Cette thèse concerne l'étude de certains problèmes elliptiques d'ordre 4 et, notamment, des propriétés qualitatives des solutions. Ces problèmes apparaissent dans de nombreux domaines, par exemple dans la théorie des plaques et dans la géométrie conforme, et, comparés à leurs homologues du deuxième ordre, ils présentent des difficultés intrinsèques, surtout liées à l'absence de principe de maximum. Premièrement on étudie la positivité des solutions dans le cas des conditions au bord de Steklov, qui sont intermédiaires entre les conditions de Dirichlet et de Navier. Elles apparaissent naturellement dans l'étude des minimiseurs de la fonctionnelle de Kirchhoff-Love, qui représente l'énergie d'une plaque encastrée soumise à l'action d'une force extérieure, en fonction d'un paramètre $\sigma$. On trouve des conditions suffisantes sur le domaine pour que les minimiseurs de la fonctionnelle soient positifs. De plus, pour ces domaines on étudie une version généralisée de la fonctionnelle. En utilisant des techniques variationnelles, on examine l'existence et la positivité des états fondamentaux, ainsi que leur comportement asymptotique pour les valeurs pertinentes de $\sigma$. Dans la deuxième partie de la thèse on établit des estimations uniformes a priori pour des problèmes semi linéaires du quatrième ordre dans $\mathbb R^4$, et donc avec des non linéarités exponentielles. On considère des conditions au bord soit de Dirichlet soit de Navier et on suppose que les non linéarités sont positives et sous-critiques. Nos arguments combinent des estimations uniformes près du bord et une analyse de blow-up. Enfin, en utilisant la théorie du degré, on obtient l'existence d'une solution. / This thesis concerns the study of fourth-order elliptic boundary value problems and, in particular, qualitative properties of solutions. Such problems arise in various fields, from plate theory to conformal geometry and, compared to their second-order counterparts, they present intrinsic difficulties, mainly due to the lack of the maximum principle. In the first part of the thesis, we study the positivity of solutions in case of Steklov boundary conditions, which are intermediate between Dirichlet and Navier boundary conditions. They naturally appear in the study of the minimizers of the Kirchhoff-Love functional, which represents the energy of a hinged thin and loaded plate in dependence of a parameter $\sigma$. We establish sufficient conditions on the domain to obtain the positivity of the minimizers of the functional. Then, for such domains, we study a generalized version of the functional. Using variational techniques, we investigate existence and positivity of the ground states, as well as their asymptotic behaviour for the relevant values of $\sigma$. In the second part of the thesis we establish uniform a-priori bounds for a class of fourth-order semi linear problems in $\mathbb R^4$, and thus with exponential non linearities. We considered both Dirichlet and Navier boundary conditions and we suppose our non linearities positive and subcritical. Our arguments combine uniform estimates near the boundary and a blow-up analysis. Finally, by means of the degree theory, we obtain the existence of a positive solution.

This thesis covers the basics in the stochastic homogenization of elliptic partial differential equations, from underlying theory up to numerical ap- proaches. In particular, we introduce and analyze a combination of the Fourier-Galerkin method in the spatial domain with a collocation method in the stochastic domain. The material coefficients are assumed to depend on a finite number of random variables. We present a comparison of the Monte Carlo method with the full tensor grid and sparse grid collocation method for two applications. The first one is the checkerboard problem with continuous random variables, the other considers the material coefficients to be described in terms of an autocorrelation function.

In a Hilbert space setting homogenization of evolutionary equations is discussed. In order to do so, a suitable topology on material laws is introduced and several properties of that topology are shown. With those properties homogenization theorems of a large class of linear evolutionary problems of classical mathematical physics can be obtained. The results are exemplified by the equations of piezo-electro-magnetism.

The topic of this thesis is the convex hull property for systems of partial differential equations, which is a natural generalisation of the maximum principle for scalar equations. The main result of this thesis is a theorem asserting the convex hull property for the solutions of a certain class of parabolic systems of nonlinear partial differential equations. It also investigates the coefficients of linear systems. The respective results are sharp which is demonstrated by counterexamples to the convex hull property for solutions of linear elliptic and parabolic systems. The general theme is that the coupling of the system is what breaks the convex hull property, not necessarily the non-linearity.

In this thesis, we will develop the fundamental properties of financial mathematics, with a focus on establishing meaningful connections between martingale theory, stochastic calculus, and measure-theoretic probability. We first consider a simple binomial model in discrete time, and assume the impossibility of earning a riskless profit, known as arbitrage. Under this no-arbitrage assumption alone, we stumble upon a strange new probability measure Q, according to which every risky asset is expected to grow as though it were a bond. As it turns out, this measure Q also gives the arbitrage-free pricing formula for every asset on our market. In considering a slightly more complicated model over a finite probability space, we see that Q once again makes its appearance. Finally, in the context of continuous time, we build a framework of stochastic calculus to model the trajectories of asset prices on a finite time interval. Under the absence of arbitrage once more, we see that Q makes its return as a Radon-Nikodym derivative of our initial probability measure. Finally, we use the properties of Q and a stochastic differential equation that models the dynamics of the assets of our market, known as the Ito formula, in order to derive the classic Black-Scholes Equation.

<p>U tezi je predstavljena implementacija metode teorija funkcionala gustine (DFT) bazirana na metodi za sklapanje naelektrisanja (CPM) koja koristi bazise gausijanskih funkcija. Metod je baziran na pretpostavci da se elektronska gustina naelektrisanja velikih sistema, može predstaviti kao suma doprinosa pojedinačnih atoma, takozvanih motiva gustine naelektrisanja, koji se dobijaju računanjem malog prototip sistema. Talasna funkcija,<br />kao i gustina naelektrisanja, se u na&scaron;oj implementaciji reprezentuju uz pomoć bazise gausijanskih funkcija, dok se motivi opisuju kori&scaron;ćenjem prostornih koordinata. Uz pomoć procedure za minimizaciju se iz motiva opisanih koordinatama, dobija gustina naelektrisanja predstavljena u bazisu Gausijana. Implementacija serijskog programa pokazuje značajno pobolj&scaron;anje u performansama u odnosu na prethodne implementacije bazirane na ravnim talasima. Ova implementacija re&scaron;ava sistem od približno 1000 atoma na jednom procesorskom jezgru za svega nekoliko sati. Paralelna implementacija uz pomoć naprednih metoda paralelizacije i distribucije podataka omogućava re&scaron;avanje sistema od vi&scaron;e desetina hiljada atoma. Najveći testirani sistem ima približno<br />20000 atoma i testiran je na 256 paralelnih procesa.</p> / <p>We present the implementation of the density functional theory (DFT) based charge patching method (CPM) using the basis of Gaussian functions. The method is based on the assumption that the electronic charge density of a large system is the sum of contributions of individual atoms, so called charge density motifs, that are obtained from calculations of small prototype systems.In our implementation wave functions and electronic charge density are represented using the basis of Gaussian functions, while charge density motifs are represented using a real space grid. A constrained minimization procedure is used to obtain Gaussian basis representation of charge density from real space representation of motifs. The code based on this&nbsp; implementation exhibits superior performance in comparison to previous implementation of the charge patching method using the basis of plane waves. It enables calculations of electronic structure of systems with around 1000 atoms on a single CPU core with computational time of just several hours. The parallel implementation enables calculations for the system with more than ten thousand atoms. The largest system tested has around 20000 atoms and was computed on 256 parallel processes.</p>

Indiana University-Purdue University Indianapolis (IUPUI) / In 1970, Serge Novikov made a statement which is now called, "The Novikov Conjecture" and is considered to be one of the major open problems in topology. This statement was motivated by the endeavor to understand manifolds of arbitrary dimensions by relating the surgery map with the homology of the fundamental group of the manifold, which becomes diffi cult for manifolds of dimension greater than two. The Novikov Conjecture is interesting because it comes up in problems in many different branches of mathematics like algebra, analysis, K-theory, differential geometry, operator algebras and representation theory. Yu later proved the Novikov Conjecture holds for all closed manifolds with discrete fundamental groups that are coarsely embeddable into a Hilbert space. The class of groups that are uniformly embeddable into Hilbert Spaces includes groups of Property A which were introduced by Yu. In fact, Property A is generally a property of metric spaces and is stable under quasi-isometry. In this thesis, a new version of Yu's Property A in the case of locally compact groups is introduced. This new notion of Property A coincides with Yu's Property A in the case of discrete groups, but is different in the case of general locally compact groups. In particular, Gromov's locally compact hyperbolic groups is of Property A.

This work deals with diffrential equations, with the possibility     of using them for analysis of the line and the possibility     of accelerating the computations in GPU using nVidia CUDA.

Mathematical models in physiology are useful to formulate and verify hypotheses, to make predictions, to estimate hidden parameters and in education. This thesis deals with modelling in physiology using the ​Modelica language. New methods for model implementation and simulator production were developed. Modelica is an ​open standard equation-based object-oriented language for modelling complex systems. It is highly convenient in physiology modelling due to its ability to describe extensive models in a lucid hierarchical way. The models are described by algebraic, ordinary differential and discrete equations. Partial differential equations are not supported by the Modelica standard yet. The thesis focuses on two main topics: 1) modelling of systems described by partial differential equations in Modelica 2) production of web-based e-learning simulators driven by models implemented in Modelica. A Modelica language extension called PDEModelica1 for 1-dimensional partial differential equations was designed (based on a previous extension). The OpenModelica modelling tool was extended to support PDEModelica1 using the method of lines. A model of countercurrent heat exchange between the artery and vein in a leg of a bird standing in water was implemented using PDEModelica1 to prove its usability. The...

This work presents implementation of a hybridized discontinuous Galerkin (DG) method for robust simulation of the hypersonic aerothermoelastic multiphysics system. Simulation of hypersonic vehicles requires accurate resolution of complex multiphysics interactions including the effects of high-speed turbulent flow, extreme heating, and vehicle deformation due to considerable pressure loads and thermal stresses. However, the state-of-the-art procedures for hypersonic aerothermoelasticity are comprised of low-fidelity approaches and partitioned coupling schemes. These approaches preclude robust design and analysis of hypersonic vehicles for a number of reasons. First, low-fidelity approaches limit their application to simple geometries and lack the ability to capture small scale flow features (e.g. turbulence, shocks, and boundary layers) which greatly degrades modeling robustness and solution accuracy. Second, partitioned coupling approaches can introduce considerable temporal and spatial inaccuracies which are not trivially remedied. In light of these barriers, we propose development of a monolithically-coupled hybridized DG approach to enable robust design and analysis of hypersonic vehicles with arbitrary geometries. Monolithic coupling methods implement a coupled multiphysics system as a single, or monolithic, equation system to be resolved by a single simulation approach. Further, monolithic approaches are free from the physical inaccuracies and instabilities imposed by partitioned approaches and enable time-accurate evolution of the coupled physics system. In this work, a DG method is considered due to its ability to accurately resolve second-order partial differential equations (PDEs) of all classes. We note that the hypersonic aerothermoelastic system is composed of PDEs of all three classes. Hybridized DG methods are specifically considered due to their exceptional computational efficiency compared to traditional DG methods. It is expected that our monolithic hybridized DG implementation of the hypersonic aerothermoelastic system will 1) provide the physical accuracy necessary to capture complex physical features, 2) be free from any spatial and temporal inaccuracies or instabilities inherent to partitioned coupling procedures, 3) represent a transition to high-fidelity simulation methods for hypersonic aerothermoelasticity, and 4) enable efficient analysis of hypersonic aerothermoelastic effects on arbitrary geometries.