Thesis (doctoral)--Université de Paris, 1867. Thèse de doctorat : Physique : Paris, Faculté des sciences : 1867. / "No. d'ordre: 293" Titre provenant de l'écran-titre. Références bibliogr.

Not available / text

In this study, novel Ray Based finite difference method for Time Domain electromagnetics(RBTD) has been developed. Instead of solving Maxwell&rsquo / s hyperbolic partial differential equations directly, Geometrical Optics tools (wavefronts, rays) and Taylor series have been utilized. Discontinuities of electromagnetic fields lie on wavefronts and propagate along rays. They are transported in the computational domain by transport equations which are ordinary differential equations. Then time dependent field solutions at a point are constructed by using Taylor series expansion in time whose coefficients are these transported distincontinuties. RBTD utilizes grid structure conforming to wave fronts and rays and treats all electromagnetic problems, regardless of their dimensions, as one dimensional problem along the rays. Hence CFL stability condition is implemented always at one dimensional eqaulity case on the ray. Accuracy of RBTD depends on the accuracy of grid generation and numerical solution of transport equations. Simulations for isotropic medium (homogeneous/inhomogeneous) have been conducted. Basic electromagnetic phenomena such as propagation, reflection and refraction have been implemented. Simulation results prove that RBTD eliminates numerical dispersion inherent to FDTD and is promising to be a novel method for computational electromagnetics.

Thesis (Ph. D.)--University of Texas at Austin, 2005. / Vita. Includes bibliographical references.

This dissertation investigates various design metrologies on designing freeform surfaces for LED illumination applications. The major goal of this dissertation is to study designing freeform optical surfaces to redistribute the radiance (which can be simplified as intensity distribution for point source) of LED sources for various applications. Nowadays many applications, such as road lighting systems, automotive headlights, projection displays and medical illuminators, require an accurate control of the intensity distribution. Freeform optical lens is commonly used in illumination system because there are more freedoms in controlling the ray direction. Design methods for systems with rotational and translational symmetry were well discussed in the 1930's. However, designing freeform optical lenses or reflectors required to illuminate targets without such symmetries have been proved to be much more challenging. For the simplest case when the source is an ideal point source, the determination of the freeform surface in a rigorous manner usually leads to the tedious Monge-Ampère second order nonlinear partial different equation, which cannot be solved with standard numerical integration techniques. Instead of solving the differential equation, ray mapping is an easier and more efficient method in controlling one or more freeform surfaces for prescribed irradiance patterns. In this dissertation, we investigate the ray mapping metrologies in different coordinate systems to meet the integrability condition for generating smooth and continuous freeform surfaces. To improve the illumination efficiency and uniformity, we propose a composite ray mapping method for designing the total internal reflective (TIR) freeform lens for non-rotational illumination. Another method called "double pole" ray mapping method is also proposed to improve system performance. The ray mapping designs developed for the point source do not work well for extended sources, we have investigated different design methodologies including optimization method, deconvolution method and feedback modification method to design freeform optical surfaces for extended sources.

[pt] Antenas refletoras para cobertura omnidirecional vêm sendo utilizadas em diversos estudos de micro-ondas e ondas milimétricas. A principal motivação para trabalhar nestas bandas, entre outras aplicações, é o desenvolvimento de sistemas de comunicação sem fio de banda larga. O presente trabalho utiliza uma lente dielétrica na abertura do alimentador da antena refletora para reduzir a largura do feixe e, simultaneamente, evitar a presença de lóbulos laterais na região de cobertura. Uma apropriada modelagem da lente reduz o tamanho da antena sem degradar as características de radiação. O trabalho é dividido em duas partes: a modelagem de lentes dielétricas; e a modelagem de reflertores. A modelagem de lentes dielétricas circularmente simétricas utilizando os princípios da Óptica Geométrica para controlar a largura de feixe do diagrama de radiação transmitido pela lente. O modelo é feito a partir do deslocamento do foco da geratriz da lente, o novo foco virtual é o ponto geométrico onde convergem os raios emergentes da lente. A rotação da geratriz da superfície refletora faz que este ponto se torne um anel cáustico virtual. As lentes modeladas têm como alimentador uma corneta coaxial que fornece um diagrama de radiação circularmente simétrico. O diagrama de radiação transmitido pela lente é calculado usando as aproximações da ótica geométrica e ótica física em campo distante e próximo. Os resultados são comparados com os resultados simulados em um software especializado de simulação eletromagnética. A modelagem de refletores baseada nas propriedades da Ótica Geométrica. O refletor é uma superfície de revolução obtida através da rotação de uma geratriz em torno ao eixo de simetria. Neste trabalho, a geratriz é descrita por uma sucessão de seções de cônicas concatenadas, esta metodologia utiliza o diagrama de radiação transmitido da lente para obter uma distribuição de campo no plano vertical previamente especificada, que nosso caso é uma distribuição constante. O diagrama de radiação em campo distante radiado pelo refletor é calculado usando as aproximações da física ótica, estes resultados são 6 comparados com os resultados simulados em um software especializado em simulação eletromagnética. / [en] Reflector antennas for omnidirectional coverage have been considered in several studies of microwave and millimeter wave. The main motivation to work in these bands, among other applications, is the development of wireless broadband communication systems. This work uses a dielectric lens in the aperture of the reflector antenna feeder to reduce the beamwidth and simultaneously avoid the presence of side lobes in the coverage area. An appropriate lens modeling reduces the antenna size without degrading the radiation characteristics. The work is divided into two parts: the modeling of lens and modeling of reflector.The modeling of lens using the principles of the Geometrical Optics to control the radiation pattern transmitted by the lens. The modeling depends on the lens focus displacement, which is the geometric point of convergence of the rays emerging from the lens. The rotation of the lens generatrix causes this point to become a virtual caustic ring. The lens has a coaxial feed horn that provides a circularly symmetrical radiation pattern. The radiation pattern transmitted by the lens is calculated using the approximation of Geometric Optics and Physical Optics in the near and far field; these results are compared with the simulated results in a specialized electromagnetic simulation software. The reflector modeling is based on the properties of Geometric Optics. The reflector is a revolution surface obtained by rotating a generatrix around the symmetry axis. In this work, the generatrix is described by a sequence of concatenated conical sections. This method uses the transmitted radiation pattern of the lens to obtain a previously specified field distribution in the vertical plane, which is a constant distribution in the present case. The far field radiation pattern of the reflector is calculated using the Physical Optics approximation. These results are compared with the simulated results in a specialized electromagnetic simulation software.

Neste trabalho, construímos novas soluções com simetria esférica ou axial para as equações do campo gravitacional induzido num cenário brana do tipo Randall-Sundrum, onde uma 3-brana com constante cosmológica está imersa num espaço-tempo 5-dimensional conhecido como bulk. Para o caso esfericamente simétrico, com constante cosmológica negativa, obtivemos uma família de soluções numa brana assintoticamente anti-de Sitter, sendo que cada membro desta é diferenciado de outro por uma constante integração C, que fixada proveu-nos soluções de buracos negros ou buracos de minhocas. Com a mesma simetria, geometrias que descrevem buracos de minhoca foram encontradas numa brana assintoticamente de Sitter. Para o caso axialmente simétrico, métricas assintoticamente anti-de Sitter e de Sitter foram construídas no contexto citado. Propriedades óticas, como a rotação do vetor de polarização, foram estudadas, mostrando-nos algumas diferenças entre os mundos brana, tratados como uma teoria de gravitação modificada, e a relatividade geral. / In this work we have constructed new solutions, with axial or spherical symmetry, to the induced gravitational field equations within a Randall-Sundrum type brane world scenario, where a 3-brane with cosmological constant is embedded in a 5-dimensional space-time called bulk. For the spherical case with negative cosmological constant, we have found a family of solutions in an anti-de Sitter brane, where each member of the family is different from another when C is fixed. With a fixed C we have black holes or wormholes. In this same case, with spherical symmetry, we have obtained wormhole solutions in a de Sitter brane. In the same context, for the axially symmetric case, metrics asymptotically anti-de Sitter and de Sitter were built. Optics features, like rotation of polarization vector, have been studied. These features showed some differences between brane worlds, a theory of modified gravity, and the General Relativity.

Ce travail de thèse est une contribution au projet national AMOCOPS, financé par l’ANR. Le thème central du projet est la diffusion de lumière par des particules de formes complexes et de grandes tailles (plusieurs dizaines de µm au moins), domaine où les méthodes de simulation numérique existantes trouvent leurs limites d’applicabilité. Nous abordons le problème par le biais des effets mécaniques de la lumière, autrement dit les forces et couples créés par la pression de radiation. Etant la conséquence du transfert d’impulsion entre l’onde et la matière, ces effets sont directement liés à la diffusion de lumière. La thèse comprend une partie expérimentale –majoritaire- concernant les réponses mécaniques de particules de polystyrène de forme ellipsoïdale et d’allongement variable sous illumination par un ou deux faisceaux laser. Les cas de faisceaux faiblement focalisés (lévitation optique) et d’un faisceau très fortement focalisé (pincette optique) sont examinés successivement. Nous caractérisons différents types d’équilibre statique, certains d’entre eux non décrits auparavant, obtenus dans les deux géométries. Par ailleurs nous confirmons l’existence de réponses purement dynamiques, où la particule oscille en permanence. Trois nouveaux modes sont observés, deux dans la géométrie lévitation optique et un autre sous pincette optique. Cette étude nous permet de distinguer les oscillations dites de Simpson-Hanna dans le régime linéaire de celles non linéaires mises en évidence avant nous par Mihiretie et al..Les résultats de nos expériences sont comparés à ceux obtenus par les simulations de J.C. Loudet, sur la base de la simple optique géométrique (OG) et limitées à 2 dimensions (2d). Nous montrons que ces simulations permettent de reproduire qualitativement et comprendre physiquement la plupart des comportements observés dans nos expériences. La principale limitation de ces calculs tient à ce que l’OG ignore le caractère ondulatoire de la lumière. Pour faire mieux et aller vers des simulations fiables quantitativement, il faut développer un modèle alliant optique géométrique et optique ondulatoire. C’est la fonction du modèle VCRM (Vectorial Complex Ray Model) développé récemment par K.F. Ren en 2d. Le but du projet Amocops est de mettre au point la version 3d de la méthode et de la valider sur la base d’expériences comme celles que nous avons conduites. La deuxième partie de la thèse est consacrée à la méthode VCRM. Nous en exposons les principes, et nous présentons quelques résultats des travaux en cours avec une version intermédiaire entre 2d et 3d, dite « 2d+ ». Quelques illustrations sont proposées sur des exemples impliquant des sphères et ellipsoïdes de grandes tailles. / This work is a contribution to the “AMOCOPS” project, funded by Agence Nationale de la Recherche. AMOCOPS is dedicated to the development of new computation schemes to simulate the light scattering patterns of large complexly shaped particles. Particle sizes are of the order of several 10s of micrometres, which is at the limit, or beyond the capabilities of currently available computation techniques.Our work indirectly deals with light scattering through the corresponding mechanical effects of light. Light scattering is the source of momentum transfer between light and matter, and therefore of the forces and torques acting on the exposed particles. The majority of Part A of this thesis is about the mechanical responses of ellipsoidal polystyrene particles of varying aspect ratios, under illumination by one or two laser beams. We investigate the case of weakly focused beams (optical levitation), and that of a single large aperture beam (optical tweezers). Different types of static equilibria, some of which are new, are observed and characterized in both geometries. We confirm the existence of dynamic states, whereby the particle permanently oscillates within the laser beam(s). Three new oscillation modes are observed, two of them in the conditions of optical levitation, and another one in the optical tweezer geometry. The study allows us to make a distinction between noise-driven oscillations in the linear regime, of the type predicted by Simpson and Hanna, and nonlinear oscillations such as those evidenced prior to this work, by Mihiretie et al..Results from our experiments are compared to simulations by J.C. Loudet, using simple ray-optics (RO) in two dimensions (2D). We show that results from 2D-RO qualitatively match most of our observations, and allow us to physically understand the main mechanisms at work in the observed phenomena. The simulations cannot be quantitatively exact, due to the 2D limitation, and because RO essentially ignores the wave nature of light. In Part B of the manuscript, we present the principles of the Vectorial Complex Ray Model (VCRM), which was recently developed by K.F. Ren in 2d. The goal of AMOCOPS is to develop a full 3D version of VCRM, able to simulate light scattering by particles of any shape with a smooth surface. We explain the basics of the model, as well as the “2D+” version, which is an extension of the basic 2D-VCRM. A few illustrative examples of light scattering patterns computed with 2d+-VCRM for large-sizes spheres and ellipsoids are presented

[pt] Este trabalho apresenta um estudo sobre lentes dielétricas projetadas para corrigir a fase e amplitude de antenas do tipo corneta. Baseado nos princípios da Ótica Geométrica (GO), foi desenvolvido um programa computacional que sintetiza uma lente dielétrica que transforma os campos da abertura da corneta na distribuição de campo da abertura especificada. Além da síntese, o programa também realiza a análise aproximada baseada na GO para calcular o diagrama de radiação de antenas do tipo corneta, considerando as perdas devido à reflexão que não foram consideradas na síntese. Isto é realizado através do rastreamento de raios desde o alimentador até uma abertura virtual em frente à lente onde é, então, aplicado o Método da Abertura para calcular o diagrama de radiação. Os resultados obtidos são comparados com estudos disponíveis na literatura técnica e com a análise rigorosa realizada no Microwave Studio CST. A primeira parte do estudo apresenta exemplos de lentes para corrigir os campos da abertura de corneta corrugada que radia um diagrama de radiação aproximadamente circular, levando a lentes com simetria circular. Finalmente, na segunda parte, foi considerado o caso onde o alimentador é uma corneta cônica, que requere lentes assimétricas. / [en] The work presents a study about dielectric lenses designed to correct the aperture phase and amplitude of horn antennas. Based on Geometrical Optics (GO) principles, it has been developed a computational tool that synthesizes a dielectric lens that transforms the aperture fields of horn into a specified aperture distribution. Besides the synthesis, the tool also presents a GO approximated analysis of the horn antenna radiation pattern, by considering the losses due to the reflection not accounted in the synthesis. It is performed by applying the ray tracing from the feed to a virtual aperture in front the lens, and by applying the Aperture Method to calculate the radiation pattern. The design performances are compared with studies found in the technical literature and with a rigorous analysis performed in Microwave Studio CST. The first part of the study, presents design examples to correct the aperture fields of corrugated horns that radiates approximated circular radiation pattern, leading to circular symmetric lenses. Finally, in the second part, it has been considered the case where the feed is a conic horn, which requires asymmetric lenses.

Reflector design stemmed from the need to shape the light emitted by candles or lamps. Over 2,000 years ago people realized that a mirror shaped as a parabola can concentrate light, and thus significantly boosts its intensity, to the point where objects can be set afire. Nowadays many applications require an accurate control of light, such as automotive headlights, streetlights, projection displays, and medical illuminators. In all cases light emitted from a light source can be shaped into a desired target distribution with a reflective surface. Design methods for systems with rotational and translational symmetry were devised in the 1930s. However, the freeform reflector shapes required to illuminate targets with no such symmetries proved to be much more challenging to design. Even when the source is assumed to be a point, the reflector shape is governed by a set of second-order partial non-linear differential equations that cannot be solved with standard numerical integration techniques. An iterative approach to solve the problem for a discrete target, known as the method of supporting ellipsoids, was recently proposed by Oliker. In this research we report several efficient implementations of the method of supporting ellipsoids, based on the point source approximation, and we propose new reflector design techniques that take into account the extent of the source. More specifically, this work has led to three major achievements. First, a thorough analysis of the method of supporting ellipsoids was performed that resulted in two alternative implementations of the algorithm, which enable a fast generation of freeform reflector shapes within the point source approximation. We tailored the algorithm in order to provide control over the parameters of interest to the designers, such as the reflector scale and geometry. Second, the shape generation algorithm was used to analyze how source flux can be mapped onto the target. We derived the condition under which a given source-target mapping can be achieved with a smooth continuous surface, referred as the integrability condition. We proposed a method to derive mappings that satisfy the integrability condition. We then use these mappings to quickly generate reflector shapes that create continuous target distributions as opposed to reflectors generated with the method of supporting ellipsoids that create discrete sets of points on the target. We also show how mappings that do not satisfy the integrability condition can be achieved by introducing step discontinuities in the reflector surface. Third, we investigated two methods to design reflectors with extended sources. The first method uses a compensation approach where the prescribed target distribution is adjusted iteratively. This method is effective for compact sources and systems with rotational or translational symmetry. The second method tiles the source images created by a reflector designed with the method of supporting ellipsoids and then blends the source images together using scattering in order to obtain a continuous target distribution. This latter method is effective for freeform reflectors and target distributions with no sharp variations. Finally, several case studies illustrate how these methods can be successfully applied to design reflectors for general illumination applications such as street lighting or luminaires. We show that the proposed design methods can ease the design of freeform reflectors and provide efficient, cost-effective solutions that avoid unnecessary energy consumption and light pollution.