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.

This thesis deals with study of dynamics of the swirling flow. The swirling flow occurs frequently in hydraulic machinery (e.g., vortex rope in draft tube of the hydraulic turbine) and often influences operation of these machines. For this reason, sufficient knowledge regarding this characteristic flow is necessary for subsequent improvement of hydraulic machines. The theoretical part of this thesis contains description of flow instabilities and their manifestations, notably Kármán vortex street and vortex rope. In the next part, two methods are applied to these two transient flows in order to identify their specific dynamic properties. The first method, Fourier transform, enables to find frequencies of transient flow. The second method, proper orthogonal decomposition (POD), enables to identify planar or spatial eigenmodes of a specific swirling flow. Proper orthogonal decomposition is used in this thesis to identify planar eigenmodes of Kármán vortex street and spatial eigenmodes of vortex rope.

La machine à réluctance variable peut être utilisée dans les véhicules électriques où pour des considérations d’autonomie, le rendement est crucial. En raison du fort champ de fuite dans la région de l’entrefer de la machine à réluctance variable due à sa géométrie particulière à pôles saillants, les pertes « cuivre » peuvent devenir conséquentes. Il est alors recommandé de ne pas placer les conducteurs au voisinage de l’entrefer. Cependant, des instructions concrètes pour la conception d’un enroulement optimal sont manquantes. Généralement, les pertes « cuivre » dans les machines électriques sont la somme des pertes Ohm DC classiques et des pertes additionnelles dites par courants de Foucault. Les pertes DC étant constantes à un point de fonctionnement donné, l’optimisation est axée alors sur la réduction des pertes par courants de Foucault en jouant sur la configuration géométrique de l’enroulement. Dans le cas de calculs répétitifs fastidieux, rencontrés par exemple lors des processus de conception et d’optimisation du bobinage des machines électriques, il y a un intérêt significatif à réduire le temps de calcul. Dans ce travail, on présente trois techniques de réduction du modèle et leurs applications dans les analyses par la méthode des éléments finis. Outre l’influence de la fréquence d’alimentation et de la section du conducteur, plusieurs facteurs liés à la configuration de l’enroulement influent sur les pertes additionnelles par courants de Foucault :i) la position du conducteur dans l’encoche au voisinage de la dent du stator ou de la zone de l’entrefer .ii) la disposition des conducteurs envers les lignes du champ magnétique bidimensionnelles de l’encoche .iii) l’utilisation d’un conducteur massif ou multi filamentaire; les filaments sont connectés en parallèle et peuvent permuter leurs positions périodiquement au sein du conducteur tout au long du bobinage. Dans cette thèse, on étudie principalement l’influence de la disposition géométrique des spires dans l’encoche et du type du conducteur utilisé s’il s’agit d’un conducteur massif, en fils de Litz ou en fils torsadés. Les pertes par courants de Foucault sont la conséquence d’un couplage fort électrique-magnétique entre la densité du courant et la variation en fonction du temps du champ magnétique. En utilisant le modèle de Maxwell, ce couplage est décrit par une équation différentielle à dérivée partielle qui ne peut être résolue simplement. La résolution de cette équation utilisant l’approche analytique n’est possible que sous certaines hypothèses simplificatrices qui peuvent dégrader la fiabilité de la solution. La modélisation par la méthode des éléments finis permet quant à elle de prendre en compte le mouvement du rotor et la non-linéarité du circuit magnétique garantissant ainsi une meilleure précision. Néanmoins, cela conduit à une large capacité de stockage et à un temps de calcul substantiel qui peut entraver tout processus de conception ou d’optimisation. Pour surmonter ce problème, on propose dans ce manuscrit trois techniques de réduction du modèle. Ces techniques assurent une réduction efficace de la taille du système matriciel associé à la modélisation par la méthode des éléments finis et diminuent par conséquent le temps de calcul : i) une réduction spatiale qui évite une modélisation en 3D des conducteurs complexes en fils torsadés et en fils de Litz et propose une modélisation 2D satisfaisante .ii) la technique de la perturbation. iii) la réduction de l’ordre du modèle utilisant la méthode de la décomposition orthogonale aux valeurs propres combinée à la méthode d’interpolation empirique discrète. La comparaison du modèle réduit à un modèle complet de référence montre l’efficacité de la réduction du modèle à réduire le temps de calcul tout en restant en deçà d’une erreur de précision acceptable. / The switched reluctance machine can be used in hybrid or electric vehicle where, for autonomy considerations, energy efficiency is crucial. Because of the strong stray field in the air-gap region of the switched reluctance machine due to its salient pole geometry, the copper losses can become substantial. It is firmly recommended therefore not to place the coil conductors near the air-gap region. Nevertheless, concrete instructions for optimal winding design are missing. The copper losses in electrical machines are subdivided into classical DC ohmic losses and additional eddy current losses occurring due to the time varying magnetic fields penetrating the copper conductors. Based on the fact that the DC losses are constant at a given operating point, the optimization is focused on reducing the eddy current losses by modifying the winding geometry configuration. In the case of tedious repetitive calculations, met for example during design and optimization processes of electrical machine windings, there is a significant interest in reducing the computation time. This work suggests three model reduction techniques and their applications in the finite element analyses.Besides the frequency of the excitation current and the cross section of the coil conductors, several factors related to the winding configuration can affect the addition al eddy current losses:i) the coil conductor position in the winding slot especially near the stator pole or close to the air gapii) the disposition of the coil conductor against the two-dimensional flux lines in the slot windingiii) the subdivision of the solid conductor into multiple parallel strands swapping their positions periodically in the conductor cross section throughout the length of the machine winding.This thesis mainly studies the influence of the geometric coils disposition in the slot windings and the type of the conductor used whether it is solid or stranded, with Litz or twisted wires.The eddy current losses exit through the strong electro-magnetic coupling between the electric current density and the time dependent magnetic flux lines penetrating the conductors; it is described mathematically by a partial differential equation that cannot be solved easily. The analytical approach, which is used practically for a quick resolution of the strong electro-magnetic coupling equation, is only possible under certain simplifying assumptions that deteriorate brutally the reliability of the copper losses calculation. The finite element modeling as for it, allows taking into account the rotor motion and the non-linear behavior of the magnetic circuit, thus ensuring a higher accuracy. However, it leads under these conditions to a substantial calculation time and requires large storage capacity. These constraints are critical and may hinder therefore any process of conception or optimization. In this thesis, we suggest three different model reduction techniques that can be effective in reducing the size of large scale complete finite element models and enable therefore to shorten the computational time:i) the spatial reduction avoiding the 3D modeling which seems required in the case of twisted and Litz wires and suggesting an alternative satisfactory 2D modeling.ii) the perturbation technique.iii) the model order reduction using the proper orthogonal decomposition combined with the discrete empirical interpolation method.The comparison between the reduced model solutions to that of the complete finite element model has proved the effectiveness of the proposed model reduction techniques; they allow shrinking the required computational time while staying below an acceptable error of accuracy.

Efficient analysis and storage of data is an integral but often challenging task when working with computation fluid dynamics mainly due to the amount of data it can output. Methods centered around the proper orthogonal decomposition were used to analyze, compress, and model various simulation cases. Two different high-fidelity, time-accurate turbomachinery simulations were investigated to show various applications of the analysis techniques. The first turbomachinery example was used to illustrate the extraction of turbulent coherent structures such as traversing shocks, vortex shedding, and wake variation from deswirler and rotor blade passages. Using only the most dominant modes, flow fields were reconstructed and analyzed for error. The reconstructions reproduced the general dynamics within the flow well, but failed to fully resolve shock fronts and smaller vortices. By decomposing the domain into smaller, independent pieces, reconstruction error was reduced by up to 63 percent. A new method of data compression that combined an image compression algorithm and the proper orthogonal decomposition was used to store the reconstructions of the flow field, increasing data compression ratios by a factor of 40.The second turbomachinery simulation studied was a three-stage fan with inlet total pressure distortion. Both the snapshot and repeating geometry methods were used to characterize structures of static pressure fluctuation within the blade passages of the third rotor blade row. Modal coefficients filtered by frequencies relating to the inlet distortion pattern were used to produce reconstructions of the pressure field solely dependent on the inlet boundary condition. A hybrid proper orthogonal decomposition method was proposed to limit burdens on computational resources while providing high temporal resolution analysis.Parametric reduced order models were created from large databases of transient and steady conjugate heat transfer and airfoil simulations. Performance of the models were found to depend heavily on the range of the parameters varied as well as the number of simulations used to traverse that range. The heat transfer models gave excellent predictions for temperature profiles in heated solids for ambitious parameter ranges. Model development for the airfoil case showed that accuracy was highly dependent on modal truncation. The flow fields were predicted very well, especially outside the boundary layer region of the flow.

No description available.

No description available.

Bats are some of the most agile flyers in nature. Their wings are highly articulated which affords them very fine control over shape and form. This thesis investigates the flight of Hipposideros Pratti. The flight pattern studied is nominally level and straight. Measured wing kinematics are used to describe the wing motion. It is shown that Proper Orthogonal Decomposition (POD) can be used to effectively to filter the measured kinematics to eliminate outliers which usually manifest as low energy higher POD modes, but which can impact the stability of aerodynamic simulations. Through aerodynamic simulations it is established that the first two modes from the POD analysis recover 62% of the lift, and reflect a drag force instead of thrust, whereas the first three modes recover 77% of the thrust and even more lift than the native kinematics. This demonstrates that mode 2, which features a combination of spanwise twisting (pitching) and chordwise cambering, is critical for the generation of lift, and more so for thrust. Based on these inferences, it is concluded that the first 7 modes are sufficient to represent the full native kinematics. The aerodynamic simulations are conducted using the immersed boundary method on 128 processors. They utilize a grid of 31 million cells and the bat wing is represented by about 50000 surface elements. The movement of the immersed wing surface is defined by piecewise cubic splines that describe the time evolution of each control point on the wing. The major contribution of this work is the decomposition of the native kinematics into canonical flapping wing physical descriptors comprising of the flapping motion, stroke-plane deviation, pitching motion, chordwise, and spanwise cambering. It is shown that the pitching mode harvests a Leading Edge Vortex (LEV) during the upstroke to produce thrust. It also stabilizes the LEV during downstroke, as a result, larger lift and thrust production is observed. Chordwise cambering mode allows the LEV to glide over and cover a large portion of the wing thus contributing to more lift while the spanwise cambering mode mitigates the intensification of LEV during the upstroke by relative rotation of outer part of the wing ( hand wing ) with respect to the inner part of the wing ( arm wing). While this thesis concerns itself with near straight-level flight, the proposed decomposition can be applied to any complex flight maneuver and provide a basis for unified comparison not only over different bat flight regimes but also across other flying insects and birds. / MS / Bats are some of the most agile flyers in nature. Their wings are highly articulated which affords them very fine control over shape and form. This thesis investigates the flight of Hipposideros Pratti. The flight pattern studied is nominally level and straight. Measured wing kinematics are used to describe the wing motion. The central motivation of the thesis is to characterize how the bat uses its wings to generate lift to counter gravity and thrust to move forward against drag forces. A mathematical filter based on Proper Orthogonal Decomposition (POD) is used to filter the measured wing motion to eliminate high frequency noise in the data but at the same time including including the important motions which produce lift and thrust. The filtered native kinematics is decomposed into flapping wing motions comprising of flapping mode, stroke-plane deviation, pitching motion, chordwise, and spanwise cambering. It is shown that the pitching mode harvests the low pressure region created by the Leading Edge Vortex (LEV) during the upstroke to produce thrust. It also stabilizes the LEV during the downstroke, as a result, larger lift and thrust production is observed. Chordwise cambering mode allows the LEV to glide over and cover a large portion of the wing thus contributing to more lift, while the spanwise cambering mode mitigates the intensification of LEV during the upstroke by relative rotation of the outer part of the wing (hand wing) with respect to the inner part of the wing (arm wing). While this thesis concerns itself with near straight-level flight, the proposed decomposition can be applied to any complex flight maneuver and provide a basis for unified comparison not only over different bat flight regimes but also across other flying insects and birds.

This thesis develops a one-dimensional version of a new data driven model of turbulence that uses the KL expansion to provide a spectral solution of the turbulent flow field based on analysis of Particle Image Velocimetry (PIV) turbulent data. The analysis derives a 2nd order random field over the whole flow domain that gives better turbulence properties in areas of non-uniform flow and where flow separates than the present models that are based on the Navier-Stokes Equations. These latter models need assumptions to decrease the number of calculations to enable them to run on present day computers or super-computers. These assumptions reduce the accuracy of these models. The improved flow field is gained at the expense of the model not being generic. Therefore the new data driven model can only be used for the flow situation of the data as the analysis shows that the kernel of the turbulent flow field of undular hydraulic jump could not be related to the surface waves, a key feature of the jump. The kernel developed has two parts, called the outer and inner parts. A comparison shows that the ratio of outer kernel to inner kernel primarily reflects the ratio of turbulent production to turbulent dissipation. The outer part, with a larger correlation length, reflects the larger structures of the flow that contain most of the turbulent energy production. The inner part reflects the smaller structures that contain most turbulent energy dissipation. The new data driven model can use a kernel with changing variance and/or regression coefficient over the domain, necessitating the use of both numerical and analytical methods. The model allows the use of a two-part regression coefficient kernel, the solution being the addition of the result from each part of the kernel. This research highlighted the need to assess the size of the structures calculated by the models based on the Navier-Stokes equations to validate these models. At present most studies use mean velocities and the turbulent fluctuations to validate a models performance. As the new data driven model gives better turbulence properties, it could be used in complicated flow situations, such as a rock groyne to give better assessment of the forces and pressures in the water flow resulting from turbulence fluctuations for the design of such structures. Further development to make the model usable includes; solving the numerical problem associated with the double kernel, reducing the number of modes required, obtaining a solution for the kernel of two-dimensional and three-dimensional flows, including the change in correlation length with time as presently the model gives instant realisations of the flow field and finally including third and fourth order statistics to improve the data driven model velocity field from having Gaussian distribution properties. As the third and fourth order statistics are Reynolds Number dependent this will enable the model to be applied to PIV data from physical scale models. In summary, this new data driven model is complementary to models based on the Navier-Stokes equations by providing better results in complicated design situations. Further research to develop the new model is viewed as an important step forward in the analysis of river control structures such as rock groynes that are prevalent on New Zealand Rivers protecting large cities.

Dans une modélisation intégrée des impacts, l'objectif est de tester plusieurs scénarios d'entrées de modèle et/ ou d'identifier l'effet de l'incertitude des entrées sur les sorties de modèle. Dans les deux cas, un grand nombre de simulations de modèle sont nécessaires. Cela reste bien évidemment infaisable avec un modèle de Chimie-Transport à cause du temps CPU demandé. Pour surmonter cette difficulté, deux approches ont été étudiées dans cette thèse : La première consiste à construire un modèle réduit. Deux techniques ont été utilisées : la première est la méthode POD (Proper Orthogonal Decomposition) liée au comportement statistique du système. La seconde méthode est une méthode efficace de prétabulation fondée sur la troncature d'un développement multivariables de la relation Entrées/ sorties associé au modèle.<br />La seconde est relative à la réduction du nombre de simulations demandé par la méthode Monte-Carlo classique de propagation d'incertitude. La technique utilisée ici est basée sur une représentation d'une sortie de modèle incertaine comme un développement de polynômes orthonormaux de variables d'entrées. Un autre point clé dans la modélisation intégrée d'impacts est de développer des stratégies de réduction des émissions en calculant des matrices de transfert sur plusieurs années de simulation. Une méthode efficace de calcul de ces matrices a été ainsi développée, notamment en définissant des scénarios "chimiquement" représentatifs.<br />L'ensemble de ces méthodes a été appliqué au modèle POLAIR3D, modèle de Chimie-Transport développé dans le cadre de cette thèse.

Les objectifs de cette thèse sont d'étudier les capacité prédictive des méthodes statistiques URANS et hybrides RANS-LES à modéliser des écoulements complexes à haut nombre de Reynolds et de réaliser l'analyse physique de la turbulence et des structures cohérentes en proche paroi. Ces travaux traitent de configurations étudiées dans le cadre des projets européens ATAAC (Advanced Turbulent Simulation for Aerodynamics Application Challenges) et TFAST (Transition Location Effect on Shock Wave Boundary Layer Interaction). Premièrement, l'écoulement décollé autour d’une configuration de cylindre en tandem, positionnés l'un derrière l’autre, est étudiée à un nombre de Reynolds de 166000. Un cas statique, correspondant schématique aux support de train d'atterrissage, est d’abord considéré. L’interaction fluide-structure est ensuite étudiée dans le cas dynamique, dans lequel le cylindre aval possède un degré de liberté en translation dans la direction perpendiculaire à l'écoulement. Une étude paramétrique est menée afin d'identifier les différents régimes d'interaction en fonction des paramètres structuraux. Dans un deuxième temps, la physique du tremblement transsonique est étudiée au moyen d’une analyse temps-fréquence et d’une décomposition orthogonale en modes propres (POD), dans l’intervalle de nombre de Mach 0.70–0.75. Les interactions entre le choc principal, la couche limite décollée par intermittence et les tourbillons se développant dans le sillage, sont analysées. Un forçage stochastique, basée sur une réinjection de turbulence synthétique dans les équations de transport de l’énergie cinétique et du taux de dissipation générée à partir de la reconstruction POD, a été introduit dans l’approche OES (organised-eddy simulation). Cette méthode introduit une modélisation de la turbulence “upscale" agissant comme un mécanisme de blocage par tourbillons capable de prendre en compte les interfaces turbulent/non-turbulent et de couches de cisaillement autour des géométries. Cette méthode améliore grandement la prédiction des forces aérodynamiques et ouvre de nouvelles perspectives quant aux approches de type moyennes d’ensemble pour modéliser les processus cohérents et aléatoires à haut nombre de Reynolds. Enfin, l'interaction onde de choc/couche limite (SWBLI) est traitée, dans le cas d’un choc oblique à nombre de Mach 1.7, contribuant aux études de "design d'ailes laminaires" au niveau européen. Les performances des modèles URANS et hybrides RANS-LES ont été analysées en comparant, avec les résultats expérimentaux, les valeurs intégrales de la couche limite (épaisseurs de déplacement et de quantité de mouvement) et les valeurs à la paroi (coefficient de frottement). Les effets de la transition dans la couche limite sur l’interaction choc/couche limite sont caractérisés. / 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.