Advances in numerical optimization have raised the possibility that efficient and novel aircraft configurations may be ``discovered'' by an algorithm. To begin exploring this possibility, a fast and robust set of tools for aerodynamic shape optimization is developed. Parameterization and mesh-movement are integrated to accommodate large changes in the geometry. This integrated approach uses a coarse B-spline control grid to represent the geometry and move the computational mesh; consequently, the mesh-movement algorithm is two to three orders faster than a node-based linear elasticity approach, without compromising mesh quality. Aerodynamic analysis is performed using a flow solver for the Euler equations. The governing equations are discretized using summation-by-parts finite-difference operators and simultaneous approximation terms, which permit nonsmooth mesh continuity at block interfaces. The discretization results in a set of nonlinear algebraic equations, which are solved using an efficient parallel Newton-Krylov-Schur strategy. A gradient-based optimization algorithm is adopted. The gradient is evaluated using adjoint variables for the flow and mesh equations in a sequential approach. The flow adjoint equations are solved using a novel variant of the Krylov solver GCROT. This variant of GCROT is flexible to take advantage of non-stationary preconditioners and is shown to outperform restarted flexible GMRES. The aerodynamic optimizer is applied to several studies of induced-drag minimization. An elliptical lift distribution is recovered by varying spanwise twist, thereby validating the algorithm. Planform optimization based on the Euler equations produces a nonelliptical lift distribution, in contrast with the predictions of lifting-line theory. A study of spanwise vertical shape optimization confirms that a winglet-up configuration is more efficient than a winglet-down configuration. A split-tip geometry is used to explore nonlinear wake-wing interactions: the optimized split-tip demonstrates a significant reduction in induced drag relative to a single-tip wing. Finally, the optimal spanwise loading for a box-wing configuration is investigated.

La prochaine génération d’avions civil sera probablement une révolution en termede configuration d’avion, différant largement de l’architecture désormais classique “fuselage- ailes- moteurs sous voilure”. Du point de vue des qualités de vol, la tendance actuelle est d’évoluer versdes avions de moins en moins stables, à la fois en longitudinal et latéral. Il est dès lors probableque les futurs avions ne seront pas directement contrôlables par un humain sans l’apport de lois decommande stabilisantes. Il devient alors nécessaire de considérer l’apport des systèmes de commandesde vol très tôt dans la conception de l’avion, notamment pour le dimensionnement desempennages, gouvernes et actionneurs, contrairement au processus actuel qui ne prend principalementen compte que des critères “boucle ouverte” d’équilibre en phase de conception préliminaire.Plutôt qu’un processus itératif de dimensionnement puis synthèse de lois de commande, nousproposons d’optimiser simultanément les tailles de gouvernes, actionneurs et commandes de volen tenant compte des instabilités longitudinales et latérales, ainsi que des contraintes industriellessur la structure de correcteurs, sur un cas d’application de type aile volante. Ce processus de“co-design” permet de dimensionner des paramètres physiques de l’avion en tenant compte desapports d’une boucle de retour pour contrer des perturbations externes telles que de la turbulenceatmosphérique, permettant un avion plus sûr et optimal. / Next generation of civil transport aircraft is likely to be a radical change in overallconfiguration compared to traditional tube-and-wing design. From a handling qualities perspective,current trend in modern airliners is to evolve towards more and more unstable aircraft, bothfrom longitudinal and lateral-directional point of view. As a consequence future aircraft may notbe controllable by human operator without stabilizing control laws. It then becomes necessaryto consider flight control systems contribution early in the design phase for control surfaces,empennages and actuators sizing, as opposed to traditional way of working dealing only withopen-loop criteria for preliminary sizing. Instead of an iterative process of sizing and controllaws synthesis, we propose to concurrently optimize control surfaces, actuators and flight controllaws taking into account longitudinal and lateral instability as well as industrial structure forcontrollers, for unstable configurations such as Blended Wing-Body (BWB). This “co-design”procedure enables sizing of physical aircraft parameters taking into account benefits from feedbackstabilization for counteracting external disturbance such as atmospheric turbulence, thus leadingto safer and more optimal aircraft configurations.

Envisagée à partir des 15 à 30 années à venir dans l'industrie aéronautique, la propulsion hybrideélectrique permet d'intégrer de nouvelles briques technologiques offrant des degrés de libertésupplémentaires pour améliorer les performances des aéronefs, limiter l'utilisation de ressourcesfossiles et réduire l’impact environnemental des avions. Aujourd'hui, la technologie hybrideélectrique est principalement appliquée aux transports terrestres, aux voitures, aux bus et auxtrains, mais aussi aux navires. La faisabilité pour le transport aérien doit encore être établie etl'amélioration des performances des aéronefs reste à démontrer. Cette thèse vise à évaluer lesgains énergétiques apportés par l'hybridation électrique d'un avion régional de 70 places. Toutd'abord, les opportunités d'économie d'énergie sont identifiées à partir de l'analyse desrendements propulsifs et aérodynamiques d'un avion bi-turbopropulsé conventionnel. Les gainspotentiels provenant de la variation de la taille des moteurs principaux et de nouvelles gestions depuissance par l'utilisation de batteries sont étudiés. De plus, les possibles améliorationsaérodynamiques émanant de nouveaux positionnements des hélices sont considérées. Pourchaque sujet, des analyses simplifiées fournissent une estimation d'économie d'énergie. Cesrésultats sont ensuite utilisés pour sélectionner quatre systèmes propulsifs électrifiés qui sontétudiés plus en détail dans la thèse: un hybride parallèle, un turboélectrique avec propulsiondistribuée, un turboélectrique partiel à hélices hypersustentatrices, et un tout-électrique.L'évaluation des avions hybrides électriques sélectionnés est d'autant plus difficile que ledimensionnement des différentes composants, les stratégies de gestion d'énergie et les profils demission que l'on peut imaginer sont nombreux et variés. En outre, le processus global deconception de l'avion et les outils d'évaluation doivent être adaptés en conséquence. L'outilinterne de conception par optimisation multidisciplinaire d'Airbus nommé XMDO, qui inclut laplupart des modifications requises, est finalement sélectionné et développé au cours de la thèse.Par exemple, de nouveaux modèles paramétriques de composants (voilure soufflée, moteurélectrique, turbine à gaz, hélice, etc...) sont créés, une formulation générique pour résoudrel'équilibre du système de propulsion est mise en place, et les modèles de simulation de décollageet d'atterrissage sont améliorés. Afin d'évaluer l'efficacité énergétique des avions hybridesélectriques, un avion de référence équipé d'un système propulsif conventionnel est d'abordoptimisé avec XMDO. Différents algorithmes d'optimisation sont testés, et la consistance de lanouvelle méthode de conception est vérifiée. Par la suite, les configurations hybrides électriquessont toutes optimisées selon les mêmes exigences de conception que l'avion de référence. Pourles composants électriques, deux niveaux de technologie sont définis selon la date d'entrée enservice de l'aéronef. Les résultats d'optimisation pour le turbo-électrique et le turbo-électriquepartiel sont utilisés pour mieux appréhender les gains aérodynamiques potentiels identifiés enpremière partie de thèse. Les optimisations pour l'hybride parallèle, comprenant différentsscénarios de recharge batterie, mettent en évidence les meilleures stratégies de gestion d'énergielorsque des batteries sont utilisées comme sources d'énergie secondaire. Tous les résultats sontfinalement comparés à la référence en termes de consommations de carburant et d'énergie, pourles deux niveaux de technologie électrique. La dernière partie de la thèse se concentre sur l'aviontout électrique. Elle vise à identifier l'énergie spécifique minimale requise pour les batteries enfonction de la distance maximale à parcourir. Une étude de sensibilité est également réalisée enfonction de la date d'entrée en service pour les autres composants électriques / Envisioned in the next 15 to 30 years in the aviation industry, hybrid-electric propulsion offers theopportunity to integrate new technology bricks providing additional degrees of freedom to improveoverall aircraft performance, limit the use of non-renewable fossil resources and reduce the aircraftenvironmental footprint. Today, hybrid-electric technology has mainly been applied to groundbased transports, cars, buses and trains, but also ships. The feasibility in the air industry has to beestablished and the improvement in aircraft performance has still to be demonstrated. This thesisaims to evaluate the energy savings enabled by electric power in the case of a 70-seat regionalaircraft. First, energy saving opportunities are identified from the analysis of the propulsion andaerodynamic efficiencies of a conventional twin turboprop aircraft. The potential benefits comingfrom the variation of the size of prime movers and the new power managements with the use ofbatteries are studied. Also, possible aerodynamic improvements enabled by new propellerintegrations are considered. For each topic, simplified analyses provide estimated potential ofenergy saving. These results are then used to select four electrified propulsion systems that arestudied in more detail in the thesis: a parallel-hybrid, a turboelectric with distributed propulsion, apartial-turboelectric with high-lift propellers and an all-electric. Evaluating the selected hybrid-electric aircraft is even more challenging that the sizing of the different components, the energymanagement strategies and the mission profiles one can imagine are many and varied. Inaddition, the overall aircraft design process and the evaluation tools need to be adaptedaccordingly. The Airbus in-house Multidisciplinary Design Optimisation platform named XMDO,which includes most of the required modifications, is eventually selected and further developedduring the thesis. For examples, new parametric component models (blown wing, electrical motor,gas turbine, propeller, etc…) are created, a generic formulation for solving the propulsion systemequilibrium is implemented, and simulation models for take-off and landing are improved. In orderto evaluate the energy efficiency of the hybrid-electric aircraft, a reference aircraft equipped with aconventional propulsion system is first optimised with XMDO. Different optimisation algorithms aretested, and the consistency of the new design method is checked. Then, all the hybrid-electricconfigurations are optimised under the same aircraft design requirements as the reference. Forthe electrical components, two levels of technology are defined regarding the service entry date ofthe aircraft. The optimisation results for the turboelectric and the partial-turboelectric are used tobetter understand the potential aerodynamic improvements identified in the first part of the thesis.Optimisations for the parallel-hybrid, including different battery recharge scenarios, highlight thebest energy management strategies when batteries are used as secondary energy sources. All theresults are finally compared to the reference in terms of fuel and energy efficiencies, for the twoelectrical technology levels. The last part of the thesis focuses on the all-electric aircraft, and aimsat identifying the minimum specific energy required for batteries as a function of the aircraft designrange. A trade study is also carried-out in accordance with the service entry date for the otherelectrical components

This master thesis work presents the development of a parameterized automated generic model for the structural design of an aircraft wing. Furthermore, in order to perform finite element analysis on the aircraft wing geometry, the process of finite element mesh generation is automated. Aircraft conceptual design is inherently a multi-disciplinary design process which involves a large number of disciplines and expertise. In this thesis work, it is investigated how high-end CAD software‟s can be used in the early stages of an aircraft design process, especially for the design of an aircraft wing and its structural entities wing spars and wing ribs. The generic model that is developed in this regard is able to automate the process of creation and modification of the aircraft wing geometry based on a series of parameters which define the geometrical characteristics of wing panels, wing spars and wing ribs.Two different approaches are used for the creation of the generic model of an aircraft wing which are “Knowledge Pattern” and “PowerCopy with Visual Basic Scripting” using the CATIA V5 software. A performance comparison of the generic wing model based on these two approaches is also performed. In the early stages of the aircraft design process, an estimate of the structural characteristic of the aircraft wing is desirable for which a surface structural analysis (using 2D mesh elements) is more suitable. In this regard, the process of finite element mesh generation for the generic wing model is automated. The finite element mesh is generated for the wing panels, wing spars and wing ribs. Furthermore, the finite element mesh is updated based on any changes in geometry and the shape of the wing panels, wing spars or wing ribs, and ensure that all the mesh elements are always properly connected at the nodes. The automated FE mesh generated can be used for performing the structural analysis on an aircraft wing.

Sonic boom annoyance is an important technical showstopper for commercial supersonic aircraft operations. It has been proposed that aircraft can be shaped to alleviate sonic boom. Choosing the right aircraft shape reflecting the design requirements is a fundamental and most important step that is usually over simplified in the conceptual stages of design by resorting to a qualitative selection of a baseline configuration based on historical designs and designers perspective. Final aircraft designs are attempted by minor shape modifications to this baseline configuration. This procedure may not yield large improvements in the objectives, especially when the baseline is chosen without a rigorous analysis procedure. Traditional analyses and implementations tend to have a complex algorithmic flow, tight coupling between tools used and computational limitations. Some of these shortcomings are overcome in this study and a diverse mix of tools is seamlessly integrated to provide a simple, yet powerful and automatic procedure for sonic boom minimization. A shape optimization procedure for supersonic aircraft design using better geometry generation and improved analysis tools has been successfully demonstrated. The geometry engine provides dynamic reconfiguration and efficient manipulation of various components to yield unstructured watertight geometries. The architecture supports an assimilation of different components and allows configuration changes to be made quickly and efficiently because changes are localized to each component. It also enables an automatic way to combine linear and non-linear analyses tools. It has been shown in this study that varying atmospheric conditions could have a huge impact on the sonic boom annoyance metrics and a quick way of obtaining probability estimates of relevant metrics was demonstrated. The well-accepted theoretical sonic boom minimization equations are generalized to a new form and the relevant equations are derived to yield increased flexibility in aircraft design process. Optimum aircraft shapes are obtained in the conceptual design stages weighing in various conflicting objectives. The unique shape optimization procedure in conjunction with parallel genetic algorithms improves the computational time of the analysis and allows quick exploration of the vast design space. The salient features of the final designs are explained. Future research recommendations are made.

This dissertation investigates the development of a methodology suitable for the evaluation of advanced propulsion concepts. At early stages of development, both the future performance of these concepts and their requirements are highly uncertain, making it difficult to forecast their future value. A systematic methodology to identify potential advanced propulsion concepts and assess their robustness is necessary to reduce the risk of developing advanced propulsion concepts. Existing advanced design methodologies have evaluated the robustness of technologies or concepts to variations in requirements, but they are not suitable to evaluate a large number of dissimilar concepts. Variations in requirements have been shown to impact the development of advanced propulsion concepts, and any method designed to evaluate these concepts must incorporate the possible variations of the requirements into the assessment. In order to do so, a methodology had to do two things. First, it had to systemically identify a probabilistic distribution for the future requirements. Such a distribution would allow decision-makers to quantify the uncertainty introduced by variations in requirements. Second, the methodology must assess the robustness of the propulsion concepts as a function of that distribution. These enabling elements have been synthesized into new methodology, the Evolving Requirements Technology Assessment (ERTA) method. The ERTA method was used to evaluate and compare advanced propulsion systems as possible power systems for a hurricane tracking, High Altitude, Long Endurance (HALE) unmanned aerial vehicle (UAV). The problem served as a good demonstration of the ERTA methodology because conventional propulsion systems will not be sufficient to power the UAV, but the requirements for such a vehicle are still uncertain.

<p>Throughout the development of this report there will be a brief explanation of what the actual Aircraft Design Process is and in which stages the methodology that the authors are proposing will be implemented as well as the tools that will interact to produce this methodology.</p><p>The proposed tool will be the first part of a methodology that, according to the authors, by integrating separate tools that are currently used in different stages of the aeronautical design, will promote a decrease in the time frame for the initial stages of the design process.</p><p>The first part of the methodology above, that is proposed in this project, starts by creating a computer generated aircraft model and analyzing its basic aerodynamic characteristics “Lift Coefficient” and “Induced Drag Coefficient”, this step will be an alternative to statistical and empirical methods used in the industry, which require vast amount of data.</p><p>This task will be done in several steps, which will transfer the parametric aircraft model to an input file for the aerodynamic analysis program. To transfer the data a “translation” program has been developed that arranges the geometry and prepares the input file for analysis.</p><p>During the course of this report the reader will find references to existing aircrafts, such as the MD-11 or Airbus 310. However, these references are not intended to be an exact computer model of the mentioned airplanes. The authors are using this as reference so the reader can relate what he/she is seeing in this paper to existing aircrafts. By doing such comparison, the author intends to demonstrate that the Parametric Model that has been created possesses the capability to simulate to some extend the shape of existing aircrafts.</p><p>Finally from the results of this project it is concluded that the methodology in question is promising. Linking the two programs is possible and the aerodynamic characteristics of the models tested fall in the appropriate range. None the less the research must continue following the line that has been discussed in this report.</p>

Throughout the development of this report there will be a brief explanation of what the actual Aircraft Design Process is and in which stages the methodology that the authors are proposing will be implemented as well as the tools that will interact to produce this methodology. The proposed tool will be the first part of a methodology that, according to the authors, by integrating separate tools that are currently used in different stages of the aeronautical design, will promote a decrease in the time frame for the initial stages of the design process. The first part of the methodology above, that is proposed in this project, starts by creating a computer generated aircraft model and analyzing its basic aerodynamic characteristics “Lift Coefficient” and “Induced Drag Coefficient”, this step will be an alternative to statistical and empirical methods used in the industry, which require vast amount of data. This task will be done in several steps, which will transfer the parametric aircraft model to an input file for the aerodynamic analysis program. To transfer the data a “translation” program has been developed that arranges the geometry and prepares the input file for analysis. During the course of this report the reader will find references to existing aircrafts, such as the MD-11 or Airbus 310. However, these references are not intended to be an exact computer model of the mentioned airplanes. The authors are using this as reference so the reader can relate what he/she is seeing in this paper to existing aircrafts. By doing such comparison, the author intends to demonstrate that the Parametric Model that has been created possesses the capability to simulate to some extend the shape of existing aircrafts. Finally from the results of this project it is concluded that the methodology in question is promising. Linking the two programs is possible and the aerodynamic characteristics of the models tested fall in the appropriate range. None the less the research must continue following the line that has been discussed in this report.

Multidisciplinary Design Optimization (MDO) is an area in engineering design which has been growing rapidly in terms of applications in the last few decades, aircraft design being no exception to that. The application of MDO to aircraft and more specifically, wing design, presents many challenges, since disciplines like aerodynamics and structures have to be combined and interact. The level to which this interaction is implemented depends only on how much one is willing to pay in terms of computational cost. The objective of the current work is therefore to develop a simplified MDO tool, suitable for the preliminary design of aircraft wings. At the same time, versatility in the definition of optimization problems (in terms of design variables, constraints and objective function) is given great attention. At the same time, modularity will ensure that this framework is upgradeable with higher-fidelity and/or more capable modules. The disciplines that were chosen for interaction were aerodynamics and structures/ aeroelasticity, though more data can be extracted from their results in order to perform other types of analyses. The aerodynamics module employs a Vortex Lattice code developed specifically for the current implementation of the tool. The structural module is based on Equivalent Plate model theory. The fluid structure interaction is simply one-way, wherein the aerodynamics loads are passed on to the structural analyzer for computation of the static deformation. Semi-empirical relations are then used to estimate the flutter speed. The optimizer, which controls the activity of the other modules, makes use of a gradient based algorithm (Sequential Quadratic Programming) to search for a local minimum of a user defined objective function. Among the myriad of MDO strategies available, two are chosen to exemplify the modularity of the tool developed: Multidiscipline Feasible (MDF) and Sequential Optimization (SO), and their results are compared. Several case studies are analyzed to cover a broad spectrum of the capabilities of the framework. Because user interaction is of prime concern in design optimization, a graphical interface (GUI) of the tool is presented. Its advantages in terms of the set up of optimization problems and post-processing of results are made clear. In conclusion, some topics for future work regarding the expansion and improvement of the features of the application are noted.

A robustness and reliability based multidisciplinary analysis and optimization framework for aircraft design is presented. Robust design optimization and Reliability Based Design Optimization are merged into a uni ed formulation which streamlines the setup of optimization problems and aims at preventing foreseeable implementation issues in uncertainty based design. Surrogate models are evaluated to circumvent the intensive computations resulting from using direct evaluation in nondeterministic optimization. Three types of models are implemented in the framework: quadratic interpolation, regression Kriging and artificial neural networks. Regression Kriging presents the best compromise between performance and accuracy in deterministic wing design problems. The performance of the simultaneous implementation of robustness and reliability is evaluated using simple analytic problems and more complex wing design problems, revealing that performance benefits can still be achieved while satisfying probabilistic constraints rather than the simpler (and not as computationally intensive) robust constraints. The latter are proven to to be unable to follow a reliability constraint as uncertainty in the input variables increases. The computational effort of the reliability analysis is further reduced through the implementation of a coordinate change in the respective optimization sub-problem. The computational tool developed is a standalone application and it presents a user-friendly graphical user interface. The multidisciplinary analysis and design optimization tool includes modules for aerodynamics, structural, aeroelastic and cost analysis, that can be used either individually or coupled.