Ce travail cherche a comprendre comment l'ajout d'une géométrie non euclidienne dans un problème de Schrödinger non linéaire influe sur l'existence et l'unicité des solutions explosives de masse critique. On s'inspire pour beaucoup des travaux de Merle et Raphaël sur la méthode de modulation des paramètres d'invariance géométrique pour une EDP qui possède de bonnes lois de conservations. On s'appuie ici plus particulièrement sur un article de Raphaël et Szeftel qui prouve l'existence et l'unicité d'une solution de masse critique en dimension 2 pour l'équation de Schrödinger non linéaire avec potentiel d'inhomogénéité devant la non-linéarité, et qui explose par ailleurs au maximum de l'inhomogénéité. Dans un premier temps, il s'agit de reprendre la méthode dans son ensemble afin de l'adapter à des cas où le Laplacien n'est plus plat, et est remplacé par un opérateur de type Laplace-Beltrami ou Laplacien généralisé. Ayant mis en avant le rôle de la courbure au point d'explosion, en termes de conditions sur les dérivées de termes métriques, on reprend dans un deuxième temps l'étude dans le cas plus général d'une variété riemannienne. Grâce à un ansatz sur la solution qui intègre maintenant la transformation induite par la métrique, on est capable d'énoncer un résultat d'existence et d'unicité en termes de conditions géométriques sur la variété elle même. Par soucis de simplicité, on se limite néanmoins au rôle local de la métrique, en la supposant globalement définie dans une certaine carte, et asymptotiquement équivalente a la métrique euclidienne. / The present work aims at investigating the effects of a non-euclidean geometry on existence and uniqueness results for critical blow up NLS solutions. We will use many ideas from the works of Merle and Raphaël, particularly ideas from modulation theory which describes a solution in terms of geometric invariants parameters. We will rely more specically on a paper from Raphaël and Szeftel for existence and uniqueness of a critical mass blow up solution in dimension two tothe nonlinear Schrödinger equation with inhomogeneous potential acting on the nonlinearity, and which blows up where the inhomogeneity reaches its maximum. At first, we consider a generalized Laplacian operator and deploy the classical ansatz method to point out difficulties inherited from the non-flat metric terms, and in particular the key role played by the curvature at the blow-up point. In a second part, we reproduce the method when modifying the geometrical ansatz on which the parametrix is constructed, and investigate more precisely what is needed for existence and then uniqueness when dealing with a Laplace-Beltrami operator associated to a riemannian manifold. For simplicity, we shall only consider the role of g locally around the blow up point we are constructing, by assuming g is globally defined in some map, and asymptotically equals the usual euclidean metric.

Neste trabalho, estudamos a existência de soluções não triviais para uma classe de equações de Schrödinger não lineares envolvendo um campo magnético com condição de Dirichlet ou condição de fronteira mista Dirichlet-Neumann. Nos dois primeiros capítulos, damos uma estimativa para o número de soluções não triviais para o problema de Dirichlet em termos da topologia do domínio. Nos dois capítulos restantes, consideramos o problema de fronteira mista e estimamos o número de soluções não triviais em termos da topologia da porção da fronteira onde é prescrita a condição de Neumann. Em ambos os casos, usamos a teoria de categoria de Ljusternik-Schnirelmann e a teoria de Morse / We study the existence of nontrivial solutions for a class of nonlinear Schrödinger equations involving a magnetic field with Dirichlet or mixed DirichletNeumann boundary condition. In the first two chapters we give an estimate for the number of nontrivial solutions for the Dirichlet boundary value problem in terms of topology of the domain. In the last two chapters we consider mixed DirichletNeumann boundary value problems and the estimation of the number of nontrivial solutions is given in terms of the topology of the part of the boundary where the Neumann condition is prescribed. In both cases, we use Lyusternik- Shnirelman category and the Morse theory

<p>Numerical methods and algorithms for solving of partial differential equations, especially parallel algorithms, are an important research topic, given the very broad applicability range in all areas of science. Rapid advances of computer technology open up new possibilities for development of faster algorithms and numerical simulations of higher resolution. This is achieved through paralleliza-tion at different levels that&nbsp; practically all current computers support.</p><p>In this thesis we develop parallel algorithms for solving one kind of partial differential equations known as nonlinear Schr&ouml;dinger equation (NLSE) with a convolution integral kernel. Equations of this type arise in many fields of physics such as nonlinear optics, plasma physics and physics of ultracold atoms, as well as economics and quantitative&nbsp; finance. We focus on a special type of NLSE, the dipolar Gross-Pitaevskii equation (GPE), which characterizes the behavior of ultracold atoms in the state of Bose-Einstein condensation.</p><p>We present novel parallel algorithms for numerically solving GPE for a wide range of modern parallel computing platforms, from shared memory systems and dedicated hardware accelerators in the form of graphics processing units (GPUs), to&nbsp;&nbsp; heterogeneous computer clusters. For shared memory systems, we provide an algorithm and implementation targeting multi-core processors us-ing OpenMP. We also extend the algorithm to GPUs using CUDA toolkit and combine the OpenMP and CUDA approaches into a hybrid, heterogeneous al-gorithm that is capable of utilizing all&nbsp; available resources on a single computer. Given the inherent memory limitation a single&nbsp; computer has, we develop a distributed memory algorithm based on Message Passing Interface (MPI) and previous shared memory approaches. To maximize the performance of hybrid implementations, we optimize the parameters governing the distribution of data&nbsp; and workload using a genetic algorithm. Visualization of the increased volume of output data, enabled by the efficiency of newly developed algorithms, represents a challenge in itself. To address this, we integrate the implementations with the state-of-the-art visualization tool (VisIt), and use it to study two use-cases which demonstrate how the developed programs can be applied to simulate real-world systems.</p> / <p>Numerički metodi i algoritmi za re&scaron;avanje parcijalnih diferencijalnih jednačina, naročito paralelni algoritmi, predstavljaju izuzetno značajnu oblast istraživanja, uzimajući u obzir veoma &scaron;iroku primenljivost u svim oblastima nauke. Veliki napredak informacione tehnologije otvara nove mogućnosti za razvoj bržih al-goritama i&nbsp; numeričkih simulacija visoke rezolucije. Ovo se ostvaruje kroz para-lelizaciju na različitim nivoima koju poseduju praktično svi moderni računari. U ovoj tezi razvijeni su paralelni algoritmi za re&scaron;avanje jedne vrste parcijalnih diferencijalnih jednačina poznate kao nelinearna &Scaron;redingerova jednačina sa inte-gralnim konvolucionim kernelom. Jednačine ovog tipa se javljaju u raznim oblas-tima fizike poput nelinearne optike, fizike plazme i fizike ultrahladnih atoma, kao i u ekonomiji i kvantitativnim finansijama. Teza se bavi posebnim oblikom nelinearne &Scaron;redingerove jednačine, Gros-Pitaevski jednačinom sa dipol-dipol in-terakcionim članom, koja karakteri&scaron;e pona&scaron;anje ultrahladnih atoma u stanju Boze-Ajn&scaron;tajn kondenzacije.<br />U tezi su predstavljeni novi paralelni algoritmi za numeričko re&scaron;avanje Gros-Pitaevski jednačine za &scaron;irok spektar modernih računarskih platformi, od sis-tema sa deljenom memorijom i specijalizovanih hardverskih akceleratora u ob-liku grafičkih procesora, do heterogenih računarskih klastera. Za sisteme sa deljenom memorijom, razvijen je&nbsp; algoritam i implementacija namenjena vi&scaron;e-jezgarnim centralnim procesorima&nbsp; kori&scaron;ćenjem OpenMP tehnologije. Ovaj al-goritam je pro&scaron;iren tako da radi i u&nbsp; okruženju grafičkih procesora kori&scaron;ćenjem CUDA alata, a takođe je razvijen i&nbsp; predstavljen hibridni, heterogeni algoritam koji kombinuje OpenMP i CUDA pristupe i koji je u stanju da iskoristi sve raspoložive resurse jednog računara.<br />Imajući u vidu inherentna ograničenja raspoložive memorije koju pojedinačan računar poseduje, razvijen je i algoritam za sisteme sa distribuiranom memorijom zasnovan na Message Passing Interface tehnologiji i prethodnim algoritmima za sisteme sa deljenom memorijom. Da bi se maksimalizovale performanse razvijenih hibridnih implementacija, parametri koji određuju raspodelu podataka i računskog opterećenja su optimizovani kori&scaron;ćenjem genetskog algoritma. Poseban izazov je vizualizacija povećane količine izlaznih podataka, koji nastaju kao rezultat efikasnosti novorazvijenih algoritama. Ovo je u tezi re&scaron;eno kroz inte-graciju implementacija sa najsavremenijim alatom za vizualizaciju (VisIt), &scaron;to je omogućilo proučavanje dva primera koji pokazuju kako razvijeni programi mogu da se iskoriste za simulacije realnih sistema.</p>

Nesse trabalho, estudamos a equação de Schrödinger não-linear com uma função potencial delta atrativa. As soluções para essa equação tem uma componente localizada e uma dispersiva. Além de estudar o comportamento das soluções dessa equação em espaços de Sobolev clássicos, mostramos algumas propriedades do grupo unitário em espaços Lp, L2 com peso, Sobolev com peso e assim obtemos alguns resultados de boa colocação local e global das soluções. O ponto central desta tese é mostrarmos a existência de uma variedade invariante centro que irá consistir de órbitas periódicas no tempo. / In this work, we study the nonlinear Schrodinger equation with an attractive delta function potential.The solutions to this equation have a localized and a dispersive component. In addition to studying the behavior of solutions of this equation in classical Sobolev space, we show some properties for the unitary group in Lp, weighted L2 and Sobolev spaces and so we get some results of local and global well-posedness of solutions. The central theme this thesis is to show the existence of a center invariant manifold, which will consist of time-periodic orbits.

In the mid-1920s, the great Albert Einstein proposed that at extremely low temperatures, a gas of bosonic particles will enter a new phase where a large fraction of them occupy the same quantum state. This state would bring many of the peculiar features of quantum mechanics, previously reserved for small samples consisting only of a few atoms or molecules, up to a macroscopic scale. This is what we today call a Bose-Einstein condensate. It would take physicists almost 70 years to realize Einstein's idea, but in 1995 this was finally achieved. The research on Bose-Einstein condensates has since taken many directions, one of the most exciting being to study their behavior when they are placed in optical lattices generated by laser beams. This has already produced a number of fascinating results, but it has also proven to be an ideal test-ground for predictions from certain nonlinear lattice models. Because on the other hand, nonlinear science, the study of generic nonlinear phenomena, has in the last half century grown out to a research field in its own right, influencing almost all areas of science and physics. Nonlinear localization is one of these phenomena, where localized structures, such as solitons and discrete breathers, can appear even in translationally invariant systems. Another one is the (in)famous chaos, where deterministic systems can be so sensitive to perturbations that they in practice become completely unpredictable. Related to this is the study of different types of instabilities; what their behavior are and how they arise. In this thesis we compare classical and quantum mechanical nonlinear lattice models which can be applied to BECs in optical lattices, and also examine how classical nonlinear concepts, such as localization, chaos and instabilities, can be transfered to the quantum world.

We numerically study the dynamics and interactions of vortex rings in the nonlinear Schrödinger equation (NLSE). Single ring dynamics for both bright and dark vortex rings are explored including their traverse velocity, stability, and perturbations resulting in quadrupole oscillations. Multi-ring dynamics of dark vortex rings are investigated, including scattering and merging of two colliding rings, leapfrogging interactions of co-traveling rings, as well as co-moving steady-state multi-ring ensembles. Simulations of choreographed multi-ring setups are also performed, leading to intriguing interaction dynamics. Due to the inherent lack of a close form solution for vortex rings and the dimensionality where they live, efficient numerical methods to integrate the NLSE have to be developed in order to perform the extensive number of required simulations. To facilitate this, compact high-order numerical schemes for the spatial derivatives are developed which include a new semi-compact modulus-squared Dirichlet boundary condition. The schemes are combined with a fourth-order Runge-Kutta time-stepping scheme in order to keep the overall method fully explicit. To ensure efficient use of the schemes, a stability analysis is performed to find bounds on the largest usable time step-size as a function of the spatial step-size. The numerical methods are implemented into codes which are run on NVIDIA graphic processing unit (GPU) parallel architectures. The codes running on the GPU are shown to be many times faster than their serial counterparts. The codes are developed with future usability in mind, and therefore are written to interface with MATLAB utilizing custom GPU-enabled C codes with a MEX-compiler interface. Reproducibility of results is achieved by combining the codes into a code package called NLSEmagic which is freely distributed on a dedicated website.

Les résultats présentés dans cette thèse concernent l'étude, dans la limite semi-classique, de systèmes d'équations de Schrödinger non-linéaires couplées. Selon le potentiel considéré, le système peut, ou non, présenterun couplage linéaire, en plus de celui induit par le terme non-linéaire. Dans ce manuscrit, c'est la propagation d'états cohérents -états localisés dans l'espace des phases, et que l'on va faire vivre dans un niveau d'énergie donné - qui va nous intéresser.Dans le cadre linéaire, plusieurs situations ont été étudiées, certaines préservant l'adiabaticité,et d'autres la brisant, faisant apparaître des transitions entre les niveaux d'énergie.Le rôle de la non-linéarité et l'interaction de ses effets avec un éventuel couplage linéaire sur ces phénomènes est une questionimportante pour comprendre des systèmes qui entrent en jeu dans des problèmes très actuels en physique quantique.Dans un premier temps, le potentiel pris en compte aura des valeurs propres bien séparées par un trou spectral,et nous montrerons un théorème adiabatique pour une non-linéarité qui présente un exposant critique pour le paramètre semi-classique devant la non-linéarité. Un point de vue équivalent est de considérer des données petites de l'ordre d'une puissance positive du paramètre semi-classique.Il s'agit d'un résultat analogue à celui de Carles et Fermanian-Kammerer mais dans un cadre sur-critique L^2.Dans un deuxième temps, nous considèrerons, pour le cas unidimensionnel, un potentiel explicite de taille 2 X 2,qui présente un croisement évité :les deux valeurs propres sont séparées par un paramètre delta - paramètre adiabatique -qui va tendre vers zéro lorsque le paramètre semi-classique va tendre vers zéro. Nous montrerons alors que des transitions entre les modes ont lieu.Il s'agit ici d'une version non-linéaire des travaux d'Hagedorn et Joyeoù une telle transition est démontrée pour des systèmes linéaires. / This thesis is devoted to the study of coupled nonlinear Schrödinger equations in the semi-classical limit.Depending on the potential we consider, the system can present a linear coupling, in addition to the nonlinear one.We will focus on the propagation of coherent states that will be polarized along a given eigenvector of the potential.In the linear setting, several situations have been analyzed; some of them lead to adiabatic theorems whereas the others implytransitions between energy levels. When one adds a nonlinearity, understanding nonlinear effects onthe propagation and the competition between them and the linear coupling becomes a very interesting issue.We first consider a potential with eigenvalues that present a spectral gap and will prove an adiabatic theoremfor a critical nonlinearity in the semi-classical sense. This is a L^2-supercritical result,similar to the one proved by Carles and Fermanian-Kammerer for the one-dimensional case, which is L^2-subcritical.The second part of the thesis deals with an explicit 2 X 2 potential that presents an avoided crossing point :the minimal gap between its eigenvalues becomes smaller as the semiclassical parameter tends to zero. We will prove that this system exhibits transitions between the modes. This result is a nonlinear version of the study performed by Hagedorn and Joye in the linear case.

Neste trabalho, estudamos a existência de soluções não triviais para uma classe de equações de Schrödinger não lineares envolvendo um campo magnético com condição de Dirichlet ou condição de fronteira mista Dirichlet-Neumann. Nos dois primeiros capítulos, damos uma estimativa para o número de soluções não triviais para o problema de Dirichlet em termos da topologia do domínio. Nos dois capítulos restantes, consideramos o problema de fronteira mista e estimamos o número de soluções não triviais em termos da topologia da porção da fronteira onde é prescrita a condição de Neumann. Em ambos os casos, usamos a teoria de categoria de Ljusternik-Schnirelmann e a teoria de Morse / We study the existence of nontrivial solutions for a class of nonlinear Schrödinger equations involving a magnetic field with Dirichlet or mixed DirichletNeumann boundary condition. In the first two chapters we give an estimate for the number of nontrivial solutions for the Dirichlet boundary value problem in terms of topology of the domain. In the last two chapters we consider mixed DirichletNeumann boundary value problems and the estimation of the number of nontrivial solutions is given in terms of the topology of the part of the boundary where the Neumann condition is prescribed. In both cases, we use Lyusternik- Shnirelman category and the Morse theory

In this thesis, we study Schrödinger equations with an external magnetic field. In the first part, we are interested in an eigenvalue problem. We work in an open, bounded and simply connected domain in dimension two. We consider a magnetic potential singular at one point in the domain, and related to the magnetic field being a multiple of a Dirac delta. Those two objects are related to the Bohm-Aharonov effect, in which a charged particle is influenced by the presence of the magnetic potential although it remains in a region where the magnetic field is zero. We consider the Schrödinger magnetic operator appearing in the Schrödinger equation in presence of an external magnetic field. We want to study the spectrum of this operator, and more particularly how it varies when the singular point moves in the domain. We prove some results of continuity and differentiability of the eigenvalues when the singular point moves in the domain or approaches its boundary. Finally, in case of half-integer circulation of the magnetic potential, we study some asymptotic behaviour of the eigenvalues close to their critical points. In the second part, we study nonlinear Schrödinger equations in a cylindrically setting. We are interested in the semiclassical limit of the equation. We prove the existence of a semiclassical solution concentrating on a circle. Moreover, the radius of that circle is determined by the electric potential, but also by the magnetic potential. This result is totally new with respect to the ones before, in which the concentration is driven only by the electric potential. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

Ce travail est une contribution à l'étude des équations de Schrödinger non-linéaires (NLS) en dimension un d'espace. De telles équations interviennent notamment comme modèles dans plusieurs domaines de la physique mathématique, tels l'optique non-linéaire, la superfluidité, la supraconductivité et la condensation de Bose-Einstein.Cette thèse contient trois thèmes connexes inclus dans les chapitres 2, 3 et 4. Dans la première partie (chapitre 2), on s'intéresse à la construction des solutions en multi-solitons de l'équation de Gross-Pitaevskii (NLS défocalisante avec non-linéarité cubique), comme une superposition approximative des ondes progressives (solitons). Cette partie contient également une description détaillée des interactions entre les solitons. Ces résultats sont obtenus en exploitant l'intégrabilité de l'équation de Gross-Pitaevskii et son système de Marchenko associé.La deuxième partie (chapitre 4) clarifie les relations entre la formulation classique et la formulation dite hydrodynamique de l'équation de Gross-Pitaevskii. Cette dernière a un sens lorsque la solution ne s'annule jamais dans le domaine spatial. La dernière partie (chapitre 3) est consacrée à l'étude du problème de Cauchy d'une famille d'équations aux dérivées partielles quasi-linéaires qui généralise l'équation du flot par courbure binormal d'une courbe dans l'espace euclidien de dimension trois. Cette dernière est liée formellement à NLS par la transformation de Hasimoto. Dans notre généralisation, la vitesse d'un point de la courbe est toujours dirigée dans la direction du vecteur binormal, mais son amplitude peut dépendre de l'abscisse curviligne ainsi de la position dans l'espace. Notre approche pour prouver l'existence est le suivant: schéma semi-discret (discret en espace et continu en temps), obtention de bornes sur les problèmes discrets et argument par compacité. Un théorème de comparaison entraîne l'unicité. / This work is a contribution to the study of nonlinear Schrödinger equations (NLS) in the one-dimensional space. Such equations arise in many physical fields, including nonlinear optics and Bose-Einstein condensation. The thesis contains three connected themes included in chapters 2, 3 and 4. The first part (chapter 2) constructs multi-soliton solutions of the Gross-Pitaevskii (or defocussing NLS) equation, as an approximate superposition of traveling waves (solitons). This part contains also a detailed description of the interactions between solitons. These results are obtained by exploiting the integrability of the the Gross-Pitaevskii equation and its associated Marchenko system. The second part (chapter 4) clarifies the relations between the classical formulation and the so-called hydrodynamical formulation that only has a meaning when the solution does not vanish anywhere in the spatial domain The last part (chapter 3) of this thesis concerns existence and uniqueness results for a family of quasi-linear partial differential equations that generalize the equation of the binormal curvature flow for a curve in the three-dimensional space. The latter equation is in connection to the focussing cubic NLS by Hasimoto transformation. In our generalization, the velocity of a point on the curve is still directed along the binormal vector (so that in particular the length of the curve is preserved) but the magnitude of the speed is allowed to depend both on the curvilinear parameter and on the position in space. Existence is proven using spatial discretization together with some a priori bounds on the approximate solutions. Uniqueness follows from a comparison theorem.