Academic literature on the topic 'Ondes propagatives'

Abstract This study deals with two teleconnection patterns and the subsequent wave train propagations during an East Asian summer. Diagnostic results are as follows: 1) a stationary wave ray with zonal wavenumber 5 approximates the arc path linking the correlation centers originating from the Caspian Sea via Lake Baikal to the sea off the southeast coast of Japan (i.e., the OKJ arc path as a focus area) in a pentad correlation map between 500-hPa geopotential height (Z500) and outgoing longwave radiation (OLR) at 30°N, 150°E in June 1979–98. Ray tracing shows that it took 8–10 days for this stationary wave to propagate from an initial position around the Caspian Sea to the focus area, which roughly coincides with the observed case in July 1998. 2) A wave train pattern (P-Ja) observed in the boreal summer propagated along the arc line in the same way as the normal poleward Rossby wave train originating from the Philippines across the North Pacific (P-J), but with a phase shift northeastward of about 90°. 3) Further correlation analyses showed that the P-J-like waves belong mainly to intraseasonal propagating ones while OKJ waves belong mainly to intraseasonal stationary ones. 4) Propagation of the newly observed wave train pattern (P-Ja) occurred following another wave train along the OKJ arc path in mid-July 1998. Both northeastward and southeastward wave propagations merged off the east coast of Japan. 5) The northeastward-propagating wave train observed in mid-July 1998 was triggered by the southeastward-propagating (OKJ) wave train that produced a deep cyclonic circulation and a strong convective activity in the focus area. The link of wave forcing and deep convection was made solely because of a strong upper-level divergence in the focus area.

Abstract The asymptotic stability of one-dimensional linear Bresse systems under infinite memories was obtained by Guesmia and Kafini [10] (three infinite memories), Guesmia and Kirane [11] (two infinite memories), Guesmia [9] (one infinite memory acting on the longitudinal displacement) and De Lima Santos et al. [6] (one infinite memory acting on the shear angle displacement). When the kernel functions have an exponential decay at infinity, the obtained stability estimates in these papers lead to the exponential stability of the system if the speeds ofwave propagations are the same, and to the polynomial one with decay rate otherwise. The subject of this paper is to study the case where only one infinite memory is considered and it is acting on the vertical displacement. As far as we know, this case has never studied before in the literature. We show that this case is deeply different from the previous ones cited above by proving that the exponential stability does not hold even if the speeds of wave propagations are the same and the kernel function has an exponential decay at infinity. Moreover, we prove that the system is still stable at least polynomially where the decay rate depends on the smoothness of the initial data. For classical solutions, this decay rate is arbitrarily close to . The proof is based on a combination of the energy method and the frequency domain approach to overcome the new mathematical difficulties generated by our system.

It is well known that the propagation of Lamb waves in elastic solid medias depends on the frequency and on the material properties, but it does not depend on the propagation direction in isotropic medias and gives rise to two basic types of modes, symmetric and antisymmetric ones. This paper presents an application of the semi-analytical finite element (SAFE) method on an isotropic aluminum plate of thickness e = 2mm. The main goal of this paper is to make a quantitative evaluation of the SAFE method by carrying out the following tasks: a) calculation of the dispersion curves (wavenumber, phase velocity, group velocity, b) calculation of the relative error and validation of the proposed algorithm, c) comparison between the curves obtained by SAFE method on those obtained by Bisection method. It should be noted here that our study is restricted on the propagative modes.

No início de 1998, o cancro bacteriano da videira, causado pela bactéria Xanthomonas campestris pv. viticola, foi detectado pela primeira vez, no Brasil, em parreirais do Submédio São Francisco, onde a doença vem ocasionando prejuízos nas cultivares suscetíveis Red Globe, Itália, Festival, Brasil, Piratininga, Patrícia, Benitaka e Catalunha. Os sintomas, nas folhas, surgem como pontos necróticos (1 a 2mm de diâmetro) com ou sem halos amarelados, algumas vezes coalescendo e causando a morte de extensas áreas do limbo foliar. Nas nervuras e pecíolos, nos ramos e ráquis dos frutos, formam-se manchas escuras alongadas que evoluem para fissuras longitudinais de coloração negra conhecidas como cancros. Descoloração vascular é também observada. As bagas são desuniformes em tamanho e cor podendo apresentar lesões necróticas. A disseminação do patógeno ocorre através de material propagativo infectado, material de colheita (contentores, tesouras de poda e raleio, luvas), tratos culturais (desbrota, poda, raleio de bagas, colheita), ventos e chuvas. Apesar da região apresentar um curto período chuvoso, a disseminação da bactéria é mais eficiente durante essa época. Em condições de umidade e temperatura elevadas, o patógeno sobrevive em restos de cultura. Para o controle da doença, recomenda-se o uso de material propagativo sadio, inspeção no campo, poda drástica de órgãos infectados, eliminação de plantas severamente infectadas, condução da época de poda de produção, desinfestação de veículos, de equipamentos e de materiais para poda, utilização de fungicidas protetores cúpricos e tiocarbamatos, e utilização de quebra-ventos para reduzir a disseminação do patógeno.

In this investigation, the fractional Fornberg–Whitham equation (FFWE) is solved and analyzed via the variational iteration method (VIM) and Adomian decomposition method (ADM) with the help of the Aboodh transformation (AT). The FFWE is an important model for describing several nonlinear wave propagations in various fields of science and plasma physics. The AT provides a powerful tool for transforming fractional-order differential equations (DEs) into integer-order ones, making them more amenable to analytical solutions. Accordingly, the main objective of this investigation is to demonstrate the effectiveness and accuracy of ADM and VIM in deriving some approximations for the FFWE. Furthermore, we highlight the advantages and potential applications of these methods in solving other fractional-order nonlinear problems in several scientific fields, especially in plasma physics and some engineering problems.

The general coupled Hirota equations are investigated, which describe the wave propagations of two ultrashort optical fields in a fibre. Firstly, we study the modulational instability for the focusing, defocusing and mixed cases. Secondly, we present a unified formula of high-order rational rogue waves (RWs) for the focusing, defocusing and mixed cases, and find that the distribution patterns for novel vector rational RWs of focusing case are more abundant than ones in the scalar model. Thirdly, the N th-order vector semirational RWs can demonstrate the coexistence of N th-order vector rational RWs and N breathers. Fourthly, we derive the multi-dark-dark solitons for the defocsuing and mixed cases. Finally, we derive a formula for the coexistence of dark solitons and RWs. These results further enrich and deepen the understanding of localized wave excitations and applications in vector nonlinear wave systems.

O objetivo deste trabalho foi caracterizar o potencial germinativo de sementes de porta-enxertos de pessegueiro. O material propagativo constituiu-se de caroços das cultivares de porta-enxertos ‘Capdeboscq’ ‘Tsukuba 1’, ‘Tsukuba 2’, ‘Tsukuba 3’, ‘Okinawa Roxo’ e dos híbridos intraespecíficos de P. persica, seleções ‘NR 0280401’ (híbrido F1 derivado do cruzamento ‘Flordaguard’ x ‘Capdeboscq’), ‘NR 0080412’ (híbrido F1 derivado do cruzamento ‘Capdeboscq’ x ‘Flordaguard’), ‘NR 0050303’ (híbrido F1 derivado do cruzamento ‘Capdeboscq’ x ‘Tsukuba 1’) e ‘NR 0300402’ (híbrido F1 derivado do cruzamento ‘Flordaguard’ x ‘Nemaguard’) foram estratificadas em BOD, na ausência de luz, onde permaneceram por até 28 dias. Verificou-se que as sementes das seleções de porta-enxertos ‘NR 0280401’, ‘NR 0080412’ e da cv. ‘Capdeboscq’ requerem menor tempo para a germinação e apresentam maior percentual de sementes germinadas. Genótipos com maturação de frutos mais precoces, o período de armazenamento deve ser menor, devido à perda viabilidade, conforme verificado para ‘Okinawa Roxo’.

Ultrashort pulses are severely distorted even by low dispersive media. While the mathematical analysis of dispersion is well known, the technical literature focuses on pulses, Gaussian and Airy pulses, which keep their shape. However, the cases where the shape of the pulse is unaffected by dispersion is the exception rather than the norm. It is the objective of this paper to present a variety of pulse profiles, which have analytical expressions but can simulate real-physical pulses with great accuracy. In particular, the dynamics of smooth rectangular pulses, physical Nyquist-Sinc pulses, and slowly rising but sharply decaying ones (and vice versa) are presented. Besides the usage of this paper as a handbook of analytical expressions for pulse propagations in a dispersive medium, there are several new findings. The main findings are the analytical expressions for the propagation of chirped rectangular pulses, which converge to extremely short pulses; an analytical approximation for the propagation of super-Gaussian pulses; the propagation of the Nyquist-Sinc Pulse with smooth spectral boundaries; and an analytical expression for a physical realization of an attenuation compensating Airy pulse.

Las propiedades electromagnéticas de todos los materiales existentes en la naturaleza pueden ser determinadas a partir dos parámetros: la permeabilidad magnética µ y la permitividad eléctrica ε. Estos dos parámetros caracterizan la respuesta del material cuando interactúa con la radiación electromagnética. En principio, no existe límite alguno para el rango de valores posibles que pueden tomar µ y ε, por lo cual podemos pensar en diseñar y construir a voluntad materiales con características de respuesta electromagnética especificas (es decir µ y ε) no encontradas en la naturaleza. Estos materiales fabricados en el laboratorio reciben el nombre genérico de metamateriales, y entre ellos se encuentran los conocidos por sus siglas en ingles LHM (Left Handed Materials), así llamados porque los vectores de campo , de las Ondas Electromagnéticas que viajan en su interior están relacionados por la regla de la mano izquierda. La característica distintiva de los LHM es que para ciertas bandas de frecuencia presentan índice de refracción negativo ( ) con modos propagativos posibles solamente si ambos parámetros µ y ε, dentro de dichas bandas de frecuencias son negativos. El propósito de este trabajo es presentar los principios y fundamentos de estos metamateriales de manera que despierte el interés de lectores no especializados. © 2016. Acad. Colomb. Cienc. Ex. Fis. Nat.<div id="spoon-plugin-kncgbdglledmjmpnikebkagnchfdehbm-2" style="display: none;"> </div>

Introdução: O termo greenwashing (“maquiagem verde”), é utilizado para denominar o modo de comunicação desenvolvido por empreendedores ao tentar mascarar as verdades que permeiam seus produtos e serviços, o que pode denotar um fraco comprometimento ambiental. Com base na análise de meios gráficos propagativos de atributos sustentáveis de condomínios residenciais no município de Caruaru, Pernambuco visou-se compreender até que ponto o greenwashing está presente na propaganda deste segmento de mercado na Região. Método: Realizou-se uma pesquisa de campo, entre janeiro e março de 2012 em pontos de vendas de empreendimentos imobiliários, utilizando-se peças publicitárias de divulgação dos empreendimentos complementada por conversas informais com corretores e outras pessoas responsáveis pela comercialização destes. Entre os materiais gráficos disponíveis destacam-se os sites web dos empreendimentos, os folderes, catálogos e outdoors, e os estandes de venda. Resultados: Todos os empreendimentos analisados apresentaram determinado grau de inserção no contexto estabelecido para os “seis pecados do greenwashing” bem como no artigo 36º. do CONAR, referente à infração da autenticidade publicitária em seus meios gráficos. Conclusão: O crescimento exponencial da cidade de Caruaru atraiu empreendimentos dos mais diversificados setores. Contudo, urge pensar que uma cidade ao ser moldada para novos olhares e conceitos de vida, deve unicamente desenvolver-se em concordância com ações de fato consolidadas em um transparente desenvolvimento sustentável, onde todos os interessados sejam verdadeiramente contemplados.

La spectrométrie est une méthode d’analyse fondamentale, omniprésente pour comprendre l’interaction lumière-matière. La principale limitation des spectromètres est leurs structures relativement complexes, difficiles à simplifier et à miniaturiser. SWIFTS (Stationary Waves Integrated Fourier Transform Spectrometer) est un concept innovateur de spectromètre intégré qui permet de dépasser cette difficulté. Le principe de ce composant est la détection de l’intensité d’une onde stationnaire contenue dans un guide. Des lignes métalliques très fines agissant comme des nano-sondes viennent échantillonner l’onde stationnaire, le signal prélevé est récupéré par des détecteurs. Le spectre est obtenu par Transformation de Fourier de l’interférogramme échantillonné. Nos travaux se sont concentrés sur la simulation des nano-sondes. L’extraction du signal contenu dans le guide est en effet une étape critique : la nano-sonde doit diriger le maximum du signal prélevé sur les détecteurs sans toutefois perturber le signal dans le guide. Pour cela, nous avons développé la méthode de simulation FMM (Fourier Modal Method) associée à des PMLs (Perfectly Matched Layer). A partir de cette méthode, nous avons développé des outils de simulation permettant de caractériser et d’optimiser la perturbation d’une nano-sonde positionnée sur un guide. Le modèle développé est également adapté pour simuler le spectromètre SWIFTS comportant un nombre important de nano-sondes. Les simulations obtenues sont confrontées à une étude expérimentale
SWIFTS (Stationary Waves Integrated Fourier Transform Spectrometer) is a new concept of Fourier Transform spectrometry that opens the way to high-resolution ultra-compact spectrometers. This integrated spectrometer is based on the direct near-field probing of the optical source’s Fourier Interferogram. To obtain the most compact structure a Guided Interferogram (GI) is created into an optical integrated waveguide thanks to a counterpropagative phenomenon. Once the stationary GI created, the challenge is to read this useful information confined into the waveguide. The idea is to sample the GI by embedding a comb of nano-probes in the waveguide’s near-field. These nano-elements scatter the GI’s near-field profile towards a detector array placed in far-field. The incident light spectrum can be recovered by Fourier Transform of the collected signal. We focused our work on the simulation of one nano-probe. This study is critical: the scattering light has to go towards the detector without disturbing the useful information confined into the waveguide. To study this phenomenon, we developed the Fourier Modal Method with PML (Perfectly Matched Layer). Then, we developed specific tools to characterize the scattering light of a nano-element embedded in the waveguide’s near-field. This numerical method is also adapted to the whole spectrometer, including many nano-probes elements. We validated simulation work by an experimental realization

Cette thèse s’intéresse à l’étude expérimentale d'ondes hydrothermales apparaissant dans une géométrie quasi-unidimensionnelle pour une couche de fluide avec surface libre soumise à des gradients horizontaux et verticaux de température. Ceux-ci sont induits par la présence d'un fil chaud au sein de la couche de fluide mais près de la surface libre. Nous mettons en évidence les propriétés générales de ces ondes avant de réaliser une étude expérimentale et théorique nous permettant d’appréhender l'aspect hydrodynamique de ce phénomène. Notamment nous nous intéressons aux rôles respectifs des effets de volume et des effets de surface dans le déclenchement de l’instabilité. Nous étudions aussi les propriétés critiques de l'apparition des ondes à l'aide de mesures près du seuil, ces dernières nous permettant de conclure à la nature convective de l’instabilité au voisinage du seuil. Enfin, par un forçage modulé dans le temps, nous faisons apparaître des ondes stationnaires a la place des ondes propagatives. Un travail théorique utilisant des équations d'amplitude et s'accordant très bien avec les observations expérimentales nous permet de comprendre le mécanisme d'apparition de ces ondes stationnaires.

La comprehension des mecanismes hemodynamiques locaux impliques dans l'apparition, la progression et les complications de l'atherosclerose necessite la mise au point de methodes de quantification des grandeurs hemodynamiques d'interet. Ces variables biomecaniques sont essentiellement representees par les contraintes de cisaillement qui s'exercent sur la paroi arterielle et par les deformations parietales induites par les ondes propagatives. Dans cette these, nous proposons une quantification non-invasive de ces variables grace au developpement de modeles biophysiques en privilegiant l'analyse de la vitesse au centre du vaisseau et du diametre arteriel, de telle facon que les explorations puissent etre atraumatiques chez des sujets asymptomatiques. De maniere comparative, plusieurs methodes analytique (womersley) et numeriques (modele non-newtonien) pour les determinations in vivo chez l'homme des taux de cisaillement et des contraintes de cisaillement endothelial sont presentees et discutees en fonction de leurs hypotheses et des informations qu'elles peuvent procurer pour l'etude de la vasomotricite arterielle dependante du debit. Nous avons pu ainsi apprecier l'incertitude apportee par l'hypothese de periodicite du modele de womersley, discuter de l'interet de considerer le sang comme un fluide non-newtonien, et montrer que la vasomotricite de l'artere humerale etait essentiellement sous la dependance des valeurs pulsatiles des taux et contraintes de cisaillement. Nous avons egalement pu montrer grace a des modeles lineaire et non-lineaire qu'il etait possible a partir de l'enregistrement de la vitesse au centre et du diametre instantane d'effectuer une separation de l'onde propagative en ondes incidente et retrograde. La comparaison de ces deux modeles nous a permis de conclure que les termes non-lineaires d'acceleration convective et de la compliance pouvaient etre negliges au niveau de l'artere humerale.

L'attraction d'ondes dans des systèmes contra-propagatifs est un phénomène général, établi initialement en Physique dans le contexte de l'attraction de polarisation entre deux ondes contra-propagatives se propageant dans des fibres optiques. Ce phénomène a été observé expérimentalement, et ses propriétés étudiées via des simulations numériques. Les modèles qui s'y rattachent sont des systèmes hyperboliques d'équations aux dérivées partielles, avec des conditions aux bords dépendant du temps sur un intervalle fini. Le mécanisme sous-jacent peut être expliqué par l'existence de tores singuliers dans les équations stationnaires correspondantes. Le but de cette thèse est d'analyser en détail l'exemple le plus simple dans cette famille de modèles. Nous montrons que la plupart des phénomènes de processus d'attraction d'ondes sont en fait existants dans un modèle linéaire avec intéraction résonnante. Nous établissons l'existence et la régularité des solutions et analysons la relaxation vers la solution stationnaire qui caractérise les propriétés de l'attraction d'ondes.

Au sens large, le terme de localisation ondulatoire fait référence à un phénomène où les ondes sont spatialement confinées dans de petites régions de l'espace sans la contrainte de barrières matérielles. Dans cette thèse, nous étudions (analytiquement, numériquement et expérimentalement) différents mécanismes physiques collectifs pour localiser spatialement, et donc pour contrôler les ondes électromagnétiques. En particulier, nous nous concentrons sur le rôle des potentiels non corrélés et corrélés, ainsi que sur des effets topologiques pour réaliser le confinement des ondes. Les études analytiques et numériques sont réalisées dans le cadre d'une approche récente de la modélisation de la localisation d'Anderson appelée théorie du paysage de localisation. D'autre part, des expériences sont réalisées à l'aide d'une plate-forme micro-ondes composée de petits cylindres diélectriques placés à l'intérieur d'une cavité constituée de deux plaques métalliques. La cavité met en œuvre un système d'ondes propagatives, où nous pouvons contrôler efficacement la permittivité locale au moyen des cylindres agissant comme des diffuseurs, ou comme un système de de liaison forte analogique, où, dans ce cas, les cylindres diélectriques jouent le rôle de résonateurs. Dans un premier temps, nous étendons le champ d'application de l'approche du paysage de localisation à une large classe de systèmes de liaison forte unidimensionnels et bidimensionnels en présence d'un désordre non corrélé, où des fonctions propres localisées apparaissent en bord de bande. Nous démontrons comment la théorie du paysage de localisation est capable de prédire avec précision non seulement les emplacements, mais aussi les énergies des fonctions propres localisées dans les régimes de basse et de haute énergie. Ensuite, en utilisant notre cavité expérimentale comme système de propagation, nous réalisons des expériences de transport de micro-ondes dans des réseaux planaires bidimensionnels. Les expériences sont réalisées sur un réseau désordonné et sur une spirale de Vogel apériodique à partir de laquelle nous caractérisons les structures modales électromagnétiques dans l'espace réel. Nos résultats révèlent que les systèmes apériodiques possèdent une grande variété de modes à longue durée de vie - avec des décroissances spatiales gaussiennes, exponentielles et en loi de puissance - qui sont capables de survivre même dans un environnement tridimensionnel. Ceci est confirmé par différentes quantités de transport telles que la densité d'états, le temps de décroissance caractéristique et la conductance de Thouless qui sont également accessibles expérimentalement. À l'inverse, nous montrons que les états propres dans les milieux désordonnés traditionnels sont toujours limités à des décroissances radiales exponentielles avec d'importantes fuites dès que les systèmes ne sont plus bidimensionnels. Enfin, nous utilisons la configuration expérimentale de liaison forte pour étudier la propagation des états hélicoïdaux topologiques. En particulier, nous analysons expérimentalement un ensemble de structures en nid d'abeille construites à l'aide d'un réseau triangulaire avec une cellule unitaire hexagonale, qui sont caractérisées par l'invariant topologique Z_2. En accédant à la structure modale dans l'espace réel et à la densité d'états, nos résultats révèlent la possibilité d'ouvrir une bande interdite topologique, peuplée d'états de bord localisés en bordure de la structure. Nous démontrons la nature unidirectionelle de la propagation de ces états de bord hélicoïdaux contre-propagatifs. Dans l'ensemble, nos résultats démontrent qu'il est possible de modéliser, de contrôler et de localiser les ondes électromagnétiques non seulement du point de vue d'Anderson, mais aussi au-delà. Grâce aux différents jalons que nous avons posés, nous ouvrons une voie vers l'hypothétique localisation d'Anderson des ondes électromagnétiques tridimensionnelles
In a broad sense, the term wave localization refers to a phenomenon where waves are spatially confined in small regions of the space without any bounding material barriers.In this Thesis, we investigate (analytically, numerically and experimentally) different physical collective mechanisms to spatially localize, and therefore, to control electromagnetic waves. Specifically, we focus on the role of uncorrelated and correlated potentials, as well as of topological effects to achieve wave confinement. Analytical and numerical studies are accomplished in the framework of a recent approach in the modeling of Anderson localization called localization landscape theory. On the other hand, experiments are performed using a microwave platform composed by small dielectric cylinders placed inside a cavity made of two metallic plates. The cavity implements a propagative wave system, where we can efficiently control the local permittivity by means of the cylinders acting as scatterers, or as an analogic tight-binding system, where, in this case, the dielectric cylinders play the role of resonators.First, we extend the scope of the localization landscape approach to a wide class of one and two dimensional tight-binding systems in the presence of uncorrelated disorder, where localized eigenfunctions appear in both band-edges. We demonstrate how the landscape theory is able to predict accurately not only the locations, but also the energies of localized eigenfunctions in the low- and high-energy regimes. Later, by using our experimental cavity as a propagative system, we perform microwave transport experiments in two dimensional planar arrays. Experiments are carried out on a disordered lattice and on an aperiodic Vogel spiral from where we characterize the electromagnetic modal structures in real space. Our results reveals that aperiodic systems can carry a rich variety of long-lived modes—with Gaussian, exponential, and power law spatial decays—which are able to survive even in a three-dimensional environment. This is supported by different transport quantities such as the density of states, the characteristic decay time, and the Thouless conductance that are also experimentally accessible. On the contrary, we show that the eigenstates in traditional disordered media are always limited to exponential radial decays with leaking features beyond two-dimensions.Finally, we use the experimental tight-binding configuration to investigate the propagation of topological helical states. Particularly, we experimentally analyze a set of honeycomb-like structures built using a triangular lattice with an hexagonal unit cell, which are characterized by the Z_2 topological invariant. By recovering the modal structure in real space and the density of states, our results reveal the possibility to open a topological gap, dwelt by edge states that lives in the border of the structure.We demonstrate the unidirectional counterpropagative features of such helical edge states.Taken together, our results demonstrate that it is possible to model, control and localize electromagnetic waves not only within, but beyond Anderson's conception. Thanks to the crossroads we have taken, we have mapped out an itinerary that brings us closer to the main avenue leading perhaps to Anderson localization of three dimensional electromagnetic waves

La dynamique d’une structure peut être vue aussi bien en termes de modes (ondes stationnaires) qu’en termes d’ondes élastiques libres. Les approches modales sont largement utilisées en mécanique et de nombreuses techniques de réduction de modèles (Model Order Reduction - MOR) ont été développées dans ce cadre. Quant à la dynamique des structures périodiques, les approches propagatives sont majoritairement utilisées, où la périodicité est exploitée en utilisant la théorie de Bloch. Pour les structures périodiques complexes, plusieurs techniques MOR sur la base d’onde ont été proposées dans la littérature. Dans ce travail, une approche couplée propagative et modale a été développée pour étudier la propagation des ondes dans les structures périodiques. Cette approche commence par la description modale d’une cellule unitaire (échelle mésoscopique) en utilisant la synthèse modale (Component Mode Synthesis - CMS). Par la suite, la méthode propagative - Wave Finite Element Method (WFEM) est appliquée sur la structure (échelle macroscopique). Cette méthode est nommée “CWFEM” pour CondensedWave Finite Element Method. Elle combine les avantages de la CMS et WFEM. La CMS permet d’analyser le comportement local d’en extraire une base réduite. La WFEM exploite la périodicité de la structure d’en extraire les paramètres de propagation. Ainsi, l’analyse de la propagation des ondes dans la structure à l’échelle macroscopique peut être réalisée en prenant en compte l’échelle mésoscopique. L’efficacité de la CWFEM est illustrée par de nombreuse applications aux structures périodiques monodimensionnelle (1D) et bidimensionnelle (2D). Le critère de réduction optimale assurant la convergence est discuté. Les caractéristiques de propagation dans les structures périodiques sont identifiées: bande passante, bande interdite, la directivité marquée (wave beaming effects), courbe de dispersion, band structure, surface des lenteurs... Ces propriétés peuvent répondre au besoin de conception des barrières vibroacoustiques, pièges à ondes. La CWFEM est ensuite appliquée pour étudier la propagation des ondes dans des plaques perforées et plaques raidies. Une méthode d’homogénéisation pour déterminer le modèle équivalent de la plaque perforée est proposée. Les comportements à haute fréquence tels que la directivité marquée sont également prédits par CWFEM. Trois modèles de plaques avec perforations différentes sont étudiées dans ce travail. Une validation expérimentale est effectuée sur deux plaques. Pour la plaque raidie, l’influence des modes internes sur la propagation globale est discutée. La densité modale est estimée, en moyenne et haute fréquences, pour une plaque raidie finie, où une bonne corrélation est obtenue en comparant les résultats à l’issue des analyses modales
Structural dynamics can be described in terms of structural modes as well as elastic wave motions. The mode-based methods are widely applied in mechanical engineering and numerous model order reduction (MOR) techniques have been developed. When it comes to the study of periodic structures, wave description is mostly adopted where periodicity is fully exploited based on the Bloch theory. For complex periodic structures, several MOR techniques conducted on wave basis have been proposed in the literature. In this work, a wave and modal coupled approach is developed to study the wave propagation in periodic structures. The approach begins with the modal description of a unit cell (mesoscopic scale) using Component Mode Synthesis (CMS). Subsequently, the wave-based method -Wave Finite Element Method (WFEM) is applied to the structure (macroscopic scale). The method is referred as “CWFEM” for Condensed Wave Finite Element Method. It combines the advantages of CMS and WFEM. CMS enables to analyse the local behaviour of the unit cell using a reduced modal basis. On the other hand, WFEM exploits fully the periodic propriety of the structure and extracts directly the propagation parameters. Thus the analysis of the wave propagation in the macroscopic scale waveguides can be carried out considering the mesoscopic scale behaviour. The effectiveness of CWFEM is illustrated via several one-dimensional (1D) periodic structures and two-dimensional (2D) periodic structures. The criterion of the optimal reduction to ensure the convergence is discussed. Typical wave propagation characteristics in periodic structures are identified, such as pass bands, stop bands, wave beaming effects, dispersion relation, band structure and slowness surfaces...Their proprieties can be applied as vibroacoustics barriers, wave filters. CWFEM is subsequently applied to study wave propagation characteristics in perforated plates and stiffened plate. A homogenization method to find the equivalent model of perforated plate is proposed. The high frequency behaviours such as wave beaming effect are also predicted by CWFEM. Three plate models with different perforations are studied. Experimental validation is conducted on two plates. For the stiffened plate, the influence of internal modes on propagation is discussed. The modal density in the mid- and high- frequency range is estimated for a finite stiffened plate, where good correlation is obtained compared to the mode count from modal analysis

L'objectif de cette thèse est de clarifier les conditions d'émergence en métaux liquides des ondes d'Alfvén dans un domaine géométriquement contraint. La première partie de ce travail de recherche est consacrée à une étude linéaire des ondes d'Alfvén dans l'approximation bas-Rm et en régime non inertiel. La seconde partie porte sur l'étude expérimentale d'un écoulement oscillant forcé électriquement, soumis à un champ magnétique axial, statique et uniforme, et confiné entre deux parois horizontales rigides, sans glissement et électriquement isolantes.Dans l'étude théorique menée, une première partie vise à discuter la relation de dispersion pour la dynamique des ondes d'Alfvén. Elle présente les conséquences liées à des gradients (mécaniques et magnétiques) perpendiculaires au champ magnétique imposé, plus particulièrement la manière dont la propagation de l'onde est ainsi modifiée. Dans la deuxième partie, un vortex axisymétrique confiné entre deux parois horizontales isolées électriquement et sans glissement est magnétiquement forcé à une fréquence donnée. Ce forçage prend en compte le rayon du vortex afin d'étudier l'impact des gradients transversaux sur la dynamique de l'écoulement. Une étude semi-analytique de la dynamique de l'écoulement est à nouveau réalisée dans un cadre bas-Rm et non inertiel. Cette étude, réalisée en faisant varier la fréquence de forçage et l'intensité du champ magnétique, met en évidence deux régimes très distincts, à savoir un premier régime oscillant-diffusif, régi par la compétition entre l'effet pseudo-diffusif de la force de Lorentz et le terme instationnaire de la quantité de mouvement, et un second régime, propagatif, régi par les ondes d'Alfvén et obtenu pour des fréquences de forçage plus élevées. L'étude met également en évidence l'impact des gradients transversaux sur ce régime propagatif. En plus de sur-amortir les ondes, les gradients transversaux modifient les fréquences naturelles des pics de résonance d'ondes, lesquels résultent de la superposition d'ondes incidentes et réfléchies entre les parois du domaine d'étude.Parallèlement à ce travail théorique, un dispositif a été conçu afin d'étudier expérimentalement la dynamique d'écoulements oscillants sous un champ magnétique (jusqu'à 10T). Un écoulement est forcé dans un récipient cubique de 15 cm x 15 cm x 10 cm au moyen d'un courant alternatif injecté à l'aide de quatre électrodes situées sur la plaque inférieure. En utilisant une instrumentation basée sur les différences locales de potentiel électrique aux niveau des plaques (d'Hartmann) supérieure et inférieure, nous validons les prédictions du modèle. Plus précisément, nous retrouvons un régime propagative modifié par les gradients transversaux ainsi que le régime oscillant-diffusif, obtenu pour des fréquences de forçage plus faibles.En plus des résultats obtenus à la fréquence de forçage, un premier aperçu des signaux obtenus à d'autres fréquences est présenté. Certains des pics de fréquence obtenus ne pouvant pas être expliqués par une approche linéaire, nous suggérons qu'ils sont générés par des interactions non linéaires d'ondes d'Alfvén. En outre, une étude préliminaire sur le pic à la première harmonique de la fréquence de forçage montre qu'il est très probablement associé à des ondes d'Alfvén
The thesis aims to clarify the conditions for Alfvén waves to propagate in a closed liquid metal domain. A first part of the research work presented is dedicated to a linear study of Alfvén waves in the low-Rm approximation and under the inertia-less limit. The second part is the experimental investigation of an electrically-induced oscillating flow subjected to an axial, static and uniform magnetic field and confined between two electrically insulating and no-slip horizontal walls.The theoretical study is itself split into two sub-parts. The first one aims to discuss the dispersion relation which contains the Alfvén wave dynamics. It presents the consequences of (mechanical and magnetic) gradients perpendicular to the imposed magnetic field. As such transverse gradients tend to impede the wave propagation. In the second sub-part an axisymmetric vortex confined between to electrically insulated and no-slip horizontal walls is magnetically forced at a given frequency. This forcing is radially dependent so as to study the impact of transverse gradients on the flow dynamics. A semi-analytical investigation of the flow dynamics is again carried out in the low-Rm approximation and under the inertia-less limit. This investigation is performed by varying the forcing frequency and the magnetic field intensity. This brings to emphasize two very distinct regimes for the oscillating vortex:- an oscillating-diffusive regime governed by the competition between pseudo-diffusive effects of the Lorentz force and the unsteady term of the momentum- a truly propagative regime, obtained for higher forcing frequencies, found definitelygoverned by Alfvén waves.The study also highlights how the propagative regime can be affected by transverse gradients. In addition to over-damping the waves, transverse gradients are found to modify the natural frequencies for which wave resonance peaks result from the superimposition of incident and reflected waves in the container.Beside this theoretical work, a setup has been designed in order to experimentally investigate the dynamics of oscillating flows under a strong magnetic field (up to 10T). A flow was forced in a cuboid vessel 15 cm x 15 cm x 10 cm by means of AC currents injected through a cartesian grid of four electrodes located at the bottom plate. Using instrumentation based on the measurement of local electric potential differences at the top and bottom horizontal (Hartmann) plates, we validate model's prediction. More precisely, a propagative dynamics in the presence of transverse gradients is recovered. The oscillating-diffusive regime is also recovered from experiments performed at small enough forcing frequency.In addition to results obtained at the forcing frequency, a first insight of signals obtained at other frequencies is shown. Frequency peaks obtained, eg the harmonics of the forcing frequency, are demonstrated not to be explained by a linear approach. We suggest that Alfvén wave non-linear interactions are a good candidate to explain these peaks. A preliminary study further shows that peaks at the first harmonic are likely to be Alfvén waves

Orientador: Jose Roberto de França Arruda
Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecanica
Made available in DSpace on 2018-08-13T08:53:33Z (GMT). No. of bitstreams: 1 Nascimento_RangelFerreirado_D.pdf: 7547341 bytes, checksum: 0793f0ff7763f81b44868f59db73aab6 (MD5) Previous issue date: 2009
Resumo: Esta tese estuda e investiga o problema de propagação de ondas em estruturas periódicas usando o método de elemento espectral, a relação entre a matriz dinâmica e a matriz de transferência é mostrada para alguns casos, tais como, viga, barra, placa de Levy e modelo de Minddlin Hermman. A partir destas teorias, o método de propagação de ondas usando um modelo de elementos finitos de uma fatia do guia de ondas, WFEM é apresentado e o problema de prever os modos de propagação e os números de onda correspondentes. O objetivo deste trabalho é mostrar que usando o método WFEM e uma fatia do guia de onda modelado com elementos finitos sólido é possível construir elementos finitos espectrais para ser usado em guias de ondas homogêneos sem precisar de malha de refinamento. Tais elementos podem ser usados para modelar guias de ondas com seção transversal constante. A matriz de rigidez dinâmica para o elemento de barra elementar e para o elemento de viga de Euler Bernoulli são obtidos usando a formulação espectral padrão e obtidas usando uma fatia do guia de onda modelado pelo método FEM, são mostrados resultados do método proposto.
Abstract: This thesis, studies and investigates wave propagation problem in periodic structures using the spectral element method, the relation between the dynamic matrix and the transfer matrix is shown for some cases, such as, beam, bar, Levy plate and Mindlin-Herrmann's model. From these theories, the Wave Finite Element Method, WFEM is presented and the problem of predicting the wave propagation modes and the respective wavenumbers. The purpose of this work is to show that using the WFEM method and a slice of the waveguide modeled with solid finite elements, it is possible to develop spectral finite elements to be used in long homogeneous waveguides without the need of mesh refinement. Such elements can be used to model waveguides with constant cross section and long spans. The dynamic stiffness matrix of a simple rod and Bernoulli Euler beam element obtained using the standard spectral formulation and obtained via the FEM model of a slice are shown to be similar, thus validating the proposed method.
Doutorado
Mecanica dos Sólidos e Projeto Mecanico
Doutor em Engenharia Mecânica

Étude des propriétés acoustiques des verres magnétiques isolants pour des concentrations en impuretés magnétiques variant entre 1,5 et 10,1% et, à très basse température (de 10 mk a 10 k) pour des champs magnétiques atteignant 60 ko. Le choix de ho comme impureté magnétique a permis de mettre en évidence des mouvements actives de faible énergie d'entités magnétiques qui ne sont pas lies a la phase verre de spin. L’étude des états tunnels élastiques a permis de mettre en évidence que la description classique des temps de relaxation n'est plus valide dans les isolants
We study the acoustical properties of amorphous magnetic insulators for varying concentration of magnetic impurities between 10,1 % at. And 1,5 % at. . We use a method of propagating wave between 10 MHz and 1000 MHz at very low temperature (from 10 mK up to 10 K) and for magnetic fields up to 60 kOe. With the choice of Ho3+ ions as magnetic impurities, we have been able to point out low energies activated movements of magnetic entities (V 5 K) which are not related at the spin glass transition. The study of tunnelling states in these glasses shows that their classical description in the insulators is no more valid here and that we must take into account the "direct" interaction between tunnelling states and magnetic impurities

Orientador: Lúcio Tunes dos Santos
Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática Estatística e Computação Científica
Made available in DSpace on 2018-08-25T10:27:58Z (GMT). No. of bitstreams: 1 Camargo_AlexandreWilliam_M.pdf: 30605851 bytes, checksum: 95a610618221b15f52fb3f96d3239196 (MD5) Previous issue date: 2014
Resumo: A base fundamental do modelamento sísmico é a equação da onda acústica, que para meios com velocidade variável exige métodos numéricos eficientes para encontrar a solução da equação. O Método de Diferenças Finitas (MDF) é muito usado para obter a solução da equação da onda, pois é de fácil implementação. Uma vez garantindo a convergência do método a solução aproximada é confiável. Contudo, os esquemas utilizados pelo MDF ou utilizam um comprimento longo para os operadores da derivada espacial ou precisam ter uma malha suficientemente densa, isto é, ter passos pequenos para garantir uma solução precisa, porém, isso requer o cálculo de muitas operações na fórmula recursiva do método. Nesta dissertação, analisamos um esquema do MDF que utiliza comprimentos adaptativos para o operador da derivada espacial assumindo uma malha computacional com o passo fixo. O critério que escolhe esses comprimentos depende da velocidade do meio. Pela relação entre o comprimento e a velocidade, o método escolhe um comprimento longo em regiões de baixa velocidade e um comprimento curto nas regiões de alta velocidade. Os testes numéricos comprovaram que o MDF com o esquema de comprimentos adaptativos obtém uma solução com uma precisão similar à solução do MDF com o esquema que usa somente comprimentos longos, mas o esquema de comprimento adaptativo realiza um número de operações menor na fórmula de recursão
Abstract: The fundamental base of seismic modeling is the acoustic wave equation that for a medium with variable velocity require efficients numerical methods to find the equation solution. The Finite Difference Method (FDM) is widely used to get the solution of the wave equation because the computational implementation is very easy. Ensuring the con- vergence of the method, the approximate solution is reliable. However, the schemes used by FDM either need a high length to the operator of the spatial derivative or take a sufficiently dense mesh, that is, the grid is small to ensure an accurate solution. But this requires the calculus of many operations in the recursion formula. In this dissertation, we analyzed a scheme of the FDM that use adaptive lengths to the operator of the spatial de- rivative assuming a fixed grid. The criterion to choose the lengths depend on the velocity of the medium. Thereby, in regions of low velocity a long length is used and in regions of high velocity a short length is enough. The numerical tests show that FDM with the adaptive length scheme obtains a solution with a similar accuracy to the solution of the FDM with the long length scheme, but performs a smaller number of operations in the recursion formula
Mestrado
Matematica Aplicada
Mestre em Matemática Aplicada

The current markets are characterized by an increasing demand for individual products, increasing product complexity and stricter safety regulations. This results in large hazard and safety analysis efforts for each individual product variant. One possible solution to reduce these efforts is the automation of analyses and a preliminary assessment of individual changes. Therefore, several approaches are published to e.g. model safety aspects or evaluate change propagations. However, the approaches fail to directly establish the connection between product changes and safety aspects by a common model. This paper develops a method to identify and assess the potential hazard impact of product changes through a common static model. It builds on a graph-based product model and graph-rewriting. From the state of the art, suitable methods tools and principles are identified and evaluated. Based on this, requirements are derived. The capabilities of existing methods are assessed and the most suitable ones adapted and integrated to the method to assess the potential hazard impact of product changes (MBHPA). The MBHPA reduces the complexity by providing two independent interconnected analyses. To evaluate the product changes, a static propagation analysis is provided. It uses defined graph-rewriting patterns to trace propagations and extract a propagation tree. The affected components in a second step can then be evaluated on their hazard potential. Again graph-rewriting patterns are applied to identify the connected hazards which are visualized in a hazard potential portfolio. The MBHPA is implemented and evaluated with the industrial case of an automated coffee machine. The evaluation underlines that the MBHPA successfully identifies possible propagations and their effect on hazards from a static perspective. It helps to improve safety awareness and traceability and reduces the required experience.

In this paper, the Kriging metamodel is applied to accelerate the model assisted probability of detection (MAPoD) analysis for surface flaws in eddy current nondestructive testing (ECNDT). The kernel degeneration (KD) algorithm accelerated boundary element method (BEM) EC simulation solver is selected as the physical model to generate the responses of the surface flaws in the conducting plate considering the propagations of uncertain parameters. The predicted PoD metrices by Kriging metamodel are compared well with the ones from pure physical model. Numerical results show that to reach the required accuracy, as compared with the pure BEM based physical model, Kriging metamodel can accelerate the MAPoD analysis for ECNDT problems.

The development of electrooptic polymers now stands at the onset of technological fruition as a result of almost two decades of intense molecular engineering studies and of the more recent maturing of semiconductor compatible integrated optics fabrication processes. Nevertheless, the full potential of organic systems for nonlinear optics(1) and related applications may not have been fully exploited so-far within the paradigmatic orientational scheme of a dipolar molecular diode structure coupled with an externally applied poling electric field at thermal equilibrium. In particular, such a configuration makes it difficult to implement (quasi)-phase matched gratings for (cascaded) quadratic NLO, guarantee polarization independant telecom device behaviour or engineer other propagative (e.g. soliton generation) or QED (e.g. microcavity) configurations. A different approach, whereby all-optical photoinduced processes are called-upon instead of thermally equilibrated ones and traditional 1-D systems are traded for more general 3-D multipolar molecules (e.g. octupolar, dipolar or a combination of these) opens-up new possibilities to adress these issues. It permits indeed to micro-pattern and control with subwavelength accuracy a continuous or pixellised spatial χ(2) tensorial distribution whereby the magnitude and ratios of macroscopic tensorial coefficients are balanced at will by phase and ellipsometric adjustment of the writing beams2.

The success of graph convolutional neural networks (GCNNs) based semi-supervised node classification is credited to the attribute smoothing (propagating) over the topology. However, the attributes may be interfered by the utilization of the topology information. This distortion will induce a certain amount of misclassifications of the nodes, which can be correctly predicted with only the attributes. By analyzing the impact of the edges in attribute propagations, the simple edges, which connect two nodes with similar attributes, should be given priority during the training process compared to the complex ones according to curriculum learning. To reduce the distortions induced by the topology while exploit more potentials of the attribute information, Dual Self-Paced Graph Convolutional Network (DSP-GCN) is proposed in this paper. Specifically, the unlabelled nodes with confidently predicted labels are gradually added into the training set in the node-level self-paced learning, while edges are gradually, from the simple edges to the complex ones, added into the graph during the training process in the edge-level self-paced learning. These two learning strategies are designed to mutually reinforce each other by coupling the selections of the edges and unlabelled nodes. Experimental results of transductive semi-supervised node classification on many real networks indicate that the proposed DSP-GCN has successfully reduced the attribute distortions induced by the topology while it gives superior performances with only one graph convolutional layer.