Дисертації з теми "Réduction d'ordre de modèles"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 дисертацій для дослідження на тему "Réduction d'ordre de modèles".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
Casas, Francisco Javier. "Techniques de réduction d'ordre des modèles de circuits haute fréquence faiblement non linéaire." Limoges, 2003. http://www.theses.fr/2003LIMO0022.
Повний текст джерелаThe work reported deals with system-level modelling techniques of microwave circuits commonly employed in emission and reception telecommunications channels. The need of reduced models for simulation of communication systems is discussed, and the state of the art of this activity is summarized in the first chapter. The second chapter focus on model reduction methods for pumped and non-pumped circuits exhibiting a linear input-output behaviour. Then, in the third chapter, a model reduction technique for pumped weakly non linear circuits with memory is presented. In the fourth chapter, the proposed methods are applied to the modelling and simulation of a transmitter and receiver. Finally, the conclusions and perspectives of this work are summarised
Durbiano, Sophie. "Vecteurs caractéristiques de modèles océaniques pour la réduction d'ordre en assimilation de données." Université Joseph Fourier (Grenoble), 2001. http://www.theses.fr/2001GRE10228.
Повний текст джерелаPetit, Daniel. "Réduction de modèles de connaissance et identification de modèles d'ordre réduit : application aux processus de diffusion thermique." Aix-Marseille 1, 1991. http://www.theses.fr/1991AIX11348.
Повний текст джерелаGuillet, Jérôme. "Etude et réduction d'ordre de modèles linéraires structurés : application à la dynamique du véhicule." Phd thesis, Université de Haute Alsace - Mulhouse, 2011. http://tel.archives-ouvertes.fr/tel-00807199.
Повний текст джерелаBel, Liliane. "Sur la réduction des modèles linéaires : analyse de données en automatique." Paris 11, 1985. http://www.theses.fr/1985PA112306.
Повний текст джерелаTwo state space model reduction methods are studied: aggregation method and the balanced state space representation method. In the case of aggregation a new method of selecting eigenvalues is proposed, which is both geometrical and sequential. Problems of robustness of aggregation are evoked and resolved in some particular cases. The balanced state space representation is approached by means of contralibility and observability degrees. The notion of perturbability degree is introduced. Then we study the application of those two methods to reduced order compensator design. The two methods are finally applied to the system representing the launch booster Ariane flying
Missoffe, Alexia. "Réduction d'ordre de modèle d'un phénomène d'amortissement non-linéaire dans le cadre des microsystèmes." Phd thesis, Ecole Centrale Paris, 2010. http://tel.archives-ouvertes.fr/tel-00552076.
Повний текст джерелаVentura, Jérôme. "Réduction de modèle en convection forcée par des méthodes d'identification." Phd thesis, Chasseneuil-du-Poitou, Ecole nationale supérieure de mécanique et d'aérotechnique, 2010. http://www.theses.fr/2010ESMA0013.
Повний текст джерелаMost engineering processes dealing with aerothermics involve numerical simulations (CFD) which need to take transient phenomena into account to be realistic. Low order models and responses surfaces offer the engineer an alternative to the high costs associated with CFD processes, providing him with surrogates for the outputs of his interest. This work is devoted to extend the Modal Identification Method (MIM) in the field of fluid dynamics. MIM is a low order modelling techniques, based on the minimization of the discrepancy between a usually large reference model, and a low order model. It solves an optimization problem whose variables are the low order model parameters. Low order models are built for several kinds of turbulent flows. The circular cylinder in a crossflow provides us with several test cases. We cope with steady flows or unsteady ones, be them in forced convection or isothermal cases. Those models are able to produce outputs as different as velocity and temperature fields, heat fluxes. . . The compressible flow around a wing profile is also considered to investigate the sampling effect. Those models give quick approximates of the pressure distribution or the force coefficients
Lestandi, Lucas. "Approximations de rang faible et modèles d'ordre réduit appliqués à quelques problèmes de la mécanique des fluides." Thesis, Bordeaux, 2018. http://www.theses.fr/2018BORD0186/document.
Повний текст джерелаNumerical simulation has experienced tremendous improvements in the last decadesdriven by massive growth of computing power. Exascale computing has beenachieved this year and will allow solving ever more complex problems. But suchlarge systems produce colossal amounts of data which leads to its own difficulties.Moreover, many engineering problems such as multiphysics or optimisation andcontrol, require far more power that any computer architecture could achievewithin the current scientific computing paradigm. In this thesis, we proposeto shift the paradigm in order to break the curse of dimensionality byintroducing decomposition and building reduced order models (ROM) for complexfluid flows.This manuscript is organized into two parts. The first one proposes an extendedreview of data reduction techniques and intends to bridge between appliedmathematics community and the computational mechanics one. Thus, foundingbivariate separation is studied, including discussions on the equivalence ofproper orthogonal decomposition (POD, continuous framework) and singular valuedecomposition (SVD, discrete matrices). Then a wide review of tensor formats andtheir approximation is proposed. Such work has already been provided in theliterature but either on separate papers or into a purely applied mathematicsframework. Here, we offer to the data enthusiast scientist a comparison ofCanonical, Tucker, Hierarchical and Tensor train formats including theirapproximation algorithms. Their relative benefits are studied both theoreticallyand numerically thanks to the python library texttt{pydecomp} that wasdeveloped during this thesis. A careful analysis of the link between continuousand discrete methods is performed. Finally, we conclude that for mostapplications ST-HOSVD is best when the number of dimensions $d$ lower than fourand TT-SVD (or their POD equivalent) when $d$ grows larger.The second part is centered on a complex fluid dynamics flow, in particular thesingular lid driven cavity at high Reynolds number. This flow exhibits a seriesof Hopf bifurcation which are known to be hard to capture accurately which iswhy a detailed analysis was performed both with classical tools and POD. Oncethis flow has been characterized, emph{time-scaling}, a new ``physics based''interpolation ROM is presented on internal and external flows. This methodsgives encouraging results while excluding recent advanced developments in thearea such as EIM or Grassmann manifold interpolation
Telescu, Mihai. "Modélisation d'ordre réduit des interconnexions de circuits VLSI." Brest, 2007. http://www.theses.fr/2007BRES2038.
Повний текст джерелаLntegrated circuit designers are showing a growing interest in the effects of interconnect structures. Taking these effects into consideration during simulations has become a major goal. The main objective of this PhD was the development new model order reduction mathematical tools. VLSI interconnect applications were our main priority. Our model order reduction strategy supposes an initial modeling of the origjnal system using either a Laguerre or a Kautz representation. This manuscript contains a synthetic presentation 0f these orthogonal function bases. The five order reduction methods studied during this PhD are then presented. We make available several examples of application of methods to interconnect lines. Weillçistate, among other aspects, the possibility of obtaining Iow complexity equivalent circuits from our models and the possibility of performing reduced order modeling directly from data provided by full-wave simulation
Pereira, Da Silva Luciano. "Atténuation du bruit et des vibrations de structures minces par dispositifs piézoélectriques passifs : modèles numériques d'ordre réduit et optimisation." Thesis, Paris, CNAM, 2014. http://www.theses.fr/2015CNAM0944/document.
Повний текст джерелаPassive structural vibration and noise reduction by means of shunted piezoelectric patches is addressed in this thesis. The first part of the work concerns the finite element modeling of shunted piezoelectric systems. Firstly, an original finite element formulation, with only a couple of electric variables per piezoelectric patch (the global charge/ voltage), is analyzed and validated. Secondly, several reduced order models based on a normal mode expansion are proposed to solve the electromechanical problem. The convergence of these reduced order models is then analyzed for a resistive and a resonant shunt circuits. In the second part of the work, the concept of topology optimization, based on the Solid Isotropic Material with Penalization method (SIMP), is employed to optimize, in terms of damping efficiency, the geometry of piezoelectric patches as well as their placement on the host elastic structure. The proposed optimization procedure consists of distributing the piezoelectric material in such a way as to maximize the modal electromechanical coupling factor of the mechanical vibration mode to which the shunt is tuned, independently of the choice of electric circuit components. Numerical examples validate and demonstrate the potential of the proposed approach for the design of piezoelectric shunt devices. Finally, the last part of the work concerns the numerical modeling of noise and vibration reduction of thin structures in the low frequency range by using shunted piezoelectric elements. An efficient approach that can be applied to any thin continuous plates in an infinite baffle, independently of the boundary conditions, is proposed. An application example of a thin plate with reinforcements is presented and analyzed
Flament, Théo. "Modèle d'ordre réduit de structures non-linéaires pour l'aéroélasticité des turbomachines." Electronic Thesis or Diss., Paris, HESAM, 2023. http://www.theses.fr/2023HESAC038.
Повний текст джерелаThis work concerns the development of a reduced order model for geometric nonlinear structures, to replace the nonlinear structure solver within the framework of partitioned coupling for the numerical resolution of fluid-structure interaction problems, and thus the prediction of aeroelasticity phenomena encountered in turbomachinery. The construction of the reduced order model is based on a projection of the equations into a basis of reduced dimension, containing both linear modes of the structure and dual modes. The purpose of the latter is to enhance the basis of linear normal modes in order to capture the non-linearity. An original method for calculating the coefficients of the non-linear forces projected into this basis is also proposed. Forces are imposed on the structure, as opposed to the usual approach of imposed displacements. The same loading cases can thus be used to determine both the dual modes and the coefficients of the projected non-linear forces.In this thesis, the methodology to build the reduced order model is detailed. It is first validated on a simple case of a nonlinear Euler-Bernoulli beam subjected to different loading conditions, including a partitioned fluid-structure coupling involving vortex-induced vibrations. The ability of this reduced order model to replace a nonlinear finite element solver is demonstrated in this last application. Validation on 3D cases is also proposed, including the complex geometry of a realistic engine fan blade subjected to unsteady aerodynamic loading
Verdon, Nicolas. "Un système dynamique d'ordre réduit basé sur une approche APR-POD pour l'étude de l'interaction écoulement turbulent-particules." Phd thesis, Université de La Rochelle, 2007. http://tel.archives-ouvertes.fr/tel-00323067.
Повний текст джерелаLa méthode APR est d'abord testée dans des cas modèles simples : l'équation de convection-diffusion 2D et les équations de Burgers 1D et 2D. Comparée aux méthodes de résolution classique, l'APR permet de diminuer fortement les temps de calcul tout en conservant une précision du même ordre de grandeur.
Les équations de Navier-Stokes sont ensuite résolues à l'aide d'un code volumes finis 2D, utilisant un découplage vitesse-pression de type Van Kahn. Un algorithme de réduction a priori adapté à l'algorithme de projection est alors présenté et appliqué pour le cas de la cavité entraînée 2D à Re=10000. Les résultats obtenus sur un court intervalle de temps sont assez encourageants. Enfin, une démarche d'avancement temporel basée sur le couplage d'APR et de systèmes dynamiques est présentée.
Pomarede, Marie. "Investigation et application des méthodes d'ordre réduit pour les calculs d'écoulements dans les faisceaux tubulaires d'échangeurs de chaleur." Phd thesis, Université de La Rochelle, 2012. http://tel.archives-ouvertes.fr/tel-00815992.
Повний текст джерелаPomarède, Marie. "Investigation et application des méthodes d'ordre réduit pour les calculs d'écoulements dans les faisceaux tubulaires d'échangeurs de chaleur." Thesis, La Rochelle, 2012. http://www.theses.fr/2012LAROS355/document.
Повний текст джерелаThe objective of this thesis is to study the ability of model reduction for investigations of flow-induced vibrations in heat exchangers tube bundle systems.These mechanisms are a cause of major concern because heat exchangers are key elements of nuclear power plants and on-board stoke-holds.In a first part, we give a recall on heat exchangers functioning and on vi-bratory problems to which they are prone. Then, complete calculations leaded with the CFD numerical code Code_Saturne are carried out, first for the flow around a single circular cylinder (fixed then elastically mounted) and then for the case of a tube bundle system submitted to cross-flow. Reduced-order method POD is ap-plied to the flow resolution with fixed structures. The obtained results show the efficiency of this technique for such configurations, using stabilization methods for the dynamical system resolution in the tube-bundle case.Multiphase-POD, which is a method enabling the adaptation of POD to fluid-structure interactions, is applied. Large displacements of a single cylinder elastically mounted under cross-flow, corresponding to the lock-in phenomenon,are well reproduced with this reduction technique. In the same way, large displace-ments of a confined moving tube in a bundle are shown to be faithfully recon-structed.Finally, the use of model reduction is extended to parametric studies. First,we propose to use the method which consists in projecting Navier-Stokes equations for several values of the Reynolds number on to a unique POD basis. The resultsobtained confirm the fact that POD predictability is limited to a range of parameter values. Then, a basis interpolation method, constructed using Grassmann mani-folds and allowing the construction of a POD basis from other pre-calculated basis,is applied to basic cases
Saigre, Thomas. "Modélisation mathématique, simulation et réduction d’ordre de flux oculaires et leurs interactions : construire le jumeau numérique de l'oeil." Electronic Thesis or Diss., Strasbourg, 2024. http://www.theses.fr/2024STRAD052.
Повний текст джерелаThe human body is a complex system, and the human eye is no exception. Despite medical advances, many pathological questions remain. Mathematical and computational models complement clinical studies by revealing complex pathophysiological mechanisms.The eye, which can be accessed non-invasively, offers useful biological markers for diagnosing diseases. Understanding its behavior, pathologies and associated treatments is therefore essential.This thesis explores the modeling and simulation of ocular flows, notably heat transfer and aqueous humor flow. These approaches require rigorous clinical validation and take into account numerous parameters, both patient-specific and external. A global sensitivity analysis assesses their impact to guide clinicians. These computationally-intensive analyses benefit from certified model reduction methods, enabling accurate and faster simulations, favoring the integration of models into clinical practice
Iassamen, Nadia. "Modélisation et simulation des connexions intra et inter systèmes électroniques." Phd thesis, Université de Bretagne occidentale - Brest, 2013. http://tel.archives-ouvertes.fr/tel-00965131.
Повний текст джерелаLe, Thanh-Ha. "Analyses et mesures avancées du rayonnement électromagnétique d'un circuit intégré." Grenoble INPG, 2007. http://www.theses.fr/2007INPG0097.
Повний текст джерелаAnalyze two methods: DifferentiaI Power Analysis (DP A) and the Correlation Power Analysis (CP A). The attacks DPA and CPA are based on a power consumption mode!. However, because of the complexity of the circuit, the fact of having a rigorous model is not realistic. We proposed a new method, Partitioning Power Analysis (PP A), which makes it possible to improve the performance of an attack without a precis consumption mode!. Side channel attacks depend closely on the quality of measured signaIs. The performance of an attack can be definitely reduced if the noise level is too high or if the signaIs are not weIl aligned. We proposed solutions based on signal processing techniques to reduce the noise and the effects generated by the signal misalignment. These solutions make it possible to improve the attack effectiveness
Luu, Thi Hieu. "Amélioration du modèle de sections efficaces dans le code de cœur COCAGNE de la chaîne de calculs d'EDF." Thesis, Paris 6, 2017. http://www.theses.fr/2017PA066120/document.
Повний текст джерелаIn order to optimize the operation of its nuclear power plants, the EDF's R&D department iscurrently developing a new calculation chain to simulate the nuclear reactors core with state of the art tools. These calculations require a large amount of physical data, especially the cross-sections. In the full core simulation, the number of cross-section values is of the order of several billions. These cross-sections can be represented as multivariate functions depending on several physical parameters. The determination of cross-sections is a long and complex calculation, we can therefore pre-compute them in some values of parameters (online calculations), then evaluate them at all desired points by an interpolation (online calculations). This process requires a model of cross-section reconstruction between the two steps. In order to perform a more faithful core simulation in the new EDF's chain, the cross-sections need to be better represented by taking into account new parameters. Moreover, the new chain must be able to calculate the reactor in more extensive situations than the current one. The multilinear interpolation is currently used to reconstruct cross-sections and to meet these goals. However, with this model, the number of points in its discretization increases exponentially as a function of the number of parameters, or significantly when adding points to one of the axes. Consequently, the number and time of online calculations as well as the storage size for this data become problematic. The goal of this thesis is therefore to find a new model in order to respond to the following requirements: (i)-(online) reduce the number of pre-calculations, (ii)-(online) reduce stored data size for the reconstruction and (iii)-(online) maintain (or improve) the accuracy obtained by multilinear interpolation. From a mathematical point of view, this problem involves approaching multivariate functions from their pre-calculated values. We based our research on the Tucker format - a low-rank tensor approximation in order to propose a new model called the Tucker decomposition . With this model, a multivariate function is approximated by a linear combination of tensor products of one-variate functions. These one-variate functions are constructed by a technique called higher-order singular values decomposition (a « matricization » combined with an extension of the Karhunen-Loeve decomposition). The so-called greedy algorithm is used to constitute the points related to the resolution of the coefficients in the combination of the Tucker decomposition. The results obtained show that our model satisfies the criteria required for the reduction of the data as well as the accuracy. With this model, we can eliminate a posteriori and a priori the coefficients in the Tucker decomposition in order to further reduce the data storage in online steps but without reducing significantly the accuracy
Ponsignon, Jean-Charles. "Réduction fuchsienne et modèles stellaires." Thesis, Reims, 2013. http://www.theses.fr/2013REIMS046/document.
Повний текст джерелаThe object of this thesis is the study of a non linear differential equation stemming from a stellar model. After reduction and unknowns changes and variables, we achieve to an analytic second member in each of the problem variables and well chosen functions. Then we show that the solutions can be described in a space of absolute convergent series. This theorem of existence will be used as an elementary brick to a nearby method of Fuchsian reduction. The objective was to obtain a development which elicits arbitrary various constants inherent to this type of equations
Da, Costa Cardoso Inês. "Disciplinary surrogates for the analysis and optimization of partitioned multidisciplinary systems : application to aeroelastic coupling in wing design." Electronic Thesis or Diss., Toulouse, ISAE, 2024. http://www.theses.fr/2024ESAE0057.
Повний текст джерелаMultidisciplinary design analysis and optimization (MDAO) is commonly used in preliminary aircraft design studies. Nevertheless, most of these studies are conducted using low-fidelity simulations of the different participating disciplines. With growing interest in new aircraft configurations for which little to no experimental data is available, the need for high-fidelity simulations increases. However, due to their computational cost and lack of derivative information, the use of high-fidelity solvers in MDAO remains scarce. To address this issue, Bayesian optimization approaches, where optimization is carried out using surrogates of some or all of the system's functions have risen as cost-saving, derivative-free approaches. One such approach is the Efficient Global Multidisciplinary Design Optimization (EGMDO) algorithm. EGMDO reduces the cost of the MDO problem by replacing the disciplinary solvers by adaptively enriched Gaussian Process (GP) surrogates. Although EGMDO successfully reduces the computational burden, it does not possess a constraint handling strategy and therefore is not directly useful for most preliminary design studies. Moreover, using disciplinary surrogates when dealing with high-fidelity solvers can lead to additional challenges. Indeed, the output of these solvers is typically some quantity discretized over a mesh, for which surrogate models cannot be directly built. Model order reduction strategies provide a possible lead to address this issue, as they allow to represent the high-dimensional disciplinary outputs in a lower dimensional space. The Disciplinary Proper Orthogonal Decomposition and Interpolation (DPOD+I) strategy developed this idea by combining disciplinary global POD bases with GP interpolation. But although the DPOD+I strategy successfully enabled the use of disciplinary surrogates with high-fidelity solvers, its performance is significantly decreased for problems where the number of basis vectors required to obtain a small projection error is large. This thesis proposes to address the shortcomings of both EGMDO and DPOD+I frameworks by developing an extension of the EGMDO algorithm to constrained optimization problems and proposing a different model reduction strategy for the DPOD+I approach, based on the interpolation of local POD bases. We equally explore the possibility of gradient-based optimization through exploitation of the analytical derivatives of the disciplinary surrogates. Application of the proposed developments to the aeroelastic coupling of a wing shows that the proposed approach is capable of reducing the computational cost of the multidisciplinary analysis and optimization problems often found in preliminary aircraft design studies
Benaceur, Amina. "Réduction de modèles en thermo-mécanique." Thesis, Paris Est, 2018. http://www.theses.fr/2018PESC1140/document.
Повний текст джерелаThis thesis introduces three new developments of the reduced basis method (RB) and the empirical interpolation method (EIM) for nonlinear problems. The first contribution is a new methodology, the Progressive RB-EIM (PREIM) which aims at reducing the cost of the phase during which the reduced model is constructed without compromising the accuracy of the final RB approximation. The idea is to gradually enrich the EIM approximation and the RB space, in contrast to the standard approach where both constructions are separate. The second contribution is related to the RB for variational inequalities with nonlinear constraints. We employ an RB-EIM combination to treat the nonlinear constraint. Also, we build a reduced basis for the Lagrange multipliers via a hierarchical algorithm that preserves the non-negativity of the basis vectors. We apply this strategy to elastic frictionless contact for non-matching meshes. Finally, the third contribution focuses on model reduction with data assimilation. A dedicated method has been introduced in the literature so as to combine numerical models with experimental measurements. We extend the method to a time-dependent framework using a POD-greedy algorithm in order to build accurate reduced spaces for all the time steps. Besides, we devise a new algorithm that produces better reduced spaces while minimizing the number of measurements required for the final reduced problem
Oulefki, Abdelhakim. "Réduction de modèles thermiques par amalgame modal." Phd thesis, Ecole Nationale des Ponts et Chaussées, 1993. http://tel.archives-ouvertes.fr/tel-00523620.
Повний текст джерелаAit, Yahia Ali. "Réduction de modèles d'état de systèmes thermiques." Marne-la-vallée, ENPC, 1998. http://www.theses.fr/1998ENPC9819.
Повний текст джерелаWahl, Jean-Baptiste. "The Reduced basis method applied to aerothermal simulations." Thesis, Strasbourg, 2018. http://www.theses.fr/2018STRAD024/document.
Повний текст джерелаWe present in this thesis our work on model order reduction for aerothermal simulations. We consider the coupling between the incompressible Navier-Stokes equations and an advection-diffusion equation for the temperature. Since the physical parameters induce high Reynolds and Peclet numbers, we have to introduce stabilization operators in the formulation to deal with the well known numerical stability issue. The chosen stabilization, applied to both fluid and heat equations, is the usual Streamline-Upwind/Petrov-Galerkin (SUPG) which add artificial diffusivity in the direction of the convection field. We also introduce our order reduction strategy for this model, based on the Reduced Basis Method (RBM). To recover an affine decomposition for this complex model, we implemented a discrete variation of the Empirical Interpolation Method (EIM) which is a discrete version of the original EIM. This variant allows building an approximated affine decomposition for complex operators such as in the case of SUPG. We also use this method for the non-linear operators induced by the shock capturing method. The construction of an EIM basis for non-linear operators involves a potentially huge number of non-linear FEM resolutions - depending on the size of the sampling. Even if this basis is built during an offline phase, we usually can not afford such expensive computational cost. We took advantage of the recent development of the Simultaneous EIM Reduced basis algorithm (SER) to tackle this issue
Gaudel, Romaric. "Paramètres d'ordre et sélection de modèles en apprentissage : caractérisation des modèles et sélection d'attributs." Phd thesis, Université Paris Sud - Paris XI, 2010. http://tel.archives-ouvertes.fr/tel-00549090.
Повний текст джерелаTrinh, Duy Khanh. "Méthodes d'homogénéisation d'ordre supérieur pour les matériaux architecturés." Paris, ENMP, 2011. https://pastel.hal.science/pastel-00677046.
Повний текст джерелаStandard homogenization with Cauchy medium has got many successes in heterogeneous material study. However, it is limited when the characterised length of applied macroscopic loadings is at the same order of material's heterogeneities; in particular in case of grand gradient solicitation, for instance in sandwich bending case. The homogenization with generalized continua is then necessary to overcome these limits and extend the validation of homogenization approach out of strict scale separation condition. There are recently much progression in this domain; the contributions have developed mainly material multi-scale modelling by Cosserat (or micropolar) medium, by second gradient medium, by couple-stress medium and recently by micromorphic continuum. There is many methods for multi-scale modelling an heterogeneous material : computational average on RVE; asymptotic developement method or empirical methods. My contribution follows the local fields average technique, with the purpose of looking for a new method which is not so complicated but systematically enough for modelling composite material by generalized continuum media. Motivation of this method is its good applicability to any micro--structure, and also relatively simple to apply non-linear behavior (elasto-plastic behavior)
Labbé, Sébastien. "Réduction paramétrée de spécifications formées d'automates communicants : algorithmes polynomiaux pour la réduction de modèles." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2007. http://tel.archives-ouvertes.fr/tel-00180174.
Повний текст джерелаL'idée que nous proposons consiste à contourner ce phénomène en appliquant des techniques de réduction paramétrée, pouvant être désignées sous le terme anglo-saxon "slicing'', en amont d'une analyse complexe. Cette analyse peut ainsi être effectuée a posteriori sur une spécification réduite, donc potentiellement moins sujette à l'explosion combinatoire. Notre méthode de réduction paramétrée est basée sur des relations de dépendances dans la spécification sous analyse, et est fondée principalement sur les travaux effectués par les communautés de la compilation et du slicing de programmes. Dans cette thèse nous établissons un cadre théorique pour les analyses statiques de spécifications formées d'automates communicants, dans lequel nous définissons formellement les relations de dépendances mentionnées ci-dessus, ainsi que le concept de "tranche" de spécification par rapport à un "critère" de réduction. Ensuite, nous décrivons et démontrons les algorithmes efficaces que nous avons mis au point pour calculer les relations de dépendances et les tranches de spécifications, et enfin nous décrivons notre mise en oeuvre de ces algorithmes dans l'outil "Carver", pour la réduction paramétrée de spécifications formées d'automates communicants.
Lang, Frédéric. "Modèles de la bêta-réduction pour les implantations." Lyon, École normale supérieure (sciences), 1998. http://www.theses.fr/1998ENSL0103.
Повний текст джерелаRavazzola, Patrice. "Approximation et réduction de modèles en traitement d'antenne." Grenoble INPG, 1990. http://www.theses.fr/1990INPG0104.
Повний текст джерелаJurczenko, Emmanuel. "Modèles d'évaluation des prix des actifs financiers et moments d'ordre supérieur." Paris 1, 2006. http://www.theses.fr/2006PA010007.
Повний текст джерелаCornillon, Emilien. "Modèles qualitatifs de réseaux génétiques : réduction de modèles et introduction d'un temps continu." Thesis, Université Côte d'Azur (ComUE), 2017. http://www.theses.fr/2017AZUR4085/document.
Повний текст джерелаFormal methods from computer science constitute a powerful tool for the modelling of gene networks, including the study of their dynamics. The discrete modelling of René Thomas allows for a proper representation of biological knowledge as well as for use of formal methods. These models have two main limitations: the underlying combinatorics does not allow one to process very large networks, and the chronometric aspects are not taken into account. This thesis offers two contributions according to these issues. The design of gene network models begins with a selectiCalibrion of the most relevant entities. The resulting networks are often too large, and we show how to reduce them without altering the important dynamic properties. Here, we define a completely formal framework, inspired by a technique from Aurélien Naldi, driving the suppression of variables or thresholds. These reductions preserve the asymptotic behaviour. We formally prove the asymptotic equivalence of different published models for the same network. In order to take into account chronometric information that are crucial in some systems (e.g. circadian cycle), we define a hybrid formalism based on the Thomas' formalism where expression levels are discrete but time is continuous. This framework allows for the construction of an abstract model of the circadian clock in mammals. The model explains with very few variables the robustness of the system when submitted to duration changes of the day/night alternation
Royer, Julien. "Processus ARCH d'ordre infini, Bêtas dynamiques et applications financières." Electronic Thesis or Diss., Institut polytechnique de Paris, 2022. http://www.theses.fr/2022IPPAG012.
Повний текст джерелаThe modeling of financial time series is made difficult by the presence of stylized facts. These empirical statistical properties led to the development of heteroskedastic nonlinear models. Infinite ARCH specifications have been introduced to allow finer modeling of these stylized facts, and in particular the phenomenon of strong persistence of volatility shocks. We present new extensions to these flexible models and study their inference. First, we consider an asymmetric infinite ARCH model. We prove the existence of a stationary solution and establish the asymptotic properties of the quasi-maximum likelihood estimator in this framework. In particular, we allow the parameter to lie on the boundary of the parameter space, precluding asymptotic normality. Moreover, we introduce a portmanteau test assessing the goodness-of-fit of the model on data. We also propose a test for the presence of memory and asymmetry. In a second time, we consider the modeling of the coefficients of a conditional linear regression. Linear factor models are key to many financial models and regression coefficients are often wrongfully assumed constant. We propose a model allowing for dynamic beta coefficients within the framework of multivariate infinite ARCH models. In particular, we allow the addition of exogenous variables in the dynamics of conditional betas and discuss potential candidates. We establish the conditions of existence of a stationary solution and discuss the existence of its moments. Finally, we consider an asset pricing exercise based on dynamic betas. To this end, we extend the results of statistical tests in the case of score-driven betas and propose a bootstrap procedure. Additionally, we introduce a two-step estimation method to measure the dynamic risk premia underlying the asset pricing model
Guillaume, Philippe. "Dérivées d'ordre supérieur en conception optimale de forme." Toulouse 3, 1994. http://www.theses.fr/1994TOU30041.
Повний текст джерелаBroze, Laurence. "Réduction, identification et estimation des modèles à anticipation rationnelles." Doctoral thesis, Universite Libre de Bruxelles, 1986. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/213559.
Повний текст джерелаGerges, Youssef. "Méthodes de réduction de modèles en vibroacoustique non-linéaire." Phd thesis, Université de Franche-Comté, 2013. http://tel.archives-ouvertes.fr/tel-00933423.
Повний текст джерелаCasagranda, Stefano. "Modélisation, analyse et réduction des systèmes biologiques." Thesis, Université Côte d'Azur (ComUE), 2017. http://www.theses.fr/2017AZUR4049/document.
Повний текст джерелаThis thesis deals with modeling, analysis and reduction of various biological models, with a focus on gene regulatory networks in the bacterium E. coli. Different mathematical approaches are used. In the first part of the thesis, we model, analyze and reduce, using classical tools, a high-dimensional transcription-translation model of RNA polymerase in E. coli. In the second part, we introduce a novel method called Principal Process Analysis (PPA) that allows the analysis of high-dimensional models, by decomposing them into biologically meaningful processes, whose activity or inactivity is evaluated during the time evolution of the system. Exclusion of processes that are always inactive, and inactive in one or several time windows, allows to reduce the complex dynamics of the model to its core mechanisms. The method is applied to models of circadian clock, endocrine toxicology and signaling pathway; its robustness with respect to variations of the initial conditions and parameter values is also tested. In the third part, we present an ODE model of the gene expression machinery of E. coli cells, whose growth is controlled by an external inducer acting on the synthesis of RNA polymerase. We describe our contribution to the design of the model and analyze with PPA the core mechanisms of the regulatory network. In the last part, we specifically model the response of RNA polymerase to the addition of external inducer and estimate model parameters from single-cell data. We discuss the importance of considering cell-to-cell variability for modeling this process: we show that the mean of single-cell fits represents the observed average data better than an average-cell fit
Krob, Michel. "Identification aveugle de modèles non linéaires à l'aide de statistiques d'ordre supérieur." Paris 11, 1994. http://www.theses.fr/1994PA112002.
Повний текст джерелаBaron, Céline. "Contribution à l'estimation de paramètres physiques à l'aide de modèles d'ordre réduit." Poitiers, 2003. http://www.theses.fr/2003POIT2349.
Повний текст джерелаThe works presented in the thesis dissertation deal with a methdology for the determination of uncertainty parametric domains and for the estimation of physical parameters using reduced order models. Firstly, we present a methodology for the determination of uncertainty parametric domains with a hypothesis of bounded error. This method is based on global identification approach defined by J. RICHALET. Uncertainty domains correspond to iso-criterion curves given by the minimization of the quadratic criterion. We also present a methodology for the estimation of physical parameters using reduced order models. Modelling error, responsible of a deterministic bias of the estimator, is taken into account using a constrained black box model. These constraints are introduced using time moments or frequency ones. Application examples are used to validate this methodology
Gerhard, Pierre. "Réduction de modèles cinétiques et applications à l'acoustique du bâtiment." Thesis, Strasbourg, 2020. https://publication-theses.unistra.fr/public/theses_doctorat/2020/pierre_gerhard_2020_ED269.pdf.
Повний текст джерелаIn this PhD thesis we are interested in the study of different numerical methods for the simulation of high-frequency acoustic problems taking place on the scale of the building. ln the high-frequency approximation the sound propagation can be modeled through a kinetic transport equation paired with the boundary conditions that describe the specular or diffuse nature of the reflections with the boundaries of the demain. ln the first part of this paper we will tackle the resolution of this model, posed in a seven-dimensional space, by the application of the discrete ordinates method. This method consists in the discretisation of the velocity space into a finite number of allowable directions and leads to a system of coupled transport equations having lest ail velocity dependence. Secondly, we will apply the method of moments with entropie closure. The resulting system, of a hyperbolic nature, allows the macroscopic dynamics to be described by only two conservative variables. ln two dimensions, the resolution of these models is performed through a finite volume scheme implemented on GPU. ln three dimensions, a discontinuous Galerkin method is used which can be executed on a hybrid GPU/CPU architecture. For comparative purposes, a particle method has aise been implemented and solved using a fully GPU-parallelised ray-tracing algorithm. Finally, we will apply and compare the developed methods on several test cases specific to room acoustics
Henri, Thibault. "Réduction de modèles par des méthodes de décomposition orthogonale propre." Rennes, INSA, 2004. http://www.theses.fr/2004ISAR0005.
Повний текст джерелаModelling complex systems leads one to solve problems of very high size numerically, for example in case of finite elements discretization. An alternative consists in using a priori known data to reduce the size of the system to solve. Given the solution u of an evolution problem, proper orthogonal decomposition (POD) methods allow one to define a basis of functions from a priori known snapshots u(ti) on which the problem is projected : one speaks of POD-Galerkin methods. This method has been especially implemented in case of fluids mechanics problems. We present a synthesis of POD methods. We show convergence results and estimates in the parabolic case and in the case of general equations in fluids mechanics. The stability of the method is proven. Numerical computations show that the size of systems to solve is very small compared to classical finite elements methods. We present an interesting application to compute the solution of optimal control problems
Boucher-Lafleur, Véronique. "Les cycles électoraux au Canada : une question d'ordre?" Master's thesis, Université Laval, 2021. http://hdl.handle.net/20.500.11794/70287.
Повний текст джерелаThis research project is studying electoral cycles in Canada. Two main theories from this field of research are mobilized in this paper. The first one is the theory of first and second order election and the second is the theory of electoral balancing. In a multi-level electoral system, these two theories suggest that electoral results from different electoral levels are related to each other. These theories assumed that national elections are more important to voters. Therefore, regional electoral results are explained by factors coming from the national electoral level. In Canada, this theory implies that provincial electoral results are influenced by the previous federal election. To study this phenomenon, all provincial and federal electoral results from 1988 until today were coded into a dataset. A first set of linear and polynomial regressions was performed using federal electoral results as the independent variable and provincial electoral results as the dependent variable. Then, another set of regressions was performed but to study the opposite relation. Electoral results were split by province and into five families of political parties: liberal, conservative, labor, ecologist, and regionalist. Results vary between provinces: federal elections have more influence on provincial results in some provinces while the opposite relation is more important in others. This paper also concludes that there is a dynamic of electoral balancing between federal and provincial results in certain provinces, but that no general tendency can be seen across Canada. This research project is a significant contribution to the literature on electoral cycles in Canada.
Drifi, Karim. "Estimation du mouvement par assimilation de données dans des modèles dynamiques d'ordre réduit." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2013. http://tel.archives-ouvertes.fr/tel-00846688.
Повний текст джерелаDrifi, Karim. "Estimation du mouvement par assimilation de données dans des modèles dynamiques d'ordre rédui." Paris 6, 2013. http://www.theses.fr/2013PA066275.
Повний текст джерелаMotion estimation is a major challenge in the field of image sequenceprocessing. The thesis is a study of the dynamics of geophysical flowsvisualized by satellite imagery. Satellite image sequences are currentlyunderused for the task of dynamics estimation. A good understanding ofgeophysical flows allows a better analysis and forecast of phenomena indomains such as oceanography and meteorology. Data assimilation provides anexcellent framework for achieving a compromise between heteorogenous data,especially numerical models and observations. Hence, in this thesis we set outto apply variational data assimilation methods, such as 4D-Var, to estimatemotion in image sequences. Asmajor drawbacks of applying 4D-Var are theconsiderable computation time and memory required, we define and use a modelreduction method in order to significantly decrease the necessary computationtime and memory. We then explore the possibilities that reduced models providefor motion estimation, particularly the possibility of strictly imposing someknown constraints on the computed solutions. Different kinds of reductions arediscussed, using a proper orthogonal decomposition, a sine basis fordivergence-free motion and a basis dedicated to a particular spatialdomain. In each case, results are presented on both synthetic and satelittedata
Balabonski, Thibaut. "La plein paresse, une certain optimalité : partage de sous-termes et stratégies de réduction en réécriture d'ordre supérieur." Paris 7, 2012. http://www.theses.fr/2012PA077198.
Повний текст джерелаThis dissertation investigates efficient evaluation strategies for functional programming languages and higher-order rewriting, with two constraints: no early evaluation of non-instantiated function bodies, and a restriction of sharing mechanisms to complete subterms. In the lambda-calculus, as well as in the more general framework of orthogonal second-order rewriting, three evaluation models are considered: optimal reduction without sharing, fully lazy reduction à la Wadsworth, and optimal family reduction à la Levy. This defines a unified notion of weak optimality, that teaches us that Levy-optimality in weak reduction can be implemented by a fully lazy strategy as well as a strategy without sharing. It is however also proved that such an optimal strategy without sharing cannot be computable, while sharing allows in a number of cases the definition of very simple optimal strategies. To reach these conclusions, this dissertation makes good use of the contraints to reduce the discussion to simpler well-known Systems. On the one hand the graphs modelling the sharing of subterms are represented by labelled terms, thanks to the axiomatic framework of Sharing-via-Labelling Systems (SLS) developped for the occasion. On the other hand, a transformation process inspired from lambda-lifting defines a bisimulation between any weak second-order System and some first-order System, said bisimulation preserving in particular the length of reduction sequences, the possible orthogonality of the Systems, and sharing. Thus graphs for higher-order rewriting are formalized and studied, without graphs nor higher-order
Sévellec, Muriel. "Réduction de l'ordre des systèmes à fonctions de transfert rationnelles." Brest, 1993. http://www.theses.fr/1993BRES2015.
Повний текст джерелаBériot, Hadrien. "Eléments finis d'ordre élévé pour l'opérateur de Galbrun en régime harmonique." Compiègne, 2008. http://www.theses.fr/2008COMP1791.
Повний текст джерелаInvestments in Computational methods for aeroacoustic have grown recently due to their application in the transportation sector. To account for the flow effects in the propagation zones, the Linearized Euler Equations (LEE) are generally preferred. If robust methods are available in the time domain, practically no methods have been proposed in the frequency domain on unstructured grids. In this manuscript, we propose an original finite element method for the operator of Galbrun, known to be physically equivalent to the LEE. To stabilize the numerical scheme and improve the performance, a mixed high-order approach is employed. A complete dispersion analysis of high-order methods in the presence of flow is performed. Validations are proposed in ducted swirling flows based on Pridmore-Brown's model. Acoustic and hydrodynamic phenomenon are examined as well as the use of non reflecting conditions
Lombard, Jean-Pierre. "Contribution à la réduction des modèles éléments finis par synthèse modale." Phd thesis, Université de Franche-Comté, 1999. http://tel.archives-ouvertes.fr/tel-00351693.
Повний текст джерелаUne nouvelle technique de synthèse modale est proposée: elle permet l'obtention de modèles réduits exempts de coordonnées de jonction. Ses performances sont comparées vis-à-vis des méthodes de sous-structuration classiques lors de la réduction de modèles de structures automobiles. Un modèle sous-structuré d'une caisse nue de véhicule de 450 000 ddl, présentant une forte connectivité, est condensé à 2000 ddl par cette méthode.
L'usage de transformations de coordonnées physiques avec la prise en compte des chargements appliqués est généralisé à l'ensemble des méthodes de réduction.
La robustesse des modèles réduits paramétrés est traitée ainsi que les techniques d'adaptation, par réductions multiplies, des données issues du modèle aux observations mesurées sur la structure dans des objectifs de recalage de modèle et d'optimisation de comportement.
Poquet, Christophe. "MODÈLES STOCHASTIQUES INTERAGISSANTS : SYNCHRONISATION ET RÉDUCTION À UN SYSTÈME DE PHASES." Phd thesis, Université Paris-Diderot - Paris VII, 2013. http://tel.archives-ouvertes.fr/tel-00966053.
Повний текст джерелаArnoux, Adrien, and Adrien Arnoux. "Réduction des modèles numériques en dynamique linéaire basse fréquence des automobiles." Phd thesis, Université Paris-Est, 2012. http://tel.archives-ouvertes.fr/tel-00806508.
Повний текст джерелаArnoux, Adrien. "Réduction des modèles numériques en dynamique linéaire basse fréquence des automobiles." Thesis, Paris Est, 2012. http://www.theses.fr/2012PEST1019/document.
Повний текст джерелаThe objective of this research is to construct a reduced-order model to predict the dynamical response, in the LF band, of the stiff parts of a complete automotive vehicle in order to facilitate the draft design. The vehicles under consideration have many elastic modes in LF due to the presence of many flexible parts and equipments. To build such a model, we introduced a non-usual basis of the admissible space of global displacements. The construction of this basis requires the decomposition of the domain of the structure. This subdomain decomposition is performed by using the Fast Marching Method that we have extended to take into account the high complexity of the mesh of an automotive vehicle. Then the matrix equations of the FE model are projected on this basis. To take into account the system parameters uncertainties, the model uncertainties induced by the modeling errors and finally, the uncertainties related to the neglecting of local contributions in the reduced-order model, a nonparametric probabilistic model of the three sources of uncertainties has been implemented on the reduced-order model constructed with the global displacements eigenvectors. The dispersion parameters of the probabilistic model are identified using the maximum likelihood method and the responses obtained from a stochastic reference model which includes experimental data resulting from previous works. This stochastic model which has been designed for the prediction of the global displacements of the rigid parts in the LF band is validated on a simple structure of an automotive model and has been successfully applied on a complete model of automotive vehicle