Cette thèse s'intéresse au développement de stratégies de décalage de charge pouvant être appliquées à un parc de chauffe-eau Joule (CEJ).On propose une modélisation entrée-sortie du système que constitue le CEJ. L'idée est de concevoir un modèle précis et peu coûteux numériquement, qui pourrait être intégré dans un CEJ intelligent. On présente notamment un modèle phénoménologique multi-période d'évolution du profil de température dans le CEJ ainsi qu'un modèle de la demande en eau chaude. On étudie des stratégies d'optimisation pour un parc de CEJ dont la résistance peut être pilotée par un gestionnaire central. Trois cas de figures sont étudiés. Le premier concerne un petit nombre de ballons intelligents et présente une méthode de résolution d'un problème d'optimisation en temps discret. Puis, on s'intéresse à un parc de taille moyenne. Une heuristique gardant indivisibles les périodes de chauffe (pour minimiser les aléas thermo-hydrauliques) est présentée. Enfin, un modèle de comportement d'un nombre infini de ballon est présenté sous la forme d'une équation de Fokker-Planck. / This thesis focuses on the development of advanced strategies for load shifting of large groups of electric hot water tanks (EHWT).The first part of this thesis is dedicated to representing an EHWT as an input-output system. The idea is to design a simple, tractable and relatively accurate model that can be implemented inside a low-power computing unit embedded in a smart EHWT, for practical applications of optimization strategies. It includes in particular a phenomenological multi-period model of the temperature profile in the tank and a realistic domestic hot water consumption model.The second part focuses on the design of optimal control strategies for a group of tanks. Three use-cases are studied. The first one deals with a small number of smart and controllable EHWT for which we propose a discrete-time optimal resolution method. The second use-case adresses a medium-scale group of controllable tanks and proposes a heuristic which keeps the heating period undivided to minimize thermo-hydraulic hazards. Finally, we present the modelling of the behavior of a infinite population of tanks under the form of a Fokker-Planck equation.

In dieser Arbeit untersuchen wir zwei Modelle, die das Laden und Entladen einer Lithium-Ionen Batterie beschreiben. Beide Modelle spiegeln eine Hysterese in dem Spannungs-Ladungs-Verlauf wider. Wir skizzieren den Modellierungsprozess von einem diskreten vielteilchen Modell sowie einem kontinuierlichen vielteilchen Modell. Das erste führt zu einer axiomatischen Beschreibung der Evolution makroskopischer Größen, während das zweite in eine nichtlineare Fokker-Planck Gleichung mündet. Wir zeigen die Existenz und Eindeutigkeit von Lösungen der nichtlinearen Fokker-Planck Gleichung und untersuchen deren qualitative Eigenschaften. Wir benutzen Interpolationsräume und Halbgruppen sektorieller Operatoren um den semilinearen Charakter der partiellen Differentialgleichung auszunutzen. Um globale Existenz zu erhalten, schätzen wir die Dissipation einer mit dem Modell verknüpften Energie ab. Diese Energie ist verwandt mit der L-log-L Norm, welche wir mithilfe einer Gagliardo-Nirenberg Ungleichung zu der L^2 Norm in Verbindung setzen können. Die notwendigen und hinreichenden Bedingungen zur globalen Existenz von Lösungen sind aus physikalischer Sicht plausibel. Der Ladezustand der Batterie muss innerhalb der Werte Voll und Leer sein. In numerischen Experimenten untersuchen wir das qualitative Verhalten von Lösungen. Wir zeigen die Konvergenz der numerischen Lösungen zu den exakten Lösungen. Dafür nutzen wir ähnliche Techniken wie bei der lokalen Existenztheorie. Wir beobachten die Tendenz von Lösungen sich um bestimmte Punkte zu konzentrieren. Unterstützt durch die formale Asymptotik zeigt dies für eine bestimmte Wahl von Parameter-Skalierungen, dass Lösungen gegen Dirac-Maße konvergieren. In diesem Grenzverhalten wird das System durch die Evolution von makroskopischen Größen beschrieben, welche wir auch in dem diskreten vielteilchen Modell wiederfinden. In diesen makroskopischen Größen lässt sich eine Hysterese beobachten. / We discuss two models which describe the charging and discharging of a lithium-ion battery and especially the hysteretical behaviour therein. We give an overview on the modelling process for a discrete many particle model and a continuous many particle model. The former results in an axiomatic description of macroscopic quantities while the latter gives a nonlinear Fokker-Planck equation. The nonlinear Fokker-Planck equation is analysed with respect to existence and uniqueness of solutions as well as qualitative behaviour of solutions. The nonlinearity in this partial differential equation stems from a coefficient which depends on the solution first non-local and second in a higher order. We use interpolation spaces and semigroups generated from sectorial operators to show the existence and uniqueness of solutions locally in time. The global existence in time relies on estimates for the dissipation of an energy. The suitable energy is related to the L-log-L norm and so a Gagliardo-Nirenberg inequality is needed to connect this back to L^2 estimates. It turns out that the conditions for global in time existence of solutions are physical reasonable. One needs that the loading state of the battery shall stay between totally empty and totally full. In numerical experiments we investigate the qualitative behaviour of solutions to the nonlinear Fokker-Planck equation. We are able to show convergence of the numerical solutions to the exact solution. We observe that solutions tend to concentrate at certain points. Supported by results from formal asymptotic expansions, we document the limiting behaviour in a certain scaling of the appearing parameters, which is the formation of Dirac measures. The evolution of the global quantities, which we observe in numerical simulations, is the same as what results from the discrete many particle model and one observes hysteretic behaviour in macroscopic quantities.

Made available in DSpace on 2015-03-04T18:50:49Z (GMT). No. of bitstreams: 1 tese.pdf: 5590910 bytes, checksum: a89ccd96ade2b696f0e5b9163dc31bf5 (MD5) Previous issue date: 2006-11-13 / Lagrangian stochastic models are a largely used tool in the study of passive substances dispersion inside the Atmospheric Boundary Layer. Its application is related to the trajectory computation of thousands of particles, that numerically simulate the dispersion of suspense substances in the atmosphere. In this study, the basic concepts related to the Lagrangian stochastic modelling are presented and discussed together with its main characteristics and its computational implementation, to the study of particles dispersion in the atmosphere. In a computational experiment, the obtained results are compared with observational data from the TRACT experiment, that took place in Europe in 1992. The input data needed for the dispersion model are extracted from simulations with the numerical weather forecast model RAMS. Dispersion over Rio de Janeiro region is also tested in a second experiment. / Modelos Lagrangianos estocásticos constituem ferramenta muito utilizada no estudo da dispersão de substâncias passivas na Camada Limite Atmosférica. Sua aplicação consiste em calcular a trajetória de milhares de partículas, que simulam numericamente a dispersão de uma substância em suspensão na atmosfera. Nesta tese, são apresentados e discutidos os conceitos básicos relacionados à Modelagem Lagrangiana Estocástica de Partículas, bem como suas principais características e sua implementação computacional, para o estudo da dispersão de partículas na atmosfera. Numa experimentação computacional, comparam-se os resultados obtidos com dados observacionais provenientes do experimento TRACT, realizado na Europa em 1992. Os dados de entrada necessários ao modelo de dispersão são extraídos de simulações do modelo de previsão numérica do tempo RAMS. A dispersão sobre o Estado do Rio de Janeiro é também testada em um segundo experimento.

Cette thèse est consacrée à l'étude mathématique de quelques modèles d'équations aux dérivées partielles issues de la physique des plasmas. On s'intéresse principalement à l'analyse théorique de différents régimes asymptotiques de systèmes d'équations cinétiques de type Vlasov-Poisson-Fokker-Planck. Dans un premier temps, en présence d'un champ magnétique extérieur on se concentre sur l'approximation des électrons sans masse fournissant des modèles réduits lorsque le rapport me{mi entre la masse me d'un électron et la masse mi d'un ion tend vers 0 dans les modèles. Suivant le régime considéré, on montre qu'à la limite les solutions vérifient des modèles hydrodynamiques de type convection-diffusion ou sont données par des densités de type Maxwell-Boltzmann-Gibbs, suivant l'intensité des collisions dans la mise à l'échelle. En utilisant les propriétés hypocoercives et hypoelliptiques des équations, on est capable d'obtenir des taux de convergence en fonction du rapport de masse. Dans un second temps, par des méthodes similaires, on montre la convergence exponentielle en temps long vers l'équilibre des solutions du système de Vlasov-Poisson-Fokker-Planck sans champ magnétique avec des taux explicites en les paramètres du modèles. Enfin, on conçoit un nouveau type de schéma volumes finis pour des équations de convection-diffusion non-linéaires assurant le bon comportement en temps long des solutions discrètes. Ces propriétés sont vérifiées numériquement sur plusieurs modèles dont l'équation de Fokker-Planck avec champ magnétique / This thesis is devoted to the mathematical study of some models of partial differential equations from plasma physics. We are mainly interested in the theoretical study of various asymptotic regimes of Vlasov-Poisson-Fokker-Planck systems. First, in the presence of an external magnetic field, we focus on the approximation of massless electrons providing reduced models when the ratio me{mi between the mass me of an electron and the mass mi of an ion tends to 0 in the equations. Depending on the scaling, it is shown that, at the limit, solutions satisfy hydrodynamic models of convection-diffusion type or are given by Maxwell-Boltzmann-Gibbs densities depending on the intensity of collisions. Using hypocoercive and hypoelliptic properties of the equations, we are able to obtain convergence rates as a function of the mass ratio. In a second step, by similar methods, we show exponential convergence of solutions of the Vlasov-Poisson-Fokker-Planck system without magnetic field towards the steady state, with explicit rates depending on the parameters of the model. Finally, we design a new type of finite volume scheme for a class of nonlinear convection-diffusion equations ensuring the satisfying long-time behavior of discrete solutions. These properties are verified numerically on several models including the Fokker-Planck equation with magnetic field

The topic of this thesis is a diffusion process on a potential landscape which is given by a smooth Hamiltonian function in the regime of small noise. The work provides a new proof of the Eyring-Kramers formula for the Poincaré inequality of the associated generator of the diffusion. The Poincaré inequality characterizes the spectral gap of the generator and establishes the exponential rate of convergence towards equilibrium in the L²-distance. This result was first obtained by Bovier et. al. in 2004 relying on potential theory. The presented approach in the thesis generalizes to obtain also asymptotic sharp estimates of the constant in the logarithmic Sobolev inequality. The optimal constant in the logarithmic Sobolev inequality characterizes the convergence rate to equilibrium with respect to the relative entropy, which is a stronger distance as the L²-distance and slightly weaker than the L¹-distance. The optimal constant has here no direct spectral representation. The proof makes use of the scale separation present in the dynamics. The Eyring-Kramers formula follows as a simple corollary from the two main results of the work: The first one shows that the associated Gibbs measure restricted to a basin of attraction has a good Poincaré and logarithmic Sobolev constants providing the fast convergence of the diffusion to metastable states. The second main ingredient is a mean-difference estimate. Here a weighted transportation distance is used. It contains the main contribution to the Poincaré and logarithmic Sobolev constant, resulting from exponential long waiting times of jumps between metastable states of the diffusion.

Quantification of stochastic or quantum systems by a joint probability density or wave function is a notoriously difficult computational problem, since the solution depends on all possible states (or realizations) of the system. Due to this combinatorial flavor, even a system containing as few as ten particles may yield as many as $10^{10}$ discretized states. None of even modern supercomputers are capable to cope with this curse of dimensionality straightforwardly, when the amount of quantum particles, for example, grows up to more or less interesting order of hundreds. A traditional approach for a long time was to avoid models formulated in terms of probabilistic functions, and simulate particular system realizations in a randomized process. Since different times in different communities, data-sparse methods came into play. Generally, they aim to define all data points indirectly, by a map from a low amount of representers, and recast all operations (e.g. linear system solution) from the initial data to the effective parameters. The most advanced techniques can be applied (at least, tried) to any given array, and do not rely explicitly on its origin. The current work contributes further progress to this area in the particular direction: tensor product methods for separation of variables. The separation of variables has a long history, and is based on the following elementary concept: a function of many variables may be expanded as a product of univariate functions. On the discrete level, a function is encoded by an array of its values, or a tensor. Therefore, instead of a huge initial array, the separation of variables allows to work with univariate factors with much less efforts. The dissertation contains a short overview of existing tensor representations: canonical PARAFAC, Hierarchical Tucker, Tensor Train (TT) formats, as well as the artificial tensorisation, resulting in the Quantized Tensor Train (QTT) approximation method. The contribution of the dissertation consists in both theoretical constructions and practical numerical algorithms for high-dimensional models, illustrated on the examples of the Fokker-Planck and the chemical master equations. Both arise from stochastic dynamical processes in multiconfigurational systems, and govern the evolution of the probability function in time. A special focus is put on time propagation schemes and their properties related to tensor product methods. We show that these applications yield large-scale systems of linear equations, and prove analytical separable representations of the involved functions and operators. We propose a new combined tensor format (QTT-Tucker), which descends from the TT format (hence TT algorithms may be generalized smoothly), but provides complexity reduction by an order of magnitude. We develop a robust iterative solution algorithm, constituting most advantageous properties of the classical iterative methods from numerical analysis and alternating density matrix renormalization group (DMRG) techniques from quantum physics. Numerical experiments confirm that the new method is preferable to DMRG algorithms. It is as fast as the simplest alternating schemes, but as reliable and accurate as the Krylov methods in linear algebra.

Centro Latino-Americano de Física / Os processos estocásticos com ruído branco multiplicativo são objeto de atenção constante em uma grande área da pesquisa científica. A variedade de prescrições possíveis para definir matematicamente estes processos oferece um obstáculo ao desenvolvimento de ferramentas gerais para seu tratamento. Na presente tese, estudamos propriedades de equilíbrio de processos markovianos com ruído branco multiplicativo. Para conseguirmos isto, definimos uma transformação de reversão temporal de tais processos levando em conta que a distribuição estacionária de probabilidade depende da prescrição. Deduzimos um formalismo funcional visando obter o funcional gerador das funções de correlação e resposta de um processo estocástico multiplicativo representado por uma equação de Langevin. Ao representar o processo estocástico neste formalismo (de Grassmann) funcional eludimos a necessidade de fixar uma prescrição particular. Neste contexto, analisamos as propriedades de equilíbrio e estudamos as simetrias ocultas do processo. Mostramos que, usando uma definição apropriada da distribuição de equilíbrio e considerando a transformação de reversão temporal adequada, as propriedades usuais de equilíbrio são satisfeitas para qualquer prescrição. Finalmente, apresentamos uma dedução detalhada da formulação supersimétrica covariante de um processo markoviano com ruído branco multiplicativo e estudamos algumas das relações impostas pelas funções de correlação através das identidades de Ward-Takahashi. / Multiplicativewhite-noise stochastic processes continuously attract the attention of a wide area of scientific research. The variety of prescriptions available to define it difficults the development of general tools for its characterization. In this thesis, we study equilibrium properties of Markovian multiplicative white-noise processes. For this, we define the time reversal transformation for this kind of processes, taking into account that the asymptotic stationary probability distribution depends on the prescription. We deduce a functional formalism to derive a generating functional for correlation and response functions of a multiplicative stochastic process represented by a Langevin equation. Representing the stochastic process in this functional (Grassmann) formalism, we avoid the necessity of fixing a particular prescription. In this framework, we analyze equilibrium properties and study hidden symmetries of the process. We show that, using a careful definition of equilibrium distribution and taking into account the appropriate time reversal transformation, usual equilibrium properties are satisfied for any prescription. Finally, we present a detailed deduction of a covariant supersymmetric formulation of a multiplicativeMarkovian white-noise process and study some of the constraints it imposes on correlation functions using Ward-Takahashi identities.

The Planck constant is one of the most important constants in nature, as it describes the world governed by quantum mechanics. However, it cannot be derived from other natural constants. We present a model from which it is possible to derive this constant without any free parameters. This is done utilizing the force between two oscillating electric dipoles described by an extension of Weber electrodynamics, based on a gravitational model by Assis. This leads not only to gravitational forces between the particles but also to a newly found Casimir-type attraction. We can use these forces to calculate the maximum point mass of this model which is equal to the Planck mass and derive the quantum of action. The result hints to a connection of quantum effects like the Casimir force and the Planck constant with gravitational ones and the origin of mass itself.

Lagrangian stochastic models are a largely used tool in the study of passive substances dispersion inside the Atmospheric Boundary Layer. Its application is related to the trajectory computation of thousands of particles, that numerically simulate the dispersion of suspense substances in the atmosphere. In this study, the basic concepts related to the Lagrangian stochastic modelling are presented and discussed together with its main characteristics and its computational implementation, to the study of particles dispersion in the atmosphere. In a computational experiment, the obtained results are compared with observational data from the TRACT experiment, that took place in Europe in 1992. The input data needed for the dispersion model are extracted from simulations with the numerical weather forecast model RAMS. Dispersion over Rio de Janeiro region is also tested in a second experiment. / Modelos Lagrangianos estocásticos constituem ferramenta muito utilizada no estudo da dispersão de substâncias passivas na Camada Limite Atmosférica. Sua aplicação consiste em calcular a trajetória de milhares de partículas, que simulam numericamente a dispersão de uma substância em suspensão na atmosfera. Nesta tese, são apresentados e discutidos os conceitos básicos relacionados à Modelagem Lagrangiana Estocástica de Partículas, bem como suas principais características e sua implementação computacional, para o estudo da dispersão de partículas na atmosfera. Numa experimentação computacional, comparam-se os resultados obtidos com dados observacionais provenientes do experimento TRACT, realizado na Europa em 1992. Os dados de entrada necessários ao modelo de dispersão são extraídos de simulações do modelo de previsão numérica do tempo RAMS. A dispersão sobre o Estado do Rio de Janeiro é também testada em um segundo experimento.

Centro Latino-Americano de Física / Os processos estocásticos com ruído branco multiplicativo são objeto de atenção constante em uma grande área da pesquisa científica. A variedade de prescrições possíveis para definir matematicamente estes processos oferece um obstáculo ao desenvolvimento de ferramentas gerais para seu tratamento. Na presente tese, estudamos propriedades de equilíbrio de processos markovianos com ruído branco multiplicativo. Para conseguirmos isto, definimos uma transformação de reversão temporal de tais processos levando em conta que a distribuição estacionária de probabilidade depende da prescrição. Deduzimos um formalismo funcional visando obter o funcional gerador das funções de correlação e resposta de um processo estocástico multiplicativo representado por uma equação de Langevin. Ao representar o processo estocástico neste formalismo (de Grassmann) funcional eludimos a necessidade de fixar uma prescrição particular. Neste contexto, analisamos as propriedades de equilíbrio e estudamos as simetrias ocultas do processo. Mostramos que, usando uma definição apropriada da distribuição de equilíbrio e considerando a transformação de reversão temporal adequada, as propriedades usuais de equilíbrio são satisfeitas para qualquer prescrição. Finalmente, apresentamos uma dedução detalhada da formulação supersimétrica covariante de um processo markoviano com ruído branco multiplicativo e estudamos algumas das relações impostas pelas funções de correlação através das identidades de Ward-Takahashi. / Multiplicativewhite-noise stochastic processes continuously attract the attention of a wide area of scientific research. The variety of prescriptions available to define it difficults the development of general tools for its characterization. In this thesis, we study equilibrium properties of Markovian multiplicative white-noise processes. For this, we define the time reversal transformation for this kind of processes, taking into account that the asymptotic stationary probability distribution depends on the prescription. We deduce a functional formalism to derive a generating functional for correlation and response functions of a multiplicative stochastic process represented by a Langevin equation. Representing the stochastic process in this functional (Grassmann) formalism, we avoid the necessity of fixing a particular prescription. In this framework, we analyze equilibrium properties and study hidden symmetries of the process. We show that, using a careful definition of equilibrium distribution and taking into account the appropriate time reversal transformation, usual equilibrium properties are satisfied for any prescription. Finally, we present a detailed deduction of a covariant supersymmetric formulation of a multiplicativeMarkovian white-noise process and study some of the constraints it imposes on correlation functions using Ward-Takahashi identities.