La modélisation des composants à semiconducteur est un grand classique. Dans ce travail, outre un rappel de l'état de l'art, la modélisation des composants à semiconducteur de puissance est abordée par l'approche des graphes de liens. Les graphes de liens fournissent une représentation unifiée de la dynamique des systèmes. Cette approche multiphysique impose une représentation explicite des transferts d'énergie qui confère à cette méthode une grande généricité. De plus l'analyse de causalité associée, permet de retrouver par exemple les règles d'association des composants de puissance, bien connue en électronique de puissance. La modélisation par graphe de liens des composants à semiconducteur repose ici sur l'analyse régionale très classique des composants à semiconducteur : les zones de charge d'espace et les zones neutres (en forte injection, en faible injection à dopage uniforme et en faible injection à gradient de dopage). Le mémoire se poursuit par la description succincte du simulateur PACTE (http://pacte.ampereforge.org ) qui a permis de simuler la diode PIN de puissance et le transistor bipolaire de puissance en réutilisant les mêmes modèles de région semiconductrice. Le mémoire se termine par quelques conclusions et prospective.

Tato diplomová práce pojednává o možnostech simulace a modelování elektrických pohonů na základě Bondových grafů. Je zde uvedena základní teorie k problematice. V práci jsou představeny modely stejnosměrného a synchronního stroje sestavené touto technikou. Dále je zde seznámení s Freeware knihovnou Bondových grafů BG V2.1 k programu MATLABSimulink. Za pomocí této knihovny byly sestaveny oba modely strojů. V poslední kapitole je uvedeno srovnání modelování na základě teorie Bondových grafů s modelováním v blokových diagramech.

Les travaux de cette thèse s’inscrivent dans le cadre de la mise en place d’une méthodologie innovante de gestion thermique globale des véhicules automobiles. Ils portent plus particulièrement sur l’analyse de la réduction des émissions de polluants et l’amélioration des performances énergétiques d’un moteur à combustion, notamment dans les charges partielles ou en régime transitoire. Le premier objectif vise la mise en évidence de l’effet de la température d’admission sur le fonctionnement du moteur. Le deuxième objectif est relatif à la prédiction de l’apparition des phénomènes de condensation inhérents aux procédés de recirculation des gaz d’échappement à l’admission moteur. Enfin le troisième objectif est la modélisation et la mise en œuvre d’un circuit d’eau refroidit par la boucle de climatisation dont la fonction est de sous refroidir les gaz d’admission du moteur. Le premier chapitre est consacré à la présentation du système thermique véhicule et de la démarche de conception en V adoptée dans ce travail. Dans le second chapitre, et après avoir mis en évidence les effets d’un sous refroidissement des gaz admission sur le rendement thermodynamique du moteur, on montre à l’aide d’une étude technologique et numérique de la boucle de climatisation qu’il est possible d’opérer ce refroidissement par un système embarqué capable de se régénérer thermiquement lors d’un freinage. Le troisième chapitre est dédié à la modélisation du système à l’aide d’une modélisation énergétique centrée sur l’utilisation du langage bond graph. Le dernier chapitre est dédié au volet expérimental afin de valider d’une part le modèle de condensation et d’autres part un démonstrateur d’hybridation thermique et ses stratégies de pilotage / The work of this thesis is part of the development of an innovative methodology in the field of global thermal management for motor vehicles. It focuses specifically on the analysis of the reduction of pollutant emissions and improving energy efficiency of a combustion engine, especially in partial load or transient operation. The first objective is to analyze the effect of inlet temperature on the engine performance. The second objective relates to the prediction of the condensation processes inherent to exhaust gas recirculation into the engine intake. Finally, the third objective is the modeling and implementation of a water circuit cooled by the air conditioning loop and whose function is to cool the gases in at the engine intake.The first chapter is devoted to the presentation of the vehicle thermal management system and the design process adopted in this work. In the second chapter, after having shown the effects of intake gas cooling on the thermodynamic efficiency of the engine, it is shown with a numerical and technological study of the air conditioning loop that is possible to operate an onboard cooling system that is capable of regenerating heat when braking. The third chapter focuses on system modeling using an energy modeling focuses on the use of bond graph language. The final chapter is dedicated to the experimental part with the objective of validating the model of gas condensation and a demonstrator of thermal hybridation and its control strategies

Os paradigmas de mercado e produção têm-se alterado sensivelmente nos últimos cinquenta anos. Nesse novo contexto, para que as empresas tenham um desempenho competitivo, deve-se considerar fatores como integração, qualidade da informação e incertezas do ambiente. Um dos objetivos principais deste trabalho é analisar as inter-relações entre esses três fatores citados, no âmbito do Planejamento e Controle da Produção (PCP), bem como avaliar seu impacto no desempenho do PCP e da empresa. O desempenho do PCP foi avaliado considerando-se o nível de reprogramações, ou seja, o nível de modificações na programação da produção, e considerando-se o atendimento das metas definidas para os indicadores dessa função. Essa pesquisa qualitativa e descritiva foi feito por meio da metodologia de estudo de casos múltiplos. Como resultados, foram observadas evidências da existência de relações positivas entre os constructos analisados, ou seja, integração, incerteza, qualidade da informação e desempenho. Além disso, foram identificados diferentes mecanismos de integração utilizados nas empresas, e foram observadas diferentes causas para as reprogramações. O segundo objetivo principal deste trabalho é o desenvolvimento de um modelo dinâmico para controle da produção de múltiplos produtos, capaz de responder às incertezas que afetam estabilidade dos sistemas produtivos. O objeto da modelagem foi um sistema com fluxo job shop destinado à produção de embalagens de polipropileno, pertencente a uma empresa dos setores têxtil e petroquímico. O modelo foi desenvolvido com base na Modelagem de Sistemas Dinâmicos e na Teoria de Controle. Para obtenção do equacionamento matemático, utilizou-se a metodologia dos Grafos de Ligação. A implementação e simulação do modelo foi realizada com o auxílio do módulo Simulink®, do software Matlab®. O objetivo de controle consiste no ajuste da frequência de operação das máquinas de forma a atender as demandas previstas e, simultaneamente, manter os níveis estabelecidos de estoque em processo. No presente trabalho, focou-se na análise da resposta do sistema com controle no regime transiente, com os estoques iniciais nulos. Os melhores resultados foram observados com a utilização de um controlador híbrido, que estabelece uma produção constante em um período inicial e passa posteriormente a atuar como um controlador proporcional. O modelo dinâmico desenvolvido neste trabalho é coerente com resultados obtidos na pesquisa qualitativa e com prática do PCP, pois está alinhado aos indicadores de desempenho desta função, promove a melhoria da qualidade das informações disponíveis ao planejamento e é capaz de responder a incertezas que afetam o fluxo de produção. / The production paradigm has considerably changed over the last fifty years. In this new context, factors such as integration, information quality and ability to respond to uncertainties must be pursued by companies that want to remain competitive. Thus, one of the main objectives of this research is to analyze the relationships among these three factors mentioned, in the level of the Production Planning and Control (PPC) function, as well as to assess their impact on the performance of the PPC function and on the overall business performance. The performance of the PPC function was measured by means of the rescheduling frequency, i.e., the frequency with which the production activities have to be rescheduled, and by means of the degree of accomplishment of organizational goals, that is, the extent to which the PPC metrics are accomplished. This qualitative and descriptive research was carried out as a multiple case study. Evidences of the existence of a positive relationship among the analyzed constructs, i.e., integration, uncertainty, information quality and performance, were found. In addition, different integration mechanisms and different rescheduling causes were observed on the studied cases. The second main objective of this work is the development of a dynamic model for the production control of multi-product systems, i.e, a model that could respond to the environmental and internal uncertainties that affect the production flow. A job shop production system of propylene bags from a manufacturing company of the textile and petrochemical industry was modeled. The model was developed using Control Theory, System Dynamics modeling, and, more specifically, the Bond Graph technique. The mathematical formulation of the system was derived from the Bond Graphs. System implementation and simulation was performed with the aid of Simulink® and Matlab® software. The control objective was to adjust the operation frequency of the machines to attend the required demand while simultaneously keeping the work in process at the desired levels. In this current work, the focus of the analysis was placed on the transient response of the controlled system, with initial inventory levels set to zero. The best results were achieved with a hybrid controller, which leads the machines to operate at constant frequency in the initial period and, later on, starts to perform as a proportional controller. The proposed dynamic model is compatible with the findings of the qualitative research and with the PPC practice, since it is aligned with the PPC metrics, it improves the quality of information available for planning and it can respond to the uncertainties that may affect the production flow.

Os paradigmas de mercado e produção têm-se alterado sensivelmente nos últimos cinquenta anos. Nesse novo contexto, para que as empresas tenham um desempenho competitivo, deve-se considerar fatores como integração, qualidade da informação e incertezas do ambiente. Um dos objetivos principais deste trabalho é analisar as inter-relações entre esses três fatores citados, no âmbito do Planejamento e Controle da Produção (PCP), bem como avaliar seu impacto no desempenho do PCP e da empresa. O desempenho do PCP foi avaliado considerando-se o nível de reprogramações, ou seja, o nível de modificações na programação da produção, e considerando-se o atendimento das metas definidas para os indicadores dessa função. Essa pesquisa qualitativa e descritiva foi feito por meio da metodologia de estudo de casos múltiplos. Como resultados, foram observadas evidências da existência de relações positivas entre os constructos analisados, ou seja, integração, incerteza, qualidade da informação e desempenho. Além disso, foram identificados diferentes mecanismos de integração utilizados nas empresas, e foram observadas diferentes causas para as reprogramações. O segundo objetivo principal deste trabalho é o desenvolvimento de um modelo dinâmico para controle da produção de múltiplos produtos, capaz de responder às incertezas que afetam estabilidade dos sistemas produtivos. O objeto da modelagem foi um sistema com fluxo job shop destinado à produção de embalagens de polipropileno, pertencente a uma empresa dos setores têxtil e petroquímico. O modelo foi desenvolvido com base na Modelagem de Sistemas Dinâmicos e na Teoria de Controle. Para obtenção do equacionamento matemático, utilizou-se a metodologia dos Grafos de Ligação. A implementação e simulação do modelo foi realizada com o auxílio do módulo Simulink®, do software Matlab®. O objetivo de controle consiste no ajuste da frequência de operação das máquinas de forma a atender as demandas previstas e, simultaneamente, manter os níveis estabelecidos de estoque em processo. No presente trabalho, focou-se na análise da resposta do sistema com controle no regime transiente, com os estoques iniciais nulos. Os melhores resultados foram observados com a utilização de um controlador híbrido, que estabelece uma produção constante em um período inicial e passa posteriormente a atuar como um controlador proporcional. O modelo dinâmico desenvolvido neste trabalho é coerente com resultados obtidos na pesquisa qualitativa e com prática do PCP, pois está alinhado aos indicadores de desempenho desta função, promove a melhoria da qualidade das informações disponíveis ao planejamento e é capaz de responder a incertezas que afetam o fluxo de produção. / The production paradigm has considerably changed over the last fifty years. In this new context, factors such as integration, information quality and ability to respond to uncertainties must be pursued by companies that want to remain competitive. Thus, one of the main objectives of this research is to analyze the relationships among these three factors mentioned, in the level of the Production Planning and Control (PPC) function, as well as to assess their impact on the performance of the PPC function and on the overall business performance. The performance of the PPC function was measured by means of the rescheduling frequency, i.e., the frequency with which the production activities have to be rescheduled, and by means of the degree of accomplishment of organizational goals, that is, the extent to which the PPC metrics are accomplished. This qualitative and descriptive research was carried out as a multiple case study. Evidences of the existence of a positive relationship among the analyzed constructs, i.e., integration, uncertainty, information quality and performance, were found. In addition, different integration mechanisms and different rescheduling causes were observed on the studied cases. The second main objective of this work is the development of a dynamic model for the production control of multi-product systems, i.e, a model that could respond to the environmental and internal uncertainties that affect the production flow. A job shop production system of propylene bags from a manufacturing company of the textile and petrochemical industry was modeled. The model was developed using Control Theory, System Dynamics modeling, and, more specifically, the Bond Graph technique. The mathematical formulation of the system was derived from the Bond Graphs. System implementation and simulation was performed with the aid of Simulink® and Matlab® software. The control objective was to adjust the operation frequency of the machines to attend the required demand while simultaneously keeping the work in process at the desired levels. In this current work, the focus of the analysis was placed on the transient response of the controlled system, with initial inventory levels set to zero. The best results were achieved with a hybrid controller, which leads the machines to operate at constant frequency in the initial period and, later on, starts to perform as a proportional controller. The proposed dynamic model is compatible with the findings of the qualitative research and with the PPC practice, since it is aligned with the PPC metrics, it improves the quality of information available for planning and it can respond to the uncertainties that may affect the production flow.

Ce mémoire d'Habilitation à Diriger des Recherches présente tout d'abord au chapitre 1 une vue globale et synthétique de mes activités depuis ma prise de poste en septembre 1998. Le chapitre 2, « Bilan des activités depuis la prise de poste », détaille le contenu de ma recherche, de mon enseignement et des responsabilités administratives. Il tente de mettre en évidence les liens forts de mes activités de recherche à celles d'enseignement. Le chapitre 3, « Développement des activités de recherche », présente de manière approfondie et technique les principaux travaux de recherche menés durant cette période. Il montre comment ces travaux ont évolué progressivement d'un contexte « post doctoral » sur des sujets en continuité de celui de ma thèse de doctorat vers une volonté d'animer un axe de recherche en proposant de nouveaux sujets au sein du laboratoire. La transition s'est effectuée par l'appropriation de l'axe de recherche en question. Le chapitre 4, « Conclusion et perspectives », présente un bilan général de l'évolution de mes activités de recherche et les met en perspectives à travers un certain nombre de propositions que je souhaite développer à l'avenir. Enfin, trois annexes récapitulent mes productions scientifiques, ma participation à la communauté et mon activité contractuelle, et finalement les textes des productions jugées les plus significatives par rapport à mes activités de recherche. Cette dernière annexe est fournie dans un document séparé pour en faciliter sa lecture.

Im Beitrag wird zunächst die konventionelle auf Signalflüssen basierte Modellbildung mit modernen leistungsflussbasierten Methoden, die auf dem Prinzip von Aktion und Reaktion aufbauen, verglichen. BG (Bond Graph), POG (Power Oriented Graph) und EMR (Energetic Macroscopic representation) sind solche modernen Methoden die den Leistungsaustausch zwischen Teilsystemen als Grundlage für den Modellbildungsansatz nutzen. Diese Werkzeuge erhalten die physikalische Struktur, erlauben es in das dynamische System hineinzuschauen und unterstützen das Verständnis des Leistungsflusses. Unterschiede werden anhand verschiedener Eigenschaften in einer Tabelle angegeben. Nach Erläuterung der Grundlagen zu POG und BG erfolgt die Vorstellung einer Freeware Zusatzbibliothek zur Simulation von Bondgraphen. Spezielle Eigenschaften werden kurz umrissen. Diese Blockbibliothek läuft unter Simulink, besteht aus nur 9 mittels Menü konfigurierbaren Blöcken und realisiert bidirektionale Verbindungen. Die Beispiele Gleichstrommotor, Pulssteller und elastische Welle demonstrieren die Vorteile der leistungsflussorientierten Modellbildung. Zustandsregelung, Energieeffizienz und Simulink LTI Analysewerkzeuge führen in die Anwendung der vorgestellten Simulink Zusatzbibliothek für Bondgraphen ein. / The paper starts with a comparison of the conventional modelling method based on signal flow and modern power flow oriented modelling methods based on the principle of action and reaction. BG (Bond Graph), POG (Power Oriented Graph) and EMR (Energetic Macroscopic representation) are such modern methods based on the power exchange between partial systems as a key element for the basic modelling approach. These tools preserve the physical structure, enable a view inside dynamical systems and support understanding the power flow. Relationships between these graphical representations will be given. After the explanation of basics for POG and BG an overview and special features of a freeware add-on library for simulation of BGs will be outlined. The block library runs under Simulink, consists of nine menu-driven customised blocks only and realises bidirectional connections. Examples DC motor, chopper and elastic shaft demonstrate the advantages of power flow oriented modelling. State space control, energy efficiency and Simulink LTI analysis tools exemplify the application of the presented Simulink add-on BG library.

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.

Le travail de thèse concerne la conception d'un système de surveillance robuste aux incertitudes paramétriques à base de modèles bond graphs sous forme LFT (Linear Fractional Transformations).<br />Une procédure de génération automatique de résidus robustes et de seuils adaptatifs de fonctionnement normal a été développée et implémentée à l'aide d'outils logiciels appropriés. Les performances du diagnostic sont contrôlées par une analyse de la sensibilité de ces résidus, permettant de définir des indices de sensibilité aux incertitudes paramétriques, et des indices de détectabilité des défauts. L'outil bond graph permet par sa nature physique d'estimer a priori les valeurs détectables des défauts physiques.<br />Outre l'implémentation informatique, les résultats de recherche sont validés par deux applications industrielles de complexité différente :<br /> un système électromécanique pour détecter et isoler des défauts tels que la variation anormale de l'amplitude du jeu dans la partie réducteur, tout en distinguant le défaut des variations paramétriques<br /> un procédé énergétique de génération de vapeur, qui constitue un système non stationnaire, où l'espace paramétrique est plus complexe.

De nos jours, les processus industriels se doivent d’être efficaces, en particulier au niveau de leur production et/ou consommation énergétique.Ce travail vise à améliorer l’efficacité des processus en analysant l’influence des perturbations sur leur comportement, de la phase de conception à la synthèse des contrôleurs/observateurs, ceci dans une approche intégrée.Le problème du Rejet de Perturbation est introduit ainsi que différents types de contrôles permettant d’atténuer et/ou rejeter ces perturbations. Le système de Barre de Torsion est présenté. Une loi de commande basée sur le concept d’état dérivé est présentée et ensuite validée avec comme application le rejet de perturbation. Il est nécessaire d’estimer les grandeurs physiques utilisées dans les différentes expressions de loi de commande. Un observateur à entrées inconnues basé sur la représentation bond graph est rappelé et ensuite étendu au cas multi-variable. C’est la première contribution théorique de ce travail de recherche.Nous comparons ensuite l’efficacité de différentes techniques de commandes pour le rejet de perturbation par simulation sur le système barre de torsion et analysons ainsi l’efficacité de la technique proposée. Une extension théorique au problème du découplage entrée-sortie nous permet de généraliser le problème du rejet de perturbation dans une même démarche intégrée d’analyse et de synthèse. Enfin, ces techniques sont exploitées et analysées sur le système réel. Nous validons ainsi expérimentalement nos résultats.Un modèle très simplifié de centrale hydroélectrique est développé afin d’appliquer les résultats de nos travaux. Un modèle bond graph simplifié est validé par simulation. / Nowadays, industrial processes must be efficient, particularly at the production level and/or energy consumption.This research work aims at improving the process efficiency by analysing the influences of disturbances on their behaviour, from the conception phase to the synthesis of controller/observer, in an integrated approach.The disturbance rejection problem is first introduced as well as different control laws allowing attenuate/reject these disturbances. A control law based on the concept of derivative state variable is presented and validated while applied as disturbance rejection.In order to reject the disturbance, different physical variables must be estimated, such as state variables, derivative state variables as disturbance variables. An unknown input observer based on the bond graph representation is recalled and extended in the multivariable case. It is the first theoretical contribution of this work.We thus compare the efficiency of different so-called «modern control laws» for the disturbance rejection problems by simulation with the Torsion-Bar system example. We analyse the efficiency of our approach. One extension to the Input-Output decoupling problem allows us to extend the disturbance rejection problem to other control law type in an integrated approach. At least, these techniques are applied on the real Torsion-Bar system and compared. We validate our approach.Since this work aims at analysing and developing efficient control laws for industrial processes, a simplified model of a hydroelectric plant is developed, in order to apply our results. A simplified bond graph model is validated with simulations.