L'objectif des travaux présentés dans ce mémoire concerne le développement d'un modèle global représentant le couplage de l'enveloppe du bâtiment avec les équipements énergétiques. Une approche systémique appelée les Bond Graphs, peu employée jusqu'ici dans la modélisation des systèmes thermiques, est utilisée. Le modèle global du bâtiment, regroupant sous le même environnement de simulation, les modèles de l'enveloppe du bâtiment, les apports solaires, les émetteurs de chauffage et de rafraîchissement et le système de ventilation, est développé pour reconstituer l'ensemble des articulations énergétiques entre l'enveloppe et les environnements intérieur et extérieur. A travers la modélisation d'un bâtiment multizone, le couplage systémique des modèles de l'enveloppe et des apports solaires est présenté. Par ailleurs, un système combinant un plancher chauffant et un plafond rafraîchissant est étudié à l'aide des modèles des émetteurs de chauffage et de rafraîchissement. Le renouvèlement d'air dans le bâtiment est également concerné par la modélisation Bond Graph. Enfin, des éléments de validation expérimentale sont présentés. Pour cela, la plateforme de tri-génération d'énergie ENERBAT est exploitée. L'objectif est d'étudier le couplage optimal enveloppe du bâtiment - équipements énergétiques pour lequel les modèles BG sont développés. Une étude paramétrique tenant compte des interactions entre les paramètres étudiés est menée sur un projet réel de rénovation. Finalement, une combinaison appropriée des paramètres étudiés a été retenue afin de réduire la consommation énergétique selon la réglementation thermique française (RT2012) / Our works focus on the setting of reliable tools able to analyze the interaction between the building envelope and HVAC systems. The developed approach is based on Bond Graphs methodology, a graphical modeling language which is particularly suitable for energy exchanges. A numerical model gathering, under the same simulation environment, sub-models representing the building envelope, the solar gains, the floor heating, the chilled ceiling and the ventilation system, is developed in order to predict the energy interactions between these sub-systems. The multi-zone building model is developed in order to simulate and analyze the overall building thermal behavior. Then, the solar gains model is also included to predict the solar radiation exchanges in a way close to reality. The model of the heating and cooling system, combining the floor heating and the chilled ceiling, is developed in order to improve the thermal comfort of the building. Afterwards, the ventilation system is modeled in order to represent the air exchange inside the building. The experimental validation is carried out on the tri-generation unit integrated with a thermal solar system (platform ENERBAT). Furthermore, the parametrical study was realized in order to gain a better understanding according to the impact of some factors in the energy performance of the single-family building located in Meurthe-et-Moselle region (France). Optimization of several measures, such as insulation of the building envelope, type of glazing, building orientation and ventilation system, is performed to respond to the requirements of the French thermal standard (RT2012)

Cette thèse a pour but l'étude de la propriété de platitude sur des modèles bond graphs (BGs) non linéaires et de contribuer à la résolution des problèmes rencontrés en pratique qui sont liés principalement à l'identification des sorties plates et le calcul de la paramétrisation différentielle. <br />Pour atteindre cet objectif, de nouveaux concepts et outils graphiques ont été introduits. En particulier, grâce à l'introduction de la notion de modèle BG tangent ou variationnel à l'aide de l'utilisation des différentielles de Kähler, il est possible de calculer les sorties plates d'un modèle BG non linéaire par intégration des bases du module qui lui est associé.<br />Par ailleurs, en définissant la notion d'anneau BG non commutatif, une nouvelle règle de gain connue sous le nom de "règle de Riegle" est introduite en BG. En montrant alors qu'un modèle BG variationnel est un cas particulier d'anneau BG non commutatif, l'obtention graphique de la paramétrisation différentielle en utilisant la règle de Riegle et la notion de bicausalité est rendue possible.<br />Enfin, pour aller plus loin dans l'introduction de l'outil d'algèbre et de modules différentiels aux BGs, le cas des modèles BGs non linéaires régis par des équations différentielles polynômiales a été abordé. Dans ce contexte, le BG permet de faire une analyse directe des propriétés principales du système telles que le choix des variables d'entrée, les dynamiques correspondant à un choix d'entrée, le calcul des degrés de transcendance (non différentiel) différentiel, etc. à partir de son modèle BG associé. Il est également montré que la règle graphique de Riegle peut être étendue à cette classe de modèles BGs.

Modelling and simulation are essential ingredients of the analysis and design process in power electronics. It helps a design engineer gain an increased understanding of circuit operation. Accordingly, for a set of specifications given, the designer will choose a particular topology, select component types and values, estimate circuit performance etc. Typically hierarchical modelling, analysis and simulation rather than full detailed simulation of the system provides a crucial insight and understanding. The combination of these insights with hardware prototyping and experiments constitutes a powerful and effective approach to design. Obtaining the mathematical model of the power electronic systems is a major task before any analysis or synthesis or simulation can be performed. There are circuit oriented simulators which uses inbuilt mathematical models for components. Simulation with equation solver needs mathematical models for simulation which are trimmed according to user requirement. There are various methods in the literature to obtain these mathematical models. However, the issues of multi-domain system modelling and causality of the energy variables are not sufficiently addressed. Further, specifically to power converter systems, the issue of switching power models with fixed causality is not addressed. Therefore, our research focuses on obtaining solutions to the above using relatively untouched bond graph method to obtain models for power electronic systems. The power electronic system chosen for the present work is Switched Mode Power Converters (SMPC’s) and in particular PWM DC-DC converters. Bond graph is a labelled and directed graphical representation of physical systems. The basis of bond graph modelling is energy/power flow in a system. As energy or power flow is the underlying principle for bond graph modelling, there is seamless integration across multiple domains. As a consequence, different domains (such as electrical, mechanical, thermal, fluid, magnetic etc.) can be represented in a unified way. The power or the energy flow is represented by a half arrow, which is called the power bond or the energy bond. The causality for each bond is a significant issue that is inherently addressed in bond graph modelling. As every bond involves two power variables, the decision of setting the cause variable and the effect variable is by natural laws. This has a significant bearing in the resulting state equations of the system. Proper assignment of power direction resolves the sign-placing problem when connecting sub-model structures. The causality will dictate whether a specific power variable is a cause or the effect. Using causal bars on either ends of the power bond, graphically indicate the causality for every bond. Once the causality gets assigned, bond graph displays the structure of state space equations explicitly. The first problem we have encountered in modelling power electronic systems with bond graph is power switching. The essential part of any switched power electronic system is a switch. Switching in the power electronic circuits causes change in the structure of the system. This results in change in dynamic equations of the circuit according to position of the switch. We have proposed the switched power junctions (SPJ) to represent switching phenomena in power electronic systems. The switched power junctions are a generalization of the already existing 0-junction and 1-junction concepts of the bond graph element set. The SPJ’s models ideal switching. These elements maintain causality invariance for the whole system for any operational mode of the system. This means that the state vector of the resulting state equation of the system does not change for any operating mode. As SPJs models ideal power switching, the problem of stiff systems and associated numerical stability problems while simulating the system is eliminated. Further, it maintains one to one correspondence with the physical system displaying all the feasible modes of operation at the same time on the same graph. Using these elements, the switched mode power converters (SMPC's) are modelled in bond graph. Bond graph of the converter is the large signal model of the converter. A graphical procedure is proposed that gives the averaged large signal, steady state and small signal ac models. The procedure is suitable for the converters operating in both Continuous Conduction Mode (CCM) and in Discontinuous Conduction Mode (DCM). For modelling in DCM, the concept of virtual switch is used to model the converter using bond graph. Using the proposed method, converters of any complexity can be modelled incorporating all the advantages of bond graph modelling. Magnetic components are essential part of the power electronic systems. Most common parts are the inductor, transformer and coupled inductors which contain both the electric and magnetic domains. Gyrator-Permeance approach is used to model the magnetic components. Gyrator acts as an interface between electric and magnetic domain and capacitor model the permeance of the magnetic circuits. Components like inductor, tapped inductor, transformer, and tapped transformer are modelled. Interleaved converters with coupled inductor, zero ripple phenomena in coupled inductor converters as well as integrated magnetic Cuk converter are also modelled. Modelling of integrated magnetic converters like integrated magnetic forward converter, integrated magnetic boost converter are also explored. To carry out all the simulations of proposed bond graph models, bond graph toolbox is developed using MATLAB/SIMULINK. The MATLAB/SIMULINK is chosen since it is general simulation platform widely available. Therefore all the analysis and simulation can be carried out using facilities available in MATLAB/SIMULINK. Symbolic equation extraction toolbox is also developed which extracts state equations from bond graph model in SIMULINK in symbolic form.

[pt] Modelagem e Controle Não Linear de um Veículo Terrestre sobre Suspensão descreve um estudo em dinâmica veicular no qual inicialmente apresenta-se um modelo analítico para representar a geração de forças longitudinais e laterais no contato do pneu com o solo. Em seguida é desenvolvido, para um automóvel de passeio terrestre sobre suspensão, um modelo não linear de 4 graus de liberdade (velocidades longitudinal, lateral, de guinada e de rolagem), e a sua linearização. Expande-se esse modelo para um de 8 graus de liberdade, no qual inclui-se o movimento de rotação axial de cada uma das quatro rodas, e consideram-se os movimentos do veículo somente no plano, sem efeitos de pitch ou bounce, mas apresentando alguma relação de distribuição de cargas devido ao roll. Descrevem-se ainda modelos em Grafos de Ligação para os três dinâmicas de um veículo terrestre (longitudinal, lateral e vertical) e seus acoplamentos, visando futuras análises mais detalhadas desse sistema. Todos os modelos em malha aberta são validados através simulações computacionais em diversas condições típicas de operação. Na segunda parte desse trabalho é apresentada a estratégia proposta para o tratamento do problema de controle direcional do veículo em uma manobra qualquer, empregando a metodologia da linearização por realimentação, tendo como base o modelo linear de 4 graus de liberdade. São analisados os resultados encontrados através de simulação computacional para a malha fechada com diferentes combinações de parâmetros, empregando os modelos não lineares de 4 e 8 graus de liberdade. Conclui-se discutindo a possibilidade de generalização deste procedimento para diferentes aplicações em Dinâmica Veicular. / [en] Modeling and Non Linear Control of a Ground Vehicle's Steering describles a study in vehicle dynamics, which presents an analytic model representing the generation of longitudinal and lateral forces at the contact patch between tire and ground. Next it is developed, for a typical passenger car, a non-linear model with four degrees of freedom (longitudinal, lateral, yaw and roll velocities), and its linearization. This model is then expanded to another one with eight degrees of freedom, which includes the axial rotation of each one of the four wheels, and considers the vehicle's movement only at a known plane, whithoud pitch and bounce effects, but including some load distribution among the wheels, due to roll. Computational simulation in varius typical operation condition validate all open loop models. The second part of this work presents the proposed strategy for directional control of a vehicle at any type of manoeuvre, using the feedback linearization methodology, directly applied to the linear four degrees of freedom model. Theresults obtained trhough computational simulation for a closed loop model with different parameters are analysed using both nonlinear four and eight degrees of freedom models. The possibility of generalizing this procedure to distinct applications in Vehicle Dynamics is, then, discussed.

[pt] A sub-divisão de um veículo em módulos é muito útil quando se quer estudar o comportamento dinâmico de um determinado subsistema e sua influência nos demais componentes. Em alguns casos, devido ao tipo de tratamento empregado para descrever os elementos, não se consegue perceber de que modo as variáveis inerentes a um subsistema interagem com as demais, e, por conseguinte, os subsistemas entre si. A abordagem modular baseada no fluxo de potência permite uma melhor identificação das relações de causa e efeito entre subsistemas, uma vez que se pode definir, de forma clara e consistente, quem são as variáveis de entrada e de saída de cada componente ou módulo, e, conseqüentemente, seus acoplamentos. Neste tipo de tratamento, aplicado aos sistemas mecânicos, uma vez estabelecida a cinemática de um subsistema, podese obter as relações entre os esforços que seus componentes produzem uns sobre os outros, a partir da caracterização da potência transmitida através dos seus diversos elementos. Este trabalho apresenta um procedimento semi-analítico de equacionamento modular aplicado à modelagem e integração dos sistemas de suspensão e direção de veículos terrestres, no qual as variáveis de entrada e saída indicam o fluxo de potência entre os elementos de todo o sistema. Tal abordagem tem como base a técnica dos Grafos de Ligação, empregada em sistemas multidomínio em geral, e usa alguns conceitos da metodologia dos Transformadores Cinemáticos, normalmente aplicada aos sistemas multicorpos. A partir da definição da geometria dos mecanismos em questão, encontram-se as matrizes que representam os vínculos cinemáticos entre seus elementos, das quais o funcionamento dos sistemas integrados pode ser simulado e analisado, e informações necessárias aos seus projetos determinadas. As equações (malhas) algébricas que existem em mecanismos com estrutura cinemática fechada são analiticamente resolvidas, evitando deste modo modelos matemáticos com equações diferenciais e algébricas simultâneas. Das relações cinemáticas, o modelo dinâmico (matrizes de inércia, rigidez e amortecimento, etc) é obtido, e novamente informações essenciais à análise e síntese dos sistemas podem ser determinadas. O comportamento no tempo desses modelos pode ser encontrado por um método de integração de equações diferenciais qualquer. Adota-se o Simulink/MatLab® para representar o modelo assim desenvolvido em diagrama de blocos, e conseqüentemente simulá-lo. Através deste tratamento, cada bloco da implementaçao em Simulink/MatLab® contém o correspondente modelo analítico de um único módulo, cujo estabelecimento depende das características dinâmicas do sistema que se deseja analisar. A vantagem de adotar tal representação, baseada no fluxo de potência, consiste no fato de que um módulo pode ser substituído por outro, descritivo de um elemento ou subsistema com a mesma função, porém com configuração física distinta, e, conseqüentemente, modelo matemático específico, sem qualquer alteração nos demais componentes do sistema. Este procedimento está sendo adotado para modelagem dos diversos sistemas veiculares, como os de suspensão, direção, transmissão e freios, e também os pneus, inseridos em um chassi, incluindo os graus de liberdade desejados do veículo, todos descritos de forma modular semi- analítica através da mesma abordagem, empregando a técnica de modelagem mais apropriada para representá-los. / [en] The sub-division of a vehicle in modules is very useful when we want to study the dynamical behavior of a certain sub-system and its influence in other components. In some cases, due to the type of treatment employed to describe the dynamic behavior of the elements, we don`t get to notice the way that inherent variables in a sub-system interacts with the others, and, consequently, the subsystems amongst themselves. The modular approach based on the power flow allows a better identification of the causal relationships among sub-systems, once it can define, in clear and consistent way, what are the input and output variables of each component or module, and, consequently, their couplings. In this type of treatment applied to the mechanical systems, once established the kinematics of a sub-system, it can be obtained the relationships among the efforts that their components produce on the other ones, from the characterization of the power transmitted through their several elements. This paper presents a semi-analytical procedure of modular modeling applied to the suspension and steering systems of a ground vehicle, in which the input and output variables indicate the power flow among the elements of the whole system. Such approach has as base the Bond Graphs technique, used in multidomain systems in general, and uses some concepts of the Kinematic Transformers methodology, usually applied to the multibody systems. From the mechanisms geometry, the matrices that represent the kinematics links between its elements are found, the operation of the integrated systems can be simulated and analyzed, and information about its design can be obtained. The algebraic loops (equations) inherent to mechanisms with closed kinematic structure are solved analytically, and there is not a mathematical model with simultaneous algebraic and differential equations. From the kinematic relations, the dynamic model (inertial, stiffness and damping matrices) is obtained, and again essential information to the systems analysis and synthesis can be determined. The models time behavior can be found by any differential equations integration method. The Simulink/Matlab is adopted to represent the model developed by block diagrams, and consequently to simulate it. Through this treatment, each block in the Simulink/Matlab implementation contains the correspondent analytical model of a single module, whose establishment depends on the dynamic characteristics of the system to be analyzed. The advantage of adopting such representation, based on the power flow, consists in the fact that a module can be substituted for other, descriptive of an element or sub-system with the same function, however with different physical configuration, and, consequently, specific mathematical model, without any alteration in the other components of the system. This procedure is being adopted for modeling all vehicular systems, like the suspension, steering, transmission and brakes systems, and also the tires, inserted in the chassis, including the desired degrees of freedom of the vehicle, all described in a semi- analytical modular way by the same approach, using the most appropriate modeling technique to represent them.

[pt] Mecanismos são essencialmente (mas não exclusivamente) compostos por vários corpos rígidos que possuem movimento relativo entre si. Cada corpo rígido está ligado através de uma junta a um ou mais corpos, sendo a sequência de corpos conectados chamada de cadeia cinemática. Cadeias cinemáticas abertas (ou em série) não têm restrições sobre uma de suas extremidades, já cadeias fechadas (ou paralelas) têm restrições em ambas as extremidades. O foco neste trabalho será dado no estudo de mecanismos com cadeias cinemáticas fechadas ou mecanismos paralelos. Assim, este trabalho apresenta a determinação da solução analítica do modelo dinâmico de um mecanismo paralelo plano com três graus de liberdade através da caracterização do fluxo de potência entre os seus componentes. A partir das relações geométricas associadas ao deslocamento dos seus graus de liberdade, as relações cinemáticas associadas às suas velocidades são determinadas. Considerando o fluxo de potência entre os graus de liberdade, e também entre estes e os elementos de atuação (atuadores lineares elétricos), as relações de equilíbrio das forças e torques são obtidas. Levando em consideração os efeitos inerciais dos componentes do sistema, a rigidez e efeitos de amortecimento, as equações de movimento ou as equações de estado são analiticamente determinadas e representadas em qualquer sistema de referência, local ou global. Além disso, as relações entre a cinemática inversa e a dinâmica direta são apresentadas. Esta abordagem adota os mesmos fundamentos, conceitos e elementos da técnica dos grafos de ligação, com a sua notação simbólica e representação gráfica. A metodologia proposta é generalizada e aplicável em qualquer tipo de mecanismo (aberto ou fechado, plano ou espacial). O modelo cinemático inverso do mecanismo de cadeia fechada, que tem uma solução fácil quando comparado com o modelo direto, pode ser desenvolvido por qualquer metodologia conhecida. Neste trabalho, a técnica da cadeia vetorial é usada para determinar o modelo geométrico inverso, e com a sua derivação, as relações cinemáticas são obtidas, e, portanto, a matriz Jacobiana inversa. Desse modo, é construída a estrutura em grafos de ligação da cinemática inversa e, a partir das relações de causa e efeito, encontra-se o modelo dinâmico direto do mecanismo. Assim, esta metodologia (grafos de ligação ou fluxo de potência) é mais eficiente e segura para determinar os modelos dinâmicos analíticos (fechados) de mecanismos paralelos. Um conjunto de simulações foi realizado para validar esta abordagem, usando os dados reais (geometria, inércia, amortecimento, forças de atuação, etc.) a partir de um mecanismo plano projetado e construído especialmente para a finalidade de comparar os resultados simulados e experimentais. Uma estratégia de controle de malha fechada usando a cinemática inversa e os modelos dinâmicos diretos é proposta. Finalmente, testes experimentais validam esta estratégia. As equações analíticas levam a um processo de simulação e controle em tempo real mais eficientes destes sistemas. / [en] Mechanisms are essentially (but not exclusively) made up of multiple rigid bodies that have relative motion between themselves. Each rigid body is connected through a joint to one or more bodies, wherein the sequence of connected bodies is called kinematic chain. Open (or serial) kinematic chains have no restrictions on one of their ends, as closed (or parallel) chains have restrictions on both ends. The focus in this work will be given on the study of mechanisms with closed kinematic chains or parallel mechanisms. Thus, this work presents the analytical form determination of the dynamic model of a parallel planar mechanism with three degrees of freedom through the characterization of the power flow between its components. From the geometrical relations associated to the displacement of their degrees of freedom, the kinematic relations associated to their speeds are determined. Considering the power flow between the degrees of freedom, and also between these and the actuating elements (linear electric actuators) the equilibrium relations of the forces and torques are obtained. Accounting for inertial effects of system components, the stiffness and damping effects, the equations of motion or the state equations are analytically determined and represented in any reference frame, local or global. Besides, the relation between the inverse kinematics and the direct dynamics is presented. This approach adopts the same fundamentals, concepts and elements of the Bond Graph Technique, with its symbolic notation and graphical representation. The proposed methodology is generalized and applicable in any type of mechanism (open or closed, planar or spatial). The inverse kinematic model of the closed chain mechanism, which has easy solution when compared to the direct model, can be developed by any known methodology. In this work, the vector loop technique is used to determine the inverse geometric model, and with its derivation, the kinematic relations are obtained, and therefore the inverse Jacobian matrix. Thereby, the inverse kinematics bond graph is built and, from the cause and effect relations, the direct dynamic model of the mechanism is found. Thus, this methodology (bond graphs or power flow) is more efficient and secure to achieve the dynamic analytical (closed) models of parallel mechanisms. A set of simulations are performed to validate this approach, using the real data (geometry, inertia, damping, actuators forces, etc.) from a planar mechanism designed and built especially for the purpose to compare the simulated and experimental results. A closed-loop control strategy using the inverse kinematic and the direct dynamic models is proposed. Finally, experimental tests validate this strategy. The analytical equations lead to a more efficient simulation process and real-time control of these systems.

Cette thèse développe des approches pour le diagnostic et le pronostic de systèmes dynamiques incertains en utilisant la technique de modélisation Bond Graph (BG). Tout d'abord, une représentation par intervalles des incertitudes paramétriques et de mesures est intégrée à un modèle BG-LFT (Linear Fractional Transformation). Une méthode de détection robuste de défaut est développée en utilisant les règles de l'arithmétique d'intervalle pour la génération de seuils robustes et adaptatifs sur les résidus nominaux. La méthode est validée en temps réel sur un système de générateur de vapeur.Deuxièmement, une nouvelle méthodologie de pronostic hybride est développée en utilisant les Relations de Redondance Analytique déduites d'un modèle BG et les Filtres Particulaires. Une estimation de l'état courant du paramètre candidat pour le pronostic est obtenue en termes probabilistes. La prédiction de la durée de vie résiduelle est atteinte en termes probabilistes. Les incertitudes associées aux mesures bruitées, les conditions environnementales, etc. sont gérées efficacement. La méthode est validée en temps réel sur un système mécatronique incertain.Enfin, la méthodologie de pronostic développée est mise en œuvre et validée pour le suivi efficace de la santé d'un sous-système électrochimique d’une pile à combustible à membrane échangeuse de protons (PEMFC) industrielle à l’aide de données de dégradation réelles. / This thesis develops the approaches for diagnostics and prognostics of uncertain dynamic systems in Bond Graph (BG) modeling framework. Firstly, properties of Interval Arithmetic (IA) and BG in Linear Fractional Transformation, are integrated for representation of parametric and measurement uncertainties on an uncertain BG model. Robust fault detection methodology is developed by utilizing the rules of IA for the generation of adaptive interval valued thresholds over the nominal residuals. The method is validated in real time on an uncertain and highly complex steam generator system.Secondly, a novel hybrid prognostic methodology is developed using BG derived Analytical Redundancy Relationships and Particle Filtering algorithms. Estimations of the current state of health of a system parameter and the associated hidden parameters are achieved in probabilistic terms. Prediction of the Remaining Useful Life (RUL) of the system parameter is also achieved in probabilistic terms. The associated uncertainties arising out of noisy measurements, environmental conditions etc. are effectively managed to produce a reliable prediction of RUL with suitable confidence bounds. The method is validated in real time on an uncertain mechatronic system.Thirdly, the prognostic methodology is validated and implemented on the electrical electro-chemical subsystem of an industrial Proton Exchange Membrane Fuel Cell. A BG of the latter is utilized which is suited for diagnostics and prognostics. The hybrid prognostic methodology is validated, involving real degradation data sets.

Les travaux présentés dans ce mémoire concernent l'étude du diagnostic de défauts simples et multiples pour des systèmes dynamiques continus et consistent à développer une stratégie de diagnostic globale pour la gestion des modes de fonctionnement en situations normale et anormale. Nous avons d'abord développé un nouveau formalisme graphique de modélisation des systèmes dynamiques émanant des BG et que nous avons appelé le BGS. Ce formalisme est très aisément interprétable grâce à un certain nombre de propriétés et de définitions que nous avons établies. L'élaboration d'un tel formalisme permet de faire appel aux propriétés structurelles et causales du BG et d'élargir leur champ d'étude pour inclure le raisonnement qualitatif. Nous avons ensuite proposé un modèle générique permettant d'intégrer les modèles Génériques de Composants (MGC) fonctionnels et les modèles BGS pour la gestion, par un automate ni, des modes de fonctionnement et des conditions de reconfiguration d'un système autonome. En fin, nous avons proposé une méthode de diagnostic des défauts simples et multiples en utilisant une approche par abduction basée sur l'étude de la propagation de défauts sur le BGS à partir des observations. La méthodologie proposée est validée par deux systèmes de complexités différentes et en l'occurrence une pile à combustible à membrane échangeuse de protons et un système électromécanique d'un véhicule électrique.

L'objectif des travaux présentés dans ce mémoire concerne le développement d'un modèle global représentant le couplage de l'enveloppe du bâtiment avec les équipements énergétiques. Une approche systémique appelée les Bond Graphs, peu employée jusqu'ici dans la modélisation des systèmes thermiques, est utilisée. Le modèle global du bâtiment, regroupant sous le même environnement de simulation, les modèles de l'enveloppe du bâtiment, les apports solaires, les émetteurs de chauffage et de rafraîchissement et le système de ventilation, est développé pour reconstituer l'ensemble des articulations énergétiques entre l'enveloppe et les environnements intérieur et extérieur. A travers la modélisation d'un bâtiment multizone, le couplage systémique des modèles de l'enveloppe et des apports solaires est présenté. Par ailleurs, un système combinant un plancher chauffant et un plafond rafraîchissant est étudié à l'aide des modèles des émetteurs de chauffage et de rafraîchissement. Le renouvèlement d'air dans le bâtiment est également concerné par la modélisation Bond Graph. Enfin, des éléments de validation expérimentale sont présentés. Pour cela, la plateforme de tri-génération d'énergie ENERBAT est exploitée. L'objectif est d'étudier le couplage optimal enveloppe du bâtiment - équipements énergétiques pour lequel les modèles BG sont développés. Une étude paramétrique tenant compte des interactions entre les paramètres étudiés est menée sur un projet réel de rénovation. Finalement, une combinaison appropriée des paramètres étudiés a été retenue afin de réduire la consommation énergétique.

In recent years, Polymer Electrolyte Membrane Fuel Cells (PEMFC) have attracted much attention due to their potential as a clean power source for many applications, including automotive, portable and stationary devices. This resulted in a tremendous technological progress, such as the development of new membranes and electro-catalysts or the improvement of electrode structures. However, in order to compete within the most attractive markets, the PEMFC technologies did not reach all the required characteristics yet, in particular in terms of cost and durability.Because of the strong coupling between different physicochemical phenomena, the interpretation of experimental observations is difficult, and analysis through modeling becomes crucial to elucidate the degradation and failure mechanisms, andto help improving both PEMFC electrochemical performance and durability.The development of a theoretical tool is essential for industrials and the scientific community to evaluate the PEMFC degradation and to predict itsperformance and durability in function of the materials properties and in a diversity of operating conditions. This manuscript summarizes my scientific research efforts in this exciting topic during the last 9 years in France, including my invention of the MEMEPhys multiscale simulation package,developed on the basis of my childhood passion for the New Technologies for Energyin Argentina. My perspectives of adapting this approach to other electrochemical systems such as water electrolyzers and batteries are also discussed.