Cette thèse porte sur la caractérisation expérimentale et la simulation numérique d’une configurationde jet rond en impact peu rencontrée dans la littérature : un jet chauffé issu d’une conduitepleinement développée à un haut nombre de Reynolds (ReD = 60 000) impacte normalement uneparoi située à trois diamètres en aval. Le premier volet de ce travail est dédié à la génération d’unebase de donnée expérimentale à l’aide de plusieurs moyens de mesure, avec pour objectif de caractériserà la fois la dynamique et la thermique de l’écoulement. Les techniques complémentaires devélocimétrie laser à franges (LDV) et vélocimétrie par image de particules (S-PIV) ont été mises àprofit pour la caractérisation du champ de vitesse et du tenseur de Reynolds tandis que les champsde température moyenne et fluctuante ont été mesurés à l’aide d’un fil froid. Enfin, les échangesthermiques au niveau de la paroi ont été obtenus par la méthode inverse de thermographie en facearrière (ThEFA). En plus de fournir une base de donnée très complète nécessaire à la validation dessimulations numériques, ces mesures ont également permis de mettre en évidence l’organisation àgrande échelle de l’écoulement, avec la présence de grandes structures tourbillonnaires dont la fréquencede passage correspond au mode colonne du jet libre et qui s’approchent de la paroi d’impactaux alentours du second maximum observé dans la distribution des échanges pariétaux. Le secondvolet concerne les simulations numériques visant à reproduire la configuration expérimentale. Deuxapproches ont été évaluées : l’approche RANS pour quantifier la pertinence des modèles utilisés parles industriels et l’approche LES, plus coûteuse, mais donnant accès aux propriétés instationnaireset tridimensionnelles de l’écoulement. Les simulations RANS ont montré que les modèles reconnuscomme les plus performants pour ce type de configuration sont incapables de prévoir correctementle niveau des échanges pariétaux. Ils sont, en revanche, bien reproduits par la simulation LES. Lesdonnées obtenues ont été mises à profit pour mieux comprendre les mécanismes liés à l’apparitiondu second maximum. Cette analyse a mis en avant le rôle des « points chauds ». Seuls certains d’entreeux ont pu être reliés à la présence de régions « décollées » tandis que la majorité est associée à desstructures allongées dans la direction de l’écoulement. / This thesis is dedicated to the experimental characterisation and the numerical simulations ofa round impinging jet configuration seldom dealt with in the literature : a heated jet issues from apipe fully developed pipe at a high Reynolds number (ReD = 60 000) and normally impinges a platelocated three diameters downstream. The first part of this work is directed towards the generationof an experimental database by means of several measurement techniques in order to characteriseboth the dynamical and thermal flow features. The complementary techniques of laser Doppler velocimetry(LDV) and particle image velocimetry (S-PIV) allowed for the velocity and Reynolds tensorfield characterisation. The mean and fluctuating temperature fields were measured through cold-wirethermometry. Finally, the plate heat transfer distribution was obtained through the inverse methodof « rear face thermography » (ThEFA). The gathered data not only provided a comprehensive databasenecessary to validate numerical simulations but also permitted to highlight the large-scale floworganisation, with the presence of large vortices shedding at the free jet preferred mode and closelyapproaching the plate in the vicinity of the secondary peak observed in the heat transfer distribution.The second part of this thesis focuses on the numerical simulations aiming at reproducing the experimentalconfiguration. Two approaches were evaluated : the RANS approach in order to quantifythe relevance of industrial turbulence models and the Large-Eddy Simulation, more expensive, butproviding the 3D unsteady flow features. The RANS simulations showed that the models recognisedas the most efficient for this kind of configuration are unable to correctly predict the heat transferlevels. They are, on the other hand, well reproduced by the LES. The generated data allowed for betterunderstanding of the mechanisms leading to the secondary peak. This analysis highlighted theprominent role of the "hot spots", where only some of them can be related to « separated » regions,while the majority are associated with streamwise elongated structures.

The recent resurgence in hypersonics is centered around the development of SCRAMJET engine technology to power future hypersonic vehicles. Successful flight trials by Australian and American scientists have created interest in the scramjet engine research across the globe. To develop scramjet engine, it is important to study heat transfer effects on the engine performance and aerodynamic forces acting on the body. Hence, the main aim of present investigation is the design of scramjet engine configuration and measurement of aerodynamic forces acting on the model and heat transfer rates along the length of the combustor. The model is a two-dimensional single ramp model and is designed based on shock-on-lip (SOL) condition. Experiments are performed in IISc hypersonic shock tunnel HST2 at two different Mach numbers of 8 and 7 for different angles of attack. Aerodynamic forces measurements using three-component accelerometer force balance and heat transfer rates measurements using platinum thin film sensors deposited on Macor substrate are some of the shock tunnel flow diagnostics that have been used in this study.

In this research, an elastic cylinder that utilized vortex-induced vibration (VIV) was applied to improve convective heat transfer rates by disrupting the thermal boundary layer. Rigid and elastic cylinders were placed across a fluid channel. Vortex shedding around the cylinder led to the periodic vibration of the cylinder. As a result, the flow-structure interaction (FSI) increased the disruption of the thermal boundary layer, and therefore, improved the mixing process at the boundary. This study aims to improve convective heat transfer rate by increasing the perturbation in the fluid flow. A three-dimensional numerical model was constructed to simulate the effects of different flow channel geometries, including a channel with a stationary rigid cylinder, a channel with a elastic cylinder, a channel with two elastic cylinders of the same diameter, and a channel with two elastic cylinders of different diameters. Through the numerical simulations, the channel maximum wall temperature was found to be reduced by approximately 10% with a stationary cylinder and by around 17% when introducing an elastic cylinder in the channel compared with the channel without the cylinder. Channels with two-cylinder conditions were also studied in the current research. The additional cylinder with the same diameter in the fluid channel only reduced the surface wall temperature by 3% compared to the channel without any cylinders because the volume of the second cylinder could occupy some space, and therefore, reduce the effect of the convective heat transfer. By reducing the diameter of the second cylinder by 25% increased the effect of the convection heat transfer and reduced the maximum wall temperature by around 15%. Compared to the channel with no cylinder, the introduction of cylinders into the channel flow was found to increase the average Nusselt number by 55% with the insertion of a stationary rigid cylinder, by 85% with the insertion of an elastic cylinder, by 58% with the insertion of two cylinders of the same diameter, and by approximately 70% with the insertion of two cylinders of different diameters (the second cylinder having the smaller diameter). Furthermore, it was also found that the maximum local Nusselt number could be enhanced by around 200%-400% at the entrance of the fluid channel by using the elastic cylinders compared to the channel without cylinders.

This thesis deals with an experimental determination of heat transfer coefficients from the surface of the thermal manikin. The main focus of the work lies on separating radiative and convective heat fluxes from the surface of the thermal manikin. Both nude and clothed, standing and seated postures were investigated respectively. The tests were conducted in a constant air temperature (cca 24°C) and a constant wind speed (cca 0,05 m.s-1) environment. The major part of the radiative heat flux was eliminated by a low emissivity coating applied to the surface of the nude thermal manikin, and in the case of clothed manikin by a low emissivity two-piece dress. Favorable results were achieved only in the case of the nude manikin measurements. The measurements were performed across 34 zones that logically represent parts of a human body. Experimental work confirms theoretical expectations in the means of a heat transfer. In addition, the results of this work were compared to results of a similar experimental work. The outcomes of this thesis provide essential information in order to create detailed computational models of a thermal environment. Such models require anatomically specific, separate values of convective and radiative heat transfer coefficients.

A three-dimensional computational tool was developed that simulates the heat and mass transfer interaction in a soil-root-stem system (SRSS) for a tree in a seasonally varying deciduous forest. The development of the SRSS model involved the modification and coupling of existing heat and mass transport tools to reproduce the three-dimensional diurnal internal and external temperatures, internal fluid distribution, and heat flow in the soil, roots, and stems. The model also required the development of a parallel Monte-Carlo algorithm to simulate the solar and environmental radiation regime consisting of sky and forest radiative effects surrounding the tree. The SRSS was tested, component-wise verified, and quantitatively compared with published observations. The SRSS was applied to simulate a tree in a dense temperate hardwood forest that included the calculations of surface heat flux and comparisons between cases with fluid flow transport and periods of zero flow. Results from the winter simulations indicate that the primary influence of temperature in the trunk is solar radiation and radiative energy from the soil and surrounding trees. Results from the summer simulation differed with previous results, indicating that sap flow in the trunk altered the internal temperature change with secondary effects attributed to the radiative energy from the soil and surrounding trees. Summer simulation results also showed that with sap flow, as the soil around the roots become unsaturated, the flow path for the roots will be changed to areas where the soil is still saturated with a corresponding increase in fluid velocity.

Dans le but de réduire la consommation en carburant et les émissions de CO2 des moteurs à combustion interne, un des leviers, qui a intéressé diffèrent acteurs dans le secteur automobile, est la récupération de l’énergie thermique disponible dans les gaz d’échappement. Malgré différents technologie ont été investigués dans le passé; les transferts de chaleur qui apparient dans les gaz d’échappement n’ont pas encore étés suffisamment étudiés. Le fait que les échanges de la chaleur apparent dans des conditions pulsatives, notamment due aux conditions de fonctionnement moteur, rende les connaissances acquis jusqu’à présent limités et ne pas exploitables. A l’état actuel on n’est pas capable de pouvoir prédire le transfert thermique convectif des écoulements pulsé. Les travaux de cette thèse s’instaurent dans la continuité de ce besoin, l’objectif principal est donc l’étude expérimentale du transfert thermique convectif des écoulements turbulent pulsés dans un conduit cylindrique. La première partie de ce travail a été consacrée à le dimensionnement d’un moyen d’essais permettant la création d’un écoulement pulsé type moteur; en suite différents méthodes de mesures ont étés développes afin de connaitre les variations instantanés de vitesse et température de l’écoulement. Plusieurs essais ont été reproduits afin de caractériser l’impact de la pulsation sur le transfert de la chaleur. Les résultats expérimentaux ont été analysés avec deux approches différentes: dans un premier temps une approche analytique 1D a permis de mettre en évidence le mécanisme principal responsable de l’amélioration du transfert thermique convectif,ainsi, il a fourni des éléments supplémentaires pour le futur développement de modèles mathématiques plus adaptés à la prédiction des transferts d’énergie. En suite une approche 2D, supporté d’une phase de modélisation numérique, a permis de caractériser le mécanisme de transport radial d’énergie thermique. / Waste Energy Recovery represents a promising way to go further in fuel saving and greenhouse emissions control for Internal Combustion Engine applications. Although several technologies have been investigated in the past few years, the convective heat transfers, playing an important role in the energy exchanges at the engine exhaust, has not receive enough attention. Heat transfers, in such applications, occur in pulsating conditions because of the engine operating conditions, making thus the actual knowledge of the heat transfer phenomena limited and not exploitable. Nowadays there is not any model capable to predict convective heat transfers for pulsating flows. In this context, the present thesis addresses the purpose to study the convective heat transfer phenomena, by an experimental approach, occurring for turbulent pulsating flows in pipes. In the first part of this work, an experimental apparatus has been designed to reproduce an exhaust type pulsating flow in fully managed conditions, as well as, several measurement techniques have been developed to know the instantaneous profiles of air temperature and velocity. Many experiments have been performed in order to characterize the impact of the flow pulsation on the convective heat transfers. In the second part of this work, the experimental results have been analyzed with two different approaches: firstly, with a 1D assumption the time-average convective heat transfers has been computed, and the major mechanism responsible of the heat transfer enhancement has been pointed out. Furthermore, it has been possible to highlight the mathematical term representative of such mechanism, which should be accounted in future to define a more adapted numerical model for the heat transfer prediction. In a second phase with a 2D assumption, and, with an energy and a fluid-mechanic computational phase, the radial transport of thermal energy has been characterized for a pulsating flow.

Despite years of research in high speed boundary layer flow, there is still a need for insightful experiments to realize key features of the flow like boundary layer response to different conditions and related transition mechanisms. Volumes of data on the these problems point to the fact that there is still much to be understood about the nature of boundary layer instability causing transition and growth of boundary layer in different conditions. Boundary layer stability experiments have been found to be more useful, in which the boundary layer is perturbed and its behavior observed to infer useful conclusions. Also, apart from the stability part, the effect of various changes in boundary layer due to the perturbation makes interesting observation to gain more insight into the understood and the not so understood facets of the same. In view of the above, the effect of a steady axisymmetric thermal bump is investigated on a hypersonic boundary layer over a 60º sharp cone cylinder model. The thermal bump, placed near tip of the cone, perturbs the boundary layer, the behavior of which is observed by recording the wall heat flux on the cone and cylinder surface using platinum thin film sensors. The state of the boundary layer is qualitatively assessed by the wall heat flux comparisons between laminar and turbulent values. The same thermal bump also acts as a heat addition source to boundary layer in which case this recorded data provides a look into the effect of the heat addition to the wall heat flux. To gain a larger view of heat addition causing changes to the flow, effects of change in enthalpy are also considered. Experiments are performed in the IISc HST2 shock tunnel facility at 2MJkg−1 stag-nation enthalpy and Mach number of 8,with and without the thermal bump to form comparisons. Some experiments are also performed in the IISc HST3 free piston driven shock tunnel facility at 6MJkg−1, to investigate the effect of change in stagnation enthalpy on the wall heat flux. To support the experimental results theoretical comparisons and computational studies have also been carried out. The results of experiments show that the laminar boundary layer over the whole model remains laminar even when perturbed by the thermal bump. The wall heat flux measurements show change on the cone part where there seems to be fluctuation in the temperature gradients caused by the thermal bump, which decrease at first and then show an increase towards the base of the cone. The cylinder part remains the same with and without the thermal bump, indicating heavy damping effects by the expansion fan at cone cylinder junction. A local peak in wall heat flux is observed at the junction which is reduced by 64% by the action of the thermal bump. The possible reason for this is attributed to the increased temperature gradients at the wall due to delayed dissipation of heat that is accumulated in the boundary layer as a result of the thermal bump action. The comparison of data for enthalpies of 2MJkg−1 and 6MJkg−1 show that there are negligible real gas effects in the higher enthalpy case and they do not affect the wall heat flux much. Also it is found that the thermal bump fails to dump heat into the flow directly though it creates heat addition virtually by mere discontinuity in the surface temperature and causes temperature gradients fluctuation in the boundary layer. Considering the thermal bump action and the change in stagnation enthalpy of the flow, there seems to be no change in both cases that can be attributed to a common observation resulting from the factor of change in heat inside the boundary layer.

In most casting applications, dendritic microstructure morphology is not desired because it leads to poor mechanical properties. Forced convection causing sufficient shearing in the mushy zone of the partially solidified melt is one of the means to suppress this dendritic growth. The dendrites formed at the solid-liquid interface are detached and carried away due to strong fluid flow to form slurry. This slurry, consisting of rosette or globular particles, provides less resistance to flow even at a high solid fraction and can easily fill the die-cavity. The stated principle is the basis of a new manufacturing technology called “semi-solid forming” (SSF), in which metal alloys are cast in the semi-solid state. This technique has numerous advantages over other existing commercial casting processes, such as reduction of macrosegregation, reduction of porosity and low forming efforts. Among all currently available methods available for large scale production of semisolid slurry, the cooling slope is considered to be a simple but effective method because of its simple design and easy control of process parameters, low equipment and running costs, high production efficiency and reduced inhomogeneity. With this perspective, the primary objective of the present research is to investigate, both experimentally and numerically, convective heat transfer and solidification on a cooling slope, in addition to the study of final microstructure of the cast billets. Some key process parameters are identified, namely pouring temperature, slope angle, slope length, and slope cooling rate. A systematic scaling analysis is performed in order to understand the relative importance of the parameters in influencing the final properties of the slurry and microstructure after solidification. A major part of the present work deals with the development of an experimental set up with careful consideration of the range of process parameters involved by treating the cooling slope as a heat exchanger. Subsequently, a comprehensive numerical model is developed to predict the flow, heat transfer, species concentration solid fraction distribution of aluminum alloy melt while flowing down the cooling slope. The model uses a variable viscosity relation for slurry. The metal-air interface at the top during the melt flow is tracked using a volume of fluid (VOF) method. Solidification is modeled using an enthalpy based approach and a volume averaged technique. The mushy region is modeled as a multi-layered porous medium consisting of fixed columnar dendrites and mobile equiaxed or fragmented grains. In addition, the solidification model also incorporates a fragmentation criterion and solid phase movement. The effects of key process parameters on flow behavior involving velocity distribution, temperature distribution, solid fractions at the slope exit, and macrosegregation, are studied numerically and experimentally for aluminium alloy A356. The resulting microstructures of the cast billets obtained from the experiments are studied and characterized. Finally the experimental results are linked to the model predictions for establishing the relations involving interdependence of the stated key process parameters in determining the quality of the final cast products. This study is aimed towards providing the necessary guidelines for designing a cooling slope and optimizing the process parameters for desirable quality of the solidified product.

The objective of this thesis research is to improve our understanding of the flow obstacle effect on heat transfer at supercritical and high subcritical pressures by experimentally studying the effect of different obstacles on heat transfer in two vertical upward-flow test sections: a 3-rod bundle and an 8 mm ID tube. The heat transfer measurements cover the region of interest of the Canadian Super-critical Water Cooled Reactor (SCWR). A thorough analysis of the obstacle effect on supercritical heat transfer (SCHT) was performed. In the 3-rod bundle, two types of obstacles were employed: wire wraps and low-impact grid spacers. Wire wraps were found to be more effective than grid spacers to enhance the SCHT. In the tubular test section, obstacles appeared to suppress the heat transfer deterioration (HTD) or decrease its severity; obstacles also generally enhanced the SCHT both in the liquid-like and the gas- like region. The experiment in the tubular test section revealed that, at certain flow conditions (low mass flux, low inlet subcooling), flow obstacles can have an adverse impact on the SCHT. A criterion to predict the onset of this adverse effect was developed. At high subcritical pressures, obstacles increased the CHF and reduced the maximum post-CHF temperature. A comparison of the experimental data with prediction methods for the SCHT, single phase heat transfer, CHF and post-dryout heat transfer was performed. Lastly, a new correlation to predict the enhancement in SCHT due to obstacles was developed for heat transfer in the liquid-like and gas-like regions.

Cette thèse est consacrée à l'optimisation de la topologie et de la forme de systèmesmultiphysiques motivés par des applications de l'industrie aéronautique. Nouscalculons les dérivées de forme de fonctions de coût arbitraires pour un modèlefluide, thermique et mécanique faiblement couplé. Nous développons ensuite unalgorithme de type gradient adapté à la résolution de problèmes d'optimisation deformes sous contraintes qui ne requiert par de réglage de paramètres nonphysiques. Nous introduisons ensuite une méthode variationnelle qui permet decalculer des intégrales le long de rayons sur un maillage par la résolution d'unproblème variationnel qui ne requiert pas la détermination explicite de ces lignessur la discrétisation spatiale. Cette méthode nous a ainsi permis d'imposer unecontrainte de non-mélange de phases pour une application à l'optimisationd'échangeurs de chaleur bi-tubes. Tous ces ingrédients ont été employés pour traiterune variété de cas tests d'optimisation de formes pour des systèmes multi-physiques2-d ou 3-d. Nous avons considéré des problèmes à une seule, deux ou bien troisphysiques couplées en 2-d, et des problèmes de tailles relativement élevées en 3-dpour la mécanique, la conduction thermique, l'optimisation de profils aérodynamiques,et de la forme de systèmes en interaction fluide-structure. Un dernier chapitred'ouverture est consacré à l'étude de modèles homogénéisées d'ordres élevés pour lessystèmes elliptiques perforés. Ces équations d'ordres élevés englobent les troisrégimes homogénéisés classiques associés à divers rapports d'échelles pour la tailledes obstacles. Elles pourraient permettre, dans de futurs travaux, de développer denouvelles méthodes d'optimisation pour les systèmes fluides caractérisés par desmotifs multi-échelles, ainsi que couramment rencontré dans la conception deséchangeurs thermiques industriels. / This work is devoted to shape and topology optimization of multiphysics systemsmotivated by aeronautic industrial applications. Shape derivatives of arbitraryobjective functionals are computed for a weakly coupled thermal fluid-structuremodel. A novel gradient flow type algorithm is then developed for solving genericconstrained shape optimization problems without the need for tuning non-physicalmetaparameters. Motivated by the need for enforcing non-mixing constraints in thedesign of liquid-liquid heat exchangers, a variational method is developed in orderto simplify the numerical evaluation of geometric constraints: it allows to computeline integrals on a mesh by solving a variational problem without requiring theexplicit knowledge of these lines on the spatial discretization. All theseingredients allowed us to implement a variety of 2-d and 3-d multiphysics shapeoptimization test cases: from single, double or three physics problems in 2-d, tomoderately large-scale 3-d test cases for structural design, thermal conduction,aerodynamic design and a fluid-structure interacting system. A final opening chapterderives high order homogenized equations for perforated elliptic systems. These highorder equations encompass the three classical regimes of homogenized modelsassociated with different obstacle's size scalings. They could allow, in futureworks, to develop new topology optimization methods for fluid systems characterizedby multi-scale patterns as commonly encountered in industrial heat exchanger designs.