Cette thèse analyse les challenges auxquels sont confrontés les composants électriques pour les systèmes de traction hybrides. L’analyse de ces composants et de leurs interactions en tant qu’entité indépendante est un sujet de recherche important afin de dimensionner de manière optimale le système au lieu de combiner des composants optimaux. Les véhicules hybrides sont un domaine de recherche qui suscite un grand intérêt parce qu’il s’agit d’une solution efficace à court terme afin de préparer la transition énergétique vers les véhicules à zéro émission. Malgré les avantages de cette solution, c’est un sujet de recherche complexe car les composants électriques doivent être intégrés dans un système de propulsion conventionnel. Ainsi le but de ce travail de recherche est axé sur la détermination de méthodes appropriées pour étudier les composants électriques et les contributions apportées par cette thèse visent à répondre à la problématique suivante : déterminer le niveau suffisant de détails pour modéliser les systèmes électriques pour les systèmes de traction pour véhicules hybrides afin d’identifier le dimensionnement idéal des composants pour différents systèmes pendant la phase de développement. Afin de résoudre cette problématique, ce rapport est divisé en quatre parties au sein de six chapitres. D’abord l’état de l’art des véhicules hybrides, des composants électriques ainsi que des méthodes d’optimisation associées sont présentés (chapitre 1). Ensuite, pour chaque composant (chapitre 2 à 4), des méthodes de modélisation appropriées sont déterminées afin de les modéliser mais aussi afin d’évaluer leur intégration dans le système de propulsion. Puis, une solution pour l’étude du système globale est déterminée à partir de l’analyse de travaux précédents (chapitre 5). Finalement, une approche d’optimisation est développée et permet d’analyser différents systèmes ainsi que l’influence de différents paramètres sur le dimensionnement (chapitre 6). Grâce à l’analyse du développement actuel et des travaux précédents sur le sujet ainsi qu’au développement d’outils de simulation, cette thèse étudie et analyse les relations entre le niveau de tension et de courant, et les performances du système dans différents cas. Les résultats permettent de déterminer l’influence de ces paramètres sur les composants ainsi que l’impact de l’environnement industriel sur les résultats. En tenant compte du cadre législatif actuel, les résultats convergent globalement tous dans la même direction : une réduction du niveau de tension, respectivement une augmentation du courant, entraine une amélioration du système global par rapport aux méthodes de dimensionnent actuelles. Ces observations sont liées à l’architecture, au cycle d’évaluation et à l’environnement considérés mais les méthodes et l’approche développée ont posé les bases pour étendre les connaissances dans le domaine de l’optimisation des véhicules hybrides. En plus de l’optimisation générale, des cas particuliers sont analysés afin de montrer la modularité des méthodes et l’influence de paramètres supplémentaires (système 48V ou convertisseur Boost). Afin de conclure, cette thèse a mis en place les bases pour l’étude des composants électriques pour les véhicules hybrides. De part un environnement fluctuant et les nombreuses technologies possibles, ce sujet suscite encore un grand intérêt et les points suivants peuvent être encore étudiés de manière plus détaillée : * Application des méthodes pour d’autres systèmes de propulsion (autre architectures hybrides, véhicule à pile à combustible ou tout électrique), * Étude de nouvelles technologies comme le carbure de silicium pour l’électronique de puissance, la machine à reluctance variable ou le sulfure de lithium pour les batteries, * Analyse d’autre cycle d’évaluation ainsi que leur cadre législatif, * Mise en place de structures additionnelles pour l’électronique de puissance, * Validations supplémentaires avec d’autres composants / This work analyses the challenges faced by the electric components for traction purpose in hybrid drivetrains. It investigates the components and their interactions as an independent entity in order to refine the scope of investigation and to find the best combinations of components instead of the best components combinations. Hybrid vehicle is currently a topic of high interest because it stands for a suitable short-term solution towards zero emission vehicle. Despite its advantages, it is a challenging topic because the components need to be integrated in a conventional drivetrain architecture. Therefore, the focus of this work is set on the determination of the right methods to investigate only the electric components for traction purpose. The aim and the contributions of this work lies thereby in the resolution of the following statement: Determine the sufficient level of details in modeling electric components at the system level and develop models and tools to perform dynamic simulations of these components and their interactions in a global system analysis to identify ideal designs of various drivetrain electric components during the design process. To address these challenges, this work is divided in four main parts within six chapters. First the current status of the hybrid vehicle, the electric components and the associated optimization methods and simulation are presented (first chapter). Then for each component, the right modeling approach is defined in order to investigate the electrical, mechanical and thermal behavior of the components as well as methods to evaluate their integration in the drivetrain (second to fourth chapter). After this, a suitable method is defined to evaluate the global system and to investigate the interactions between the components based on the review of relevant previous works (chapter five). Finally, the last chapter presents the optimization approach considered in this work and the results by analyzing different system and cases (chapter six). Thanks to the analysis of the current status, previous works and the development of the simulations tools, this work investigates the relationships between the voltage, the current and the power in different cases. The results enable, under the considered assumptions of the work, to determine the influence of these parameters on the components and of the industrial environment on the optimization results. Considering the current legislative frame, all the results converge toward the same observation referred to the reference systems: a reduction of the voltage and an increase of the current leads to an improvement of the integration and the performance of the system. These observations are linked with the considered architecture, driving cycle and development environment but the developed methods and approaches have set the basis to extend the knowledge for the optimization of the electric system for traction purpose. Beside the main optimization, special cases are investigated to show the influence of additional parameters (increase of the power, 48V-system, machine technology, boost-converter…) In order to conclude, this work have set the basis for further investigations about the electric components for traction purpose in more electrified vehicle. Due to the constantly changing environment, the new technologies and the various legislative frame, this topic remains of high interest and the following challenges still need to be deeper investigated: * Application of the methods for other drivetrain architecture (series hybrid, power-split hybrid, fuel-cell vehicle, full electric vehicle), * Investigation of new technologies such as silicon-carbide for the power electronics, lithium–sulfur battery or switch reluctance machine, * Investigation of other driving cycle, legislative frame, * Integration of additional power electronics structure, * Further validation of the modeling approaches with additional components

Este estudo tem por objetivo estabelecer os principais critérios de seleção de uma arquitetura elétrica adequada a mercados emergentes, considerando as necessidades de mercado, de legislação, de inovações e/ou novas funções a serem empregadas durante o ciclo de vida de um veículo. Para tanto, o resultado da pesquisa bibliográfica deste trabalho, apresentada na revisão teórica, primeiramente define o significado de arquitetura elétrica, e conta a sua evolução desde os primórdios da indústria automotiva até os dias de hoje, apontando quais são os principais sistemas elétricos existentes atualmente, deixando clara a fundamental importância da eletrônica embarcada na evolução do automóvel. Mostram, ainda, suas principais tendências tecnológicas, além de um descritivo sobre técnicas, critérios e metodologia de desenvolvimento de projetos de arquitetura elétrica empregados hoje na indústria. Na seqüência, o trabalho define os conceitos de mercado emergente e mercado desenvolvido, apontando suas particularidades, volumes de negócio (histórico e projetado), deixando evidente a importância dos mercados emergentes nos dias de hoje para a indústria automotiva mundial e a necessidade de se estabelecer estratégias específicas para estes mercados. Com base na própria pesquisa bibliográfica, somada à análise documental e a entrevistas não-estruturadas com membros ativos da indústria automotiva, foram concebidos os critérios para seleção de arquiteturas elétricas veiculares para mercados emergentes, assim como seus descritores e taxas de substituição (pesos). A definição de pesos e descritores é necessária para a utilização do método de agregação aditiva, método este de tomada de decisão empregado devido a sua capacidade de avaliar simultaneamente diversas alternativas, através de parâmetros qualitativos. Tal método, discutido na metodologia, mostrou-se muito eficiente por permitir uma avaliação dinâmica da análise proposta perante modificações em quaisquer parâmetros existentes. Por fim, o método e os critérios foram finalmente aplicados em dois estudos de caso, selecionados com base em casos reais vivenciados pela indústria automotiva no desenvolvimento de dois projetos: um veículo hatch popular e um veículo sedan médio de luxo, ambos destinados a mercados emergentes. Através do estudo de caso, foi possível confrontar soluções de arquiteturas adotadas no passado, no presente e para um futuro próximo, além de confrontar também soluções adotadas para mercados desenvolvidos e para mercados emergentes. O resultado deste estudo comprovou a necessidade da existência de critérios específicos para mercados emergentes, a coerência destes critérios e a robustez do modelo apresentado. Convém ressaltar que, para o tema em questão (ou estudos de caso), a robustez e consistência do modelo de avaliação obtido e a importância comprovada dos critérios selecionados são pertinentes e válidos, considerando-se o fato de que a definição dos pesos e notas atribuídas ao modelo serem feitas pelo avaliador, nas quais são diretamente dependentes do conhecimento adquirido pela empresa (seja ela montadora ou fornecedora) e vivência do responsável pela avaliação, sujeitando-as, desta forma, a seus próprios conceitos. / The aim of this study is to establish the criteria of election of an electrical architecture tailor-made for the emergent markets, considering the necessities of market, legislation, innovations and/or new functions to be applied during the whole vehicles life-cycle. Due to this, the result of the bibliographical research of this study, defines the meaning of electrical architecture, and shows its evolution since the beginning of the automotive industry up to now, also presenting the main electrical systems available today, making clear the huge importance of the embedded electronics for the evolution of the automobile. This study also presents the technological trends and a brief description of techniques, criteria and development methodology for electrical architecture projects available today in the industry. Here after, the study defines the concepts of emergent market and developed market, showing their characteristics, market share and production volumes (historical and forecast), showing the importance of the emergent markets for the world-wide automotive industry nowadays and the needs for development of specific strategies for these markets. Based on the bibliographical research, documentary analysis and the unstructured interviews with active members of the automotive industry, the criteria for election of electrical architectures for emergent markets was established, as well as their describers and swing-weights. The definition of weights and describers is necessary for the use of the additive aggregation method a decision taking method which has the capacity to evaluate several alternatives simultaneously, through qualitative parameters. Such method, presented in the methodology chapter, was very efficient for dynamic evaluation of the proposals even before changes in its parameters. Finally, the method and the criteria were finally applied in two studies of case, based on real cases of automotive industry when developing two projects: a low-cost hatch vehicle and a mid-size luxury sedan vehicle, both designed for emergent markets. By the study of cases, it was possible to compare solutions of architectures adopted in the past, in the present and for a near future, and also to compare solutions adopted for developed markets and emergent markets. The results of this study proved the necessity of specific criteria for emergent markets, the coherence of the established criteria and the robustness of the adopted model. Regarding the studies of cases, the robustness and consistency of the evaluation model and the proved importance of the criteria are relevant and valid, even considering that the definition of the weights values and grades are always made by the evaluators. It means that they directly depend on the companys know-how and best practices (from carmakers or suppliers) and the professional experience of the evaluator, resulting, someway, in their own values for different companies.

La faisabilité des applications à piles à combustible (PAC) a été largement démontrée à travers le monde. Les efforts de recherche portent actuellement sur l'amélioration de la durée de vie des PAC et la diminution de leur coût. A ce jour, une PAC ne tolère pas les variations rapides de charges qui entraînent très souvent son vieillissement prématuré. Pour pallier cette faiblesse, une hybridation avec un composant électrochimique de stockage (typiquement des supercondensateurs) est généralement proposée via un ou deux convertisseurs statiques, nécessitant l'implantation d'une gestion énergétique. Une partie de ces travaux se situe dans le contexte aéronautique et fait suite au projet européen CELINA piloté par AIRBUS. Le projet européen CELINA (2005-2008) a posé la problématique du remplacement de l'éolienne (RAT) actuelle par une pile à combustible pour le réseau électrique de dernier secours sollicité en cas de perte totale des moteurs ou de la génération électrique. Il alimente les charges essentielles : auxiliaires de puissances presque constantes (calculateurs de bord, …) et les actionneurs de vol (EHA, EMA) qui constituent les principaux consommateurs à caractères très intermittents. Cette étude a permis une classification de trois architectures, dont la validation expérimentale se situant dans le cadre du projet français ISS ayant débuté en 2010 sera exposée. L'hybridation directe entre une PAC et des supercondensateurs présente les avantages de ne pas mettre en jeu de convertisseur statique et d'une autogestion énergétique naturelle. Partant du constat que toutes les applications embarquées utilisant des PAC sont hybridées et qu'un développement d'architecture et de stratégies est effectué pour chaque cas (mise au point de convertisseurs, lois de commande, etc.). Ceci représente un travail considérable et systématique, ce qui freine l'implantation des systèmes PAC dans les applications embarquées. L'objectif est alors d'étudier la faisabilité d'un composant hybride unique jouant le rôle de source de puissance et d'énergie dont la gestion énergétique est transparente pour l'utilisateur et ne nécessitant pas l'ajout d'une hybridation supplémentaire pour ces applications. Cette thématique fait l'objet de ces travaux de thèse en collaboration avec la société française HELION Hydrogen Power. Après une présentation de l'introduction des systèmes PAC en aéronautique centrée autour de l'hybridation directe, la suite des travaux regroupe deux grandes thématiques : la première concerne l'étude des interactions entre PAC et supercondensateurs lors d'une association directe selon trois approches : théorique, expérimentale et par simulation. La seconde concerne la validation expérimentale de trois architectures d'hybridation d'un système PAC retenues pour un contexte aéronautique lors d'études précédentes au laboratoire : une architecture indirecte pour laquelle le stockage possède son convertisseur, une architecture indirecte avec stockage sur le bus DC et une architecture directe. L'objectif de ces travaux étant d'augmenter le niveau de maturité technologique de ces concepts, ainsi que de comparer les différents moyens retenus pour parvenir à l'hybridation d'un système PAC suivant des critères précis. / The feasibility of fuel cell (FC) applications has been demonstrated throughout the world. Research efforts are currently focused on improving the lifetime of the FC and reducing their cost. Until today, a FC does not tolerate rapid variations of load that cause in most cases lifetime reducing. To reduce this defect, hybridization with electrochemical storage component (typically ultracapacitors) is generally suggested via one or two static converters, requiring the implementation of an energy management. Aeronautic applications constitute the framework of these studies. They are the prolongation of studies initiated within the European project CELINA piloted by AIRBUS. The CELINA project (2005-2008) dealt with the replacement of the Ram Air Turbine (RAT) which is currently used for the last emergency electrical network in the case of total losses of engines or electrical generation. This emergency network has to supply the essential loads: the piloting auxiliaries (calculators…) consume a quasi-constant power, and the flight actuators (EHA, EMA) which are the main loads whose consumption is very intermittent. This study resulted in a classification of three architectures for which experimental validation in the framework of the French ISS project started since 2010 will be exposed. The direct hybridization between a FC and ultracapacitors has the advantages of not involving static converter and provide a natural energy management. The statement of facts is that all embedded applications using FC are hybridized, architecture and strategies development is performed for each case (development of converters, control laws, etc.). This represents a significant and systematic work, which limits the implementation of FC in embedded applications. In this work, the objective is to study the feasibility of a single hybrid component acting as a power and energy source for which energy management is transparent to the user and does not require the addition of another hybridization. This work is part of collaboration with the French company HELION Hydrogen Power. After a presentation of the insertion of FC in aeronautics centered on the direct hybridization, two major themes are approached: The first concerns the study of interactions between FC and ultracapacitors in a direct association according to three approaches: theoretical, experimental and simulation. The second concerns the experimental validation of three hybridization architectures for FC considered in previous studies in the laboratory: an architecture for which the indirect storage has its converter, an architecture with indirect storage on the DC bus and a direct hybridization architecture. The objective of this work is to increase the level of technological Readiness level of these concepts, and to compare the different ways considered to achieve the hybridization of a fuel cell system according to specific standards.

Este estudo tem por objetivo estabelecer os principais critérios de seleção de uma arquitetura elétrica adequada a mercados emergentes, considerando as necessidades de mercado, de legislação, de inovações e/ou novas funções a serem empregadas durante o ciclo de vida de um veículo. Para tanto, o resultado da pesquisa bibliográfica deste trabalho, apresentada na revisão teórica, primeiramente define o significado de arquitetura elétrica, e conta a sua evolução desde os primórdios da indústria automotiva até os dias de hoje, apontando quais são os principais sistemas elétricos existentes atualmente, deixando clara a fundamental importância da eletrônica embarcada na evolução do automóvel. Mostram, ainda, suas principais tendências tecnológicas, além de um descritivo sobre técnicas, critérios e metodologia de desenvolvimento de projetos de arquitetura elétrica empregados hoje na indústria. Na seqüência, o trabalho define os conceitos de mercado emergente e mercado desenvolvido, apontando suas particularidades, volumes de negócio (histórico e projetado), deixando evidente a importância dos mercados emergentes nos dias de hoje para a indústria automotiva mundial e a necessidade de se estabelecer estratégias específicas para estes mercados. Com base na própria pesquisa bibliográfica, somada à análise documental e a entrevistas não-estruturadas com membros ativos da indústria automotiva, foram concebidos os critérios para seleção de arquiteturas elétricas veiculares para mercados emergentes, assim como seus descritores e taxas de substituição (pesos). A definição de pesos e descritores é necessária para a utilização do método de agregação aditiva, método este de tomada de decisão empregado devido a sua capacidade de avaliar simultaneamente diversas alternativas, através de parâmetros qualitativos. Tal método, discutido na metodologia, mostrou-se muito eficiente por permitir uma avaliação dinâmica da análise proposta perante modificações em quaisquer parâmetros existentes. Por fim, o método e os critérios foram finalmente aplicados em dois estudos de caso, selecionados com base em casos reais vivenciados pela indústria automotiva no desenvolvimento de dois projetos: um veículo hatch popular e um veículo sedan médio de luxo, ambos destinados a mercados emergentes. Através do estudo de caso, foi possível confrontar soluções de arquiteturas adotadas no passado, no presente e para um futuro próximo, além de confrontar também soluções adotadas para mercados desenvolvidos e para mercados emergentes. O resultado deste estudo comprovou a necessidade da existência de critérios específicos para mercados emergentes, a coerência destes critérios e a robustez do modelo apresentado. Convém ressaltar que, para o tema em questão (ou estudos de caso), a robustez e consistência do modelo de avaliação obtido e a importância comprovada dos critérios selecionados são pertinentes e válidos, considerando-se o fato de que a definição dos pesos e notas atribuídas ao modelo serem feitas pelo avaliador, nas quais são diretamente dependentes do conhecimento adquirido pela empresa (seja ela montadora ou fornecedora) e vivência do responsável pela avaliação, sujeitando-as, desta forma, a seus próprios conceitos. / The aim of this study is to establish the criteria of election of an electrical architecture tailor-made for the emergent markets, considering the necessities of market, legislation, innovations and/or new functions to be applied during the whole vehicles life-cycle. Due to this, the result of the bibliographical research of this study, defines the meaning of electrical architecture, and shows its evolution since the beginning of the automotive industry up to now, also presenting the main electrical systems available today, making clear the huge importance of the embedded electronics for the evolution of the automobile. This study also presents the technological trends and a brief description of techniques, criteria and development methodology for electrical architecture projects available today in the industry. Here after, the study defines the concepts of emergent market and developed market, showing their characteristics, market share and production volumes (historical and forecast), showing the importance of the emergent markets for the world-wide automotive industry nowadays and the needs for development of specific strategies for these markets. Based on the bibliographical research, documentary analysis and the unstructured interviews with active members of the automotive industry, the criteria for election of electrical architectures for emergent markets was established, as well as their describers and swing-weights. The definition of weights and describers is necessary for the use of the additive aggregation method a decision taking method which has the capacity to evaluate several alternatives simultaneously, through qualitative parameters. Such method, presented in the methodology chapter, was very efficient for dynamic evaluation of the proposals even before changes in its parameters. Finally, the method and the criteria were finally applied in two studies of case, based on real cases of automotive industry when developing two projects: a low-cost hatch vehicle and a mid-size luxury sedan vehicle, both designed for emergent markets. By the study of cases, it was possible to compare solutions of architectures adopted in the past, in the present and for a near future, and also to compare solutions adopted for developed markets and emergent markets. The results of this study proved the necessity of specific criteria for emergent markets, the coherence of the established criteria and the robustness of the adopted model. Regarding the studies of cases, the robustness and consistency of the evaluation model and the proved importance of the criteria are relevant and valid, even considering that the definition of the weights values and grades are always made by the evaluators. It means that they directly depend on the companys know-how and best practices (from carmakers or suppliers) and the professional experience of the evaluator, resulting, someway, in their own values for different companies.