Дисертації з теми "Aerodynamic lens"

Thesis: S.B., Massachusetts Institute of Technology, Department of Earth, Atmospheric, and Planetary Sciences, 2017.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 11-12).
Atmospheric aerosols have an important role in cloud formation and, by extension, in the overall climate system. Field studies are required to refine the uncertainty associated with the net radiative effect of atmospheric aerosols. Two pre-existing cloud sampling devices, the pumped counterflow virtual impactor (PCVI) and aerodynamic lens concentrator (ADL), were modelled using computer aided design software and printed using stereolithography printing. These devices were compared against their industrial counterparts. The printed PCVI was proven to be as effective as the industrial PCVI in a smaller working range. The printed concentrator effectively concentrated particles, but at a lower concentration factor than the industrial concentrator. This study revealed potential for further refinement in design features for both devices and it served as an essential pre-study for future field campaigns that will use these 3D printed devices.
by Libby Koolik.
S.B.

Un procédé combinant la pulvérisation cathodique magnétron avec un jet divergent de nanoparticules transporté par une lentille aérodynamique est proposé pour la synthèse de films nanocomposites. Ce procédé permet l'incorporation de nanoparticules pendant la croissance du revêtement, avec un contrôle séparé du dépôt de nanoparticules et de la matrice. De plus, l’incorporation de nanoparticules génère la formation de défauts de croissance aidant au développement de la surface du revêtement. Pour ces facultés, il est proposé d’appliquer ce procédé à la synthèse des films nanocomposites photocatalytiques. Le TiO2, reconnu pour ses propriétés photocatalytiques, a été retenu comme matrice. L'objectif de cette thèse est d’étudier les potentialités de ce procédé à partir de trois perspectives principales. La première porte sur le contrôle de la concentration en nanoparticules et leur conséquence sur les propriétés de la matrice. La deuxième perspective concerne la manière d’incorporer les nanoparticules dans la matrice en jouant sur le moment et le temps d’incorporation. La troisième perspective traite de la nature des particules, cinq types différents de nanoparticules (SiO2, Au, Bi2O3, Cu2O, P25) ont été incorporés avec succès, et une comparaison de ces différents types de nanoparticules a été effectuée. Il est démontré qu’une architecturation judicieuse des revêtements peut être facilement implémentée et conduire à des résultats prometteurs
A process combining magnetron sputtering with a divergent jet of nanoparticles transported by an aerodynamic lens is proposed for the synthesis of nanocomposite films. This process enables the incorporation of nanoparticles during coating growth, with separate control of nanoparticle and matrix deposition. In addition, the incorporation of nanoparticles generates the formation of growth defects that aids coating surface development. This is why it is proposed to apply this process to the synthesis of photocatalytic nanocomposite films. TiO2, recognized for its photocatalytic properties, was chosen as the matrix. The aim of this thesis is to investigate the potential of this process from three main perspectives. The first concerns the control of nanoparticle concentration and its impact on the matrix properties. The second concerns how to incorporate the nanoparticles into the matrix, by adjusting the time sequence and time of incorporation. The third perspective deals with the nature of the particles: five different types of nanoparticles (SiO2, Au, Bi2O3, Cu2O, P25) have been successfully incorporated, and a comparison of these different types of nanoparticles has been carried out. It is demonstrated that judicious coating architecture can be easily implemented and lead to promising results

This research is situated in the field of Computational AeroAcoustics (CAA). The thesis focuses on the computation of the aerodynamic noise generated by turbulent flows around wing, fan, or propeller airfoils. The computation of the noise radiated from a device is the first step for designers to understand the acoustical characteristics and to determine the noise sources in order to modify the design toward having acoustically efficient products. As a case study, the broadband or trailing-edge noise emanating from a CD (Controlled-Diffusion) airfoil, belonging to a fan is studied. The hybrid methods of aeroacoustic are applied to simulate and predict the radiated noise. The necessary tools were researched and developed. The hybrid methods consist in two steps simulations, where the determination of the aerodynamic field is decoupled from the computation of the acoustic waves propagation to the far field, so the first part of this thesis is devoted to an aerodynamic study of the considered airfoil. In this part of the thesis, a complete aerodynamic study has been performed. Some aspects have been developed in the used in-house solver SFELES, including the implementation of a new SGS model, a new outlet boundary condition and a new transient format which is used to extract the noise sources to be exported to the acoustic solver, ACTRAN. The second part of this thesis is concerned with the aeroacoustic study where four methods have been applied, among them two are integral formulations and the two others are partial-differential equations. The first method applied is Amiet’s theory, implemented in Matlab, based on the wall-pressure spectrum extracted in a point near the trailing edge. The second method is Curle’s formulation. It is applied proposing two approaches; the first approach is the implementation of the volume and surface integrals in SFELES to be calculated simultaneously with the flow in order to avoid the storage of noise sources which requires a huge space. In the second approach, the fluctuating aerodynamic forces, already obtained during the aerodynamics simulation, are used to compute the noise considering just the surface sources. Finally, Lighthil and Möhring analogies have been applied via the acoustic solver ACTRAN using sources extracted via SFELES. Maps of the radiated noise are demonstrated for several frequencies. The refraction effects of the mean flow have been studied.
Doctorat en Sciences de l'ingénieur et technologie
info:eu-repo/semantics/nonPublished

L'aérodynamique interne est un élément fondamental pour améliorer la combustion dans les moteurs à allumage commandé. Une meilleure maitrise des écoulements internes est permise grâce aux outils de simulation CFD qui sont de plus en plus utilisés dans le processus de développement des moteurs à allumage commandé. Cette thèse avait pour objectif d’étendre l'approche hybride RANS/LES-temporelle dite HTLES, initialement dédiée pour des écoulements statistiquement stationnaires, aux écoulements moteurs avec des parois mobiles et des modes opératoires cycliques, puis de la valider dans des configurations représentatives des écoulements moteurs. Cette approche vise à modéliser les régions proches parois par approche statistique RANS et tend continûment vers la LES temporelle loin des parois si la discrétisation spatiale et temporelle est suffisamment résolue. Le formalisme temporel permet une hybridation RANS/LES consistante dans un écoulement statistiquement stationnaire, les deux méthodes se basant sur des opérateurs temporels (respectivement la moyenne temporelle et le filtrage temporel). Une première amélioration de l’approche HTLES a été proposée en ajoutant une fonction de protection qui impose le mode RANS dans la région proche paroi, indépendamment de la discrétisation locale (spatiale et temporelle). Dans les écoulements cycliques, l’approche HTLES modélise les échelles turbulentes non-résolues en se basant sur des moyennes de phase des grandeurs résolues qui sont inconnues lors de la simulation. La moyenne glissante exponentielle (EWA) a été utilisée afin d’approximer ces moyennes de phase. Une formule pour définir la largeur de la moyenne glissante a été proposée de sorte que les fluctuations turbulentes (hautes fréquences) soient filtrées des quantités résolues, tout en conservant les composantes cycliques (basses fréquences). Cette approche a été implémentée dans le code de calcul industriel CONVERGE CFD. Elle a d'abord été validée dans deux configurations stationnaires : un canal plan infini et un banc volute. A cet effet, les résultats ont été comparés aux données de référence et aux résultats RANS et LES. Dans les régions proches parois où le maillage est sous résolu pour la LES, EWA-HTLES a mieux prédit l’écoulement grâce à l'utilisation du mode RANS, permettant une meilleure prédiction des pertes de charge. La résolution des grandes échelles dans la région centrale a permis d'obtenir des prédictions aussi précises qu’une simulation LES en termes de vitesses moyennes et des fluctuations. La validation de l'EWA-HTLES a également été effectuée dans deux configurations moteurs : le tumble compressé et le moteur Darmstadt, tous deux présentant des caractéristiques aérodynamiques typiques aux moteurs à allumage commandé telles que la génération et la compression du mouvement de tumble et la variabilité cyclique. Pour chaque configuration, un nombre total de 40 cycles consécutifs simulés à l'aide de EWA-HTLES a été utilisé pour calculer les deux premiers moments statistiques. Les résultats ont été comparés aux données de la PIV, et aux résultats donnés par les simulations RANS et LES. Les résultats ont montré que le modèle développé arrive à contrôler correctement la transition RANS-LES dans des configurations complexes avec des conditions d'écoulement non stationnaires et des déformations géométriques importantes, assurant le mode RANS aux parois et la LES au centre du cylindre. La résolution des grandes échelles a permis une bonne prédiction des phénomènes instationnaires, particulièrement l'évolution des caractéristiques du mouvement de tumble et des phénomènes associés aux variabilités cycliques, tels que l'augmentation locale de vitesses fluctuantes. Les résultats de l'EWA-HTLES sont similaires à ceux prédits par la LES et meilleurs que ceux donnés par les simulations RANS. Ces résultats montrent des perspectives encourageantes pour l'application de cette méthode dans de nombreuses configurations industrielles
Internal aerodynamics is a key element for improving the combustion efficiency in Spark-Ignition (SI) engines. Within this context, CFD tools are increasingly used to investigate in-cylinder flows and to support the design of fuel-efficient engines. The present research aimed at extending and validating a non-zonal hybrid Reynolds-Averaged Navier-Stokes / Temporal Large-Eddy Simulation (HTLES) approach, initially formulated for stationary flows, to cyclic SI engine flows with moving walls. The aim was to model the near-wall regions and coarse mesh regions in RANS, while solving the turbulent scales in core regions with sufficient mesh resolution using temporal LES, in a seamless approach with no a priori user input. HTLES was retained as it proposed a consistent hybridization combining time-averaging in RANS regions with temporal filtering in TLES.A first development consisted in implementing a smooth shielding function that enforces the RANS mode in near-wall regions, regardless of the local temporal and spatial resolution. The extension of HTLES to cyclic flows was then achieved via the formulation of a method allowing approximating the phase averages of resolved flow quantities based on an Exponentially Weighted Average (EWA). A dynamic expression for the width of the weighted average was proposed, in order to ensure that the high frequency turbulent fluctuations be filtered out from the resolved quantities, while keeping the low frequency cyclic components of the flow variables. The resulting EWA-HTLES model was implemented in the commercial CONVERGE CFD code. The developed EWA-HTLES model was first applied to the simulation of two steady flow configurations: a minimal turbulent channel and a steady flow rig. Predictions were confronted with reference data, as well as with those from RANS and LES. All simulations relied on the use of standard wall laws and coarse grids at walls. Imposing the RANS mode at walls yielded EWA-HTLES predictions of pressure losses much closer to DNS and experimental findings than with LES. At the same time, it allowed yielding results in terms of mean and RMS velocities s in the core regions of the same quality than LES, and superior to RANS.Finally, EWA-HTLES was applied to the simulation of two cyclic flows representative of SI engines: the compressed tumble and the Darmstadt single-cylinder pentroof 4valve engine. For each configuration, a total number of 40 consecutive cycles were simulated. The results were confronted to PIV data, and to RANS and LES predictions obtained using the same numerical set-up. It was shown that EWA-HTLES successfully drives the RANS-to-LES transition in such complex configurations exhibiting unsteady flow features and important cyclic geometrical deformations. It switched from the RANS mode at the walls to LES in the core region of the cylinder, allowing a better prediction of unsteady phenomena including the evolution of the overall tumble characteristics and phenomena associated to cyclic variability. The EWA-HTLES results were shown to be comparable to those predicted by LES, and superior to RANS.The performed developments and obtained results open encouraging perspectives for the application of this hybrid RANS/LES method in industrial configurations involving non-stationary conditions and in particular moving boundaries

Wind energy industry thrived in the last three decades, environmental concerns and government regulations stimulate studies on wind farm location selection and wind turbine design. Full-scale experiments and high-fidelity simulations are restrictive due to the prohibitively high cost, while the model-scale experiments and low-fidelity calculations miss key flow physics of unsteady high Reynolds number flows. A hybrid RANS/LES turbulence model integrated with transition formulation is developed and tested by a surrogate model problem through joint experimental and computational fluid dynamics approaches. The model problem consists of a circular cylinder for generating coherent unsteadiness and a downstream airfoil in the cylinder wake. The cylinder flow is subcritical, with a Reynolds number of 64,000 based upon the cylinder diameter. The quantitative dynamics of vortex shedding and Reynolds stresses in the cylinder near wake were well captured, owing to the turbulence-resolving large eddy simulation method that was invoked in the wake. The power spectrum density of velocity components showed that the flow fluctuations were well-maintained in cylinder wake towards airfoil and the hybrid model switched between RANS/LES mode outside boundary layer as expected. According to the experimental and simulation results, the airfoil encountered local flow angle variations up to ±50 degrees, and the turbulent airfoil boundary layer remained attached. Inspecting the boundary layer profiles over one shedding cycle, the oscillation about mean profile resembled the Stokes layer with zero mean. Further processing the data through phase-averaging technique found phase lags along the chordwise locations and both the phase-averaged and mean profiles collapsed into the Law of Wall in the range of 0 < y+ < 50. The features of high blade loading fluctuations due to unsteadiness and transitional boundary layers are of interest in the aerodynamic studies of full-scale wind turbine blades, making the model problem a comprehensive benchmark case for future model development and validation.
Ph. D.

This thesis aims at analysing the predictive capabilities of statistical URANS and hybrid RANS-LES methods to model complex flows at high Reynolds numbers and carrying out a physical analysis of the near-region turbulence and coherent structures. This study handles configurations included in the European research programmes ATAAC (Advanced Turbulent Simulation for Aerodynamics Application Challenges) and TFAST (Transition Location Effect on Shock Wave Boundary Layer Interaction). First, the detached flow in a configuration of a tandem of cylinders, positionned behind one another, is investigated at Reynolds number 166000. A static case, corresponding to the layout of the support of a landing gear, is initially considered. The fluid-structure interaction is then studied in a dynamic case where the downstream cylinder, situated in the wake of the upstream one, is given one degree of freedom in translation in the crosswise direction. A parametric study of the structural parameters is carried out to identify the various regimes of interaction. Secondly, the physics of the transonic buffet is studied by means of time-frequency analysis and proper orthogonal decomposition (POD), in the Mach number range 0.70–0.75. The interactions between the main shock wave, the alternately detached boundary layer and the vortices developing in the wake are analysed. A stochastic forcing, based on reinjection of synthetic turbulence in the transport equations of kinetic energy and dissipation rate by using POD reconstruction, has been introduced in the so-called organised-eddy simulation (OES) approach. This method introduces an upscale turbulence modelling, acting as an eddy-blocking mechanism able to capture thin shear-layer and turbulent/non-turbulent interfaces around the body. This method highly improves the aerodynamic forces prediction and opens new ensemble-averaged approaches able to model the coherent and random processes at high Reynolds number. Finally, the shock-wave/boundary-layer interaction (SWBLI) is investigated in the case of an oblique shock wave at Mach number 1.7 in order to contribute to the so-called "laminar wing design" studies at European level. The performance of statistical URANS and hybrid RANS-LES models is analysed with comparison, with experimental results, of integral boundary-layer values (displacement and momentum thicknesses) and wall quantities (friction coefficient). The influence of a transitional boundary layer on the SWBLI is featured.

Separated flows about single and multi-element airfoils are featured in many scenarios of practical interest, including: stall of fixed wing aircraft, dynamic stall of rotorcraft blades, and stall of compressor and turbine elements within jet engines. In each case, static and/or dynamic stall can lead to losses in performance. More importantly, modeling and analysis tools for stalled flows are relatively poorly evolved and designs must completely avoid stall due to a lack of understanding. The underlying argument is that advancements are necessary to facilitate understanding of and applications involving static and dynamic stall. The state-of-the-art in modeling stall involves numerical solutions to the governing equations of fluids. These tools often either lack fidelity or are prohibitively expensive. Ever-increasing computational power will likely lead to increased application of numerical solutions. The focus of this thesis is improvements in numerical modeling of stall, the need of which arises from poorly evolved analysis tools and the spread of numerical approaches. Technical barriers have included ensuring unsteady flow field and vorticity reproduction, transition modeling, non-linear effects such as viscosity, and convergence of predictions. Contributions to static and dynamic stall analysis have been been made. A hybrid Reynolds-Averaged Navier-Stokes/Large-Eddy-Simulation turbulence technique was demonstrated to predict the unsteadiness and acoustics within a cavity with accuracy approaching Large-Eddy-Simulation. Practices to model separated flows were developed and applied to stalled airfoils. Convergence was characterized to allow computational resources to be focused only as needed. Techniques were established for estimation of integrated coefficients, onset of stall, and reattachment from unconverged data. Separation and stall onset were governed by turbulent transport, while the location of reattachment depended on the mean flow. Application of these methodologies to oscillating flapped airfoils revealed flow through the gap was dominated by the flap angle for low angles of attack. Lag between the aerodynamic response and input flap scheduling was associated with increased oscillation frequency and airfoil/flap gap size. Massively separated flow structures were also examined.

The aerodynamics of modern rotorcraft is highly complex and has proven to be an arduous challenge for computational fluid dynamics (CFD). Flow features such as massively separated boundary layers or transition to turbulence are common in engineering applications and need to be accurately captured in order to predict the vehicle performance. The recent advances in numerical methods and turbulence modeling have resolved each of these issues independent of the other. First, state-of-the-art hybrid RANS-LES turbulence closures have shown great promise in capturing the unsteady flow details and integrated performance quantities for stalled flows. Similarly, the correlation-based transition model of Langtry and Menter has been successfully applied to a wide range of applications involving attached or mildly separated flows. However, there still lacks a unified approach that can tackle massively separated flows in the transitional flow region. In this effort, the two approaches have been combined and expended to yield a methodology capable of accurately predicting the features in these highly complex unsteady turbulent flows at a reasonable computational cost. Comparisons are evaluated on several cases, including a transitional flat plate, circular cylinder in crossflow and NACA 63-415 wing. Cost and accuracy correlations with URANS and prior hybrid URANS-LES approaches with and without transition modeling indicate that this new method can capture both separation and transition more accurately and cost effectively. This new turbulence approach has been applied to the study of wings in the reverse flow regime. The flight envelope of modern helicopters has increased significantly over the last few decades, with design concepts now reaching advance ratios up to μ = 1. In these extreme conditions, the freestream velocity exceeds the rotational speed of the blades, and a large region of the retreating side of the rotor disk experiences reverse flow. For a conventional airfoil with a sharp trailing edge, the reverse flow regime is generally characterized by massive boundary layer separation and bluff body vortex shedding. This complex aerodynamic environment has been utilized to evaluate the new hybrid transitional approach. The assessment has proven the efficiency of the new hybrid model, and it has provided a transformative advancement to the modeling of dynamic stall.

L’aile rigide navale est le moyen de propulsion qui se substitue à la grande voile souple sur les catamarans de classe Coupe de l’América et Class-C. Ce gréement est similaire à une aile aéronautique, composée de deux éléments, avec le volet séparé de l’élément principal par une fente. Comparée à une voile souple, l’aile rigide permet d’améliorer les performances du bateau en naviguant à des vitesses plus grandes que celle du vent. Cependant, le décrochage brutal qui caractérise l’aile et sa sensibilité à l’instationnarité du vent rendent difficile la correcte maîtrise de l’aile pendant la navigation. La modification des forces aérodynamiques qui agissent sur l’aile, dû à l’action d’une rafale ou au dépassement de la limite du décrochage, peuvent compromettre la stabilité du catamaran avec un possible risque de chavirage. L’aile doit donc être dessinée et réglée correctement pour éviter cette possibilité de chavirage, mais il est nécessaire de connaître l’enveloppe aérodynamique
Wingsail is a propulsion system substituting the conventional main soft sail on the America’s Cup and C-class catamarans. This rig is similar to a slotted-flap aeronautical wing, made by two elements divided by a slot. With respect to soft sails, the wingsail improves the performance of the yachts allowing navigating faster than the wind in both the upwind and downwind points of sail. However, the abrupt stall characteristics of the wing and its sensitiveness to the wind unsteadiness make difficult its management during the navigation. The modification of the strength of the aerodynamic forces acting on the wingsail, due to a gust or to the achievement of the stall limit, can compromise the stability of the catamaran. Thus, the wingsail has to be designed and trimmed to avoid the possibility of a capsize but, to do this, the aerodynamic envelop of the wingsail must be known. The aim of the Ph.D. project is, hence, to characterize the flow around the wingsail investigating the influence of the geometric and trim parameters on the wing performance

O presente trabalho busca através do uso de túnel de vento e de análises computacionais via CFD (Computacional Fluid Dynamics) avaliar o comportamento aerodinâmico que determinadas modificações nas arestas vivas de uma edificação retangular propiciam. No que tange a avaliação em túnel de vento, confeccionou-se modelos rígidos com diversas tomadas de pressão distribuídas nas fachadas dos modelos com o propósito de determinar a distribuição das isolinhas médias de pressão e os coeficientes aerodinâmicos. As simulações computacionais foram feitas a partir do uso do método de Taylor-Galerkin de 2 passos em sua forma explícita. Os modelos numéricos foram discretizados segundo o Método dos Elementos Finitos (MEF) utilizando a técnica de integração reduzida e controle de modos espúrios. A turbulência foi tratada utilizando o modelo de turbulência LES (Large Eddy Simulation), um simulador sintético de turbulência e a viscosidade turbulenta segundo a forma dinâmica. Ao final, concluiu-se que as modificações nas arestas vivas de um edifício alto, inicialmente retangular, são capazes de propiciar reduções significativas nas cargas de arrasto e laterais às quais a edificação estaria sujeita sem as modificações propostas. As isolinhas de pressão determinadas mostraram que há uma grande diferença na distribuição de pressões, sendo as modificações nas arestas capazes de diminuir os coeficientes de pressão experimentados pela estrutura. E que o uso integrado de ferramentas experimentais e numéricas pode propiciar um maior conhecimento e confiabilidade nos resultados obtidos na investigação da resposta aerodinâmica de uma estrutura. Além disso, através da comparação entre resultados experimentais e numéricos, viu-se que ambos apresentaram resultados próximos, demonstrando assim, a evolução dos métodos numéricos em avaliações de problemas de interesse da Engenharia do Vento.
The present work aims to evaluate the aerodynamics behavior that certain types of corner modifications in a rectangular building produce by using wind tunnel and computational analysis by CFD. Regarding the wind tunnel tests, rigid models were built using several pressure taps on their facades in order to determine the average pressure isolines distribution and the aerodynamic coefficients of the reduced models. Computational simulations were made using the two-step Taylor-Galerkin method in its explicit form. The numerical models were discretized according to the Finite Element Method (FEM) using the reduced integration technique and hourglassing control. The turbulence was treated using the Large Eddy Simulation (LES) methodology, a synthetic turbulence simulator and the turbulent viscosity according to the dynamic approach. At the end, it was concluded that the corner modifications in a tall building, initially rectangular, are able to produce significant reductions in drag and lift loads to which the building would be subject without the proposed modifications. The determined pressure isolines showed there is a great difference in the pressure distribution, being the corner modifications able to reduce the pressure coefficients experienced by the structure. And that the integrated use of numerical and experimental tools can provide greater knowledge and reliability in the results obtained in the investigation of the aerodynamic response of a structure. In addition, through the comparison between experimental and numerical results, it was observed that both presented close results, thus demonstrating the evolution of numerical methods in evaluations of problems of Wind Engineering interest.

Dans les chambres de combustion aéronautiques, le refroidissement par micro-percage est la technique privilégiée pour protéger les parois contre les gaz chauds. L’air frais provenant du contournement traverse des milliers de perforations inclinées et for- ment des micro-jets. Ces derniers coalescent en un film qui protège les parois du tube a flamme. Avec les moyens informatiques actuels, effectuer une simulation aux grandes échelles d’un moteur réel est impossible. En effet, le nombre de micro-trous est beaucoup trop important pour permettre une résolution détaillée de chacun. Des modèles numériques sont donc nécessaires. Le modèle homogène, développé en 2008, permet de simuler des plaques multiperforees avec des maillages dont la résolution est supérieure a celle du trou. Il ne permet cependant pas de représenter la pénétration ni le mélange des jets avec les gaz chauds. Pour remédier a cela, une approche hétérogène, appelée modèle a trou épaissi, a été développée au cours de cette thèse. La précision étant toujours relative au maillage, une méthode de maillage adaptatif augmentant automatiquement la résolution dans les zones clés a été propose afin d’obtenir de meilleurs résultats pour un faible surcoût. Predire la température des parois du tube a flamme est l’objectif final des ingénieurs. A cet effet, une méthodologie appelée Adiab2colo, permettant d’évaluer la température de paroi a partir d’un calcul adiabatique non résolu, a également été développée. Ces trois techniques sont maintenant couramment utilisées par Safran Helicopter Engine pour la conception des moteurs de demain
Numerical simulation is progressively taking importance in the design of an aero- nautical engine. However, concerning the particular case of cooling devices, the high number of sub-millimetric cooling holes is an obstacle for computational sim- ulations. A classical approach goes through the modelling of the effusion cooling by homogenisation. It allows to simulate a full combustor but failsin representing the jet penetration and mixing. A new approach named thickened-hole model was developed during this thesis to overcome this issue. A work on improving the mesh resolution onkey areas thanks to an automatic adaptive method is also presented, leading to a clear breakthrough. In parallel, as the flame tube temperature is a cornerstone for the combustor durability,a low-cost approach is proposed to predict it. To meet the time-constraints of design, it is based on thermal modelling instead of a direct thermal resolution

The flow around bluff-bodies is characterized by specific flow features (e.g. flow separations, flow re-attachments and vortex shedding) which clearly differentiate it from flows over aerodynamic bodies, such as aerofoils or turbine blades. Due to the lack of analytical solutions, bluff-body aerodynamics relies on experimental measurements and, increasingly, on Computational Fluid Dynamics (CFD), which involves the solution of fluid flow equations by means of numerical methods. However, since the computational cost for Direct Numerical Simulations (DNS) is often prohibitive, the use of turbulence models is essential for the application of CFD to engineering problems. Such models can be developed in various ways. In particular, Reynolds Averaged Navier-Stokes (RANS) models are derived by Favre-averaging Navier-Stokes equations, whereas Large Eddy Simulations (LES) rely on their spatial filtering. The major drawback in the use of such models is the loss of flow details, which may however have a strong influence on the flow field development. The resulting flow can, indeed, be remarkably different from DNS and experimental measurements. This is particularly true for RANS models, especially when used in a two-dimensional (2D) framework. In fact, in cases for which the time-averaged flow is approximately two-dimensional, the use of 2D RANS models appears to be an extremely convenient and widely adopted solution. However, such simplification can lead to strong inaccuracies. The aim of this thesis is to explore and assess the accuracy of 2D RANS models for the prediction of the aerodynamic coefficients for the wind loading of elongated structures characterized by bluff sections. Results show an overall tendency to overestimate drag due to the impossibility to generate three-dimensional structures in the wake zone. Additional LES simulations confirm the obtained results.

A numerical study using Large Eddy Simulation Coherent Structure Model (LES-CSM), of the flow around a simplified Ahmed body, has been done in this work of thesis. The models used are two salient geometries from the experimental investigation performed in [1], and consist, in particular, in two notch-back body geometries. Six simulation are carried out in total, changing Reynolds number and back-light angle of the model’s rear part. The Reynolds numbers used, based on the height of the models and the free stream velocity, are Re = 10000, Re = 30000 and Re = 50000. The back-light angles of the slanted surface with respect to the horizontal roof surface, that characterizes the vehicle, are taken as B = 31.8◦ and B = 42◦ respectively. The experimental results in [1] have shown that, depending on the parameter B, asymmetric and symmetric averaged flow over the back-light and in the wake for a symmetric geometry can be observed. The aims of the present work of master thesis are principally two. The first aim is to investigate and confirm the influence of the parameter B on the presence of the asymmetry of the averaged flow, and confirm the features described in the experimental results. The second important aspect is to investigate and observe the influence of the second variable, the Reynolds number, in the developing of the asymmetric flow itself. The results have shown the presence of the mentioned asymmetry as well as an influence of the Reynolds number on it.

Les diaphragmes utilisés comme organes de perte de charge à l'intérieur des tuyauteries de centrales électriques ont été mis en cause dans la création de sifflement. Les conséquences de ces phénomènes sont des niveaux de bruit et de vibration pouvant dépasser les valeurs admissibles. L'objectif de la thèse est d'étudier le sifflement sur la base d'expérimentations et de calculs numériques afin de proposer des outils de compréhension et de prédiction. Un résultat de la thèse correspond à l'identification expérimentale et numérique des conditions d'amplification acoustique au niveau de diaphragmes, phénomène nécessaire au sifflement. Les expériences montrent que les plages de sifflement, exprimées sous la forme d'un nombre de Strouhal fonction de l'épaisseur du diaphragme et de la vitesse dans l'orifice, s'étendent de 0,2 à 0,4 et de 0,7 à 0,9 et sont indépendantes du nombre de Reynolds. Le potentiel de sifflement de diaphragmes est également caractérisé à l'aide de simulations numériques. Deux approches sont utilisées avec des calculs U-RANS incompressibles et des simulations LES compressibles. Il apparaît que la simulation numérique permet de reproduire l'effet d'amplification acoustique à l'origine du sifflement, pour des pas de discrétisation spatial au coin amont de l'orifice suffisamment petit. Un autre résultat de la thèse est la définition des paramètres contrôlant les caractéristiques du sifflement en présence de réflexions acoustiques. Une analyse de stabilité linéaire prédit l'apparition d'un sifflement et sa fréquence. L'amplitude de sifflement est maximum pour un nombre de Strouhal autour de 0,25 et augmente avec le taux de réflexion autour du diaphragme.

La complexité de l’écoulement amont sur les éoliennes peut être la cause de décollement et/ou de décrochage sur les pales d’éoliennes. Ces phénomènes sont à l’origine de variations des efforts aérodynamiques et provoquent ainsi un vieillissement accéléré des pales. Aujourd’hui il n’existe pas de capteur capable de détecter localement le décollement sur les pales d’éoliennes, utilisable sur les éoliennes en production. Le e-Penon a été développé pour pallier ce manque. L’objectif de cette thèse a été de caractériser l’impact de la présence du e-Penon sur l’aérodynamique de la pale et d’évaluer la performance du e-Penon, dans sa capacité à détecter l’apparition du décollement/décrochage et du réattachement de l’écoulement. Pour cela des mesures ont été réalisées dans deux souffleries. La soufflerie NSA du CSTB avec un profil 2D de pale de taille réaliste et un e-Penon à échelle 1 a été le siège d’essais avec des angles d’incidences fixes. La soufflerie aérodynamique du LHEEA, a été utilisée avec le même profil et un e-Penon, tous deux à échelle réduite avec des oscillations dynamiques de l’angle d’attaque. Il a été montré que seules les fluctuations locales des pressions pariétales et la signature spectrale du sillage proche du profil ont été modifiées par la présence du e-Penon, alors que les efforts aérodynamiques moyens globaux ne le sont pas significativement. En ce qui concerne la performance du e-Penon, il a été montré, avec les essais à l’échelle 1 qu’un e-Penon assez long et suffisamment souple est capable de détecter le décollement au bord de fuite et le décrochage, lorsque ce capteur est placé au bord de fuite. Enfin, la capacité de la languette d’un capteur e-Penon à detecter les instants de décrochage et de réattachement de l’écoulement a été démontrée pour le cas d’un profil à l’échelle réduite avec un angle d’incidence oscillant
Wind turbine inflow complexity can cause flow separation and stall on wind turbine blades. These phenomena are responsible of aerodynamic load fluctuations and thus a faster aging of the blades. Today there is no sensor able to detect locally the flow separation on blades of productive wind turbines. The e-Penon was developped to fill this lack. The aim of this thesis PHD was to characterise the impact of the presence of the e-Penon on blade aerodynamics and to assess the performance of the sensor about its capacity to measure flow separation, stall and reattachment. For this work, wind tunnel measurements were performed in two different wind tunnels. The NSA wind tunnel at CSTB with a 2D blade profile and a full scale e-Penon was used to perform measurments with static angles of attack. The aerodynamic wind tunnel of the LHEEA was used with the same 2D profile and a e-Penon, both at reduced scale, with angle of attack oscillations. It was shown that the presence of the e-Penon only impacts the wall pressure fluctuations around the sensor and the near wake spectral signature, while the global and mean aerodynamic forces are not significantly modified. It was also shown that an adequate flexibility and length of the e-Penon strip makes it able to detect both the flow separation at the trailing edge and the stall when positioned near the trailing edge. It was finally demonstrated that the strip of a downscaled e-Penon is able to detect stall, flow separation and reattachment instants when placed at the trailing edge of an oscillating airfoil at a reduced scale

The context of this thesis is the numerical simulation of turbulent flows at moderate Reynolds numbers and the improvement of the capabilities of an in-house 3D unsteady and incompressible flow solver called SFELES to simulate such flows.

In addition to this abstract, this thesis includes five other chapters.

The second chapter of this thesis presents the numerical methods implemented in the two CFD solvers used as part of this work, namely SFELES and PHASTA.

The third chapter concentrates on the implementation of a new library called FlexMG. This library allows the use of various types of iterative solvers preconditioned by algebraic multigrid methods, which require much less memory to solve linear systems than a direct sparse LU solver available in SFELES. Multigrid is an iterative procedure that relies on a series of increasingly coarser approximations of the original 'fine' problem. The underlying concept is the following: low wavenumber errors on fine grids become high wavenumber errors on coarser levels, which can be effectively removed by applying fixed-point methods on coarser levels.

Two families of algebraic multigrid preconditioners have been implemented in FlexMG, namely smooth aggregation-type and non-nested finite element-type. Unlike pure gridless multigrid, both of these families use the information contained in the initial fine mesh. A hierarchy of coarse meshes is also needed for the non-nested finite element-type multigrid so that our approaches can be considered as hybrid. Our aggregation-type multigrid is smoothed with either a constant or a linear least square fitting function, whereas the non-nested finite element-type multigrid is already smooth by construction. All these multigrid preconditioners are tested as stand-alone solvers or coupled with a GMRES (Generalized Minimal RESidual) method. After analyzing the accuracy of the solutions obtained with our solvers on a typical test case in fluid mechanics (unsteady flow past a circular cylinder at low Reynolds number), their performance in terms of convergence rate, computational speed and memory consumption is compared with the performance of a direct sparse LU solver as a reference. Finally, the importance of using smooth interpolation operators is also underlined in this work.

The fourth chapter is devoted to the study of subgrid scale models for the large eddy simulation (LES) of turbulent flows.

It is well known that turbulence features a cascade process by which kinetic energy is transferred from the large turbulent scales to the smaller ones. Below a certain size, the smallest structures are dissipated into heat because of the effect of the viscous term in the Navier-Stokes equations.

In the classical formulation of LES models, all the resolved scales are used to model the contribution of the unresolved scales. However, most of the energy exchanges between scales are local, which means that the energy of the unresolved scales derives mainly from the energy of the small resolved scales.

In this fourth chapter, constant-coefficient-based Smagorinsky and WALE models are considered under different formulations. This includes a classical version of both the Smagorinsky and WALE models and several scale-separation formulations, where the resolved velocity field is filtered in order to separate the small turbulent scales from the large ones. From this separation of turbulent scales, the strain rate tensor and/or the eddy viscosity of the subgrid scale model is computed from the small resolved scales only. One important advantage of these scale-separation models is that the dissipation they introduce through their subgrid scale stress tensor is better controlled compared to their classical version, where all the scales are taken into account without any filtering. More precisely, the filtering operator (based on a top hat filter in this work) allows the decomposition u' = u - ubar, where u is the resolved velocity field (large and small resolved scales), ubar is the filtered velocity field (large resolved scales) and u' is the small resolved scales field.

At last, two variational multiscale (VMS) methods are also considered.

The philosophy of the variational multiscale methods differs significantly from the philosophy of the scale-separation models. Concretely, the discrete Navier-Stokes equations have to be projected into two disjoint spaces so that a set of equations characterizes the evolution of the large resolved scales of the flow, whereas another set governs the small resolved scales.

Once the Navier-Stokes equations have been projected into these two spaces associated with the large and small scales respectively, the variational multiscale method consists in adding an eddy viscosity model to the small scales equations only, leaving the large scales equations unchanged. This projection is obvious in the case of a full spectral discretization of the Navier-Stokes equations, where the evolution of the large and small scales is governed by the equations associated with the low and high wavenumber modes respectively. This projection is more complex to achieve in the context of a finite element discretization.

For that purpose, two variational multiscale concepts are examined in this work.

The first projector is based on the construction of aggregates, whereas the second projector relies on the implementation of hierarchical linear basis functions.

In order to gain some experience in the field of LES modeling, some of the above-mentioned models were implemented first in another code called PHASTA and presented along with SFELES in the second chapter.

Finally, the relevance of our models is assessed with the large eddy simulation of a fully developed turbulent channel flow at a low Reynolds number under statistical equilibrium. In addition to the analysis of the mean eddy viscosity computed for all our LES models, comparisons in terms of shear stress, root mean square velocity fluctuation and mean velocity are performed with a fully resolved direct numerical simulation as a reference.

The fifth chapter of the thesis focuses on the numerical simulation of the 3D turbulent flow over a re-entry Apollo-type capsule at low speed with SFELES. The Reynolds number based on the heat shield is set to Re=10^4 and the angle of attack is set to 180º, that is the heat shield facing the free stream. Only the final stage of the flight is considered in this work, before the splashdown or the landing, so that the incompressibility hypothesis in SFELES is still valid.

Two LES models are considered in this chapter, namely a classical and a scale-separation version of the WALE model. Although the capsule geometry is axisymmetric, the flow field in its wake is not and induces unsteady forces and moments acting on the capsule. The characterization of the phenomena occurring in the wake of the capsule and the determination of their main frequencies are essential to ensure the static and dynamic stability during the final stage of the flight.

Visualizations by means of 3D isosurfaces and 2D slices of the Q-criterion and the vorticity field confirm the presence of a large meandering recirculation zone characterized by a low Strouhal number, that is St≈0.15.

Due to the detachment of the flow at the shoulder of the capsule, a resulting annular shear layer appears. This shear layer is then affected by some Kelvin-Helmholtz instabilities and ends up rolling up, leading to the formation of vortex rings characterized by a high frequency. This vortex shedding depends on the Reynolds number so that a Strouhal number St≈3 is detected at Re=10^4.

Finally, the analysis of the force and moment coefficients reveals the existence of a lateral force perpendicular to the streamwise direction in the case of the scale-separation WALE model, which suggests that the wake of the capsule may have some

preferential orientations during the vortex shedding. In the case of the classical version of the WALE model, no lateral force has been observed so far so that the mean flow is thought to be still axisymmetric after 100 units of non-dimensional physical time.

Finally, the last chapter of this work recalls the main conclusions drawn from the previous chapters.
Doctorat en Sciences de l'ingénieur
info:eu-repo/semantics/nonPublished

L’objectif de ce travail de thèse consiste à analyser les solutions de contrôle permettant de réduire la traînée aérodynamique et donc de diminuer la consommation d’un véhicule. Les véhicules ciblés dans cette étude sont ceux se rapprochant d’une géométrie à culot droit telles que les versions break, monospace, SUV, utilitaires, ou même les remorques de camions. Pour s’affranchir des variantes de style, ces travaux sont concentrés sur la géométrie académique du corps de Ahmed à culot droit. La vitesse de l’écoulement est de 30m/s afin de retrouver des caractéristiques d’un écoulement de sillage fortement turbulent, proche des vitesses d’un véhicule sur autoroute. Ce travail à dominante numérique se décompose en deux parties : la première a pour objectif de valider les résultats de calculs avec et sans solution de contrôle avec des mesures expérimentales identiques, la seconde d’explorer numériquement des configurations de contrôle mixant des solutions de jets périodiques et de déflecteurs agissant sur le sillage du corps de Ahmed à culot droit. Les solutions les plus efficaces apportent des réductions de la traînée de l’ordre de 10%
This present work is focused on the analysis of control solutions that reduce the aerodynamic drag and therefore the fuel consumption of vehicles. The selected vehicle geometries are closed to a bluff body such as Estate, van, SUV, commercial vehicles or even truck trailers. This work is then focused on the academic geometry of Ahmed body with square back in order to avoid style diversity. The reference velocity flow is equal to 30m/s, which is closed to a vehicle speed on a highway, and induces a highly turbulent wake flow. This work mainly numerical is divided in two parts. The first one is dedicated to the validation of the numerical model with experimental wind tunnel measurements. The second part looks for numerical configurations of flow control solution, mixing periodic jet and deflector both acting on the wake. Most effective solutions lead to drag reduction of about 10%

Sub-grid scale (SGS) models are required in order to model the influence of the unresolved small scales on the resolved scales in large-eddy simulations (LES), the flow at the smallest scales of turbulence. In the following work two SGS models are presented and deeply analyzed in terms of accuracy through several LESs with different spatial resolutions, i.e. grid spacings. The first part of this thesis focuses on the basic theory of turbulence, the governing equations of fluid dynamics and their adaptation to LES. Furthermore, two important SGS models are presented: one is the Dynamic eddy-viscosity model (DEVM), developed by \cite{germano1991dynamic}, while the other is the Explicit Algebraic SGS model (EASSM), by \cite{marstorp2009explicit}. In addition, some details about the implementation of the EASSM in a Pseudo-Spectral Navier-Stokes code \cite{chevalier2007simson} are presented. The performance of the two aforementioned models will be investigated in the following chapters, by means of LES of a channel flow, with friction Reynolds numbers $Re_\tau=590$ up to $Re_\tau=5200$, with relatively coarse resolutions. Data from each simulation will be compared to baseline DNS data. Results have shown that, in contrast to the DEVM, the EASSM has promising potentials for flow predictions at high friction Reynolds numbers: the higher the friction Reynolds number is the better the EASSM will behave and the worse the performances of the DEVM will be. The better performance of the EASSM is contributed to the ability to capture flow anisotropy at the small scales through a correct formulation for the SGS stresses. Moreover, a considerable reduction in the required computational resources can be achieved using the EASSM compared to DEVM. Therefore, the EASSM combines accuracy and computational efficiency, implying that it has a clear potential for industrial CFD usage.

Les tourbillons hélicoïdaux générés derrière les rotors sont étudiés. Pour les générer, une méthode basée sur le couplage entre la technique de la ligne active et un solveur des équations de Navier-Stokes (ENS), incompressibles et tridimensionnelles, a été développée. Elle consiste à modéliser la pâle par son équivalent de forces volumiques. Les équations, écrites en coordonnées cylindriques, sont résolues par un schéma de différences finies, écrit en parallèle. La méthode est d'ordre deux en temps et en espace. Le solveur des ENS a été validé par la reproduction des taux de croissance d'un écoulement de jet, instable, trouvés par la théorie d'instabilité linéaire. La comparaison avec des données expérimentales a montré que la méthode prédit bien l'aérodynamique de la pâle. Ensuite, le tourbillon de bout de pâle a été, en particulier, caractérisé. La vorticité et la vitesse azimutale ont été trouvées auto-similaire et la taille du coeur suit asymptotiquement la loi de diffusion linéaire 2D. Un modèle simple du coeur du tourbillon a été proposé. La présence d'une vitesse axiale dans le coeur du tourbillon a été montrée et a été caractérisée en fonction du rapport de vitesse au bout de la pâle. Finalement, une étude de stabilité du tourbillon a été faite en utilisant une vitesse angulaire variable pour perturber l'écoulement. Les taux de croissances des modes les plus instables sont en bon accord avec celui de l'instabilité d'appariement 2D des tourbillons. Trois types de modes ont été identifiés en fonction de la fréquence des perturbations et ont été trouvés similaires aux modes décrits par la théorie et aussi trouvés, précédemment, par l'expérience
This present work is aimed to study helical vortices encountered in the wakes of rotating elements. For this, the generation of a helical wake of a one-bladed-rotor in a laminar velocity field, is simulated by the actuator line method. This method is a coupling of a Navier-Stokes (NS) solver with the Actuator Line Method where the blade is replaced by the body forces. This method has been implemented in a finite difference code, that we have written in parallel to solve the 3D incompressible NS equations written in cylindrical coordinates. The order of accuracy of the method is two both in time and space. The NS solver was validated comparing growth rates of an unstable jet, found numerically, and those of linear instability theory. A good agreement was found. A good agreement was also found comparing numerical results to analytical formulations and experimental data. It was shown that the method predicts well the blade aerodynamics . Then, the helical tip vortex is characterized for different Reynolds numbers and Tip Speed Ratios. The vorticity and the azimuthal velocity were found self-similar and the vortex core follows asymptotically the linear 2D diffusion law. A simple model for the helical vortex core was proposed. The presence of an axial velocity inside the vortex core was highlighted. Then, a stability study of the helical tip vortex was done using an angular velocity dependent on time to perturb the flow. The largest growth rates were found in good agreement with those of the (2D) pairing instability. Three types of modes were identified based on the perturbation frequency. The results are similar to those found in previous analytical and experimental works

Les mécanismes d’évolution spatio-temporelle des structures turbulentes instationnaires tridimensionnelles, et en particulier ceux rencontrés aux plus grandes échelles, sont à l’origine de phénomènes d’instabilité qui conduisent très souvent à une diminution de la performance des systèmes énergétiques. C’est le cas des variations cycle-à-cycle dans le moteur à combustion interne. Malgré les progrès substantiels réalisés par la simulation numérique en mécanique des fluides, les approches expérimentales demeurent essentielles pour l’analyse et la compréhension des phénomènes physiques ayant lieu. Dans ce travail de thèse, deux types de vélocimétrie par image de particules (PIV) ont été appliqués et adaptés au banc moteur optique du laboratoire Coria pour étudier l’écoulement en fonction de six conditions de fonctionnement du moteur. La PIV Haute Cadence 2D2C a permis d’abord d’obtenir un suivi temporel de l’écoulement dans le cylindre durant un même cycle moteur ainsi qu’identifier ces variations cycliques. La PIV Tomographique 3D3C a permis ensuite d’étendre les données mesurées vers l’espace tridimensionnel. La Tomo-PIV fait intervenir 4 caméras en position angulaire visualisant un environnement de géométrie complexe, confinée, ayant un accès optique restreint et introduisant des déformations optiques importantes. Cela a nécessité une attention particulière vis-à-vis du processus de calibration 3D des modèles de caméras. Des analyses conditionnées 2D et 3D de l’écoulement sont effectuées en se basant principalement sur la décomposition propre orthogonale (POD) permettant de séparer les différentes échelles de structure et le critère Γ permettant l’identification des centres des tourbillons
The unsteady evolution of three-dimensional large scale flow structures can often lead to a decrease in the performance of energetic systems. This is the case of cycle-to-cycle variations occurring in the internal combustion engine. Despite the substantial advancement made by numerical simulations in fluid mechanics, experimental measurements remain a requirement to validate any numerical model of a physical process. In this thesis, two types of particle image velocimetry (PIV) were applied and adapted to the optical engine test bench of the Coria laboratory in order to study the in-cylinder flow with respect to six operating conditions. First, the Time-Resolved PIV (2D2C) allowed obtaining a temporal tracking of the in-cylinder flow and identifying cyclic variabilities. Then tomographic PIV (3D3C) allowed extending the measured data to the three-dimensional domain. The Tomo-PIV setup consisted of 4 cameras in angular positioning, visualizing a confined environment with restricted optical access and important optical deformations. This required a particular attention regarding the 3D calibration process of camera models. 2D and 3D conditional analyses of the flow were performed using the proper orthogonal decomposition (POD) allowing to separate the different scales of flow structures and the Γ criterion allowing the identification of vortices centres

Cette thèse présente une étude expérimentale sur un voilier instrumenté, menée pour décrire le comportement aéro-élastique des voiles et du gréement pour des navigations au portant. Les formes des voiles utilisées sont des surfaces non développables avec de fortes courbures provoquant une séparation massive de l’écoulement. De plus, les spinnakers sont des voiles fines et souples rendant l’interaction fluide-structure fortement couplée. A cause du non-respect de certaines règles de similitude, le comportement dynamique d’un spinnaker se prête mal à l’étude en soufflerie et nécessite une comparaison avec des mesures in-situ. Les simulations numériques instationnaires modélisant le comportement aéro-élastique des voiles et du gréement doivent être qualifiées et demandent également des validations. C’est pourquoi un système d’instrumentation embarquée est mis en place sur un J/80, un voilier de huit mètres de long. Il s’agit de mesurer dynamiquement la forme en navigation du spinnaker, les efforts dans les gréements dormant et courant, la répartition de pression sur la voile ainsi que le vent et les attitudes du bateau. La forme du spinnaker en navigation est obtenue grâce à un système de mesure photogrammétrique développé pendant la thèse. La précision de ce système, meilleure que 1,5%, permet de mesurer la forme générale de la voile ainsi que les déformations importantes telles que celles liées au faseyement du guindant. L’effort aérodynamique produit par le spinnaker est obtenu grâce à la mesure de l’intensité des efforts et de leurs directions aux trois extrémités (drisse, amure, écoute) ainsi que par la mesure des pressions sur la voile. Le comportement général du spinnaker est analysé en fonction de l’angle du vent apparent. Une nouvelle représentation utilisant les surfaces de Bézier triangulaires est développée pour décrire la forme tridimensionnelle du spinnaker. Quelques points de contrôles suffisent pour représenter la voile et caractériser le type de voile. Un comportement dynamique propre au spinnaker est également étudié. Le réglage supposé optimal d’un spinnaker est à la limite du faseyement, en laissant le guindant se replier légèrement. Cependant ce réglage n’a jamais été scientifiquement étudié auparavant. Nous avons montré qu’il s’agit d’une forte interaction fluide-structure tridimensionnelle où une importante dépression apparaît au bord d’attaque, qui augmente temporairement les efforts, ce qui n’est pas observé avec un réglage plus bordé
A full-scale experimental study on an instrumented sailing yacht is conducted to better assess the aero-elastic behaviour of the sails and rigging in downwind navigations. The downwind sail shape is a non-developable surface with high curvature leading to massive flow separation. In addition, spinnakers are thin and flexible sails leading to a strongly coupled Fluid-Structure Interaction. Due to the non-respect of some rules of similitude, the unsteady behaviour of downwind sails cannot be easily investigated with wind tunnel tests that would need comparison with full-scale experiments. Moreover unsteady numerical simulations modelling the aero-elastic behaviour of the sails and rigging require validations. An inboard instrumentation system has been developed on a 8 meter J/80 sailboat to simultaneously and dynamically measure the flying shape of the spinnaker, the aerodynamic loads transmitted to the rigging, the pressure distribution on the sail as well as the boat and wind data. The shape of the spinnaker while sailing is acquired by a photogrammetric system developed during this PhD. The accuracy of this new system, better than 1.5%, is used to measure the global shape and the main dynamic deformations, such as the flapping of the luff. The aerodynamic load produced by the spinnaker is assessed by the measurements of the load magnitudes and directions on the three corners of the sail (head, tack and clew), and also by the pressure distribution on the spinnaker. The global behaviour of the spinnaker is analysed according to the apparent wind angle. A new representation using Bézier triangular surfaces defines the spinnaker 3D shape. A few control points enable to represent the sail and can easily characterise the type of sail. A typical unsteady behaviour of the spinnaker is also analysed. Letting the luff of the sail flap is known by sailors as the optimal trim but has never been scientifically studied before. It is found that it is a complex three dimensional fluid-structure interaction problem where a high suction near the leading edge occurs, producing a temporary increase of the force coefficient that would not be possible otherwise

Le but de ces travaux est de créer un modèle aérothermique, avec un temps de calcul court, d’un alternateur automobile intégré au sein d’une simulation complète d’un environnement sous-capot d’un véhicule. La prise en compte de leur influence au sein d'un comportement moteur est recherchée. Un modèle simplifié permettant la simulation du comportement aéraulique et thermique d’un alternateur a été développé. Il utilise une approche nodale afin de simuler les phénomènes thermiques et aérauliques du système. Différents algorithmes et une interface homme-machine permettent un paramétrage facile et rapide, et une implémentation automatique du modèle. En effet, le paramétrage du réseau nodal est fait automatiquement, l’utilisateur doit seulement rentrer les différents paramètres du système : dimensions, caractéristiques matériaux, pertes thermiques... Cela nous permet d’avoir aussi un modèle adaptable facilement à tout type d’alternateur. Le comportement aéraulique de l’alternateur est simulé via des coefficients de convection, intégrés au réseau nodal. Ces coefficients sont déterminés via des corrélations en fonction du nombre de Reynolds de l’écoulement. Pour chaque zone de l’écoulement d’air dans l’alternateur, ces corrélations ont été identifiées via une modélisation CFD de l’alternateur, lui-même validé par des essais aérauliques sur banc expérimental. Le modèle a été vérifié et validé via des essais expérimentaux thermiques. Il présente une erreur moyenne inférieure à 10%, et fonctionne sur l’ensemble des régimes d’utilisation. Il présente un temps de calcul de l’ordre de 2 minutes. Le modèle a été intégré dans une simulation simplifiée d’un environnement sous-capot. Une méthodologie de couplage a été développée, permettant l’intégration des données du modèle simplifié, au sein des simulations sous-capot. Ces simulations modélisent le comportement thermique des environnements sous-capot ainsi que le compartiment aéraulique. Les flux d’air sont simulés et le couplage du modèle simplifié permet d’intégrer l’influence thermique de l’alternateur, au sein de l’environnement sous-capot, ainsi que l’impact aéraulique de ce dernier. La méthodologie de couplage permet d’importer les valeurs de températures et de débits, estimées par le modèle simplifié, au sein du maillage d’une CAO de l’alternateur. Ces travaux sont en cours d’intégration dans les processus numériques du Groupe PSA. Différentes perspectives sont en cours d’étude, afin d’utiliser ce modèle sur d’autres éléments du sous-capot, ou d’autres machines tournantes, comme des moteurs électriques présents sur les véhicules hybrides et électriques
The objective of the thesis is to create a thermal model of an alternator, with a quickly time run. This model will integrate the influence of the alternator inside an under-hood simulation. A simplify model able to simulate the aerodynamic and thermal behaviour has developed. It use a nodal approach to simulate the aerodynamic and thermal behaviour. Different algorithms and an user’s interface able to a quickly set up and a automatically implementation. Indeed, the nodal, approach was realized automatically by the model, the user inform the dimensions of the alternator, the materials characteristics and the thermal losses. Thanks to we have a model that use with any automobile alternator. The aerodynamic of the alternator is simulate with convection coefficient via the nodal approach. These coefficients are estimated with correlations based on Reynolds of the flow. The CFD simulation of the alternator identified these correlations. The CFD model has been validate with an aerodynamics tests. The model is checked and validate by thermal tests. It has an average error lower than 10% and work to any regime of the use. The time run is equal to 2 minutes. The modal has been integrate inside an under-hood simulation. A coupling methodology has been developed to allow the integration of the data, like the temperatures and the flowrate was estimate by the simplify model, inside an under-hood simulation. The under-hood simulation modelling the aerodynamic and thermal behaviour of the engine compartment. Therefore, the coupling methodology allow integrating the aerodynamic and thermal influence of the alternator inside the compartment. The work is actually in progress inside the numerical processes of the PSA group. Many perspectives are studied, to use the model on other under-hood elements, or other electric machine, like the electric engines used inside the hybrid vehicles

Dans un contexte de croissance du trafic aérien et dans le but de réduire la consommation de carburant ainsi que les émissions de polluants dans l'atmosphère, l'avion de demain se doit d'être plus respectueux de l'environnement. Dans un objectif d'optimisation de ses performances aérodynamiques,d'importantes activités de recherche sont menées dans le monde pour étudier de nouveaux dispositifs de contrôle actif des écoulements en temps réel. Depuis une dizaine d'années, l'utilisation de la décharge à barrière diélectrique surfacique comme actionneur plasma pour le contrôle d'écoulements suscite un intérêt grandissant. Ce type d'actionneur permet de créer un plasma non-thermique capable de générer un écoulement basse vitesse, appelé vent ionique, qui interagit avec l'écoulement naturel en proche paroi pour l'amener dans un état souhaité. Les études expérimentales présentées dans cette thèse portent, d'une part, sur la caractérisation de l'actionneur plasma sous atmosphère contrôlée pour étudier le rôle de l'azote et de l'oxygène sur le comportement de la décharge et d'autre part, sur l'évaluation des potentialités de cet actionneur à contrôler le décollement massif naissant au bord d'attaque d'un profil d'aile placé à forte incidence. Les résultats mettent en évidence l'importance du rôle joué par O2 dans l'amorçage des filaments de plasma et dans la production de vent ionique. Le taux de production d'ozone de l'actionneur plasma a été quantifié en fonction de la puissance électrique. Les essais en soufflerie, réalisés dans le cadre du projet européen PLASMAERO, montrent l'effet de la fréquence de pulsation du signal d'alimentation haute tension sur la réponse de l'écoulement décollé et des ses instabilités naturelles. Il est ainsi possible, pour le profil placé à des incidences au-delà de l'incidence de décrochage naturel, d'augmenter la portance du profil en supprimant le décollement ou en favorisant la formation de tourbillons portants à l'extrados du profil.

Neste trabalho exploram-se alguns dos actuais recursos de computação científica no contexto da optimização estática e dinâmica. Começa-se por propor um conjunto de procedimentos computacionais algébricos que permitem automatizar todo o processo de obtenção de simetrias e leis de conservação, quer no contexto clássico do cálculo das variações, quer no contexto mais abrangente do controlo óptimo. A utilidade do package de funções desenvolvido é demonstrada com a identificação de novas leis de conservação para alguns problemas do controlo óptimo conhecidos na literatura. Estabelece-se depois uma relação entre as simetrias variacionais do controlo óptimo e as simetrias de equações diferenciais ordinárias. A partir dessa relação, deduz-se um método construtivo, alternativo aos já existentes, para obtenção de simetrias nesta segunda classe de problemas. Numa segunda parte do trabalho, investigam-se, com recurso a simulações computacionais, formas de corpos não convexos que maximizem a sua resistência aerodinâmica quando se desloquem em meios rarefeitos e, simultaneamente, exibam um ligeiro movimento rotacional. É obtido um importante resultado original para o caso bidimensional. Trata-se de uma forma geométrica que confere ao corpo uma resistência muito próxima do seu limite teórico (R=1.4965<1.5). In this thesis some of the scientific computational resources are explored in the context of static and dynamic optimization. A set of analytical computational tools is proposed in order to allow the identification, in an automatic way, of variational symmetries and conservation laws in the calculus of variations and optimal control. The usefulness of the developed routines is showed with the identification of new conservation laws to concrete optimal control problems found in the literature. A relationship between the variational symmetries of optimal control and the symmetries of ordinary differential equations is established. Based in this relationship, a constructive method is created for the purpose of getting the symmetries in this second class of problems. Finally, we investigate, by means of computational simulations, shapes of nonconvex bodies that maximize resistance to its motion on a rarefied medium, considering that bodies are moving forward and at the same time slowly rotating. An important result is obtained for the two-dimensional case which consists of a geometric shape that confers to the body a resistance very close to the supremum value (R = 1.4965 < 1.5). Some results of the thesis are available in the English language in the following references: the research reports [29, 35, 37, 79], the poster [36], the conference proceedings with referee [34] and the refereed journals [31, 32, 38, 80].

Cette thèse traite de l'étude numérique et expérimentale du transport et de la déposition d'aérosols dans les poumons. La partie numérique du travail porte sur des simulations uni-, bi- et tridimensionnelles du comportement des aérosols dans la structure pulmonaire. Les simulations unidimensionnelles (1D) sont effectuées dans des modèles trompettes et multibranche similaires à ceux utilisés dans les études de transport et de mélange gazeux dans les poumons. Le dépôt total, le profil des dépôts le long des différentes générations de l'arbre bronchique ainsi que la dispersion de boli d'aérosols sont calculés en fonction de la taille des particules et du protocole respiratoire. Un bolus consiste en un faible volume d'aérosols inhalé sous la forme d'un pic de concentration au cours d'une inspiration d'air pur. Les résultats montrent les limitations intrinsèques liées aux modèles 1D quant à la description du transport des aérosols dans les poumons et suggèrent l'utilisation d'équations multidimensionnelles pour décrire le transport de particules. Des simulations bidimensionnelles (2D) sont alors développées pour décrire le comportement des aérosols dans un modèle représentatif de la zone alvéolaire du poumon humain. Les simulations montrent que les particules ne se déposent pas uniformément sur les parois alvéolaires des conduits mais qu'elles sont principalement localisées près de l'entrée des alvéoles et ceci principalement dans le cas de petites particules (diamètre inférieure à 0.5 mm). De plus, les résultats montrent que le traditionnel coefficient de dispersion utilisé dans l'approche unidimensionnelle ne peut pas être extrapolé dans la zone alvéolaire du poumon.

Finalement, des simulations tridimensionnelles (3D) sont réalisées dans un modèle d'un conduit pulmonaire entouré d'alvéoles et confirment la déposition largement hétérogène des aérosols calculée dans l'étude bidimensionnelle suggérant que les concentrations locales et moyennes en aérosols peuvent être substantiellement différentes.

Parallèlement, des données expérimentales de déposition totale et de dispersion de boli d'aérosols sont obtenues et comparées aux résultats numériques. Des indices tels que la dispersion du bolus expiré, la déposition totale ou le déplacement du mode entre les courbes de concentration des boli inspiré et expiré mesurés au niveau de la bouche ont été évalués. Des simulations numériques similaire aux tests expérimentaux sont également effectuées. Bien qu'une approche relativement simplifiée soit utilisée, il apparaît que les simulations décrivent raisonnablement bien les résultats expérimentaux.

Doctorat en sciences appliquées
info:eu-repo/semantics/nonPublished

L’industrie automobile fournie de plus en plus d’effort pour optimiser l’aérodynamique externe des véhicules afin de réduire son empreinte écologique. Dans ce cadre, l’objectif de ce projet est d’examiner les structures tourbillonnaires responsables de la dégradation de traînée et de proposer une solution de contrôle actif permettant d’améliorer l’efficacité aérodynamique d’un véhicule SUV. Après une étude expérimentale de la maquette POSUV échelle réduite, une analyse modale croisée permet d’identifier les structures périodiques corrélées de l’écoulement qui pilotent la dépression sur le hayon. Une solution de contrôle optimale par jets pulsés sur le parechoc arrière, est obtenue avec un algorithme génétique. Celle-ci permet de réduire la dépression du hayon de 20% et l’analyse croisée des résultats instationnaires avec contrôle montre un changement significatif de la distribution spectrale. Après deux études préliminaires sur la rampe inclinée à 25° et sur le Corps d’Ahmed à 47°, la simulation de POSUV à partir d’un solveur LES, en éléments finis, est validé par rapport aux résultats expérimentaux. L’approfondissement des résultats 3D permet de comprendre les pertes aérodynamiques. La simulation de l’écoulement contrôlé permet également d’identifier les mécanismes du contrôle d’écoulements
The automotive industry dedicates a lot of effort to improve the aerodynamical performances of road vehicles in order to reduce its carbon footprint. In this context, the target of the present work is to analyze the origin of aerodynamic losses on a reduced scale generic Sport Utility Vehicle and to achieve a drag reduction using an active flow control strategy. After an experimental characterization of the flow past the POSUV, a cross-modal DMD analysis is used to identify the correlated periodical features responsible for the tailgate pressure loss. Thanks to a genetic algorithm procedure, 20% gain on the tailgate pressure is obtained with optimal pulsed blowing jets on the rear bumper. The same cross-modal methodology allows to improve our understanding of the actuation mechanism. After a preliminary study of the 25° inclined ramp and of the Ahmed Body computations, the numerical simulation of the POSUV is corroborated with experiments using the cross-modal method. Deeper investigations on the three-dimensional flow characteristics explain more accurately the wake flow behavior. Finally, the controlled flow simulations propose additional insights on the actuation mechanisms allowing to reduce the aerodynamic losses

L'aérodynamique latérale des véhicules automobiles suscite de nos jours de plus en plus d'intérêt de la part des constructeurs. L'automobiliste est en effet soumis quotidiennement à de forts courants d'air latéraux, que ce soit lors du dépassement d'un autre véhicule, ou alorsen passant dans un couloir de vent du à la topographie du terrain (passage devant un espace entre deux immeubles par exemple). Les efforts aérodynamiques mis en jeu dans ces situations peuvent provoquer des mouvements non désirés du véhicule, pouvant avoir des conséquences dramatiques si le conducteur se laisse surprendre. Des études expérimentales reproduisant les effets d'un dérapage dynamique ont mis en évidence des phénomènes transitoires importants mettant à défaut les modèles stationnaires généralement pratiqués par les constructeurs pour qualifier le comportement de leurs véhicules en présence de dérapage. Les mécanismes responsables de ces phénomènes transitoires sont encore mal connus de la communauté scientifique. Ce travail propose d'approfondir ce sujet au travers de l'étude de l'aérodynamique d'un véhicule terrestre fixe soumis à un vent longitudinal et à une rafale de vent latéral. Le but principal est d'identifier les structures tourbillonnaires au moyen de mesures PIV et de calculs numériques des champs de vitesse autour d'une maquette automobile et de les corréler aux efforts aérodynamiques. Un accord entre l'ISAT, composante de l'Université de Bourgogne, et l'Institut Supérieurde l'Aéronautique et de l'Espace (ISAE) de Toulouse a permis de mener l'étude avec les ressources de cet établissement. Le moyen d'essai principal, créé par l'ISAE, est le banc" rafale latérale ", constitué d'une soufflerie principale et d'une soufflerie secondaire, dont la sortie à volet déferlants (" Mexican Wave ") est inspirée de l'approche proposée par Ryan et Dominy (2000). L'analyse expérimentale a été effectuée à l'aide de la PIV résolue en temps et stéréoscopique, et d'une balance dard instationnaire à cinq composantes. Un banc" numérique " identique a été constitué à l'aide du logiciel FLUENT©, pour des calculs 3D. De plus, un modèle 2D annexe, basé sur la méthode " meshless ", a été développé pour de futures investigations, en raison de sa robustesse pour des problèmes à fortes discontinuités et sa bonne adaptabilité aux problèmes avec frontières mobiles.Une première phase de ce travail a consisté en la mise au point des bancs expérimental et numérique, avec génération d'un champ de dérapage homogène, de 21° dans la zone de mesure. L'évolution du dérapage en chaque point respecte bien la forme d'un créneau imposé par la rafale. Pour l'analyse des efforts, deux géométries de maquette ont été étudiées, à savoir un corps de Windsor à culot droit générant, pour un écoulement longitudinal, des structures de sillage bidimensionnelles, et son homologue à culot incliné de 25°, générant des tourbillons " cigare ". Des pics d'efforts ont été observés à l'arrivée de la rafale, tout comme la littérature le prédit. Pour ce qui est du coefficient du moment de lacet, les sursauts sont de 29 % et 19 % respectivement pour la maquette à culot droit et celle à culot incliné, par rapport aux valeurs stationnaires. Concernant le coefficient de force de dérive, ils sont de 10 % et 14 %, respectivement. Lors de nos essais, ces efforts se sont établis après 5.5 longueurs de maquette. Afin d'expliquer la différence de comportement entre les deux maquettes en termes d'efforts, l'évolution temporelle des tourbillons nommés, dans ce mémoire, ΓA, ΓB, ΓC et Γ1 à été discutée. Il en est ressorti une forte corrélation entre la circulation du tourbillon ΓA, le plusénergétique, naissant à l'avant du flanc sous le vent de la maquette, et les efforts latéraux, de sorte que ce tourbillon serait le meilleur témoin des phénomènes instationnaires mis en jeu dans l'étude de l'effet du vent latéral. [...]

La prédiction des vibrations induites par un écoulement est essentielle dans la conception des conduits de nombreuses installations industrielles, en particulier dans l’industrie du gaz. Notre étude concerne la prévision du bruit et la vibration des conduits soumis à un écoulement turbulent à faible nombre de Mach. Notre objectif est de présenter une étude numérique et expérimentale permettant aux ingénieurs de mieux comprendre le couplage entre l’excitation aléatoire et le conduit pour deux géométries (circulaire ou rectangulaire). Une approche expérimentale est développée et utilisée pour valider les prévisions numériques. Deux cas sont étudiés : (i) un conduit droit sans singularité, où les modes acoustiques du conduit sont excités par une couche limite turbulente (TBL) et (ii) un conduit droit avec un diaphragme inséré en amont qui génère une source acoustique localisée. La contribution acoustique est déterminée soit par des méthodes de mesure d’interspectres, soit à l’aide des outils de mécanique des fluides numérique (CFD) et d’analogies aéroacoustiques. La réponse de la structure est estimée par une approche dite de « couplage faible » qui utilise des fonctions de transfert modale d’un conduit fini simplement appuyé. Les mesures conduiront à évaluer et suggérer des améliorations de modèles empiriques existants de densité interspectrale de puissance (CPSD) dans un contexte d’écoulements internes turbulents. Une analyse modale expérimentale d’un conduit rectangulaire finie est confrontée à des méthodes de calcul pour évaluer l’effet des conditions aux limites, du rayonnement acoustique et de l’amortissement aérodynamique. Le couplage fluide structure est analysé par la fonction de « joint acceptance » à la fois dans le domaine spatial et dans le domaine des nombres d’onde. L’excitation comprend à la fois les contributions acoustiques et hydrodynamiques à l’aide des CPSD exprimées sur la base des fonctions de cohérence de type Corcos, champ diffus et modes acoustiques d’ordre élevé. Enfin, les études numériques et expérimentales de cette thèse ont été utilisées pour développer un cadre d’étude et de modélisation du bruit et des vibrations dans les conduites, qui relie la dynamique des fluides, les modèles analytiques et empiriques à des techniques efficaces d’analyse aléatoire
Pipeline and duct vibrations can cause a range of issues from unplanned shutdownsto decreased equipment life time. Thus, the prediction of flow-induced vibrations is essential in piping design in many industrial plants, especially, for Gas industry. This study deals with the prediction of pipe flow noise and vibration at low Mach number. We aim to present a numerical and experimental study which can offer engineers a better understanding of the coupling between random excitation and duct section for two geometries (circular or rectangular). An experimental facility and measurement approach is developed and used to validate numerical predictions. Two cases are investigated: (i) a straight duct with no singularity, duct acoustic modes are excited by the Turbulent Boundary Layer (TBL) and (ii) a straight duct with a diaphragm inserted upstream generating a localized acoustic source. The acoustic contribution is either measured via cross-spectra based methods or calculated using Computational Fluid Dynamics (CFD) and aeroacoustic analogies. The response of the structure is estimated via a ‘blocked’ approach using analytical modal Frequency Response Functions (FRFs) of a simply supported finite duct. Measurements will lead to evaluate and suggest improvements to existing Cross Power Spectral Density (CPSD) empirical models in a context of internal turbulent flows. Experimental modalanalysis of a finite rectangular duct are confronted to computational methods to assess the effect of the Boundary Conditions (BCs), the resistive damping from coupling with the internal acoustic medium and aerodynamic damping. The fluid-structure coupling is analyzed through the joint acceptance function both in the spatial and wave number domain. The excitation includes both the acoustic and hydrodynamic contributions using CPSD written on the basis of Corcos, Diffuse Acoustic Field (DAF) and acoustic duct mode coherence functions. Finally, the numerical and experimental studies in this thesis were used to develop a framework for studying and modelling pipe flow noise and vibration which links CFD, analytical and empirical models to efficient random analysis techniques

This thesis aims at improving our knowledge on swirl combustors. The work presented here is based on Large Eddy Simulations (LES) coupled to an advanced combustion model: the Conditional Moment Closure (CMC). Numerical predictions have been systematically compared and validated with detailed experimental datasets. In order to analyze further the physics underlying the large numerical datasets, Proper Orthogonal Decomposition (POD) has also been used throughout the thesis. Various aspects of the aerodynamics of swirling flames are investigated, such as precession or vortex formation caused by flow oscillations, as well as various combustion aspects such as localized extinctions and flame lift-off. All the above affect flame stabilization in different ways and are explored through focused simulations. The first study investigates isothermal air flows behind an enclosed bluff body, with the incoming flow being pulsated. These flows have strong similarities to flows found in combustors experiencing self-excited oscillations and can therefore be considered as canonical problems. At high enough forcing frequencies, double ring vortices are shed from the air pipe exit. Various harmonics of the pulsating frequency are observed in the spectra and their relation with the vortex shedding is investigated through POD. The second study explores the structure of the Delft III piloted turbulent non-premixed flame. The simple configuration allows to analyze further key combustion aspects of combustors, with further insights provided on the dynamics of localized extinctions and re-ignition, as well as the pollutants emissions. The third study presents a comprehensive analysis of the aerodynamics of swirl flows based on the TECFLAM confined non-premixed S09c configuration. A periodic component inside the air inlet pipe and around the central bluff body is observed, for both the inert and reactive flows. POD shows that these flow oscillations are due to single and double helical vortices, similar to Precessing Vortex Cores (PVC), that develop inside the air inlet pipe and whose axes rotate around the burner. The combustion process is found to affect the swirl flow aerodynamics. Finally, the fourth study investigates the TECFLAM configuration again, but here attention is given to the flame lift-off evident in experiments and reproduced by the LES-CMC formulation. The stabilization process and the pollutants emission of the flame are investigated in detail.

This thesis details the development, verification and validation of an unsteady unstructured high order (≥ 3) h/p Discontinuous Galerkin - Fourier solver for the incompressible Navier-Stokes equations on static and rotating meshes in two and three dimensions. This general purpose solver is used to provide insight into cross-flow (wind or tidal) turbine physical phenomena. Simulation of this type of turbine for renewable energy generation needs to account for the rotational motion of the blades with respect to the fixed environment. This rotational motion implies azimuthal changes in blade aero/hydro-dynamics that result in complex flow phenomena such as stalled flows, vortex shedding and blade-vortex interactions. Simulation of these flow features necessitates the use of a high order code exhibiting low numerical errors. This thesis presents the development of such a high order solver, which has been conceived and implemented from scratch by the author during his doctoral work. To account for the relative mesh motion, the incompressible Navier-Stokes equations are written in arbitrary Lagrangian-Eulerian form and a non-conformal Discontinuous Galerkin (DG) formulation (i.e. Symmetric Interior Penalty Galerkin) is used for spatial discretisation. The DG method, together with a novel sliding mesh technique, allows direct linking of rotating and static meshes through the numerical fluxes. This technique shows spectral accuracy and no degradation of temporal convergence rates if rotational motion is applied to a region of the mesh. In addition, analytical mappings are introduced to account for curved external boundaries representing circular shapes and NACA foils. To simulate 3D flows, the 2D DG solver is parallelised and extended using Fourier series. This extension allows for laminar and turbulent regimes to be simulated through Direct Numerical Simulation and Large Eddy Simulation (LES) type approaches. Two LES methodologies are proposed. Various 2D and 3D cases are presented for laminar and turbulent regimes. Among others, solutions for: Stokes flows, the Taylor vortex problem, flows around square and circular cylinders, flows around static and rotating NACA foils and flows through rotating cross-flow turbines, are presented.

碩士
淡江大學
土木工程學系碩士班
101
This study uses LES to simulate the aerodynamic characteristics of hemispherical dome in a smooth approaching flow field. The accuracy of numerical simulation was verified firstly by comparing with the wind tunnel measurements. Then the details of the aerodynamics of the dome were presented in this thesis. Prior to study the dome aerodynamics, several schemes of the grid system and numerical parameters were examined to determine the optimal ones for this study. The numerical simulation in this thesis can be categorized into two parts: the aerodynamics of dome in two Reynolds numbers, Re=6.6×104 and 2×106. In the case of Re=6.6×104, the mean and RMS pressure coefficients on the center meridian are noticeably deviated from experiment data. The numerical error may be caused by two reasons. The first probable source of error is that the separation bubble in the wake region extent beyond the mesh refined area. The second one is more subtle. At subcritical Reynolds number, the boundary layer developed over the dome surface is of laminar nature; it transits to become turbulent flow after separated from dome surface. Whether the basic setting of CFD tool, ANSYS-FLUENT, is apt to such a complex numerical simulation is to be confirmed. As for the second case, Re=2×106, the mean and RMS pressure coefficients on the center meridian agree well with experiment data except near the front stagnation area. The power spectral densities of the numerical simulated pressure fluctuations also agree with wind tunnel measurements satisfactory. Only the probability densities of the numerical simulation exhibit deviations from the wind tunnel data. It indicates that although the current numerical simulation scheme can reproduce the hemi-spherical dome’s aerodynamic quite well; it is still insufficient to generate the small scale turbulence that contributes to the pressure peaks. The lift force spectrum exhibits multiple peaks; which indicate the complexity of the vortex shedding in the horizontal plane. The time history of the vorticity further demonstrate that there exists no clear interaction between two separated free shear layers of the two opposite side of the dome; however, the wake flow show rather periodic sway synchronized with the variation of lift force coefficient.

It is often desirable to study simulated turbulent flows at steady state even if the flow has no inherent source of turbulence kinetic energy. Doing so requires a numerical forcing scheme and various methods have been studied extensively for turbulence that is isotropic and homogeneous in three dimensions. A review of these existing schemes is used to form a framework for more general forcing methods. In this framework, the problem of developing a forcing scheme in Fourier space is abstracted into the two problems of (1) prescribing the spectrum of the input power and (2) specifying a force that has the desired characteristics and that adds energy to the flow with the correct spectrum. The framework is used to construct three forcing schemes for horizontally homogeneous and isotropic, vertically stratified turbulence. These schemes are implemented in large-eddy simulations and their characteristics analyzed. Which method is “best” depends on the purpose of the simulations, but the framework for specifying forcing schemes enables a systematic approach for identifying a method appropriate for a particular application.

Least Square Kinetic Upwind Method (LSKUM) belongs to the class of mesh-less method that solves compressible Euler equations of gas dynamics. LSKUM is kinetic theory based upwind scheme that operates on any cloud of points. Euler equations are derived from Boltzmann equation (of kinetic theory of gases) after taking suitable moments. The basic update scheme is formulated at Boltzmann level and mapped to Euler level by suitable moments. Mesh-less solvers need only cloud of points to solve the governing equations. For a complex configuration, with such a solver, one can generate a separate cloud of points around each component, which adequately resolves the geometric features, and then combine all the individual clouds to get one set of points on which the solver directly operates. An obvious advantage of this approach is that any incremental changes in geometry will require only regeneration of the small cloud of points where changes have occurred. Additionally blanking and de-blanking strategy along with overlay point cloud can be adapted in some applications like store separation to avoid regeneration of points. Naturally, the mesh-less solvers have advantage in tackling complex geometries and moving components over solvers that need grids. Conventionally, higher order accuracy for space derivative term is achieved by two step defect correction formula which is computationally expensive. The present solver uses low dissipation single step modified CIR (MCIR) scheme which is similar to first order LSKUM formulation and provides spatial accuracy closer to second order. The maximum time step taken to march solution in time is limited by stability criteria in case of explicit time integration procedure. Because of this, explicit scheme takes a large number of iterations to achieve convergence. The popular explicit time integration schemes like four stages Runge-Kutta (RK4) are slow in convergence due to this reason. The above problem can be overcome by using the implicit time integration procedure. The implicit schemes are unconditionally stable i.e. very large time steps can be used to accelerate the convergence. Also it offers superior robustness. The implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) scheme is very attractive due to its low numerical complexity, moderate memory requirement and unconditional stability for linear wave equation. Also this scheme is more efficient than explicit counterparts and can be implemented easily on parallel computers. It is based on the factorization of the implicit operator into three parts namely lower triangular matrix, upper triangular matrix and diagonal terms. The use of LU-SGS results in a matrix free implicit framework which is very economical as against other expensive procedures which necessarily involve matrix inversion. With implementation of the implicit LU-SGS scheme larger time steps can be used which in turn will reduce the computational time substantially. LU-SGS has been used widely for many Finite Volume Method based solvers. The split flux Jacobian formulation as proposed by Jameson is most widely used to make implicit procedure diagonally dominant. But this procedure when applied to mesh-less solvers leads to block diagonal matrix which again requires expensive inversion. In the present work LU-SGS procedure is adopted for mesh-less approach to retain diagonal dominancy and implemented in 2-D and 3-D solvers in matrix free framework. In order to assess the efficacy of the implicit procedure, both explicit and implicit 2-D solvers are tested on NACA 0012 airfoil for various flow conditions in subsonic and transonic regime. To study the performance of the solvers on different point distributions two types of the cloud of points, one unstructured distribution (4074 points) and another structured distribution (9600 points) have been used. The computed 2-D results are validated against NASA experimental data and AGARD test case. The density residual and lift coefficient convergence history is presented in detail. The maximum speed up obtained by use of implicit procedure as compared to explicit one is close to 6 and 14 for unstructured and structured point distributions respectively. The transonic flow over ONERA M6 wing is a classic test case for CFD validation because of simple geometry and complex flow. It has sweep angle of 30° and 15.6° at leading edge and trailing edge respectively. The taper ratio and aspect ratio of the wing are 0.562 and 3.8 respectively. At M∞=0.84 and α=3.06° lambda shock appear on the upper surface of the wing. 3¬D explicit and implicit solvers are tested on ONERA M6 wing. The computed pressure coefficients are compared with experiments at section of 20%, 44%, 65%, 80%, 90% and 95% of span length. The computed results are found to match very well with experiments. The speed up obtained from implicit procedure is over 7 for ONERA M6 wing. The determination of the aerodynamic characteristics of a wing with the control surface deflection is one of the most important and challenging task in aircraft design and development. Many military aircraft use some form of the delta wing. To demonstrate the effectiveness of 3-D solver in handling control surfaces and small gaps, implicit 3-D code is used to compute flow past clipped delta wing with aileron deflection of 6° at M∞ = 0.9 and α = 1° and 3°. The leading edge backward sweep is 50.4°. The aileron is hinged from 56.5% semi-span to 82.9% of semi-span and at 80% of the local chord from leading edge. The computed results are validated with NASA experiments

Least Square Kinetic Upwind Method (LSKUM) belongs to the class of mesh-less method that solves compressible Euler equations of gas dynamics. LSKUM is kinetic theory based upwind scheme that operates on any cloud of points. Euler equations are derived from Boltzmann equation (of kinetic theory of gases) after taking suitable moments. The basic update scheme is formulated at Boltzmann level and mapped to Euler level by suitable moments. Mesh-less solvers need only cloud of points to solve the governing equations. For a complex configuration, with such a solver, one can generate a separate cloud of points around each component, which adequately resolves the geometric features, and then combine all the individual clouds to get one set of points on which the solver directly operates. An obvious advantage of this approach is that any incremental changes in geometry will require only regeneration of the small cloud of points where changes have occurred. Additionally blanking and de-blanking strategy along with overlay point cloud can be adapted in some applications like store separation to avoid regeneration of points. Naturally, the mesh-less solvers have advantage in tackling complex geometries and moving components over solvers that need grids. Conventionally, higher order accuracy for space derivative term is achieved by two step defect correction formula which is computationally expensive. The present solver uses low dissipation single step modified CIR (MCIR) scheme which is similar to first order LSKUM formulation and provides spatial accuracy closer to second order. The maximum time step taken to march solution in time is limited by stability criteria in case of explicit time integration procedure. Because of this, explicit scheme takes a large number of iterations to achieve convergence. The popular explicit time integration schemes like four stages Runge-Kutta (RK4) are slow in convergence due to this reason. The above problem can be overcome by using the implicit time integration procedure. The implicit schemes are unconditionally stable i.e. very large time steps can be used to accelerate the convergence. Also it offers superior robustness. The implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) scheme is very attractive due to its low numerical complexity, moderate memory requirement and unconditional stability for linear wave equation. Also this scheme is more efficient than explicit counterparts and can be implemented easily on parallel computers. It is based on the factorization of the implicit operator into three parts namely lower triangular matrix, upper triangular matrix and diagonal terms. The use of LU-SGS results in a matrix free implicit framework which is very economical as against other expensive procedures which necessarily involve matrix inversion. With implementation of the implicit LU-SGS scheme larger time steps can be used which in turn will reduce the computational time substantially. LU-SGS has been used widely for many Finite Volume Method based solvers. The split flux Jacobian formulation as proposed by Jameson is most widely used to make implicit procedure diagonally dominant. But this procedure when applied to mesh-less solvers leads to block diagonal matrix which again requires expensive inversion. In the present work LU-SGS procedure is adopted for mesh-less approach to retain diagonal dominancy and implemented in 2-D and 3-D solvers in matrix free framework. In order to assess the efficacy of the implicit procedure, both explicit and implicit 2-D solvers are tested on NACA 0012 airfoil for various flow conditions in subsonic and transonic regime. To study the performance of the solvers on different point distributions two types of the cloud of points, one unstructured distribution (4074 points) and another structured distribution (9600 points) have been used. The computed 2-D results are validated against NASA experimental data and AGARD test case. The density residual and lift coefficient convergence history is presented in detail. The maximum speed up obtained by use of implicit procedure as compared to explicit one is close to 6 and 14 for unstructured and structured point distributions respectively. The transonic flow over ONERA M6 wing is a classic test case for CFD validation because of simple geometry and complex flow. It has sweep angle of 30° and 15.6° at leading edge and trailing edge respectively. The taper ratio and aspect ratio of the wing are 0.562 and 3.8 respectively. At M∞=0.84 and α=3.06° lambda shock appear on the upper surface of the wing. 3¬D explicit and implicit solvers are tested on ONERA M6 wing. The computed pressure coefficients are compared with experiments at section of 20%, 44%, 65%, 80%, 90% and 95% of span length. The computed results are found to match very well with experiments. The speed up obtained from implicit procedure is over 7 for ONERA M6 wing. The determination of the aerodynamic characteristics of a wing with the control surface deflection is one of the most important and challenging task in aircraft design and development. Many military aircraft use some form of the delta wing. To demonstrate the effectiveness of 3-D solver in handling control surfaces and small gaps, implicit 3-D code is used to compute flow past clipped delta wing with aileron deflection of 6° at M∞ = 0.9 and α = 1° and 3°. The leading edge backward sweep is 50.4°. The aileron is hinged from 56.5% semi-span to 82.9% of semi-span and at 80% of the local chord from leading edge. The computed results are validated with NASA experiments