Les turbomachines aéronautiques sont composées de plusieurs roues aubagées dont la fonction estde transférer l’énergie de l’air au rotor. Les roues aubagées des modules compresseur et turbine sontdes pièces particulièrement sensibles car elles doivent répondre à des impératifs de performanceaérodynamique, de tenue mécanique, de tenue thermique et de performance acoustique. L’optimisation aéro-méca-acoustique ou aéro-thermo-mécanique des aubages consiste à chercher, pourun ensemble de formes aérodynamiques paramétrées (par plusieurs dizaines de variables), celleassurant le meilleur compromis entre la performance aérodynamique du moteur et la satisfactionde plusieurs dizaines de contraintes souvent contradictoires. Cette thèse introduit une méthode d’optimisation basée sur les métamodèles et adaptée à la grande dimension pour répondre à la problématique industrielle des aubages. Les contributions de cettethèse portent sur deux aspects : le développement de modèles de krigeage, et l’adaptation d’unestratégie d’optimisation pour la gestion du grand nombre de variables et de contraintes.La première partie de ce travail traite des modèles de krigeage. Nous avons proposé une nouvelleformulation du noyau de covariance permettant de réduire le nombre de paramètres du modèleafin d’accélérer sa construction. Une des limitations connues du modèle de krigeage concernel’estimation de ses paramètres. Cette estimation devient de plus en plus difficile lorsque nousaugmentons la dimension du phénomène à approcher. En particulier, la base de données nécessitedavantage de points et par conséquent la matrice de covariance du modèle du krigeage est de plusen plus coûteuse à inverser. Notre approche consiste à réduire le nombre de paramètres à estimer en utilisant la méthode de régression des moindres carrés partiels (PLS pour Partial Least Squares). Cette méthode de réduction dimensionnelle fournit des informations sur la relation linéaire entre les variables d’entrée et la variable de sortie. Ces informations ont été intégrées dans les noyaux du modèle de krigeage tout en conservant les propriétés de symétrie et de positivité des noyaux. Grâce à cette approche, la construction de ces nouveaux modèles appelés KPLS est très rapide étant donné le faible nombre de paramètres nécessaires à estimer. La validation de ces modèles KPLS sur des cas test académiques ou industriels a démontré leur qualité de prédiction équivalente voire même meilleure que celle des modèles de krigeage classiques. Dans le cas de noyaux de covariance de type exponentiel, laméthode KPLS peut être utilisée pour initialiser les paramètres du krigeage classique, afin d’accélérerla convergence de l’estimation des paramètres du modèle. La méthode résultante, notée KPLS+K, a permis d’améliorer la qualité des modèles dans le cas de fonctions fortement multimodales. La deuxième contribution de la thèse a consisté à développer une stratégie d’optimisation globale sous contraintes pour la grande dimension, en s’appuyant sur les modèles KPLS ou les modèlesKPLS+K. En effet, nous avons étendu la méthode d’optimisation auto-adaptative connue dans lalittérature sous le nom "Efficient Global Optimisation, EGO" pour gérer les problèmes d’optimisationsous contraintes en grande dimension. Différents critères d’enrichissement adaptatifs ont pu êtreexplorés. Cette stratégie a permis de retrouver l’optimum global sur des problèmes académiquesjusqu’à la dimension 50. La méthode proposée a été confrontée à deux types de problèmes industriels, le cas test MOPTA issu de l’industrie automobile (124 variables d’entrée et 68 fonctions contraintes) et le cas test Snecma des aubes de turbomachines (50 variables d’entrée et 31 fonctions contraintes). Les résultats ont permis de montrer la validité de la démarche ainsi que les limites de la méthode pour une application dans un cadre industriel. / Aerospace turbomachinery consists of a plurality of blades. Their main function is to transfer energybetween the air and the rotor. The bladed disks of the compressor are particularly important becausethey must satisfy both the requirements of aerodynamic performance and mechanical resistance.Mechanical and aerodynamic optimization of blades consists in searching for a set of parameterizedaerodynamic shape that ensures the best compromise solution between a set of constraints.This PhD introduces a surrogate-based optimization method well adapted to high-dimensionalproblems. This kind of high-dimensional problem is very similar to the Snecma’s problems. Ourmain contributions can be divided into two parts : Kriging models development and enhancementof an existing optimization method to handle high-dimensional problems under a large number ofconstraints. Concerning Kriging models, we propose a new formulation of covariance kernel which is able toreduce the number of hyper-parameters in order to accelerate the construction of the metamodel.One of the known limitations of Kriging models is about the estimation of its hyper-parameters.This estimation becomes more and more difficult when the number of dimension increases. Inparticular, the initial design of experiments (for surrogate modelling construction) requires animportant number of points and therefore the inversion of the covariance matrix becomes timeconsuming. Our approach consists in reducing the number of parameters to estimate using the Partial LeastSquares regression method (PLS). This method provides information about the linear relationshipbetween input and output variables. This information is integrated into the Kriging model kernelwhile maintaining the symmetry and the positivity properties of the kernels. Thanks to this approach,the construction of these new models called KPLS is very fast because of the low number of newparameters to estimate. When the covariance kernel used is of an exponential type, the KPLS methodcan be used to initialize parameters of classical Kriging models, to accelerate the convergence of theestimation of parameters. The final method, called KPLS+K, allows to improve the accuracy of themodel for multimodal functions. The second main contribution of this PhD is to develop a global optimization method to tacklehigh-dimensional problems under a large number of constraint functions thanks to KPLS or KPLS+Kmethod. Indeed, we extended the self adaptive optimization method called "Efficient Global Optimization,EGO" for high-dimensional problems under constraints. Several enriching criteria have been tested. This method allows to estimate known global optima on academic problems up to 50 inputvariables. The proposed method is tested on two industrial cases, the first one, "MOPTA", from the automotiveindustry (with 124 input variables and 68 constraint functions) and the second one is a turbineblade from Snecma company (with 50 input variables and 31 constraint functions). The results showthe effectiveness of the method to handle industrial problems.We also highlight some importantlimitations.

The Aurignacian is a contentious time period in paleoanthropology. The myriad social changes which accompany the Upper Paleolithic transition have often become associated with the physical tools which Aurignacian people left behind. One result of this is the current tendency of professionals to use blade technology as an indicator of "modernity," rather than examining how changes accompanying the Upper Paleolithic transition made blades a useful adaptation. Of particular importance is the fact that the adoption of blades coincides with a long distance shift in the system used to procure and transport the lithic raw materials. This suggests that before we can use blades to answer anthropological questions about the Aurignacian, we need to establish the relationship between blade production and the acquisition of exotic raw materials. This dissertation combines an analysis of the lithic collection from the French archaeological site of Abri Cellier with the experimental fracture of lithic raw material samples in order to examine the impact of raw material quality on Aurignacian blade production. The analysis of the assemblage from Abri Cellier demonstrates that Aurignacian blades manufactured on exotic materials were of higher quality than those produced locally. The experimental fracture of raw material samples reveals that the differences in the quality of the exotic and local materials do not sufficiently account for the differences in the quality of the blades produced on them. This implies that the differential transport of high quality final products accounts for the increased quality of exotic blades at Abri Cellier. This research examines a number of new ways to evaluate quality in the archaeological record. More importantly, however, it firmly demonstrates that the acquisition of long distance raw materials was not a prerequisite for blade production in the Perigord. This work will conclude by arguing that blades played a role in increasing the maintainability of a hafted toolkit geared towards meeting the requirements of an increasingly mobile and collaborative Aurignacian population.

Stratification of water in large reservoirs occurs in summer, or at anytime in hot climates where the water surface is exposed long-term to sunlight and the water surface is heated. Natural mixing will not occur due to the cooler and denser water always staying at the lower levels. Therefore, mechanical circulators are designed to prevent water quality problems related to stratification and depletion of dissolved oxygen. Impellers that produce the flow in mechanical circulators are available in different sizes and these impellers are designed to produce different flow rates. Due to hydraulic loadings, impellers have to be strong and durable. Loadings on impellers depend on their geometries and therefore, a durable impeller is a good combination of the use of correct materials and good geometry. Long and slender impellers are prone to failure when subjected to high hydrodynamic loadings. Nowadays, designers have very limited information on predicting the stresses on impellers and the deflection patterns of impellers because there are no design rules in designing these impeller blades and there is no such thing as "best geometry". A good impeller blade design is by guesswork and experience. In order to design the geometry that suits this application, trial-and-error finite element analyses have been conducted in this project to minimize stress levels on the blades. This research involves the use of finite element analysis (FEA) to predict stress and deflection of impeller blades used on large (5m diameter) ducted axial flow impellers as the first step in the design process. Then, based on the results, improvements have been done to the models until the final design was made. As far as the author has been able to determine, this has not been researched before. Finite Element Analysis has been used on wind turbine blades, rudders and hulls of boats but not on axial flow impeller blades of the type used in this project. For the purpose of this project, commercial finite element computer program packages STRAND6 and STRAND7 were used as the main analysis tools. A static line load increasing linearly with radius along the blade has been used to simulate the assumed hydrodynamic loading, and applied to all FEA blade models. The analysis results proved the stresses on blades are largely dependant on the blade geometry. From the analysis results, the author modified the stacking arrangement of the FEA elements in order to minimize both the tensile stresses and the displacements of the blades at the tip. Parametric studies have been done in order to obtain the best FEA impeller blade model.

For optimal rotor performance in a tiltrotor aircraft the difference in the inflow and the rotor speeds between the hover and cruise flight modes suggests different blade twist and chord distributions. The blade twist rates in current tiltrotor applications are defined based upon a compromise between the figure of merit in hover and propeller efficiency in airplane mode. However, when each operation mode is considered separately the optimum blade distributions are found to be considerably different. Passive blade twist control, which uses the inherent variation in centrifugal forces on a rotor blade to achieve optimum blade twist distributions in each flight mode through the use of extension-twist coupled composite rotor blades, has been considered for performance improvement of tiltrotor aircraft over the last two decades. The challenge for this concept is to achieve the desired twisting deformations in the rotor blade without altering the aeroelastic characteristics of the vehicle. A concept referred to as the sliding mass concept is proposed in this work in order to increase the twist change with rotor speed for a closed-cell composite rotor blade cross-section to practical levels for performance improvement in a tiltrotor aircraft. The concept is based on load path changes for the centrifugal forces by utilizing non-structural masses readily available on a conventional blade, such as the leading edge balancing mass. A multilevel optimization technique based on the simulated annealing method is applied to improve the performance of the XV15 tiltrotor aircraft. A cross-sectional analysis tool, VABS together with a multibody dynamics code, DYMORE are integrated into the optimization process. The optimization results revealed significant improvements in the power requirement in hover while preserving cruise efficiency. It is also shown that about 21% of the improvement is provided through the sliding mass concept pointing to the additional flexibility the concept provides for tailoring of the structure without any additional weight penalty on the system.

Rotorcraft optimization is a challenging problem due to its conflicting requirements among many disciplines and highly coupled design variables affecting the overall design. Also, the design process for a composite rotor blade is often ambiguous because of its design space. Furthermore, analytical tools do not produce acceptable results compared with flight test when it comes to aerodynamics and aeroelasticity unless realistic models are used, which leads to excessive computer time per iteration. To comply these requirements, computationally efficient yet realistic tools for rotorcraft analysis, such as VABS and DYMORE were used as analysis tools. These tools decompose a three-dimensional problem into a two-dimensional cross-sectional and a one-dimensional beam analysis. Also, to eliminate the human interaction between iterations, a previously VABS-ANSYS macro was modified and automated. The automated tool shortened the computer time needed to generate the VABS input file for each analysis from hours to seconds. MATLAB was used as the wrapper tool to integrate VABS, DYMORE and the VABS-ANSYS macro into the methodology. This methodology uses Genetic Algorithm and gradient-based methods as optimization schemes. The baseline model is the rotor system of generic Georgia Tech Helicopter (GTH), which is a three-bladed, soft-in-plane, bearingless rotor system. The resulting methodology is a two-level optimization, global and local. Previous studies showed that when stiffnesses are used as design variables in optimization, these values act as if they are independent and produce design requirements that cannot be achieved by local-level optimization. To force design variables at the global level to stay within the feasible design space of the local level, a surrogate model was adapted into the methodology. For the surrogate model, different ``design of experiments" (DOE) methods were tested to find the most computationally efficient DOE method. The response surface method (RSM) and Kriging were tested for the optimization problem. The results show that using the surrogate model speeds up the optimization process and the Kriging model shows superior performance over RSM models. As a result, the global-level optimizer produces requirements that the local optimizer can achieve.

The dynamic characteristics of shroud blade group played a significant role in steam turbine design. However, the complex shape and periodical structure of shroud blades make it so hard to find its dynamic characteristics under high speed operation. The complicate shape, periodic structure, and tedious computation limit the application of finite element method in the design analysis of shroud group blades. In order to design the shroud blade group, the component mode synthesis method was employed to derive the system dynamic equation of the grouped periodical blades. For simplicity, a pre-twisted and tapered cantilever beam is used to derive the approximate analytic solution of a rotating turbo blade. Then the approximated eigen solution of single blade is synthesized in company with the constrain condition by using the component mode synthesis method. In order to confirm the feasibility of the proposed simulation method, a real size turbine blade is used to discuss in the study. Through a comparison between the results solved from the proposed method and finite element method of single blade and shroud blade group to prove the reliability of the proposed method. The effect of blade parameters on the dynamic characteristic of shroud blade group has investigated in this work. Numerical results indicate the proposed method is feasible and effective in dynamic design analyses of the shroud blade group.

Conventional three dimensional structural analysis methods prove to be expensive for the preliminary design of wind turbine blades. However, wind turbine blades are large slender members with complex cross sections. They can be accurately modeled using beam models. The accuracy in the predictions of the structural behavior using beam models depends on the accuracy in the prediction of their effective section properties. Several techniques were proposed in the literature for predicting the effective section properties. Most of these existing techniques have limitations because of the assumptions made in their approaches. Two generalized beam theories, Generalized Timoshenko and Generalized Euler-Bernoulli, for the static analysis based on the principles of the simple 1D-theories are developed here. Homogenization based on the strain energy equivalence principle is employed to predict the effective properties for these generalized beam theories. Two efficient methods, Quasi-3D and Unit Cell, are developed which can accurately predict the 3D deformations in beams under the six fundamental deformation modes: extension, two shears, torsion and two flexures. These methods help in predicting the effective properties using the homogenization technique. Also they can recover the detailed 3D deformations from the predictions of 1D beam analysis. The developed tools can analyze two types of slender members 1) slender members with invariant geometric features along the length and 2) slender members with periodically varying geometric features along the length. Several configurations were analyzed for the effective section properties and the predictions were validated using the expensive 3D analysis, strength of materials and Variational Asymptotic Beam Section Analysis (VABS). The predictions from the new tools showed excellent agreement with full 3D analysis. The predictions from the strength of materials showed disagreement in shear and torsional properties. Explanations for the same are provided recalling the assumptions made in the strength of materials approach.

The necessity to develop an automated process for the design of micro-hydro power systems was based on the increasing demand for hydropower as a renewable energy source and to develop cost effective power supplies to rural areas. The application of the formula for the design of these systems is then to simplify the selection of the turbine sizing and is made possible by the similarity laws that exist within turbine and pump families. in addition the sizing of the supply and exhaust piping is also a matter of scaling. No selection process of turbine type is included due to the limitations of cost effectiveness and the category of size into which the turbine was specified. Furthermore. a new approach to turbine design was separately undertaken to satisfy low head and low flow-rate conditions. However, it was only designed up to a cost analysis with no manufacturing having been undertaken. The axial flow turbine. which was most suited for micro-hydro was designed and built as a prototype with a standardized mounting frame. The initial conditions used to generare the velocity vectors and angles were specific to the installation site and used to computationally generate the rotor and stator blades. This required an analysis of the different profiles available as well as research into their design. Once the blade profile stacking had been determined, this was translated into a software program that developed the blades from site-specific initial conditions. However, the design of the blades was interdependent on the dimensioning of the rest of the turbine components and designing these in parallel proved to be an intricate task. With the design complete, the turbine was then installed and testing proceeded with the use of pressure gauges and the results of torque and rpm obtained from a dynamometer. Analysis of the results was undertaken and presented in graphical format with comments on both the design and results. / Thesis (M.Sc.Eng.) -University of Natal, Durban, 2000.

Designing horizontal-axis wind turbine (HAWT) blades to achieve satisfactory levels of performance starts with knowledge of the aerodynamic forces acting on the blades. In this thesis, HAWT blade design is studied from the aspect of aerodynamic view and the basic principles of the aerodynamic behaviors of HAWTs are investigated. Blade-element momentum theory (BEM) known as also strip theory, which is the current mainstay of aerodynamic design and analysis of HAWT blades, is used for HAWT blade design in this thesis. Firstly, blade design procedure for an optimum rotor according to BEM theory is performed. Then designed blade shape is modified such that modified blade will be lightly loaded regarding the highly loaded of the designed blade and power prediction of modified blade is analyzed. When the designed blade shape is modified, it is seen that the power extracted from the wind is reduced about 10% and the length of modified blade is increased about 5% for the same required power. BLADESIGN which is a user-interface computer program for HAWT blade design is written. It gives blade geometry parameters (chord-length and twist distributions) and design conditions (design tip-speed ratio, design power coefficient and rotor diameter) for the following inputs / power required from a turbine, number of blades, design wind velocity and blade profile type (airfoil type). The program can be used by anyone who may not be intimately concerned with the concepts of blade design procedure and the results taken from the program can be used for further studies.

A modular structural design methodology for composite blades is developed. This design method can be used to design composite rotor blades with sophisticate geometric cross-sections. This design method hierarchically decomposed the highly-coupled interdisciplinary rotor analysis into global and local levels. In the global level, aeroelastic response analysis and rotor trim are conduced based on multi-body dynamic models. In the local level, variational asymptotic beam sectional analysis methods are used for the equivalent one-dimensional beam properties. Compared with traditional design methodology, the proposed method is more efficient and accurate. Then, the proposed method is used to study three different design problems that have not been investigated before. The first is to add manufacturing constraints into design optimization. The introduction of manufacturing constraints complicates the optimization process. However, the design with manufacturing constraints benefits the manufacturing process and reduces the risk of violating major performance constraints. Next, a new design procedure for structural design against fatigue failure is proposed. This procedure combines the fatigue analysis with the optimization process. The durability or fatigue analysis employs a strength-based model. The design is subject to stiffness, frequency, and durability constraints. Finally, the manufacturing uncertainty impacts on rotor blade aeroelastic behavior are investigated, and a probabilistic design method is proposed to control the impacts of uncertainty on blade structural performance. The uncertainty factors include dimensions, shapes, material properties, and service loads.