I construct anti-de Sitter boson stars in Einstein-Gauss-Bonnet gravity coupled to a (D-1)/(2)-tuplet of complex massless scalar field both perturbativelyand numerically in D=5,7,9,11 dimensions. Due to the choice of scalar fields, these solutions possess just a single helical Killing symmetry. For each choice of the Gauss-Bonnet parameter &alpha;&#8800;&alpha;_cr, the central energy density at the center of the boson star, q_0 completely characterizes the one parameter family of solutions. These solutions obey the first law of thermodynamics, in the case of the numerics, to within 1 part in 10^6. I describe the dependence of the boson star mass, angular momentum and angular velocity on &alpha; and on the dimensionality. For &alpha;<&alpha;_cr and D>5, these quantities exhibit damped oscillations about finite central values and the central energy density tends to infinity. The Kretschmann invariant at the center of the boson star diverges in the limit of diverging central energy. This contrasts the D=5 case, where the Kretschmann invariant diverges at a finite value of the central energy density. Solutions where &alpha;<&alpha;_cr, correspond to negative mass boson stars, and the for all dimensions the boson star mass and angular momentum decrease exponentially as the central energy density tends toward infinity with the Kretschmann invariant diverging only when in the limit the central energy density diverges. I also briefly discuss the difficulties of numerically obtaining single Killing vector hairy black hole solutions and present the explicit boundary conditions for both Einstein gravity and Einstein-Gauss-Bonnet gravity.

<p>The central object of study of this thesis is inverse mean curvature vector flow of two-dimensional surfaces in four-dimensional spacetimes. Being a system of forward-backward parabolic PDEs, inverse mean curvature vector flow equation lacks a general existence theory. Our main contribution is proving that there exist infinitely many spacetimes, not necessarily spherically symmetric or static, that admit smooth global solutions to inverse mean curvature vector flow. Prior to our work, such solutions were only known in spherically symmetric and static spacetimes. The technique used in this thesis might be important to prove the Spacetime Penrose Conjecture, which remains open today. </p><p>Given a spacetime $(N^{4}, \gbar)$ and a spacelike hypersurface $M$. For any closed surface $\Sigma$ embedded in $M$ satisfying some natural conditions, one can ``steer'' the spacetime metric $\gbar$ such that the mean curvature vector field of $\Sigma$ becomes tangential to $M$ while keeping the induced metric on $M$. This can be used to construct more examples of smooth solutions to inverse mean curvature vector flow from smooth solutions to inverse mean curvature flow in a spacelike hypersurface.</p> / Dissertation

In this thesis, Newman-Penrose techniques are used to obtain some new exact solutions to Einstein's field equations of general relativity and to assist in the physical interpretation of some exact radiative space-times. Attention is restricted to algebraically special space-times with a twist-free, repeated principal null congruence. In particular, the Robinson-Trautman type N solutions, which describe expanding gravitational waves, are investigated for all possible values of the cosmological constant A and the Gaussian curvature parameter E. The wave surfaces are always (hemi-)spherical, with successive surfaces displaced along time-like, space-like or null lines, depending on E. Explicit sandwich waves of this class are studied in Minkowski, de Sitter or anti-de Sitter backgrounds and a particular family of such solutions, which can be used to represent snapping or decaying cosmic strings, is considered in detail. The singularity and global structure of the solutions is also presented. In the remaining part of the thesis, the complete family of space-times with a non-expanding, shear-free, twist-free, geodesic principal null congruence (Kundt waves), that are of algebraic type III and for which the cosmological constant (Ac) is non-zero, is presented. The possible presence of an aligned pure radiation field is also assumed. These space-times generalise the known vacuum solutions of type N with arbitrary Ac and type III with Ac = O. It is shown that there are two, one and three distinct classes of solutions when Ac is respectively zero, positive and negative and, in these cases, the wave surfaces are plane, spherical or hyperboloidal in Minkowski, de Sitter or anti-de Sitter backgrounds respectively. The singularities which occur in these space-times are interpreted in terms of envelopes of these wave surfaces. Again, by considering functions of the retarded time which "cross-over" between canonical types, sandwich waves are also studied. The limiting cases of these, giving rise to shock or impulsive waves, are also considered.

Στην παρούσα εργασία μελετώνται οι ταλαντώσεις των αστέρων νετρονίων με ιδιαίτερη έμφαση στις ακτινικές ταλαντώσεις τους. Σκοπός αυτής της μελέτης είναι ο υπολογισμός των συχνοτήτων των ακτινικών ταλαντώσεων των αστέρων νετρονίων. Στο πρώτο, κεφάλαιο κάνουμε μία μικρή εισαγωγή για τους αστέρες νετρονίων και τους ταχέως περιστρεφόμενους αστέρες νετρονίων (pulsars) καθώς και για τον ρόλο, που διαδραματίζουν αυτοί και τα διπλά συστήματα που σχηματίζουν, στην σύγχρονη Αστροφυσική. Ακόμα αναφερόμαστε στην εσωτερική δομή των αστέρων νετρονίων και σε κάποιες από τις καταστατικές εξισώσεις, που μπορεί να περιγράφουν την ύλη στο εσωτερικό του, δίνοντας έμφαση στην πολυτροπική καταστατική εξίσωση την οποία και υιοθετούμε στην παρούσα εργασία. Στο δεύτερο κεφάλαιο, παραθέτουμε τις εξισώσεις Oppenheimer–Volkoff(OV) που περιγράφουν την ισορροπία ενός αδιατάρακτου αστέρα νετρονίων. Στη συνέ- χεια, θεωρώντας τις ακτινικές ταλαντώσεις 1) ως απειροστού πλάτους αδιαβατικές ταλαντώσεις που διατηρούν τον βαρυονικό αριθμό και 2) ως αποτέλεσμα της αργής περιστροφής του αστέρα, καταλήγουμε σε μία δεύτερης τάξης διαφορική εξίσωση που διέπει τις ακτινικές ταλαντώσεις των αστέρων νετρονίων. Η εξί- σωση αυτή γράφεται στη μορφή Sturm– Liouville. Επιπροσθέτως, συνεχίζουμε παραθέτοντας τον διορθωτικό όρο, λόγω περιστροφής, για την τιμή της συχνότη- τας και τις εξισώσεις που διέπουν τις μη ακτινικές ταλαντώσεις. Τέλος κλείνουμε το κεφάλαιο αυτό με μία ανάλυση των διαφόρων τρόπων ταλάντωσης. Στο τρίτο κεφάλαιο, αρχικά επιλύουμε, με τη χρήση ενός πρωτότυπου επα- ναληπτικού αλγορίθμου, το σύστημα διαφορικών εξισώσεων OV για την εύρεση των φυσικών παραμέτρων του αστέρα. Στη συνέχεια, αφού αρχικά αναλύσουμε τις βασικότερες μεθόδους επίλυσης της διαφορικής εξίσωσης των ακτινικών ταλα- ντώσεων, που εμφανίζονται στην βιβλιογραφία, μετατρέπουμε τη μορφή Sturm– Liouville σε ένα σύστημα δύο διαφορικών εξισώσεων πρώτης τάξης, το οποίο επιλύουμε με την βοήθεια της μεθόδου σκόπευσης (shooting method). Στη βιβλιογραφία, υπάρχουν δύο διαφορετικές τάσεις αντιμετώπισης της πο- λυτροπικής καταστατικής εξίσωσης, ανάλογα με το αν στην θέση της πυκνότητας εισέρχεται η πυκνότητα μάζας ηρεμίας ή η πυκνότητα της ολικής μάζας–ενέργειας. Ακόμα, δύο είναι και οι διαφορετικοί τρόποι αντιμετώπισης του αδιαβατικού δεί- κτη, ο οποίος εισέρχεται στην εξίσωση που περιγράφει τις ακτινικές ταλαντώσεις, ανάλογα με το αν είναι σταθερός ή μεταβάλλεται. Από τις τέσσερις αυτές βασικές υποθέσεις για την πολυτροπική καταστατική εξίσωση και τον αδιαβατικό δείκτη, προκύπτουν τέσσερα διαφορετικά πρωτότυπα μοντέλα για τις ακτινικές ταλαντώ- σεις, τα οποία και επιλύουμε. Στο τελευταίο κεφάλαιο, υπολογίζουμε και παρουσιάζουμε τις τρεις πρώτες συχνότητες των τεσσάρων πρωτότυπων μοντέλων για τις ακτινικές ταλαντώσεις των αστέρων νετρονίων για τρεις διαφορετικές τιμές του πολυτροπικού δείκτη και αναλύουμε τις αριθμητικές μεθόδους, τις οποίες χρησιμοποιούμε, καθώς και τις αντίστοιχες υπορουτίνες της βιβλιοθήκης SLATEC. Εν κατακλείδι, τα αποτελέσματα αυτής της εργασίας είναι η ανάπτυξη ενός πρωτότυπου επαναληπτικού αλγορίθμου για την εύρεση της ακτίνας του αστέρα με μεγάλη ακρίβεια και η παρουσίαση αποτελεσμάτων για τέσσερα πρωτότυπα μοντέλα που περιγράφουν τις ακτινικές ταλαντώσεις των αστέρων νετρονίων. / In the present Thesis we study the oscillations of neutron stars emphasizing on the radial oscillations. The Thesis is organized in four chapters. In the first chapter, we introduce the theoretical background of neutron stars and pulsars. We then discuss the importance of the role that the binary neutron stars play in modern Astrophysics. Next, we refer to the structure of these stars and introduce some of the equations of state (EOS) which try to describe the matter occupying the inner layers of neutron stars, emphasizing on the polytropic EOS which is adopted here. In the second chapter we, first introduce the Oppenheimer–Volkoff (OV) system of differential equations, describing the hydrostatic equilibrium of a non rotating, non pulsating neutron star, and considering the radial oscillations 1) as infinitesimal, baryon-number conserving, adiabatic oscillations 2) as the result of the slow rotation of the neutron star, we derive the second order differential equation governing the radial oscillations of a neutron star. We then rewrite this equation in the Sturm–Liouville form. The expression of the change of frequency of the radial oscillations due to slow rotation and the equations of state is obtained. Finally, we conclude this chapter with a mode analysis of oscillations of neutron stars in general. In the third chapter, we first solve the OV system of differential equations, implementing an original iterative algorithm, and thus calculate the physical parameters of the star. Next, some of the methods used for solving the equations describing the radial oscillations are discussed. Finally, we transform the Sturme–Liouville form to a set of two first order differential equations, which are computed by implementation of the shooting method. In the bibliography, the polytropic EOS is considered in two different ways, depending on which density (rest mass or total mass–energy) is involved in the polytropic EOS. In a similar manner, we have two different ways for considering the adiabatic exponent which enters the equation describing the radial oscillations (constant or variable). Considering these four different assumptions for the polytropic EOS and the adiabatic exponent, we construct four different models of pulsating neutron stars. In the final chapter, we compute and present the first three frequencies of each basic model concerning radial oscillations of neutron stars for three values of the polyropic index. We discuss the numerical methods implemented here and the involved subroutines, which can be found in the SLATEC Library. The main issues of the present Thesis are the development of an iterative algorithm for accurately computing the radius of the star and the computation of the frequencies for the four basic models describing th radial oscillations of neutron stars.

The fuzzball model of a black hole is an attempt to resolve the many paradoxes and puzzles of black hole physics that have revealed themselves over the last century. These badly behaved solutions of general relativity have given physicists one of the few laboratories to test candidate quantum theories of gravity. Though little is known about exactly what lies beyond the event horizon, and what the ultimate fate of matter that falls in to a black hole is, we know a few intriguing and elegant semi-classical results that have kept physicists occupied. Among these are the known black hole entropy and the Hawking radiation process.

ix, 111 p. / This dissertation contains the results obtained from a study of two subjects in mathematical general relativity. The first part of this dissertation is about the existence of Killing symmetries in spacetimes containing a compact Cauchy horizon. We prove the existence of a nontrivial Killing symmetry in a large class of analytic cosmological spacetimes with a compact Cauchy horizon for any spacetime dimension. In doing so, we also remove the restrictive analyticity condition and obtain a generalization to the smooth case. The second part of the dissertation presents our results on the global stability problem for a class of cosmological models. We investigate the power law inflating cosmological models in the presence of electromagnetic fields. A stability result for such cosmological spacetimes is proved. This dissertation includes unpublished co-authored material. / Committee in charge: James Brau, Chair; James Isenberg, Advisor; Paul Csonka, Member; John Toner, Member; Peng Lu, Outside Member

O impacto que estruturas algébricas podem exercer em teorias físicas e bem ilustrado pela Mecânica Quântica, onde os números complexos são inquestionavelmente a escolha mais adequada para desenvolver a teoria. A Relatividade Geral pseudo-complexa avalia a possibilidade da interação gravitacional assumir sua descrição mais natural quando construída tendo como base os números pseudo-complexos, que consistem em uma das três possibilidades de números complexos abelianos com uma unica unidade imaginária. Esse conjunto numérico e dotado de elementos não nulos cujo produto e zero, tais números recebem o nome de zeros generalizados ou divisores de zero. A presença de zeros generalizados permite a introdução de um princípio variacional modificado do qual um termo adicional, ausente na Relatividade Geral, emerge nas equações de campo. Esse termo adicional e interpretado como uma energia escura, cuja origem física está relacionada com flutuações no vácuo. A inclusão desse efeito e legítima pois flutuações no vácuo a priori devem gravitar como qualquer outra forma de energia. Das equações de campo podemos resumir a principal ideia conceitual da teoria, na Relatividade Geral pseudo-complexa massa não apenas curva o espaçotempo como também e capaz de alterar a estrutura do espaço-tempo ao redor da massa. As diferenças com relação a Relatividade Geral se manifestam em situações físicas extremas, no regime de campos gravitacionais intensos. Como aplicação analisamos sob o ponto de vista teórico um objeto compacto exótico composto por matéria escura fermiônica. / The impact that algebraic structures can exert on physical theories is well illustrated by Quantum Mechanics, where complex numbers are unquestionably the most appropriate choice to develop the theory. Pseudo-complex General Relativity evaluates the possibility that the gravitational interaction acquires its most natural description when constructed upon pseudo-complex numbers, which consist of one of the three possibilities of abelian complex numbers with a single imaginary unit. This numerical set is endowed with nonzero elements whose product is zero, such numbers are called generalized zeros or divisors of zero. The presence of generalized zeros allows the introduction of a modi ed variational principle from which an additional term, absent in General Relativity, emerges in the eld equations. This additional term is interpreted as a dark energy, whose physical origin is related to vacuum uctuations. The inclusion of this e ect is legitimate because a priori vacuum uctuations must gravitate as any other form of energy. From the eld equations we can summarize the main conceptual idea of the theory, in pseudo-complex General Relativity mass not only curves spacetime but also is able to change the structure of the spacetime around the mass. The di erences with respect to General Relativity are manifested in extreme physical situations in the regime of intense gravitational elds. As an application we analyze from the theoretical point of view an exotic compact object composed of fermionic dark matter.

Made available in DSpace on 2014-12-17T15:14:52Z (GMT). No. of bitstreams: 1 CrislaneSS_DISSERT.pdf: 1091298 bytes, checksum: 831e6bef52e8fad49a4683ec16886d4d (MD5) Previous issue date: 2011-04-14 / In the Einstein s theory of General Relativity the field equations relate the geometry of space-time with the content of matter and energy, sources of the gravitational field. This content is described by a second order tensor, known as energy-momentum tensor. On the other hand, the energy-momentum tensors that have physical meaning are not specified by this theory. In the 700s, Hawking and Ellis set a couple of conditions, considered feasible from a physical point of view, in order to limit the arbitrariness of these tensors. These conditions, which became known as Hawking-Ellis energy conditions, play important roles in the gravitation scenario. They are widely used as powerful tools for analysis; from the demonstration of important theorems concerning to the behavior of gravitational fields and geometries associated, the gravity quantum behavior, to the analysis of cosmological models. In this dissertation we present a rigorous deduction of the several energy conditions currently in vogue in the scientific literature, such as: the Null Energy Condition (NEC), Weak Energy Condition (WEC), the Strong Energy Condition (SEC), the Dominant Energy Condition (DEC) and Null Dominant Energy Condition (NDEC). Bearing in mind the most trivial applications in Cosmology and Gravitation, the deductions were initially made for an energy-momentum tensor of a generalized perfect fluid and then extended to scalar fields with minimal and non-minimal coupling to the gravitational field. We also present a study about the possible violations of some of these energy conditions. Aiming the study of the single nature of some exact solutions of Einstein s General Relativity, in 1955 the Indian physicist Raychaudhuri derived an equation that is today considered fundamental to the study of the gravitational attraction of matter, which became known as the Raychaudhuri equation. This famous equation is fundamental for to understanding of gravitational attraction in Astrophysics and Cosmology and for the comprehension of the singularity theorems, such as, the Hawking and Penrose theorem about the singularity of the gravitational collapse. In this dissertation we derive the Raychaudhuri equation, the Frobenius theorem and the Focusing theorem for congruences time-like and null congruences of a pseudo-riemannian manifold. We discuss the geometric and physical meaning of this equation, its connections with the energy conditions, and some of its several aplications. / Na teoria da Relatividade Geral de Einstein as equa??es de campo relacionam a geometria do espa?o-tempo com o conte?do de mat?ria e de energia, fontes do campo gravitacional. Esse conte?do ? descrito por um tensor de segunda ordem, conhecido como tensor energia-momento. Por outro lado, os tensores energia-momento que possuem significado f?sico n?o s?o especificados por essa teoria. Na d?cada de 70, Hawking e Ellis estabeleceram algumas condi??es, consideradas plaus?veis do ponto de vista f?sico, com o intuito de limitar as arbitrariedades desses tensores. Essas condi??es ficaram conhecidas como condi??es de energia de Hawking-Ellis, desempenham pap?is importantes no cen?rio da gravita??o. Elas s?o largamente usadas como poderosas ferramentas de an?lise, desde a demonstra??o de importantes teoremas relativos ao comportamento de campos gravitacionais e geometrias associadas, comportamento qu?ntico da gravita??o, at? as an?lises de modelos cosmol?gicos. Nesta disserta??o apresentamos uma dedu??o rigorosa das v?rias condi??es de energia em voga atualmente na literatura cient?fica, tais como: Condi??o de Energia Nula (NEC), Condi??o de Energia Fraca (WEC), Condi??o de Energia Forte (SEC), Condi??o de Energia Dominante (DEC) e Condi??o de Energia Dominante Nula (NDEC). Tendo em mente as aplica??es mais corriqueiras em Gravita??o e Cosmologia, as dedu??es foram feitas inicialmente para um tensor energia-momento de um fluido perfeito generalizado e depois estendidas aos campos escalares com acoplamento m?nimo e n?o-m?nimo ao campo gravitacional. Apresentamos tamb?m um estudo sobre as poss?veis viola??es de algumas dessas condi??es de energia, visando o estudo da natureza singular de algumas solu??es exatas da Relatividade Geral de Einstein, em 1955, o f?sico indiano Raychaudhuri derivou uma equa??o que hoje ? considerada fundamental para o estudo da atra??o gravitacional da mat?ria, a qual ficou conhecida como equa??o de Raychaudhuri. Essa c?lebre equa??o ? considerada o alicerce da compreens?o da atra??o gravitacional em Astrof?sica e Cosmologia e dos teoremas de Singularidades, como por exemplo, o teorema de Hawking e Penrose sobre a singularidade do colapso gravitacional. Nesta disserta??o derivamos a equa??o de Raychaudhuri, o teorema de Frobenius e o teorema da Focaliza??o para congru?ncias tipo-tempo e tipo-nulas de uma variedade pseudo-riemanniana. Discutimos o significado geom?trico e f?sico dessa equa??o, sua conex?o com as condi??es de energia, e algumas de suas in?meras aplica??es.

O impacto que estruturas algébricas podem exercer em teorias físicas e bem ilustrado pela Mecânica Quântica, onde os números complexos são inquestionavelmente a escolha mais adequada para desenvolver a teoria. A Relatividade Geral pseudo-complexa avalia a possibilidade da interação gravitacional assumir sua descrição mais natural quando construída tendo como base os números pseudo-complexos, que consistem em uma das três possibilidades de números complexos abelianos com uma unica unidade imaginária. Esse conjunto numérico e dotado de elementos não nulos cujo produto e zero, tais números recebem o nome de zeros generalizados ou divisores de zero. A presença de zeros generalizados permite a introdução de um princípio variacional modificado do qual um termo adicional, ausente na Relatividade Geral, emerge nas equações de campo. Esse termo adicional e interpretado como uma energia escura, cuja origem física está relacionada com flutuações no vácuo. A inclusão desse efeito e legítima pois flutuações no vácuo a priori devem gravitar como qualquer outra forma de energia. Das equações de campo podemos resumir a principal ideia conceitual da teoria, na Relatividade Geral pseudo-complexa massa não apenas curva o espaçotempo como também e capaz de alterar a estrutura do espaço-tempo ao redor da massa. As diferenças com relação a Relatividade Geral se manifestam em situações físicas extremas, no regime de campos gravitacionais intensos. Como aplicação analisamos sob o ponto de vista teórico um objeto compacto exótico composto por matéria escura fermiônica. / The impact that algebraic structures can exert on physical theories is well illustrated by Quantum Mechanics, where complex numbers are unquestionably the most appropriate choice to develop the theory. Pseudo-complex General Relativity evaluates the possibility that the gravitational interaction acquires its most natural description when constructed upon pseudo-complex numbers, which consist of one of the three possibilities of abelian complex numbers with a single imaginary unit. This numerical set is endowed with nonzero elements whose product is zero, such numbers are called generalized zeros or divisors of zero. The presence of generalized zeros allows the introduction of a modi ed variational principle from which an additional term, absent in General Relativity, emerges in the eld equations. This additional term is interpreted as a dark energy, whose physical origin is related to vacuum uctuations. The inclusion of this e ect is legitimate because a priori vacuum uctuations must gravitate as any other form of energy. From the eld equations we can summarize the main conceptual idea of the theory, in pseudo-complex General Relativity mass not only curves spacetime but also is able to change the structure of the spacetime around the mass. The di erences with respect to General Relativity are manifested in extreme physical situations in the regime of intense gravitational elds. As an application we analyze from the theoretical point of view an exotic compact object composed of fermionic dark matter.

Le Centre Galactique abrite en son cœur un objet compact de plusieurs millions de masses solaires. L'hypothèse faite à l'heure actuelle est que cet objet serait un trou noir supermassif décrit par la relativité générale. L'instrument de seconde génération du Very Large Telescope Interferometer, GRAVITY, va permettre d'apporter des réponses quant à la réelle nature de cet objet. Grâce à sa précision astrométrique de 10 microsecondes d'angle, il va pouvoir sonder l'espace-temps en champ fort via l'observation des étoiles et du gaz situés à proximité de l'objet central.Au cours de ma thèse j'ai mis au point un modèle permettant de simuler les observations d'orbites d'étoiles de GRAVITY, l'objectif étant d'extraire à l'aide de celui-ci les paramètres fondamentaux du candidat trou noir central ainsi que les effets relativistes. Pour cela, j'ai utilisé le code de tracé de rayons GYOTO développé à l'Observatoire de Paris. Ce code permet de calculer des trajectoires d'étoiles et de photons obtenues en présence d'un objet compact. Il est alors possible de simuler les positions apparentes d'étoiles en orbite autour du Centre Galactique en calculant leur image relativiste.J'ai d'abord validé le calcul des trajectoires des photons effectué dans GYOTO. Grâce à des tests effectués en déflexion faible et forte, j'ai pu démontrer que GYOTO était hautement satisfaisant pour simuler les observations GRAVITY. En effet, j'ai montré que l'erreur sur le calcul des géodésiques de genre lumière était inférieure à environ 10^-2 microseconde d'angle, et cela même pour de grandes distances d'intégration.Je me suis ensuite intéressée à l'étude d'une étoile appelée S2 qui a contribué à fortement contraindre la masse de l'objet central. Sa proximité au Centre Galactique fait d'elle une cible idéale pour sonder l'espace-temps en champ fort. En particulier, j'ai estimé quels étaient les temps minimaux d'observation nécessaires pour détecter des effets relativistes à l'aide de mesures astrométriques et spectroscopiques obtenues sur l'étoile S2. Pour cela, j'ai mis en place plusieurs modèles d'orbites prenant en compte chacun un certain nombre d'effets relativistes. Le modèle le plus précis est obtenu en relativité générale complète avec le code GYOTO. Néanmoins, puisque l'étoile S2 est suffisamment éloignée de l'objet compact, ce modèle néglige certains effets de lentilles gravitationnelles telles que les images secondaires et l'amplification des images primaires. Par ailleurs, je me suis également intéressée à la contraindre du moment cinétique du candidat trou noir central avec cette étoile. En particulier, j'ai déterminé, grâce au modèle le plus précis mis en place ici, qu'il était possible de contraindre la norme et la direction du moment cinétique avec une incertitude d'environ 0,1 et 20 degrés, respectivement, et cela en considérant des observations obtenues sur trois périodes de S2 et des précisions de 10 microsecondes d'angle et 10 km/s.En vue de la possible détection d'étoiles plus proches du Centre Galactique par GRAVITY, j'ai développé un modèle prenant en compte les effets de lentilles négligés dans le modèle précédent. Néanmoins, afin de minimiser le temps de calcul demandé par celui-ci, j'ai déterminé une zone de l'espace dans laquelle il était tout de même possible d'utiliser ce dernier.Enfin, j'ai étudié l'influence de corps du Système Solaire sur les mesures astrométriques de GRAVITY, c'est-à-dire sur la séparation angulaire entre deux sources du Centre Galactique. Cette étude a montré que ces mesures différentielles n'étaient déviées que de quelques microsecondes d'angle par la perturbation gravitationnelle engendrée par le Soleil. Cependant, celles-ci sont modifiées de plusieurs centaines de microsecondes d'angle par l'effet d'aberration induit par le mouvement de la Terre par rapport aux sources du Centre Galactique. Il sera donc nécessaire de prendre en compte cet effet lors de l'interprétation des données obtenues par GRAVITY. / Decades of studies have demonstrated the presence of a compact object of several million solar masses at the center of the Galaxy. Nowadays, the assumption is that this compact object is probably a supermassive black hole described by general relativity. The second generation instrument at the Very Large Telescope Interferometer, GRAVITY, is expected to better constrain the nature of this central object. By using its astrometric accuracy of about 10 microarcseconds, it will probe spacetime in strong gravitational fields by observing stars and gas located near the compact object.During my PhD I have developed a stellar-orbit model in order to interpret the future GRAVITY observations. By using this model it will be possible to extract the central black hole candidate parameters and relativistic effects. To implement the model, I used the ray-tracing code GYOTO developed at Observatoire de Paris. This code allows computing star and photon trajectories obtained in the vicinity of a compact object. It is thus possible to simulate apparent positions of stars orbiting the Galactic Center by computing relativistic images.My work started by validating the photon trajectories computed in GYOTO. By doing tests in both weak- and strong-deflection limits, I have shown that the GYOTO code is highly qualified to simulate GRAVITY observations. Indeed, the error made on the photon trajectories is inferior to 10^-2 microarcsecond, even when integrating over large distances.Then, I was interested in studying a star called S2 that contributed to importantly constrain the mass of the central object. This star is the second closest star to the Galactic Center and has an orbital period of about 16 years. Nowadays, we do not know whether closer-in stars will be discovered by GRAVITY. It is thus important to extract as much information as possible from this star. In particular, I have estimated the minimal observation times needed to detect relativistic effects by using astrometric and spectroscopic measurements of S2. To do so, I have developed different stellar-orbit models taking into account a certain number of relativistic effects. The more accurate model is obtained by using the ray-tracing code GYOTO and considering all relativistic effects. However, as the S2 star is sufficiently far from the compact object, this model neglects certain gravitational lensing effects such as the secondary images and the primary images amplification. Besides, I was also interested in the possibility of constraining the angular momentum of the central black hole candidate with the S2 star. In particular, I have shown that with a model which does not use ray-tracing, the norm and the direction of the angular momentum can be constrained with an uncertainty of about 0.1 and 20 degrees, respectively, by using observations obtained during three periods of S2 and with accuracies reaching 10 microarseconds and 10 km/s.Since closer-in stars could be detected by GRAVITY, I have developed a more accurate stellar-orbit model taking into account the lensing effects neglected in the previous model. However, in order to minimize the computing time required by this model, I determined a volume in which it is possible to neglect both the secondary images and the primary images amplification.Finally, I have studied the impact of different components of the Solar System on astrometric positions measured by GRAVITY. This study has shown that those measurements are deviated by an amount of a few microarcseconds by the gravitational perturbation generated by the Sun. However, those apparent positions are shifted by several hundred microarcseconds by the aberration effect due to the movement of the Earth with respect to the Galactic Center. It is thus necessary to take into account this effect in future interpretations of GRAVITY observations.