d'obtenir une estimation de l'erreur de simulations RANS basées sur le modèle de turbulence de Launder-Sharma k-e. Nous avons recherché en particulier à estimer des incertitudes pour les coefficients du modele, pour des écoulements de parois en gradients favorable et défavorable. Dans le but d'estimer la propagation des coefficients qui reproduisent le plus précisemment ces types d'écoulements, nous avons étudié 13 configurations différentes de calibrations Bayesienne. Chaque calibration était associée à un gradient de pression spécifique gràce à un modèle statistique. Nous representont la totalite des incertitudes dans la solution avec une boite-probabilite (p-box). Cette boîte-p représente aussi bien les paramètres de variabilité de l'écoulement que les incertitudes epistemiques de chaque calibration. L'estimation d'un nouvel écoulement de couche-limite est faite pour des valeurs d'incertitudes générées par cette information sur l'incertitude elle-même. L'erreur d'incertitude qui en résulte est consistante avec les mesures expérimentales.Cependant, malgré l'accord avec les mesures, l'erreur obtenue était encore trop large. Ceci est dû au fait que la boite-p est une prédiction non pondérée. Pour améliorer cela, nous avons développé une autre approche qui repose également sur la variabilité des coefficients de fermeture du modèle, au travers de multiples scénarios d'écoulements et de multiples modèles de fermeture. La variabilité est là encore estimée par le recours à la calibration Bayesienne et confrontée aux mesures expérimentales de chaque scénario. Cependant, un scénario-modèle Bayesien moyen (BMSA) est ici utilisé pour faire correspondre les distributions a posteriori à un scénario (prédictif) non mesuré. Contrairement aux boîtes-p, cette approche est une approche pondérée faisant appel aux probabilités des modèles de turbulence, déterminée par les données de calibration. Pour tous les scénarios de prédiction considérés, la déviation standard de l'estimation stochastique est consistante avec les mesures effectuées.Les résultats de l'approche BMSA expriment des barres d'erreur raisonnables. Cependant, afin de l'appliquer à des topologies plus complexes et au-delà de la classe des écoulements de couche-limite, des techniques de modeles de substitution doivent être mises en places. La méthode de la collocation Stochastique-Simplex (SSC) est une de ces techniques et est particulièrement robuste pour la propagation de distributions d'entrée incertaines dans un code de calcul. Néanmois, son utilisation de la triangulation Delaunay peut entrainer un problème de coût prohibitif pour les cas à plus de 5 dimensions. Nous avons donc étudié des moyens pour améliorer cette faible scalabilité. En premier lieu, c'est dans ce but que nous avons en premier proposé une technique alternative d'interpolation basée sur le probleme 'Set-Covering'. Deuxièmement, nous avons intégré la méthode SSC au cadre du modèle de réduction à haute dimension (HDMR) dans le but d'éviter de considérer tous les espaces de haute dimension en même temps.Finalement, avec l'utilisation de notre technique de modelisation de substitution (surrogate modelling technique), nous avons appliqué le cadre BMSA à un écoulement transsonique autour d'un profil d'aile. Avec cet outil nous sommes maintenant capable de faire des simulations prédictives d'écoulements auparavant trop coûteux et offrant des incertitudes quantifiées selon les imperfections des différents modèles de turbulence. / The goal of this thesis is to make predictive simulations with Reynolds-Averaged Navier-Stokes (RANS) turbulence models, i.e. simulations with a systematic treatment of model and data uncertainties and their propagation through a computational model to produce predictions of quantities of interest with quantified uncertainty. To do so, we make use of the robust Bayesian statistical framework.The first step toward our goal concerned obtaining estimates for the error in RANS simulations based on the Launder-Sharma k-e turbulence closure model, for a limited class of flows. In particular we searched for estimates grounded in uncertainties in the space of model closure coefficients, for wall-bounded flows at a variety of favourable and adverse pressure gradients. In order to estimate the spread of closure coefficients which reproduces these flows accurately, we performed 13 separate Bayesian calibrations. Each calibration was at a different pressure gradient, using measured boundary-layer velocity profiles, and a statistical model containing a multiplicative model inadequacy term in the solution space. The results are 13 joint posterior distributions over coefficients and hyper-parameters. To summarize this information we compute Highest Posterior-Density (HPD) intervals, and subsequently represent the total solution uncertainty with a probability box (p-box). This p-box represents both parameter variability across flows, and epistemic uncertainty within each calibration. A prediction of a new boundary-layer flow is made with uncertainty bars generated from this uncertainty information, and the resulting error estimate is shown to be consistent with measurement data.However, although consistent with the data, the obtained error estimates were very large. This is due to the fact that a p-box constitutes a unweighted prediction. To improve upon this, we developed another approach still based on variability in model closure coefficients across multiple flow scenarios, but also across multiple closure models. The variability is again estimated using Bayesian calibration against experimental data for each scenario, but now Bayesian Model-Scenario Averaging (BMSA) is used to collate the resulting posteriors in an unmeasured (prediction) scenario. Unlike the p-boxes, this is a weighted approach involving turbulence model probabilities which are determined from the calibration data. The methodology was applied to the class of turbulent boundary-layers subject to various pressure gradients. For all considered prediction scenarios the standard-deviation of the stochastic estimate is consistent with the measurement ground truth.The BMSA approach results in reasonable error bars, which can also be decomposed into separate contributions. However, to apply it to more complex topologies outside the class of boundary-layer flows, surrogate modelling techniques must be applied. The Simplex-Stochastic Collocation (SSC) method is a robust surrogate modelling technique used to propagate uncertain input distributions through a computer code. However, its use of the Delaunay triangulation can become prohibitively expensive for problems with dimensions higher than 5. We therefore investigated means to improve upon this bad scalability. In order to do so, we first proposed an alternative interpolation stencil technique based upon the Set-Covering problem, which resulted in a significant speed up when sampling the full-dimensional stochastic space. Secondly, we integrated the SSC method into the High-Dimensional Model-Reduction framework in order to avoid sampling high-dimensional spaces all together.Finally, with the use of our efficient surrogate modelling technique, we applied the BMSA framework to the transonic flow over an airfoil. With this we are able to make predictive simulations of computationally expensive flow problems with quantified uncertainty due to various imperfections in the turbulence models.

A panel method derived from inviscid irrotational flow theory and utilizing hyperboloid panels is developed and applied to the simulation of steady fully wetted flows inside water-jet pumps and rocket engine inducers. The source and dipole influence coefficients of the hyperboloid panels are computed using Gauss quadrature. The present method solves the boundary value problem subject to a uniform inflow directly by discretizing the blade, casing/shroud and hub geometries with panels. The Green's integral equation and the influence coefficients for the water-jet/inducer problem are defined and solved by allocating constant strength sources and dipoles on the blade, hub and casing surfaces and constant strength dipoles on the shed wake sheets from the rotor/ stator blades. The rotor- stator interaction is accomplished using an iterative procedure which considers the effects between the rotor and the stator, via circumferentially averaged induced velocities. Finally, the hydrodynamic performance predictions for the water-jet pump and the inducer from the present method are validated against existing experimental data and numerical results from Reynolds Averaged Navier- Stokes (RANS) solvers. / text

Since its introduction in 1997, the use of Detached Eddy Simulation (DES) and similar hybrid turbulence techniques has become increasingly popular in the field of CFD. However, with increased use some of the limitations of the DES model have become apparent. One of these is the dependence of DES on grid construction, particularly regarding the point of transition between the Reynolds-Averaged Navier-Stokes and Large Eddy Simulation models. An additional issue that arises with unstructured grids is the definition of the grid spacing in the implementation of a DES length scale. To lay the ground work to study these effects the Spalart-Allmaras one-equation turbulence model, SA based DES hybrid turbulence model, and the Scale Adaptive Simulation hybrid turbulence model are implemented in an unstructured grid CFD code, UNCLE. The implemented SA based DES model is validated for flow over a three-dimensional circular cylinder for three different turbulent Reynolds numbers. Validation included studying the pressure, skin friction coefficient, centerline velocity distributions averaged in time and space. Tools to output the mean velocity profiles and Reynolds stresses were developed. A grid generation code was written to generate a two/three dimensional circular cylinder grid to simulate flow over the cylinder in UNCLE. The models implemented and validated, and the additional tools mentioned will be used in the future.

abstract: Reynolds-averaged Navier-Stokes (RANS) simulation is the industry standard for computing practical turbulent flows -- since large eddy simulation (LES) and direct numerical simulation (DNS) require comparatively massive computational power to simulate even relatively simple flows. RANS, like LES, requires that a user specify a “closure model” for the underlying turbulence physics. However, despite more than 60 years of research into turbulence modeling, current models remain largely unable to accurately predict key aspects of the complex turbulent flows frequently encountered in practical engineering applications. Recently a new approach, termed “autonomic closure”, has been developed for LES that avoids the need to specify any prescribed turbulence model. Autonomic closure is a fully-adaptive, self-optimizing approach to the closure problem, in which the simulation itself determines the optimal local, instantaneous relation between any unclosed term and the simulation variables via solution of a nonlinear, nonparametric system identification problem. In principle, it should be possible to extend autonomic closure from LES to RANS simulations, and this thesis is the initial exploration of such an extension. A RANS implementation of autonomic closure would have far-reaching impacts on the ability to simulate practical engineering applications that involve turbulent flows. This thesis has developed the formal connection between autonomic closure for LES and its counterpart for RANS simulations, and provides a priori results from FLUENT simulations of the turbulent flow over a backward-facing step to evaluate the performance of an initial implementation of autonomic closure for RANS. Key aspects of these results lay the groundwork on which future efforts to extend autonomic closure to RANS simulations can be based. / Dissertation/Thesis / Masters Thesis Aerospace Engineering 2017

Dam spillways and downstream areas are used to guide large flows of water during for example heavy rainfall. The large flows give way to turbulent pattern sand velocities that may damage the river banks or the dam structure. Investigation of these water patterns at certain flows are therefore done to examine at risk areas. In this study CFD simulations were performed for different flows with different boundary conditions for varying surface roughness level. Results were then compared to a previous model study from 2009. The ANSYS ecosystem was used in production of the 3D model, construction of mesh and running of simulations.The flow for the maximum discharge capacity of the sluices was simulated as well as the design flow which is the highest flow the dam is supposed to be able to withstand. In this report the flow has been modeled using RANS with the SST kω-model in a VOF transient setup. Results showed that for both the design flow and the maximum discharge capacity flow the energy conversion is functioning poorly and that a considerable backward circulation exists on the right riverside. This behavior could possibly injure the right dam structure by moving debris upwards against the stream.

<p>A study has been undertaken to investigate the flow structure in the vicinity of discontinuities in the surface of a high-speed air vehicle. The effect of gaps and steps on aerodynamic heating is of particular interest. The present    thesis presents Reynolds-averaged Navier Stokes  (RANS) calculations of this class of flow. This thesis consists of two studies: a parametric study of cavity flow at Mach 2 and a study to compare with wind tunnel experiments at Mach 6. The calculations for the parametric study used the Menter two-equation SST turbulence model at fully turbulent conditions. These are two-dimensional cavity flows that were carried out to identify the influence of cavity geometry on flow structure and heating distribution inside the cavity, and to categorize cavity flow regimes. The second study employed RANS calculations for conditions  corresponding to Mach 10 wind tunnel experiments carried out by Nestler et al. (AIAA Paper 1968-673) for Mach 6 boundary layer edge conditions. The SST model used in the parametric study was paired with the Menter oneequation transition model and the two-equation realizable κ-ϵ model in CFD++ was used for the computations. The results showed that, even with adjustment of model parameters, the Menter transition model cannot match the location of laminar to turbulent transition, but it demonstrated good agreement with the experimental data in fully turbulent conditions. The two-equation realizable κ-ϵ model, available in CFD++, was able to accurately model transition and showed favorable agreement for fully turbulent conditions as well.</p>

The ability of Reynolds Averaged Navier Stokes (RANS) models to predict the characteristics of a non-Body of Revolution (non-BOR) Ellipsoidal model is studied to establish the feasibility of utilizing RANS as a non-BOR concept design tool. Data unable to be obtained experimentally, such as streamwise and spanwise pressure gradients and yaw turn boundary layer characteristics, are also established. A range of conditions are studied including ahead, pitched up, steady 10 and 15 degree yaw turns, and unsteady 10 and 15 degree yaw turns. Simulation results show good agreement for ahead and pitched forces and moments. Straight ahead skin friction values also showed good agreement, providing even improved agreement over an LES model which utilized wall functions. Yaw turn conditions also showed good agreement for roll angles up to 10 degrees. Steady maneuvering forces and moments showed good agreement up to 10 degrees roll and separation calculations also showed good agreement up to 10 degrees roll. Unsteady maneuvering characteristics showed mixed results, with the normal force and pitching moment trends generally agreeing with experimental data, whereas the unsteady rolling moment did not tend to follow experimental trends. Two primary conditions, the change in curvature between the mid-body and elliptical ends and the accuracy of modeling of 3D flows with RANS, are discussed as sources of discrepancies between the experimental data and steady simulations greater than 10 degrees roll and unsteady rolling simulations. / Master of Science

<P> The work presented m this thesis focused on using the hybrid Unsteady Reynolds-Averaged Navier-Stokes (URANS)/Large Eddy Simulation (LES) methodology to investigate the flow pulsation phenomenon in compound rectangular channels for isothermal flows. The specific form of the hybrid URANS/LES approach that was used is the Strelets (2001) version of the Detached Eddy Simulation (DES). It is of fundamental interest to study the problem of flow pulsations, as it is one of the most important mechanisms that directly affect the heat transfer occurring in sub-channel geometries such as those in nuclear fuel bundles. The predictions associated with the heat transfer and fluid flow in sub-channel geometry can be used to develop simplified physical models for sub-channel mixing for use in broader safety analysis codes. The primary goal of the current research work was to determine the applicability of the DES approach to predict the flow pulsations in sub-channel geometries. It was of interest to see how accurately the dynamics associated with the flow pulsations can be resolved from a spatial-temporal perspective using the specific DES model. The research work carried out for this thesis was divided into two stages. </p> <p> In the first stage of the research work, effort was concentrated to primarily understand the field of sub-channel flow pulsations and its implications from both an experimental and numerical point of view. It was noted that unsteady turbulence modeling approaches have great potential in providing insights into the fundamentals of sub-channel flow pulsations. It was proposed that for this thesis work, the Shear Stress Transport (SST) based DES model be used to understand the dynamics associated with sub-channel flow pulsations. To the author's knowledge the DES-SST based turbulence model has never been used for resolving the effects of sub-channel flow pulsations. Next, the hybrid URANS/LES turbulence modeling technique was reviewed in great detail to understand the philosophy of the hybrid URANS/LES technique and its ability to resolve fundamental flows of interest. Effort was directed to understand the switching mechanism (which blends the URANS region with the LES region) in the DES-SST model for fully wall bounded turbulent flows without boundary layer separation. To the author's knowledge, the DES-SST model has never been used on a fully wall bounded turbulent flow problem without boundary layer separation. Thus, the DES-SST model was first completely validated for a fully developed turbulent channel flow problem without boundary layer separation. </p> <p> In the second stage of the research work, the DES-SST model was used to study the flow pulsation phenomena on two rectangular sub-channels connected by a gap, on which extensive experiments were conducted by Meyer and Rehme (1994). It was found that the DES-SST model was successful in resolving significant portion of the flow field in the vicinity of the gap region. The span-wise velocity contours, velocity vector plots, and time traces of the velocity components showed the expected cross flow mixing between the sub-channels through the gap. The predicted turbulent kinetic energy showed two clear peaks at the edges of the gap. The dynamics of the flow pulsations were quantitatively described through temporal auto-correlations, spatial cross-correlations and power spectral functions. The numerical predictions were in general agreement with the experiments in terms of the quantitative aspects. From an instantaneous time scale point of view, the DES-SST model was able to identify different flow mixing patterns. The pulsating flow is basically an effect of the variation of the pressure field which is a response to the instability causing the fluid flow pulsations. Coherent structures were identified in the flow field to be comprised of eddies, shear zones and streams. Eddy structures with high vorticity and low pressure cores were found to exist near the vicinity of the gap edge region. A three dimensional vorticity field was identified and found to exist near the gap edge region. The instability mechanism and the probable cause behind the quasi-periodic fluid flow pulsations was identified and related to the inflectional stream-wise velocity profile. Simulations were also performed with two different channel lengths in comparison to the reference channel length. Different channel length studies showed similar statistical description of the flow field. However, frequency independent results were not obtained. In general, simulations performed using the DES-SST model were successful in capturing the effects of the fluid flow pulsations. This modeling technique has great potential to be used for actual rod bundle configurations. </p> / Thesis / Doctor of Philosophy (PhD)

The thesis deals with the flow in pipe bends and radial turbines geometries that are commonly found in an Internal Combustion Engine (ICE). The development phase of internal combustion engines relies more and more on simulations as an important complement to experiments. This is partly because of the reduction in development cost and the shortening of the development time. This is one of the reasons for the need of more accurate and predictive simulations. By using more complex computational methods the accuracy and predictive capabilities are increased. The disadvantage of using more sophisticated tools is that the computational time is increasing, making such tools less attractive for standard design purposes. Hence, one of the goals of the work has been to contribute to assess and improve the predictive capability of the simpler methods used by the industry. By comparing results from experiments, Reynolds Averaged Navier-Stokes (RANS) computations, and Large Eddy Simulations (LES) the accuracy of the different computational methods can be established. The advantages of using LES over RANS for the flows under consideration stems from the unsteadiness of the flow in the engine manifold. When such unsteadiness overlaps the natural turbulence the model lacks a rational foundation. The thesis considers the effect of the cyclic flow on the chosen numerical models. The LES calculations have proven to be able to predict the mean field and the fluctuations very well when compared to the experimental data. Also the effects of pulsatile exhaust flow on the performance of the turbine of a turbocharging system is assessed. Both steady and pulsating inlet conditions are considered for the turbine case, where the latter is a more realistic representation of the real flow situation inside the exhaust manifold and turbine. The results have been analysed using different methods: single point Fast Fourier Transforms (FFT), probe line means and statistics, area and volume based Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD). / Denna avhandling behandlar flödet i rörkrökar och radiella turbiner som vanligtvis återfinns i en förbränningsmotor. Utvecklingsfasen av förbränningsmotorer bygger mer och mer på att simuleringar är ett viktigt komplement till experiment. Detta beror delvis på minskade utvecklingskostnader men även på kortare utevklningstider. Detta är en av anledningarna till att man behöver mer exakta och prediktiva simuleringsmetoder. Genom att använda mer komplexa beräkningsmetoder så kan både nogrannheten och prediktiviteten öka. Nackdelen med att använda mer sofistikerade metoder är att beräkningstiden ökar, vilket medför att sådana verktyg är mindre attraktiva för standardiserade design ändamål. Härav, ett av målen med projektet har varit att bidra med att bedöma och förbättra de enklare metodernas prediktionsförmåga som används utav industrin. Genom att jämföra resultat från experiment, Reynolds Averaged Navier-Stokes (RANS) och Large Eddy Simulations (LES) så kan nogrannheten hos de olika simuleringsmetoderna fastställas. Fördelarna med att använda LES istället för RANS när det gäller de undersökta flödena kommer ifrån det instationära flödet i grenröret. När denna instationäritet överlappar den naturligt förekommande turbulensen så saknar modellen en rationell grund. Denna avhandling behandlar effekten av de cykliska flöderna på de valda numeriska modellerna. LES beräkningarna har bevisats kunna förutsäga medelfältet och fluktuationerna väldigt väl när man jämför med experimentell data. Effekterna som den pulserande avgasströmning har på turboladdarens turbin prestanda har också kunnat fastställas. Både konstant och pulserande inlopps randvillkor har används för turbinfallet, där det senare är ett mer realistiskt representation av den riktiga strömningsbilden innuti avgasgrenröret och turbinen. Resultaten har analyserats på flera olika sätt: snabba Fourier transformer (FFT) i enskilda punkter, medelvärden och statistik på problinjer, area och volumsbaserade metoder så som Proper Orthogonal Decomposition (POD) samt Dynamic Mode Decomposition (DMD). / <p>QC 20140919</p>

La demande croissante pour alléger les moteurs d’avions et diminuer les émissions polluantes de la propulsion aéronautique réclame à rendre plus compact le système de compression des moteurs, qui représente environ 40%-50% de la masse totale. Or, à taux de compression global égal, la réduction du nombre d’étage exige d’un point de vue aérodynamique une augmentation de la charge des aubes de compresseur par étage. La charge d’aube est aujourd’hui limitée car elle induit différents mécanismes de pertes tridimensionnelles très pénalisant. L’un des plus importants est le décollement de coin qui se forme à la jonction entre l’extrados de l’aube et le moyeu ou le carter. Bien que des travaux existent sur les mécanismes et paramètres intervenant dans le décollement de coin, il est encore difficile de proposer une méthode de contrôle efficace. Cela est principalement dû à deux raisons : (i) le manque de compréhension fine des mécanismes physiques, (ii) l’utilisation pour la conception de modèles de turbulence classiques de type RANS (Reynolds-averaged Navier-Stokes) qui ne sont pas capables de prédire précisément le décollement de coin, car ils ne peuvent pas décrire correctement les mécanismes de transport turbulent. Des simulations de type RANS et LES (large-eddy simulation = simulation des grandes échelles) sont présentées dans cette thèse sur une configuration de grille d’aubes de compresseur, et comparées avec les données expérimentales obtenues au LMFA (issues de travaux séparés). L’approche RANS surestime globalement le décollement de coin. Une amélioration significative est obtenue par la méthode LES, en particulier pour le coefficient de pression statique sur l’aube et les pertes de pression totale. Ces résultats montrent que la zone de décollement de coin, qui est la source principale des pertes, génère des tourbillons de grande échelle associés à de forts niveaux d’énergie. Les histogrammes bimodaux de la vitesse tangentielle qui ont été observés expérimentalement semblent confirmés par les résultats LES. En ce qui concerne les amplitudes des fluctuations de vitesse tangentielle, les résultats expérimentaux et ceux de la LES mettent en évidence deux pics sur certains profils perpendiculaires aux parois. Enfin, grâce à l’approche LES, les bilans de l’énergie cinétique turbulente sont calculés et analysés. Ils décrivent l’équilibre entre les termes de production, de dissipation et de transport. Une des perspectives de cette analyse est d’aider à améliorer la modélisation de la turbulence en approche RANS. / The increasing demand to reduce the mass of aircraft jet engines and emissions of aircraft propulsion requires to make the compression system of engines more compact, since this component accounts for about 40%-50% of the total mass. However, at a given overall pressure ratio, decreasing the number of stages will raise the compressor blade loading per stage. The blade loading is extremely restricted by different three-dimensional flow loss mechanisms. One of them is the corner separation that forms between the blade suction side and the hub or shroud. Although some works previously investigated the mechanisms and the parameters of corner separation, it is still difficult to propose an effective control method of the corner separation. That is mainly due to two reasons: (i) the lack of knowledge of the physical mechanisms, (ii) the nowadays classical RANS (Reynolds-averaged Navier-Stokes) turbulence models are not capable to accurately predict the corner separation, since they cannot correctly describe the turbulent transport mechanisms. RANS (Reynolds-averaged Navier-Stokes) and LES (large-eddy simulation) simulations are here presented on a compressor cascade configuration, in comparison with experimental data obtained at LMFA (from separate works). The RANS approach globally over-estimates the corner separation, whereas a significant improvement is achieved with the LES, especially for the blade surface static pressure coefficient and the total pressure losses. The corner separation region, which is the main source of the total pressure losses, is shown to generate large-scale energy-containing eddies. The bimodal histograms of the streamwise velocity that were observed experimentally seem to be confirmed by the LES results. Concerning the streamwise velocity fluctuations (RMS), both the experiment and the LES show some profiles with two peaks. Finally, thanks to the LES approach, the turbulent kinetic energy budget, which represents the balance between the production, dissipation and transport terms, are computed and analyzed. This may help the improvement of RANS turbulence modeling.