Les turbines à axe horizontal ont souvent été préférées à celles à axe vertical au vu de leurs meilleures performances. Les pistes d’amélioration des turbines à axe vertical incluent notamment l’augmentation de l’efficacité globale en utilisant des pales à calage variable, et la réduction des fluctuations de couple en utilisant des configurations 3D particulières. Un modèle de mécanique des fluides numérique 2D et 3D adaptable est développé et validé dans cette thèse. Il est basé sur le modèle de turbulence RANS k-w SST, dans sa forme instationnaire. Le logiciel FLUENT® est utilisé pour simuler l’écoulement et prédire les performances d’une turbine. Une étude paramétrique 2D des turbines à pales droites est menée dans un premier temps afin de sélectionner le meilleur candidat pour l’étude d’optimisation, tout en rafraichissant l’état de l’art. L’effet de la solidité, du nombre de pales, du rapport de vitesse, du nombre de Reynolds, de l’angle de calage fixe et de l’épaisseur des pales sur l’efficacité aérodynamique de la turbine est évalué afin de déterminer ce qui est la meilleure configuration aérodynamique et le meilleur point d’opération dans des conditions données. L’impact des effets 3D associés à l’allongement des pales et à l’utilisation ou non de plaques de bout est aussi évalué. Les simulations montrent que la solidité optimale basée sur le rayon est autour de o = 0:2. En 3D, un faible allongement de 7 implique une chute d’efficacité relative de 60% par rapport aux prédictions 2D. Des pales plus allongées améliorent radicalement l’efficacité. Les plaques de bouts ont un effet positif sur les performances, en autant que leur taille est limitée. L’étude d’optimisation a montré un potentiel d’amélioration de l’efficacité des turbines à solidité aux alentours de o = 0:5, en utilisant un angle de calage dynamique. Les efficacités 2D atteignent la limite de Betz, et les efficacités 3D suivent la tendance observée sur les turbines à pales fixes. Il est confirmé que les turbines hélicoïdales ont une efficacité réduite comparées à une même turbine avec des pales droites, associé cependant à un lissage du couple. C’est le résultat de la propagation du décrochage le long de la pale, qui fait décrocher certaines portions qui verraient autrement un écoulement attaché. Une réduction de l’angle d’hélice permet de limiter ce phénomène. Des turbines multi-rotors permettent de garder les avantages de turbines à pales droites, comme la possibilité d’un angle de calage variable, tout en diminuant les fluctuations de couple de la même manière qu’augmenter le nombre de pales, sans toutefois changer la solidité. Des éléments de pales plus petits amènent cependant à des pertes dues aux effets 3D, limitant le nombre possible de rotors déphasés dans une fenêtre d’extraction donnée. / Horizontal axis turbines were always preferred to vertical axis turbines in the past due to better characteristics. Areas for improvement of the vertical axis turbine concept include an increase of the global efficiency using variable pitch control, and the reduction of torque fluctuations on the shaft by means of multiple 3D configurations. A 2D and 3D adaptable Computational Fluid Dynamics model is developed and validated in this thesis, using the k-w SST RANS turbulence model in its unsteady form. The proven commercial software FLUENT® is used to simulate the flow and predict as reliably as possible the turbine performance and characteristics. An extensive parametric study of vertical axis turbines of the H-Darrieus type in 2D is first conducted in order to select the best candidate for optimization, as well as to refresh the state-of-the-art in terms of turbines without pitch control. The effects of solidity, number of blades, tip speed ratio, Reynolds number, fixed blade pitch angle and blade thickness on the aerodynamic efficiency of the turbine are evaluated in order to determine what are the best aerodynamic configurations and operation parameters in a given application. The impact of 3D effects associated to the blade aspect ratio and the use of end-plates is also investigated. Optimal radius-based solidity is found to be around o = 0:2. In 3D, a small blade aspect ratio (AR = 7) leads to a relative efficiency drop of nearly 60% compared to the 2D prediction. Longer blades improve the 3D efficiency greatly. End-plates are found to have a positive effect on power extraction performances, as long as their size and thus their drag is limited. The optimization study showed a great potential of efficiency improvement for turbines with “average” solidities around o = 0:5, using variable blade pitch. 2D efficiencies almost reached Betz’ limit, and the 3D efficiency reductions were consistent with the observations on fixed-pitch turbines. Helical turbines which effectively smooth the torque fluctuations are shown to have a decreased efficiency compared to the same turbine with straight blades. This is the result of the spanwise propagation of the separation bubble on one part of the blade to the other parts that wouldn’t otherwise encounter stall. Reducing the helical angle can help lower this propagation effect. Multiple-section turbines retain the advantages of straight blade turbines, including the ability of using a dynamic pitch control, and can be configured to lower the torque fluctuations in a similar manner as increasing the number of blades but without increasing the turbine solidity. However, smaller blade elements lead to larger 3D losses, thus limiting the number of dephased turbine elements possible in a given extraction window.

Ce projet de maîtrise vise à faire une caractérisation préliminaire de l'écoulement d'une turbine Francis de moyenne chute au régime sans charge dans le cadre du projet Tr-Francis au Laboratoire de machines hydrauliques. Concrètement, le projet a validé la méthodologie numérique utilisée à l'aide de mesures expérimentales préliminaires. De plus, les simulations numérique sont permis de fournir les chargements fluides afin que des simulations structurelles puissent être réalisées. Finalement, les résultats ont permis de cibler les structures dominantes dans l'écoulement qui causent les plus importantes fluctuations de pression et ainsi aider l'équipe expérimentale à positionner les capteurs de pression et à choisir leurs plans de mesures. Pour y arriver, la méthodologie est basée sur des interpolations de mesures effectuées par Hydro-Québec au régime sans charge sur le prototype. Les données sont prises telles quelles pour les simulations effectuées à l'échelle prototype et elles sont mise à l'échelle grâce aux lois de similitude de la norme IEC 60193 pour les simulations à l'échelle modèle. L'effet des conditions d'entrée dans le domaine est étudié sur des domaines partiels avec des maillages de différentes densités ainsi que sur le domaine complet. Les pressions obtenues numériquement dans l'aspirateur sont comparées aux mesures expérimentales préliminaires et les résultats sont concordants. De plus, les poches de cavitations observées numériquement au bord de fuite remontant sur le côté pression des aubes sont, de plus, confirmées par l'expérimentation. Des analyses fréquentielles à partir des signaux de pression sont utilisées dans ce projet afin de caractériser les phénomènes. En plus d'utiliser différentes méthodes de visualisation numériques afin d'isoler et d'analyser les structures principales dans l'écoulement. Imposer un profil de vitesse uniforme comparativement à un profil de vitesse tiré de simulations de la bâche et de la conduite d'amenée ne modifie pas les résultats significativement à l'échelle modèle avec le maillage le plus fin. Lors des simulations avec le domaine complet, plusieurs phénomènes d'importances ont été observés. Notamment des tourbillons inter-aubes, qui ont une modulation à f/n = 1, une importante zone de recirculation dans l'aspirateur qui remonte jusque dans la roue, des tourbillons à l'interface entre la roue et l'aspirateur qui se développent dans la couche cisaillée, un débalancement de l'aspirateur causé par le coude ainsi qu'une zone cavitante au bord de fuite. / This thesis presents the preliminary characterization of the flow in a medium head Francis turbine at speed no-load within the scope of the Tr-Francis project at the Hydraulic Machinery. Laboratory. Concretely, the project aims at validating the numerical methodology used with preliminary experimental measurements. Moreover, the numerical simulations will provide the fluid load for the FEA simulations. Ultimately, the results will allow identifying the dominant structures in the flow causing important fluctuations. Thus helping the experimental team to find the optimal location for the pressure sensors and the measuring planes. To do so, the numerical methodology is based on the measurements interpolation performed by Hydro-Québec on the prototype turbine at speed no-load. The data are applied, as they are, as the initial conditions for the prototype scale simulations. For the model scale simulations, they are scaled down using the similitude laws from the IEC 60193 standard. The effects of the inlet conditions are studied on partial domains with different mesh densities as well as on the complete domain. To validate the simulations, the pressure measurements obtained numerically in the draft tube are compared with the preliminary measurements and the results are in good agreement. Furthermore, the trailing edge cavitation observed numerically is also visible in experimental flow visualizations. Spectral analyses of pressure signals are used to help to characterize the phenomena. Different numerical visualization techniques are also used to isolate and analyse the main flow structures. Imposing a uniform velocity profile compared to the one obtained from the penstock and spiral case simulation does not significantly affect the results at model scale with the finest grid. Many important phenomena such as modulated inter-blades vortices, an important backflow region in the draft tube coming up to the runner, vortices generated in the draft tube in the shear layer, static pressure imbalance in the draft tube caused by the elbow and trailing blade cavitation were observed on the complete domain model scale simulations.

Thesis (M.Eng.)--Memorial University of Newfoundland, 2005. / Bibliography: leaves 112-116.

In the development of modern rotorcraft vehicles, many unique challenges emerge due to the highly coupled nature of individual rotorcraft design disciplines therefore, the use of an integrated product and process development (IPPD) methodology is necessary to drive the design solution. Through the use of parallel design and analysis, this approach achieves the design synthesis of numerous product and process requirements that is essential in ultimately satisfying the customers demands. Over the past twenty years, Georgia Techs Center for Excellence in Rotorcraft Technology (CERT) has continuously focused on refining this IPPD approach within its rotorcraft design course by using the annual American Helicopter Society (AHS) Student Design Competition as the design requirement catalyst. Despite this extensive experience, however, the documentation of this preliminary rotorcraft design approach has become out of date or insufficient in addressing a modern IPPD methodology. In no design discipline is this need for updated documentation more prevalent than in propulsion system design, specifically in the area of gas turbine technology. From an academic perspective, the vast majority of current propulsion system design resources are focused on fixed-wing applications with very limited reference to the use of turboshaft engines. Additionally, most rotorcraft design resources are centered on aerodynamic considerations and largely overlook propulsion system integration. This research effort is aimed at bridging this information gap by developing a preliminary turboshaft engine design methodology that is applicable to a wide range of potential rotorcraft propulsion system design problems. The preliminary engine design process begins by defining the design space through analysis of the initial performance and mission requirements dictated in a given request for proposal (RFP). Engine cycle selection is then completed using tools such as GasTurb and the NASA Engine Performance Program (NEPP) to conduct thorough parametric and engine performance analysis. Basic engine component design considerations are highlighted to facilitate configuration trade studies and to generate more detailed engine performance and geometric data. Throughout this approach, a comprehensive engine design case study is incorporated based on a two-place, turbine training helicopter known as the Georgia Tech Generic Helicopter (GTGH). This example serves as a consistent propulsion system design reference highlighting the level of integration and detail required for each step of the preliminary turboshaft engine design methodology.

Thesis (Ph. D.)--Florida State University, 2005. / Advisor: Dr. Anjaneyulu Krothapalli, Florida State University, College of Engineering, Dept. of Mechanical Engineering. Title and description from dissertation home page (viewed June 14, 2005). Document formatted into pages; contains xx, 224 pages. Includes bibliographical references.

Thesis (Ph.D)--Aerospace Engineering, Georgia Institute of Technology, 2008. / Committee Chair: Seitzman, Jerry; Committee Member: Filatyev, Sergei; Committee Member: Jagoda, Jechiel; Committee Member: Lieuwen, Timothy; Committee Member: Shelton, Samuel; Committee Member: Zinn, Ben. Part of the SMARTech Electronic Thesis and Dissertation Collection.

Thesis (M.S.)--Florida State University, 2006. / Advisor: Juan C. Ordonez, Florida State University, College of Engineering, Dept. of Mechanical Engineering. Title and description from dissertation home page (viewed Sept. 15, 2006). Document formatted into pages; contains viii, 78 pages. Includes bibliographical references.

Small scale wind turbines operating in an urban environment produce dismal amounts of power when compared to their expected output [1-4]. This is largely due to the gusty wind conditions found in an urban environment, coupled with the fact that the wind turbines are not designed for these conditions. A new concept of a Segmented Variable Pitch (SVP) wind turbine has been proposed, which has a strong possibility to perform well in gusty and variable wind conditions. This dissertation explains the concept of a SVP wind turbine in more detail and shows analytical and experimental results relating to this concept. Also, the potential benefits of the proposed concept are mentioned. The results from this dissertation show that this concept has potential with promising results on possible turbine blade aerofoil configurations. Scaled model tests were completed and although further design optimisation is required, the tests showed good potential for the SVP concept. Lastly a proof-of-concept full scale model was manufactured and tested to prove scalability to full size from concept models. Along with the proof-of-concept full scale model, a wireless control system (to control the blade segments) was developed and tested.

The typical methodology for analytically designing a wind turbine blade is by means of blade element momentum (BEM) theory, whereby the aerofoil angle of attack is optimized to achieve a maximum lift-to-drag ratio. This research aims to show that an alternative optimisation methodology could yield better results, especially in gusty and turbulent wind conditions. This alternative method looks at increasing the aerofoil Reynolds number by increasing the aerofoil chord length. The increased Reynolds number generally increases the e_ectiveness of the aerofoil which would result in a higher or similar lift-to-drag ratio (even at the decreased angle of attacked require to maintain the turbine thrust coe_cient). The bene_t of this design is a atter power curve which causes the turbine to be less sensitive to uctuating winds. Also, the turbine has more torque at startup, allowing for operatation in lower wind speeds. This research is assumed to only be applicable to small wind turbines which operated in a low Reynolds number regime (<500 000), where Reynolds number manipulation is most advantageous.

Externally applied loads from high winds or impacts may cause structural damage to the wind-turbine blade, and this may further affect the aerodynamic performance of the blade. Wind-turbine blades experience high vibration levels or amplitudes under high winds. Vibrations negatively affect the wind flow on the blade. This project considers the structural dynamic analysis of a small-scale wind turbine with a particular focus on the blade; it involves the finite element model development, model validation and structural analysis of the validated model. The analysis involves a small-scale wind-turbine structural response when subjected to different loading inputs. The analysis is specifically focused on on-shore systems. The use of small-scale wind-turbine systems is common however, apart from initial structural analysis during design stages, these systems have not been studied sufficiently to establish their behaviour under a variation of real-life loading conditions. On-shore wind turbines are often designed for low-wind speeds and their structural strength may be compromised. In addition, these systems experience widely-varying wind speeds from one location to another to an extent that it is extremely difficult to achieve a uniform structural performance. The main reason for solving this problem is to evaluate the structural response of the blade, with special emphasis on an 800 W Kestrel e230i. This involves the calculation of the distribution of blade deflections and stresses over the wind-turbine blade under different loading conditions. To solve the problem, a three-dimensional model of a Kestrel e230i blade was firstly developed in Autodesk Inventor Professional using geometrical measurements that were taken in the mechanical engineering laboratory. A 3D finite element model was developed in ANSYS using approximate material properties for fiberglass obtained from the literature. The model was then validated by comparing its responses with those from a number of static tests, plus a simple impact test for comparison of the first natural frequency. Finally, a number of numerical tests were conducted on the validated finite element model to determine its structural responses. The purpose of the numerical analysis was to obtain the equivalent von Mises stress and deformation produced in the blade. It was determined that under the examined different loading conditions, a higher stress contour was found to occur around the mid-span of the blade. The calculated maximum flexural stress on the blade was observed to be less than the allowable flexural stress for fiberglass which is 1,770 MPa. As expected, the highest deformation occurred at blade tip. The first critical speed of the assembled three-bladed wind turbine was found to be at 4.3 rpm. The first mode shape was observed to be in the flap-wise bending direction and for a range of rotor speeds between zero and 608 rpm, three out of a total of five mode shapes were in the flap-wise bending direction. Future studies should address issues relating blade vibrations with generated power, validation of dynamic tests, fluid-structural interaction and introduction of bio-inspired blade system. Although the performance of the bioinspired blade has not been studied in great detail, preliminary studies indicate that this system has a superior performance. / Mechanical and Industrial Engineering / M. Tech. (Electrical and Mining Engineering)