Spelling suggestions: "subject:"singular perturbation"" "subject:"singular erturbation""
61 |
Portfolio liquidation under small market impactKivman, Evgueni 24 July 2024 (has links)
Wir beweisen neuartige Konvergenz- und Approximationsresultate für die Lösungen einer Klasse von Modellen optimaler Portfolioliquidierung mit sofortigem Preiseinfluss und stochastischer Resilienz. Jedes betrachtete Liquidierungsproblem erlaubt nur absolut stetige Handelsstrategien, und die optimale Strategie ist durch ein voll gekoppeltes mehrdimensionales quadratisches BSDE-System mit einer singulären Endbedingung gegeben. Innerhalb unseres Modellierungsrahmens beweisen wir, dass wenn der Parameter des sofortigen Preiseinflusses gegen null konvergiert, der absolut stetige optimale Portfolioprozess gegen einen stochastischen Prozess konvergiert, der durch die eindeutige Lösung einer regulären eindimensionalen quadratischen BSDE gegeben ist. Es stellt sich heraus, dass dieser Grenzwert die Lösung eines Modells optimaler Portfolioliquidierung ohne sofortigen Preiseinfluss, aber mit allgemeiner Semimartingalkontrolle mit Sprüngen ist. Unser Resultat liefert einen vereinheitlichten Rahmen, in den die zwei am häufigsten gebrauchten Modellierungsrahmen der Literatur über optimale Liquidierung eingebettet werden können, und liefert eine Grundlage für die Nutzung von Semimartingalen als Liquidierungsstrategien und für die Nutzung von Portfolioprozessen von unbeschränkter Variation. Unsere Resultate beruhen auf neuartigen Konvergenzresultaten für BSDEs mit singulären Endbedingungen und auf einem neuartigen Resultat der Darstellung von Lösungen von BSDEs durch gleichmäßig stetige Funktionen von Vorwärtsprozessen. Wir beweisen außerdem, dass die optimale Lösung in der deterministischen Version des ursprünglichen Liquidierungsmodells gleichmäßig approximiert werden kann, indem das Taylor-Polynom erster Ordnung verwendet wird, das um den singulären Punkt, an dem der sofortige Preiseinfluss verschwindet, entwickelt wird. Diese Approximation ist explizit darstellbar, was im Allgemeinen nicht für die optimale Lösung gilt. / We establish novel convergence and approximation results for the solutions to a class of optimal portfolio liquidation problems with instantaneous price impact and stochastic resilience. Each considered liquidation problem only allows for absolutely continuous trading strategies, and the optimal strategy is given in terms of a fully coupled multi-dimensional quadratic BSDE system with a singular terminal condition. Within our modeling framework, we prove that, when the instantaneous price impact parameter converges to zero, the absolutely continuous optimal portfolio process converges to a stochastic process that is given in terms of the unique solution to a regular one-dimensional quadratic BSDE. This limit turns out to be the solution to an optimal liquidation problem without instantaneous price impact, but with general semimartingale controls with jumps. Our result provides a unified framework within which to embed the two most commonly used modeling frameworks in the optimal liquidation literature and provides a foundation for the use of semimartingale liquidation strategies and the use of portfolio processes of unbounded variation. Our results are based on novel convergence results for BSDEs with singular terminal conditions and a novel representation result of BSDE solutions in terms of uniformly continuous functions of forward processes. We also prove that the optimal solution in the deterministic version of the original pre-limit optimal liquidation model can be approximated uniformly by using the first order Taylor polynomial expanded around the singular point where the instantaneous price impact vanishes. This approximation is explicitly computable, while the optimal solution generally is not.
|
62 |
Large-scale layered systems and synthetic biology : model reduction and decompositionPrescott, Thomas Paul January 2014 (has links)
This thesis is concerned with large-scale systems of Ordinary Differential Equations that model Biomolecular Reaction Networks (BRNs) in Systems and Synthetic Biology. It addresses the strategies of model reduction and decomposition used to overcome the challenges posed by the high dimension and stiffness typical of these models. A number of developments of these strategies are identified, and their implementation on various BRN models is demonstrated. The goal of model reduction is to construct a simplified ODE system to closely approximate a large-scale system. The error estimation problem seeks to quantify the approximation error; this is an example of the trajectory comparison problem. The first part of this thesis applies semi-definite programming (SDP) and dissipativity theory to this problem, producing a single a priori upper bound on the difference between two models in the presence of parameter uncertainty and for a range of initial conditions, for which exhaustive simulation is impractical. The second part of this thesis is concerned with the BRN decomposition problem of expressing a network as an interconnection of subnetworks. A novel framework, called layered decomposition, is introduced and compared with established modular techniques. Fundamental properties of layered decompositions are investigated, providing basic criteria for choosing an appropriate layered decomposition. Further aspects of the layering framework are considered: we illustrate the relationship between decomposition and scale separation by constructing singularly perturbed BRN models using layered decomposition; and we reveal the inter-layer signal propagation structure by decomposing the steady state response to parametric perturbations. Finally, we consider the large-scale SDP problem, where large scale SDP techniques fail to certify a system’s dissipativity. We describe the framework of Structured Storage Functions (SSF), defined where systems admit a cascaded decomposition, and demonstrate a significant resulting speed-up of large-scale dissipativity problems, with applications to the trajectory comparison technique discussed above.
|
63 |
Sobre Regularização e Perturbação Singular / On Regularization and Singular PerturbationCASTRO, Ubirajara José Gama de 24 February 2011 (has links)
Made available in DSpace on 2014-07-29T16:02:17Z (GMT). No. of bitstreams: 1
UBIRAJARA JOSe GAMA DE CASTRO.pdf: 516477 bytes, checksum: c0a8c62202b2da19be1a2dc69a29e416 (MD5)
Previous issue date: 2011-02-24 / The main goal of this work is to study the behavior of Discontinuous Vector Fields in a
neighborhood of a tipical singularity (tangency) using for this the regularization process
developed by Teixeira and Sotomayor [9] and, using also, some technics of the Geometric
Singular Perturbation Theory [2]. / O principal objetivo deste trabalho é estudar o comportamento numa vizinhança de uma
singularidade típica (tangência) dos campos vetoriais descontínuos utilizando o processo
de regularização desenvolvido por Teixeira e Sotomayor [9] e perturbações singulares [2].
|
64 |
Systèmes d'équations différentielles linéaires singulièrement perturbées et développements asymptotiques combinés / Systems of singularly pertubed linear differential equations and composite asymptotic expansionsHulek, Charlotte 12 June 2014 (has links)
Dans ce travail nous démontrons un théorème de simplification uniforme concernant les équations différentielles ordinaires du second ordre singulièrement perturbées au voisinage d’un point dégénéré, appelé point tournant. Il s’agit d’une version analytique d’un résultat formel dû à Hanson et Russell, qui généralise un théorème connu de Sibuya. Pour traiter ce problème, nous utilisons les développements asymptotiques combinés Gevrey introduits par Fruchard et Schäfke. Dans une première partie nous rappelons les définitions et théorèmes principaux de cette récente théorie. Nous établissons trois résultats généraux que nous utilisons ensuite dans la seconde partie de ce manuscrit pour démontrer le théorème principal de réduction analytique annoncé. Enfin nous considérons des équations différentielles ordinaires d’ordre supérieur à deux, singulièrement perturbées à point tournant, et nous démontrons un théorème de réduction analytique. / In this thesis we prove a theorem of uniform simplification for second order and singularly perturbed differential equations in a full neighborhood of a degenerate point, called a turning point. This is an analytic version of a formal result due to Hanson and Russell, which generalizes a well known theorem of Sibuya. To solve this problem we use the Gevrey composite asymptotic expansions introduced by Fruchard and Schäfke. In the first part we recall the main definitions and theorems of this recent theory. We establish three general results used in the second part of this thesis to prove the main theorem of analytic reduction. Finally we consider ordinary differential equations of order greater than two, which are singularly perturbed and have a turning point, and we prove a theorem of analytic reduction.
|
65 |
Layer structure and the galerkin finite element method for a system of weakly coupled singularly perturbed convection-diffusion equations with multiple scalesRoos, Hans-Görg, Schopf, Martin 17 April 2020 (has links)
We consider a system of weakly coupled singularly perturbed convection-diffusion equations with multiple scales. Based on sharp estimates for first order derivatives, Linß [T. Linß, Computing 79 (2007) 23–32.] analyzed the upwind finite-difference method on a Shishkin mesh. We derive such sharp bounds for second order derivatives which show that the coupling generates additional weak layers. Finally, we prove the first robust convergence result for the Galerkin finite element method for this class of problems on modified Shishkin meshes introducing a mesh grading to cope with the weak layers. Numerical experiments support our theory.
|
66 |
On the use of singular perturbation based model hierarchies of an electrohydraulic drive for virtualization purposesZagar, Philipp, Scheidl, Rudolf 25 June 2020 (has links)
Virtualization of products means the representation of some of their properties by models. In a stronger digitalized world, these models will gain a much broader use than models had in engineering so far. Even for one modelling aspect different models of the same product will be used, depending on the specific need of the model user. That need may change in the course of product life, between first product concepts till over the different phases of development, to product use, maintenance, or even recycling. Since a digitalized world use of these diverse models will not be limited to experts model consistency will play a much stronger role. Model hierarchies will play a stronger role and can serve also as means for teaching product users a deeper understanding of product properties. A consistent model hierarchy leading from a simple to a more advanced property representation can support this learning process. In this paper perturbation methods are analyzed as a means for setting up model hierarchies in a consistent manner. This is studied by models for the behavior of a electrohydraulic drive, which consists of a variable speed motor, a pump, a double stroke cylinder and a counterbalance valve. Model hierarchy is achieved by model reduction in the sense of perturbation theory. The use of these different models for different questions in a system design context and their interrelations are exemplified.
|
67 |
High Accuracy Fitted Operator Methods for Solving Interior Layer ProblemsSayi, Mbani T January 2020 (has links)
Philosophiae Doctor - PhD / Fitted operator finite difference methods (FOFDMs) for singularly perturbed
problems have been explored for the last three decades. The construction of
these numerical schemes is based on introducing a fitting factor along with the
diffusion coefficient or by using principles of the non-standard finite difference
methods. The FOFDMs based on the latter idea, are easy to construct and they
are extendible to solve partial differential equations (PDEs) and their systems.
Noting this flexible feature of the FOFDMs, this thesis deals with extension
of these methods to solve interior layer problems, something that was still outstanding.
The idea is then extended to solve singularly perturbed time-dependent
PDEs whose solutions possess interior layers. The second aspect of this work is
to improve accuracy of these approximation methods via methods like Richardson
extrapolation. Having met these three objectives, we then extended our
approach to solve singularly perturbed two-point boundary value problems with
variable diffusion coefficients and analogous time-dependent PDEs. Careful analyses
followed by extensive numerical simulations supporting theoretical findings
are presented where necessary.
|
68 |
The Jormungand Climate ModelRackauckas, Christopher V. 11 July 2013 (has links)
No description available.
|
69 |
Réjection de perturbation sur un système multi-sources - Application à une propulsion hybride / Disturbance rejection of hybrid energy sources applied in hybrid electric vehiclesDai, Ping 19 January 2015 (has links)
Ce mémoire porte sur l'étude d'un système de gestion d'énergie électrique dans un système multi-sources soumis à des perturbations exogènes. L'application visée est l'alimentation d'une propulsion hybride diesel/électrique équipée d'un système d'absorption des pulsations de couple. Les perturbations exogènes considérées peuvent être transitoires ou persistantes. Une perturbation transitoire correspond à une variation rapide du couple de charge, due par exemple à une accélération ou une décélération du véhicule. Une perturbation persistante provient du système de compensation des pulsations de couple générées par le moteur thermique. Le premier objectif du contrôle est de maintenir constante la tension du bus continu. Le deuxième objectif est d'absorber dans un système de stockage rapide constitué de super condensateur ces perturbations qui peuvent à terme provoquer une usure prématurée de la batterie. Le troisième objectif est de compenser l'auto-décharge dans le super condensateur en maintenant constante sa tension nominale. Les deux sources (batterie et super condensateur) sont reliées au bus continu par l'intermédiaire de deux convertisseurs boost DC/DC. La commande consiste à piloter les rapports cycliques de chaque convertisseur. C'est un système non linéaire où la commande est multiplicative de l'état. L'approche classique consistant à résoudre les équations Francis-Byrnes-Isidori ne s'applique pas directement dans ce cas où la sortie et la matrice d'interconnection dépendent de la commande. De plus, si cette approche est bien adaptée au rejet de perturbations persistantes, elle montre ces limites pour le rejet de perturbations non persistantes combiné à des objectifs de régulation. Notre approche a consisté à écrire le système sous un formalisme Port-Controlled Hamiltonian et à s'affranchir de la contrainte de la dépendance de la matrice d'interconnection avec la commande en utilisant la théorie des perturbations singulières. La commande du système dégénéré peut ensuite être calculée par une approche passive. Les performances de cette commande ont été testées en simulation et à l'aide d'un banc d'essai expérimental. Les résultats montrent l'efficacité du système d'absorption des différents types de perturbation tout en respectant les deux objectifs de régulation. / This thesis presents the research of energy management in a battery/ultracapacitor hybrid energy storage system with exogenous disturbance in hybrid electric vehicular application. Transient and harmonic persistent disturbances are the two kinds of disturbances considered in this thesis. The former is due to the transient load power demand during acceleration and deceleration, and the latter is introduced from the process of the internal combustion engine torque ripples compensation. Our control objective is to absorb the disturbances causing battery wear via the ultracapacitor, and meanwhile, to maintain a constant DC voltage and to compensate the self-discharge in the ultracapacitor to maintain it operating at the nominal state of charge. The object system is nonlinear due to the multiplicative relation between the input and the state. The traditional approach to solve Francis-Byrnes-Isidori equations cannot be directly applied in this case since the interconnect matrix depends on the control input. Besides, even if this approach is well suited to the rejection of persistent disturbances, it shows the limits for the case of non-persistent disturbances which is also our object. Our contributed control method is realized through a cascade control structure based on the singular perturbation theory. The ultracapacitor current with the fastest motion rate is controlled in the inner fast loop through which we impose the desired dynamic to the system. The reduced system controlled in the outer slow loop is a Hamiltonian system and the controller is designed via interconnection and damping assignment. Simulations and experiments have been carried out to evaluate the control performance. A contrast of the system responses with and without the control algorithm shows that, with the control algorithm, the ultracapacitor effectively absorbs the disturbances; and verifies the effectiveness of the control algorithm.
|
70 |
Equations d'évolution non locales et problèmes de transition de phase / Non local evolution equations and phase transition problemsNguyen, Thanh Nam 29 November 2013 (has links)
L'objet de cette thèse est d'étudier le comportement en temps long de solutions d'équations d'évolution non locales ainsi que la limite singulière d'équations et de systèmes d'équations aux dérivées partielles, où intervient un petit paramètre epsilon. Au Chapitre 1, nous considérons une équation de réaction-diffusion non locale avec conservation au cours du temps de l'intégrale en espace de la solution; cette équation a été initialement proposée par Rubinstein et Sternberg pour modéliser la séparation de phase dans un mélange binaire. Le problème de Neumann associé possède une fonctionnelle de Lyapunov, c'est-à-dire une fonctionnelle qui décroit selon les orbites. Après avoir prouvé que la solution est confinée dans une région invariante, nous étudions son comportement en temps long. Nous nous appuyons sur une inégalité de Lojasiewicz pour montrer qu'elle converge vers une solution stationnaire quand t tend vers l'infini. Nous évaluons également le taux de la convergence et calculons précisément la solution stationnaire limite en dimension un d'espace. Le Chapitre 2 est consacré à l'étude de l'équation différentielle non locale que l'on obtient en négligeant le terme de diffusion dans l'équation d'Allen-Cahn non locale étudiée au Chapitre 1. Sans le terme de diffusion, la solution ne peut pas être plus régulière que la fonction initiale. C'est la raison pour laquelle on ne peut pas appliquer la méthode du Chapitre 1 pour l'étude du comportement en temps long de la solution. Nous présentons une nouvelle méthode basée sur la théorie des réarrangements et sur l'étude du profil de la solution. Nous montrons que la solution est stable pour les temps grands et présentons une caractérisation détaillée de sa limite asymptotique quand t tend vers l'infini. Plus précisément, la fonction limite est une fonction en escalier, qui prend au plus deux valeurs, qui coïncident avec les points stables d'une équation différentielle associée. Nous montrons aussi par un contre-exemple non trivial que, quand une hypothèse sur la fonction initiale n'est pas satisfaite, la fonction limite peut prendre trois valeurs, qui correspondent aux points instable et stables de l'équation différentielle associée. Nous étudions au Chapitre 3 une équation différentielle ordinaire non locale qui a éte proposée par M. Nagayama. Une difficulté essentielle est que le dénominateur dans le terme de réaction non local peut s'annuler. Nous appliquons un théorème de point fixe lié a une application contractante pour démontrer que le problème à valeur initiale correspondant possède une solution unique qui reste connée dans un ensemble invariant. Ce problème possède une fonctionnelle de Lyapunov, qui est un ingrédient essentiel pour démontrer que la solution converge vers une solution stationnaire constante par morceaux quand t tend vers l'infini. Au Chapitre 4, nous considérons un modèle d'interface diffuse pour la croissance de tumeurs, où intervient une équation d'ordre quatre de type Cahn Hilliard. Après avoir introduit un modèle de champ de phase associé, on étudie formellement la limite singulière de la solution quand le coefficient du terme de réaction tend vers l'infini. Plus précisément, nous montrons que la solution converge vers la solution d'un problème à frontière libre. AMS subject classifications. 35K57, 35K50, 35K20, 35R35, 35R37, 35B40, 35B25. / The aim of this thesis is to study the large time behavior of solutions of nonlocal evolution equations and to also study the singular limit of equations and systems of parabolic partial differential equations involving a small parameter epsilon. In Chapter 1, we consider a nonlocal reaction-diffusion equation with mass conservation, which was originally proposed by Rubinstein and Sternberg as a model for phase separation in a binary mixture. The corresponding Neumann problem possesses a Lyapunov functional, namely a functional which decreases in time along solution orbits. After having proved that the solution is conned in an invariant region, we study its large time behavior and apply a Lojasiewicz inequality to show that it converges to a stationary solution as t tends to infinity. We also evaluate the rate of convergence and precisely compute the limiting stationary solution in one space dimension. Chapter 2 is devoted to the study of a nonlocal evolution equation which one obtains by neglecting the diffusion term in the nonlocal Allen-Cahn equation studied in Chapter 1. Without the diffusion term, the solution can not be expected to be more regular than the initial function. Moreover, because of the absence of the diusion term, the method of Chapter 1 can not be applied to study the large time behavior of the solution. We present a new method based up on rearrangement theory and the study of the solution profile. We show that the solution stabilizes for large times and give a detailed characterization of its asymptotic limit as t tends to infinity. More precisely, it turns out that the limiting function is a step function, which takes at most two values, which are stable points of a corresponding ordinary dierential equation. We also show by means of a nontrivial counterexample that, when a certain hypothesis on the initial function does not hold, the limiting function may take three values. One of them is the unstable point and the two others are the stable points of the ordinary dierential equation. We study in Chapter 3 a nonlocal ordinary dierential equation which has been proposed by M. Nagayama. The nonlocal term involves a denominator which may vanish. We apply a contraction fixed point theorem to prove the existence of a unique solution which stays confined in an invariant region. We also show that the corresponding initial value problem possesses a Lyapunov functional and prove that the solution stabilizes for large times to a step function, which takes at most two values. In Chapter 4, we consider a diffuse-interface tumor-growth model which involves a fourth order Cahn-Hilliard type equation. Introducing a related phase-field model, we formally study the singular limit of the solution as the reaction coecient tends to infinity. More precisely, we show that the solution converges to the solution of a moving boundary problem. AMS subject classifications. 35K57, 35K50, 35K20, 35R35, 35R37, 35B40, 35B25.
|
Page generated in 0.1316 seconds