Submitted by (ana.araujo@ufrpe.br) on 2016-07-07T11:52:38Z No. of bitstreams: 1 Jose Rodrigo Santos Silva.pdf: 13129069 bytes, checksum: b427ff42ec7918c3d0cf7f63798ed648 (MD5) / Made available in DSpace on 2016-07-07T11:52:38Z (GMT). No. of bitstreams: 1 Jose Rodrigo Santos Silva.pdf: 13129069 bytes, checksum: b427ff42ec7918c3d0cf7f63798ed648 (MD5) Previous issue date: 2014-09-19 / The objective of this study was to uncloak the dynamic of climate of Brazil, seeking to measure the regularity and the long range autocorrelation of daily climate series of temperature of air (average, maximum, minimum, and temperature range), relative humidity of air average and wind speed average. The data were obtained by Instituto Nacional de Meteorologia (INMET), at 264 meteorological stations, in the period from January 1990 to December 2012. We use the Detrended Fluctuation Analysis to realize the estimation of the Hurst exponent, the Multiscale Sample Entropy to estimating the entropy of series and the Kriging to interpolate the estimates made. We observed that higher latitudes tend to attenuate the mean of temperatures of air maximum, minimum and average, but increase the variability of the same. This inversion of the magnitudes of the mean and standard deviation is also observed in the relative humidity of air. The means of the estimated Hurst exponents estimated for Brazil were 0.81, 0.79, 0.81, 0.77, 0.83 and 0.64, and the estimated Sample Entropy, 1.39, 1.78, 1.46, 1.41, 1.56 and 1.66, respectively for average, maximum and minimum temperatures of air, temperature range, relative humidity of air average and wind speed average. The values of the estimated Hurst exponents showed a positive correlation with latitude in the temperature variables studied. Such a correlation was not observed in other variables. This a correlation was not observed in other variables. The regularities of climate series in Brazil were medians. Spatially, the greatest changes occurred in estimates of entropies in the scale 1 to 2 of , in the Multiscale Sample Entropy. As from ≥2 the changes observed were more subtle. We observe the influence of the Equatorial Continental air mass in entropy of temperatures daily average and maximum of air. The climatic factor of altitude influenced with more frequently in the observed results, mainly on temperature variables. In some cases, the continentality and the air masses were also identified as important factors in characterizing the spatial distribution of estimates made. / O objetivo deste estudo foi desvendar a dinâmica climática do Brasil, buscando mensurar a regularidade e a autocorrelação de longo alcance em séries climáticas diárias de temperatura do ar (média, máxima, mínima, e amplitude térmica), umidade relativa média do ar e velocidade média diária do vento. Os dados foram obtidos pelo Instituto Nacional de Meteorologia, em 264 estações meteorológicas, no período de janeiro de 1990 a dezembro de 2012. Utilizamos o Detrended Fluctuation Analysis para realizar a estimativa do expoente de Hurst, o Multiscale Sample Entropy para as estimativas da entropia das séries e o Kriging para a interpolação das estimativas realizadas. Observamos que maiores latitudes tendem a atenuar as médias das temperaturas máxima, mínima e média do ar, porém aumentam a variabilidade das mesmas. Esta inversão entre as magnitudes da média e do desvio padrão também é observado na umidade relativa média do ar. As médias dos expoentes de Hurst estimados para todo o Brasil foram 0,81; 0,79; 0,81; 0,77; 0,83 e 0,64; e do Sample Entropy estimado, 1,39; 1,78; 1,46; 1,41; 1,56 e 1,66, respectivamente para séries diárias de temperatura média, máxima e mínima do ar, amplitude térmica do ar, umidade relativa média do ar e velocidade média do vento. Os valores do expoentes de Hurst estimados apresentaram uma correlação positiva com a latitude nas variáveis de temperatura do ar estudadas. Tal correlação não foi observada nas demais variáveis. As regularidades das séries climáticas no Brasil foram medianas. Espacialmente, as maiores alterações nas estimativas das entropias ocorreram na escala 1 para a 2 de , no Multiscale Sample Entropy. A partir de ≥2 as mudanças observadas foram mais sutis. Observamos influência da massa de ar Equatorial Continental na entropia das temperaturas do ar média e máxima diárias. O fator climático da altitude atuou com maior frequência sob os resultados observados, principalmente nas variáveis de temperatura. Em alguns casos, a continentalidade e as massas de ar também foram apontados como fatores importantes na caracterização da distribuição espacial das estimativas realizadas.

Orientador: Edison Zacarias da Silva / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Fisica Gleb Wataghin / Made available in DSpace on 2018-08-09T02:46:53Z (GMT). No. of bitstreams: 1 Faccin_GiovaniManzeppi_M.pdf: 18605908 bytes, checksum: 754a46d31802e007eede531e959a8d17 (MD5) Previous issue date: 2007 / Resumo: A presente dissertação tem como proposta iniciar o desenvolvimento de um sistema de softwares para a simulação multiescala de materiais, com foco na nanociência. Para esta finalidade um sistema de dinâmica molecular foi construído partindo-se do zero. A partir deste sistema, diversas aplicações foram feitas em problemas envolvendo o estudo de propriedades térmicas e estruturais de pequenos aglomerados metálicos, assim como a evolução de fraturas dinâmicas. Uma vez validado o sistema através destes cálculos, alguns componentes necessários ao acoplamento dinâmico entre a escala de estrutura atômica, modelada através da dinâmica molecular, e a escala de meios contínuos, modelada através do método dos elementos finitos, foram desenvolvidos. A partir destes componentes, algumas simulações ilustrativas do funcionamento do acoplamento entre escalas foram então realizadas / Abstract: The present work's objetive is the development of a software system for the multiscale simulation of materials focused on nanoscience applications. To achieve this goal, a molecular dynamics system was built from scratch. Using this system, the structural and thermal properties of small metallic clusters and also the problem of dynamic fracture were studied. Once the system was validated with these problems, some components necessary to the coupling between the atomic scale, modelled through molecular dynamics, and the continuum scale, modelled through the finite element method, were developed. Using these components, simulations that illustrate the coupling between scales were then performed. / Mestrado / Física da Matéria Condensada / Mestre em Física

O corpo humano é um sistema complexo composto por vários subsistemas interdependentes, que interagem entre si em várias escalas. Sabe-se que essa complexidade fisiológica tende a diminuir com a presença de doenças e com o avanço da idade, reduzindo a capacidade de adaptação dos indivíduos. No sistema cardiovascular, uma das maneira de se avaliar sua dinâmica regulatória é através da análise da variabilidade da frequência cardíaca (VFC). Os métodos clássicos de análise da VFC são baseados em modelos lineares, como é o caso da análise espectral. Contudo, como os mecanismos fisiológicos de regulação cardíaca apresentam características não lineares, as análises utilizando tais modelos podem ser limitadas. Nos últimos tempos, várias propostas de métodos não lineares têm surgido. Porém, não se sabe de uma medida consistente com o conceito de complexidade fisiológica, onde tanto os regimes periódicos como aleatórios são caracterizados como perda de complexidade. Baseado no conceito de complexidade fisiológica, esta tese propõe novos métodos de análise não lineares para séries de VFC. Os métodos consistem da generalização de medidas de entropia já existentes, utilizando a mecânica estatística não aditiva de Tsallis e a técnica de geração de dados substitutos. Foi definido um método, chamado de qSDiff, que calcula a diferença entre a entropia de um sinal e a entropia média de suas séries substitutas. O método de entropia utilizado consiste de uma generalização da entropia amostral (SampEn), utilizando o paradigma não aditivo. Das séries qSDiff foram extraídos três atributos, que foram avaliados como possíveis índices de complexidade fisiológica. A entropia multiescala (MSE) também foi generalizada seguindo o paradigma não aditivo, e os mesmos atributos foram calculados em várias escalas. Os métodos foram aplicados em séries reais de VFC de humanos e de ratos, bem como em um conjunto de sinais simulados, formado por ruídos e mapas, este último em regimes caótico e periódico. O atributo qSDiffmax demonstrou ser consistente para baixas escalas ao passo que os atributos qmax e qzero para escalas maiores, separando e classificando os grupos quanto à complexidade fisiológica. Observou-se ainda uma possível relação entre estes q-atributos com a presença de caos, que precisa ser melhor estudada. Os resultados ainda apontam a possibilidade de que, na insuficiência cardíaca, ocorre maior degradação nos mecanismos de baixa escala, de curto período, ao passo que na fibrilação atrial o prejuízo se estenderia para escalas maiores. As medidas baseadas em entropia propostas são capazes de extrair informações importantes das séries de VFC, sendo mais consistentes com o conceito de complexidade fisiológica do que a SampEn (clássica). Reforçou-se a hipótese de que a complexidade se revela em múltiplas escalas de um sinal. Acreditamos que os métodos propostos podem contribuir bastante na análise da VFC e também de outros sinais biomédicos. / Human body is a complex system composed of several interdependent subsystems, interacting at various scales. It is known that physiological complexity tends to decrease with disease and aging, reducing the adaptative capabilities of the individual. In the cardiovascular system, one way to evaluate its regulatory dynamics is through the analysis of heart rate variability (HRV). Classical methods of HRV analysis are based on linear models, such as spectral analysis. However, as the physiological mechanisms regulating heart rate exhibit nonlinear characteristics, analyzes using such models may be limited. In the last years, several proposals nonlinear methods have emerged. Nevertheless, no one is known to be consistent with the physiological complexity theory, where both periodic and random regimes are characterized as complexity loss. Based on physiological complexity theory, this thesis proposes new methods for nonlinear HRV series analysis. The methods are generalization of existing entropy measures, through Tsallis nonadditive statistical mechanics and surrogate data. We defined a method, called qSDiff, which calculates the difference between the entropy of a signal and its surrogate data average entropy. The entropy method used is a generalization of sample entropy (SampEn), through nonadditive paradigm. From qSDiff we extracted three attributes, which were evaluated as potential physiological complexity indexes. Multiscale entropy (MSE) was also generalized following nonadditive paradigm, and the same attributes were calculated at various scales. The methods were applied to real human and rats HRV series, as well as to a set of simulated signals, consisting of noises and maps, the latter in chaotic and periodic regimes. qSDiffmax attribute proved to be consistent for low scales while qmax and qzero attributes to larger scales, separating and ranking groups in terms of physiological complexity. There was also found a possible relationship between these q-attributes with the presence of chaos, which must be further investigated. The results also suggested the possibility that, in congestive heart failure, degradation occurs rather at small scales or short time mechanisms, while in atrial fibrillation, damage would extend to larger scales. The proposed entropy based measures are able to extract important information of HRV series, being more consistent with physiological complexity theory than SampEn (classical). Results strengthened the hypothesis that complexity is revealed at multiple scales. We believe that the proposed methods can contribute to HRV as well as to other biomedical signals analysis.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-03-03T14:15:37Z No. of bitstreams: 1 barbarademeloquintela.pdf: 17938706 bytes, checksum: 9ab0cb4d4226bdefe7051c92e73feec9 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-03-06T20:15:36Z (GMT) No. of bitstreams: 1 barbarademeloquintela.pdf: 17938706 bytes, checksum: 9ab0cb4d4226bdefe7051c92e73feec9 (MD5) / Made available in DSpace on 2017-03-06T20:15:36Z (GMT). No. of bitstreams: 1 barbarademeloquintela.pdf: 17938706 bytes, checksum: 9ab0cb4d4226bdefe7051c92e73feec9 (MD5) Previous issue date: 2011-01-31 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / CNPq - Conselho Nacional de Desenvolvimento Científico e Tecnológico / FAPEMIG - Fundação de Amparo à Pesquisa do Estado de Minas Gerais / A Homogeneização por Expansão Assintótica (HEA) é uma técnica multiescala empregada ao cálculo de propriedades efetivas de meios contínuos com estrutura periódica. As principais vantagens desta técnica são a redução do tamanho do problema a resolver e a possibilidade de se empregar uma propriedade homogeneizada que guarda informações da microestrutura heterogênea. Quando associada ao Método dos Elementos Finitos (MEF), a HEA demanda o emprego de malhas que permitam a imposição de condições de contorno periódicas – sendo portanto necessário especificar tal particularidade quando da geração dos modelos em MEF. Tais modelos representam as células periódicas, que são volumes representativos do meio heterogêneo e, em alguns casos, apresentam uma complexidade geométrica e física que torna imprescindível o emprego de malhas com alto grau de refinamento – levando a um custo computacional significativo. Este trabalho tem por objetivo a obtenção de um programa em Elementos Finitos para a aplicação da HEA à Elasticidade em 3D, empregando técnicas de programação paralela. Foram desenvolvidas versões do programa em 2D: uma sequencial em C e duas paralelas empregando OpenMP e CUDA. Foi implementado com sucesso o programa HEA3D em uma versão sequencial, em linguagem FORTRAN e uma paralela, empregando OpenMP. Para validação dos programas, foram analisadas células periódicas bifásicas e os resultados apresentaram boa concordância com valores experimentais e numéricos disponíveis na literatura. A versão paralela obteve expressivos ganhos de desempenho, com acelerações de desempenho de até 5.3 vezes em relação a versão sequencial. / The Asymptotic Expansion Homogenization (AEH) is a multiscale technique applied to estimate the effective properties of heterogeneous media with periodical structure. The main advantages of this technique are the reduction of the problem size to be solved and the ability to employ an homogenized property that keeps information from the heterogeneous microstructure. In association with the Finite Element Method (FEM), the AEH requires the application of periodic boundary conditions, which must be taken into account during the generation of FE meshes. Such models represent periodic cells, which are representative volumes for heterogeneous media and, in some cases, present a geometric and physics complexity that demands refined meshes, leading to a significant computational cost. The aim of this work is to develop a parallel program that applies both FEM and AEH to estimate the elasticity properties of 3D bodies. A sequential version of the 2D program using C, and parallel versions using OpenMP and CUDA were implemented. A sequential version of the program, called HEA3D, was successfully implemented using FORTRAN. Also, a parallel version of the code was implemented using OpenMP. The validation of the codes consisted of comparisons of the numerical results obtained, with numerical and experimental data available in the literature, showing good agreement. Significant speedups were obtained by the parallel version of the code, achieving speedups up to 5.3 times over its sequential version.

Depuis des décennies, le processus d'enlèvement de matière des composites à matrices polymères (CMP) ne cesse de susciter des interrogations. La complexité et la multitude des phénomènes physiques activés par la coupe constituent encore un défi d'actualité pour la compréhension et la maitrise du comportement des structures composites. Ce travail propose une analyse multiéchelle fine des phénomènes élémentaires émanant du comportement de chacune des phases constituantes du matériau afin de modéliser leurs couplages multiphysiques potentiels conduisant à la formation du copeau. L'étude est alors hybride conjuguant l'approche expérimentale exprimée par l'essai instrumenté et l'approche numérique exprimée par la modélisation par éléments finis (EF). La formulation du couplage multiphysique a fait l'objet d'une routine VUMAT alliant la mécanique de l'endommagement continu à la mécanique de la rupture par le biais du triptyque élasticité-endommagement-rupture. A la différence des approches binaires de la littérature, le modèle développé dans ce travail s'appuie sur un concept d'endommagement progressif pour prédire la rupture physique des phases, et par conséquent, la formation du copeau. Les mécanismes d'initiation et de propagation de la fissure sont pilotés par les énergies de rupture des phases identifiées selon les normes en vigueur. La gestion du contact par une routine VFRIC a permis d'assurer la synergie entre les propriétés locales de l'interface et les frottements générés. Les calculs ont démontré la pertinence du modèle tridimensionnel proposé dans la simulation des mécanismes de formation du copeau sensiblement à l'orientation et la nature des fibres. La bonne concordance entre les mesures et les prédictions d'efforts de coupe a mis en évidence l'intérêt d'un pilotage rigoureux du contact outil-pièce pour la simulation multiphysique de la coupe. / Since several decades, the material removal process of Fiber Reinforced Polymers (FRP) continues to raise technical and scientific queries. The understanding of the multiple and complex phenomena generated when cutting still remains challenging for controlling the behavior of composite structures. This study addresses a multiscale analysis of elementary phenomena associated to each of the composite constituents in order to model the chip formation mechanisms owing to the multiphysical coupling. An investigation combining the experimental approach resulting in the instrumented test and numerical approach allowing to the finite element (FE) development was hence conducted. A VUMAT subroutine was built to express the constitutive formulation coupling the continuum damage mechanics to the failure mechanics by means of the triptych elasticity-damage-failure. Unlike to the binary approaches proposed by the open literature, the model proposed herein bases on the progressive damage concept for predicting the physical failure allowing to the formation of the chip. The crack initiation and growth mechanisms are controlled by the failure energies determined experimentally for each material phase. The efficiency of the VFRIC subroutine to managing the contact properties, i.e. friction, at the tool-material interface was confirmed. The numerical results proved the reliability of the model to simulate the chip formation mechanisms with respect of fiber orientation. The good agreement between the measured and predicted forces proved the interest of the rigorous modeling of the tool-material interface.

Le rodage mécanique demeure encore le procédé industriel de référence pour la finition des cylindres de moteurs dans le cadre des productions de grande série. La voie actuelle de fabrication en rodage procède par frottement et abrasion à vitesse réduite pour imprimer une texture anisotrope et multi-échelle sur la surface du cylindre. La signature de cette texture a la particularité de satisfaire des exigences multifonctionnelles du cylindre (frottement, lubrification, usure, consommation d'huile, etc.). Elle est cependant générée par un rodage stratifié en production industrielle qui consiste en trois étapes successives avec effet d'échelle: ébauche (échelle macroscopique), finition (échelle mésoscopique) et superfinition (échelle microscopique). Plusieurs méthodes de rodage ont émergé cette dernière décennie et dont la différentiation technologique et économique se fait essentiellement par le choix des attributs texturaux et topographiques de surface de rodage. La présente étude analyse les processus de rodage industriel, du choix d'anisotropie possibles et paramètres texturaux dans une optique de fonctionnalisation de la surface rodée du cylindre. Une approche méthodologique sur la tribo-fonctionnalité multi-échelle de la surface de rodage du cylindre a été développée, puis validée par simulation numérique dans le cas du tribo-contact segment-cylindre. Elle s'appuie sur le triptyque suivant :- le couplage entre le processus de rodage et la fonctionnalité via la caractérisation de sa signature texturale multi-échelle,- l'exploitation de l'anisotropie contrôlée par simulation numérique qui conduit à la texture de fonction,- l'interaction surface – procédé de rodage qui devient indissociable, notamment dans la fabrication de la texture de fonction et sa qualification tribo-fonctionnelle par des essais tribométriques.Cette démarche méthodologique a été appliquée aux procédés de rodage industriel tel que le rodage plateau (PH), glissant (SH) et hélico-glissant (HSH). Cela a permis de démontrer qu'une texture anisotrope fine de type PH composée de plateaux et vallées de faible amplitude présente les meilleurs attributs fonctionnelles en frottement. Cependant, une texture anisotrope de type HSH avec un angle de striation de 130° est moins sensible en frottement à l'aspect plateau obtenu par écrêtage des pics de rugosité. Ce résultat ouvre une évolution potentielle du process HSH où la dernière étape de rodage peut être réduite voire supprimée. Par la suite, la démarche développée a été étendue aux textures innovantes à bas frottement de rodage en développement avec une anisotropie circulaire, ondulatoire, stratifiée mixte, ou assistée en trajectoire. Les résultats ont montré que des textures HSH optimisées en trajectoire (avec suppression de stries d'inversion au niveau des points mort haut (PMH) et point mort bas (PMB) du cylindre) ainsi que des textures stratifiées mixtes 45°-130°-45° (stries à 45° au PMB et PMH et 130° au milieu du cylindre) permettent une réduction significative du frottement segment-fût en régime de lubrification mixte en comparaison par rapport au texture de rodage HSH conventionnel. / Mechanical honing process is still the reference industrial process for high production engine liners. The current manufacturing way use reduced velocity friction and abrasion mechanisms to print a multiscale and anisotropic texture on the liner surface. The texture signature characteristics satisfy multifunctional requirements of the liner (friction, lubrication, wear, oil consumption etc.). It is generated by a stratified honing process for industrial production which consists of three stages with scale effect: rough (macroscopic scale), finish (mesoscopic scale) and super-finish honing (microscopic scale). Different honing methods emerged during last decade in which the technological and economic differentiation is essentially based on textural attributes of honed surfaces. The present study analyses the industrial honing processes, the possible surface anisotropy choices and texture parameters with a view to honed liner surface functionalisation. A methodological approach about multiscale honed surface tribofunctionality has been developed and then validated by numerical simulation in the case of ring-liner tribocontact. It builds on the following triptych:- the coupling between honing process and functionality through its multiscal textural signature,- the exploitation of the controlled anisotropy by numerical simulation which conducts to fonctionality,- the surface-process interaction which becomes indivisible, particularly in texture manufacturing and its tribofunctional qualification though tribometric trials.The methodology has been applied to industrial honing processes (plateau honing (PH), slide honing (SH) and helical slide honing (HSH)). The results show that smooth texture with lower plateau roughness and valley depth contributes to reduce frictional performances of honed surfaces. Nevertheless, Helical slide honed surfaces are less sensitive in friction to the plateaudness i.e., to superficial roughness comparatively to PH textures. This is promising for HSH process optimization, in which the third stage can be reduced or deleted. Then the developed approach has been extended to honing development for innovative texture anisotropy (circular, undulatory, mixed, trajectory assisted) for low-friction performances. The results show that assisted trajectory (without inversion grooves at top dead (TDC) and bottom dead centers (BDC)) and 45-130-45 mixed textures (with 45° cross-hatched grooves at TDC and BDC, 130° cross-hatched grooves at mid-height) enhance significantly frictional performances in comparison to HSH conventional process.

Ce travail de thèse a contribué au développement d'un outil numérique multi-échelle capable de restituer l'aspect visuel de revêtements de peinture de carrosserie automobile à partir des caractéristiques physico-chimiques et des propriétés diélectriques des différents constituants de base ; l'objectif est la prédiction et la complète maîtrise de l'aspect visuel des matériaux. Dans cette étude, des nanoparticules d'hématite (variété α-Fe2O3), d'oxyde et d'hydroxyde de nickel de différentes formes et tailles ont été synthétisées par un procédé hydrothermal. Les particularités de ces particules en termes de morphologie, de taille et de structure ont été soulignées car elles sont susceptibles d'affecter ses propriétés diélectriques, et donc la couleur obtenue. La réponse diélectrique des particules a été mesurée par spectroscopie de pertes d'énergie des électrons (EELS). Dans un revêtement de peinture, les pigments, qui interagissent les uns avec les autres pour former des floculats, modifient la diffusion de la lumière et donc la couleur perçue. Au cours d'une deuxième phase, des revêtements de peinture « modèles », constitués de particules d'hématite dispersées dans une résine polymère, ont été élaborés. L'auto-organisation spatiale de ces nanoparticules au sein du revêtement a été décrite à partir d'images 2D enregistrées à différentes échelles d'observation. Sur la base des mesures diélectriques locales réalisées sur les nanoparticules et peintures « modèles » synthétisées dans le présent travail, des modèles aléatoires de matériaux hétérogènes ont alors été établis par le Centre de Morphologie Mathématique, pour générer une « microstructure diélectrique» 3D de ces revêtements. Sur la base de ces modèles morphologiques, les propriétés diélectriques effectives des revêtements ont été prédites; les fonctions diélectriques alors obtenues sont en bon accord avec les mesures macroscopiques réalisées par ellipsométrie. / This work contributed to the development of a multi-scale numerical tool able to reproduce the visual appearance of coatings for motor vehicle bodywork from the physico-chemical characteristics and dielectric properties of various base constituents; the goal is the prediction and complete control of the visual appearance of materials. In this study, hematite nanoparticles (the α-Fe2O3 variety), nickel oxides and hydroxides of different shapes and sizes were synthesised using a hydrothermal process. The characteristics of these particles in terms of morphology, size and structure have been particularly emphasised since they are likely to affect the overall dielectric properties of the coating, and therefore the color obtained. The dielectric response of the particles was measured by electron energy loss spectroscopy (EELS). In a coating of automobile paint, pigments which interact with each other to form flocs alter the light scattering and therefore the perceived color. During a second phase, model paint coatings consisting of hematite particles dispersed in a polymer resin were developed. The volume self-organisation of nanoparticles in the coating has been described from 2D images recorded at different scales of observation. Based on our local measurements of the dielectric function at different scales of the nanoparticules and model paint that were synthetized in the present work, random models of heterogeneous materials were then established by a partner within the consortium to generate a "dielectric 3D microstructure" of these coatings. Based on these morphological models, effective dielectric properties of the coatings were predicted, and then satisfactorily compared with macroscopic measurements from ellipsometry.

A l'heure actuelle, les nouvelles contraintes de la réglementation thermique en vigueur ne cessent de s'adapter au contexte économique global pour lequel la recherche d'une efficacité énergétique dans le bâtiment est devenue incontournable. Pour répondre à ces défis, des Matériaux intelligents à Changement de Phase (MCP) ont fait leur apparition sur le marché de la construction. Grâce à leur capacité de stockage de l'énergie, les MCP sont de plus en plus associés aux matériaux de construction classiques (béton, plâtre, etc.) afin d'améliorer leur inertie thermique et apporter un meilleur confort aux usagers. Pour ce faire, les propriétés thermophysiques intrinsèques aux MCP doivent être suffisamment maitrisées afin de pouvoir contrôler les propriétés du produit composite final. Dans ce contexte, cette thèse est une contribution ayant pour objectif de développer des méthodologies spécifiques pour une meilleure caractérisation des MCP et des béton-MCP. Une panoplie d'approches expérimentales a été présentée pour l'identification des propriétés thermophysiques des MCP et pour identifier l'effet d'incorporation et de l'endommagement de ces matériaux sur les propriétés thermiques et mécaniques de béton. Plusieurs modèles d'homogénéisation ont été utilisés afin de prédire le comportement thermique des bétons-MCP en utilisant les propriétés thermiques moyennées obtenues expérimentalement. Une étude probabiliste paramétrique a été menée afin de prendre en compte les incertitudes liées à la dispersion aléatoire des mesures expérimentales de propriétés thermiques du béton-MCP. Les résultats issus des essais expérimentaux ont été intégrés dans le cadre d'une étude numérique par la Méthode des Volumes finis (MVF) afin d'étudier le mécanisme de transfert de chaleur à travers une paroi en béton-MCP / The thermal policies have been kept to fit the new economic in a global context particularly in terms of buildings energy efficiency. To meet these challenges, different technologies have been used such as the Phase Change Materials (PCMs) which have the ability to store and release energy. PCMs are generally used with conventional building materials in order to improve their thermal inertia and provide better comfort to users. To enhance the properties of the final composite, the PCMs thermo-physical properties must be sufficiently controlled. In this context, this thesis is a contribution aimed to develop specific methodologies for better characterization of PCM and PCM-concrete. Different experimental approaches will be presented for the identification of PCMs thermophysical properties and to identify the effect of the incorporation and the damage of these materials on the thermal and mechanical properties of concrete. A multiscale modelling considering the average of experimental thermal properties was applied to predict the thermal behaviour of PCMs-concrete. A probabilistic study of experimental uncertainties will be also conducted to assess the level of confidence of the impact of PCM on the thermodynamic properties of PCM-concrete. A numerical study was conducted using experimental data to study the heat transfer through a PCM-concrete wall

À Paris environ 44% des bâtiments ont été construits avant 1914 avec des murs épais non isolés et des matériaux poreux, caractérisés par une forte inertie thermique et des propriétés hygroscopiques. Les propriétés hygrothermiques des matériaux utilisés dans les constructions anciennes ont des effets qui : (i) confèrent (aux bâtiments) de bonnes qualités thermiques en période estivale et (ii) contribuent à réguler la température et l’humidité relative intérieure. En France les politiques d’économie d’énergie et la régulation thermique ont abouti à la mise en place de l’isolation thermique afin de réduire les consommations d'énergie pendant l'hiver. L'installation de l'isolation thermique dans la rénovation des bâtiments existants pose cependant des difficultés. D’autre part, il est prévu que des conditions extrêmes de chaleur deviennent plus fréquentes dans des scénarios du climat futur. Ainsi, les évolutions possibles du climat futur doivent être intégrées dans l'évaluation des stratégies de rénovation dans le bâtiment ancien. Ce travail de thèse porte sur l'évaluation du comportement hygrothermique de constructions anciennes rénovées à Paris, dans des conditions météorologiques actuelles et de vague de chaleur. A l’échelle des logements un modèle a été construit sur un outil de simulation thermique dynamique, calé et validé à travers des données enregistrées dans une campagne de mesure lancée en 2014 dans les logements étudiés. À l’échelle de la paroi un modèle macroscopique en 2D est proposé afin d'étudier les transferts de chaleur et de masse dans un mur poreux constitué de plusieurs couches avec de l’isolation thermique par intérieur et par l’extérieur. / In Paris 44% of the dwellings were built before 1914 with uninsulated thick walls made of porous materials, characterized by high thermal inertia and hygroscopic properties. The hygrothermal properties of existing buildings materials have effects that: (i) give (to these buildings) good thermal qualities in summer and (ii) help to regulate indoor temperature and relative humidity. In France the energy saving policies and thermal regulation have resulted in the implementation of thermal regulation to reduce energy consumption during winter. Though the installation of thermal insulation in existing buildings poses a number of difficulties. Moreover, it is expected that extreme heat conditions become more frequent in future climate scenarios. Thus, the possible evolutions of future climate must be integrated into the evaluation of renovation strategies in old buildings. This thesis focuses on the assessment of the hygrothermal behavior of old renovated Parisian buildings, in current and heat wave weather. At the building scale (housing), a model was built in dynamic thermal simulation tool. The model was calibrated and validated through recorded data from a measurement campaign launched in 2014 on the studied housings. At the wall scale, a macroscopic model in 2D is proposed to study the heat and mass transfer through a multilayered porous wall, renovated with internal thermal insulation and external thermal insulation.

Ce travail porte sur le développement de méthodes d'homogénéisation numériques pour les matériaux non linéaires élastoplastiques et élasto-visco plastiques avec pour finalité le calcul de structures hétérogènes faite de ces matériaux sous chargements cycliques. Les techniques proposées se basent sur des approches incrémentales. Dans une première partie, nous développons une méthodologie pour déterminer la taille du VER dans le cas non linéaire pour les types de comportements cités précédemment. Pour cela, une étude de la convergence statistique de paramètres d'une loi semi-analytique incrémentale est menée. Cette méthode permet de diminuer les temps de calcul pour l'identification des paramètres et ainsi qu'une meilleure approximation de la loi de comportement effective, qui peut ensuite être utilisée dans un calcul de structure. Dans une deuxième partie, nous proposons une méthode d'homogénéisation numérique incrémentale dans laquelle l'originalité est de calculer le module tangent effectif par superposition de solutions numériques calculées par éléments finis sur le VER, en exploitant la linéarisation du problème à caque incrément. Un schéma alternatif aux techniques telles que la méthode Eléments finis multi niveaux (FE2) est ainsi développé, avec pour avantage un nombre réduit de calculs éléments finis à effectuer sur le VER. La technique est appliquée au calcul de structures hétérogènes non linéaires élastoplastiques, pour des microstructures anisotropes ou de morphologies complexes, et pour des lois de comportement locales élastoplastiques avec écrouissage isotrope et cinématique / In this work, computational homogenization methods for nonlinear materials with elasto-plastic and elasto-visco-plastic phases are developed, with application to the computation of heterogeneous structures made of these nonlinear materials under cyclic loading. The proposed techniques are based on incremental approaches. In a first part, we develop a methodology to determine the size of an RVE in the nonlinear case for the types of nonlinear behavior mentioned above. For this purpose, a convergence study of the parameters of a semi-analytical incremental constitutive law is carried out. This method allows reducing computational times related to the identification of these parameters and provides a better approximation of the effective constitutive relationship, which can then be used in a structure calculation once identified. In a second part, we propose an incremental computatioal homogenization method in which the originality is to compute directly the effective tangent tensor by superposition of numerical solutions computed by finite elements on an RVE, by taking advantage of the linearization of the problem at each increment. An alternative scheme to classical multilevel finite element techniques (FE2) is then developed, with the advantage of a reduced number of computations to perform on the RVE. The technique is applied to the computation of heterogeneous, nonlinear structures, for anisotropic microstructures or with complex morphologies and for phase with elastoplastic behavior with isotropic and kinematic hardening