Philosophiae Doctor - PhD / In recent years, there has been a great interest towards the higher order numerical methods for singularly perturbed problems. As compared to their lower order counterparts, they provide better accuracy with fewer mesh points. Construction and/or implementation of direct higher order methods is usually very complicated. Thus a natural choice is to use some convergence acceleration techniques, e.g., Richardson extrapolation, defect correction, etc. In this thesis, we will consider various classes of problems described by singularly perturbed ordinary and partial differential equations. For these problems, we design some novel numerical methods and attempt to increase their accuracy as well as the order of convergence. We also do the same for existing numerical methods in some instances. We find that, even though the Richardson extrapolation technique always improves the accuracy, it does not perform equally well when applied to different methods for certain classes of problems. Moreover, while in some cases it improves the order of convergence, in other cases it does not. These issues are discussed in this thesis for linear and nonlinear singularly perturbed ODEs as well as PDEs. Extrapolation techniques are analyzed thoroughly in all the cases, whereas the limitations of the defect correction approach for certain problems is indicated at the end of the thesis. / South Africa

Philosophiae Doctor - PhD / Options are a special type of derivative securities because their values are derived from the value of some underlying security. Most options can be grouped into either of the two categories: European options which can be exercised only on the expiration date, and American options which can be exercised on or before the expiration date. American options are much harder to deal with than European ones. The reason being the optimal exercise policy of these options which led to free boundary problems. Ever since the seminal work of Black and Scholes [J. Pol. Econ. 81(3) (1973), 637-659], the differential equation approach in pricing options has attracted many researchers. Recently, numerical singular perturbation techniques have been used extensively for solving many differential equation models of sciences and engineering. In this thesis, we explore some of those methods which are based on spline approximations to solve the option pricing problems. We show a systematic construction and analysis of these methods to solve some European option problems and then extend the approach to solve problems of pricing American options as well as some exotic options. Proposed methods are analyzed for stability and convergence. Thorough numerical results are presented and compared with those seen in the literature. / South Africa

Philosophiae Doctor - PhD / There is a growing interest in the dynamics of the co-infection of these two diseases. In this thesis, firstly we focus on studying the effect of a distributed delay representing the incubation period for the malaria parasite in the mosquito vector to possibly reduce the initial transmission and prevalence of malaria. This model can be regarded as a generalization of SEI models (with a class for the latently infected mosquitoes) and SI models with a discrete delay for the incubation period in mosquitoes. We study the possibility of occurrence of backward bifurcation. We then extend these ideas to study a full model of HIV and malaria co-infection. To get further inside into the dynamics of the model, we use the geometric singular perturbation theory to couple the fast and slow models from the full model. Finally, since the governing models are very complex, they cannot be solved analytically and hence we develop and analyze a special class of numerical methods to solve them. / South Africa

Philosophiae Doctor - PhD / In this thesis, we consider some nonlinear di erential models that govern unsteady magneto-hydrodynamic convective ow and mass transfer of viscous, incompressible, electrically conducting uid past a porous plate with/without heat sources. The study focusses on the e ect of a combination of a number of physical parameters (e.g., chem- ical reaction, suction, radiation, soret e ect, thermophoresis and radiation absorption) which play vital role in these models. Non-dimensionalization of these models gives us sets of di erential equations. Reliable solutions of such di erential equations can- not be obtained by standard numerical techniques. We therefore resorted to the use of the singular perturbation approaches. To proceed, each of these model problems is discretized in time by using a suitable time-stepping method and then by using a tted operator nite di erence method in spatial direction. The combined methods are then analyzed for stability and convergence. Aiming to study the robustness of the proposed numerical schemes with respect to change in the values of the key parame-ters, we present extensive numerical simulations for each of these models. Finally, we con rm theoretical results through a set of speci c numerical experiments.

Les lipides des microalgues et les glucides de cyanobactéries peuvent être transformés en biodiesel et en bioéthanol, respectivement. L'amélioration de la production de ces molécules doit prendre en compte les entrées périodiques (principalement la lumière) forçant le réseau métabolique de ces organismes photosynthétiques. Il est donc nécessaire de tenir compte de la dynamique du réseau métabolique en réduisant sa dimension pour assurer la maniabilité mathématique. Le but de ce travail est de concevoir une approche originale pour réduire les réseaux métaboliques dynamiques tout en conservant la dynamique de base. Cette méthode est basée sur une séparation en échelles de temps. Pour une classe de modèles de réseaux métaboliques décrits par des ODE, la dynamique des systèmes réduits est calculée à l'aide du théorème de Tikhonov pour les systèmes singulièrement perturbés. Cette approximation quasi-stationnaire coïncide avec la dynamique du réseau d'origine, avec une erreur bornée. L'approche est d'abord développée pour les systèmes de réaction pouvant être linéarisés autour d'un point de travail et forcés par des entrées continues. Ensuite, une généralisation de cette méthode est donnée pour les réseaux à réactions rapides de cinétiques de Michaelis-Menten et tout type de cinétiques lentes, prenant également en compte un nombre fini d'entrées continues externes. La méthode de réduction met en évidence une relation entre la grandeur de la concentration des métabolites et la gamme des vitesses de réaction : les métabolites consommés par les réactions rapides ont une concentration inférieure d'un ordre de grandeur à celle des métabolites consommés à faible vitesse. Cette propriété est satisfaite pour les métabolites à dynamique rapide ne se trouvant pas dans un piège de flux, concept introduit dans ce travail. Le système réduit peut être calibré avec des données expérimentales à l'aide d'une procédure d'identification dédiée basée sur la minimisation. L'approche est illustrée par un réseau métabolique de microalgues autotrophes, comprenant le métabolisme central et représentant la dynamique des glucides et des lipides. Cette approche permet de bien ajuster les données expérimentales de Lacour et al. (2012) avec la microalgue Tisochrysis lutea. Enfin, un schéma visant à optimiser la production de molécules cibles est proposé en utilisant le système réduit. / Lipids from microalgae and carbohydrates from cyanobacteria can be transformed into biodiesel and bioethanol, respectively. Enhancing the production of these molecules must account for the periodic inputs (mainly light) forcing the metabolic network of these photosynthetic organisms. It is therefore necessary to account for the dynamics of the metabolic network, while reducing its dimension to ensure mathematical tractability. The aim of this work is to design an original approach to reduce dynamic metabolic networks while keeping the core dynamics. This method is based on time-scale separation. For a class of metabolic network models described by ODE, the dynamics of the reduced systems are computed using the theorem of Tikhonov for singularly perturbed systems. This Quasi Steady State Approximation accurately coincides with the original network dynamics, with a bounded error. The approach is first developed for reaction systems that can be linearized around a working point and that are forced by external continuous inputs. Then, a generalization of this method is given for networks with fast reactions of Michaelis-Menten kinetics and any type of slow kinetics, also considering a finite number of external continuous inputs. The reduction method highlights a relation between the concentration magnitude of the metabolites and the range of the reaction rates: the metabolites that are consumed by fast reactions have concentration one order of magnitude lower than metabolites consumed at slow rates. This property is satisfied for metabolites with fast dynamics that are not in a flux trap, a concept introduced in this work. The reduced system can be calibrated with experimental data using a dedicated identification procedure based on minimization. The approach is illustrated with an autotrophic microalgae metabolic network, including the core metabolism and representing the carbohydrates and lipids dynamics. The approach efficiently fits the experimental data from Lacour et al. (2012) with the microalgae Tisochrysis lutea. Finally, a scheme to optimize the production of target molecules is proposed using the reduced system.

Philosophiae Doctor - PhD / Options are a special type of derivative securities because their values are derived from the value of some underlying security. Most options can be grouped into either of the two categories: European options which can be exercised only on the expiration date, and American options which can be exercised on or before the expiration date. American options are much harder to deal with than European ones. The reason being the optimal exercise policy of these options which led to free boundary problems. Ever since the seminal work of Black and Scholes [J. Pol. Bean. 81(3) (1973), 637-659], the differential equation approach in pricing options has attracted many researchers. Recently, numerical singular perturbation techniques have been used extensively for solving many differential equation models of sciences and engineering. In this thesis, we explore some of those methods which are based on spline approximations to solve the option pricing problems. We show a systematic construction and analysis of these methods to solve some European option problems and then extend the approach to solve problems of pricing American options as well as some exotic options. Proposed methods are analyzed for stability and convergence. Thorough numerical results are presented and compared with those seen in the literature.

>Magister Scientiae - MSc / This thesis deals with the design and implementation of some novel numerical methods for nonlinear singular perturbations problems (NSPPs). We provide a survey of asymptotic and numerical methods for some NSPPs in past decade. By considering two test problems, rigorous asymptotic analysis is carried out. Based on this analysis, suitable numerical methods are designed, analyzed and implemented in order to have some relevant results of physical importance. Since the asymptotic analysis provides only qualitative information, the focus is more on the numerical analysis of the problem which provides the quantitative information.

No description available.

In this research, an attempt is made to evaluate alternative model-predictive microgrid control approaches and to understand the trade-offs that emerge between model complexity and the ability to achieve real-time optimized system performance. Three alternative controllers are considered and their computational and optimization performance compared. In the first, nonlinearities of the generators are included within the optimization. Subsequently, an approach is considered wherein alternative (non-traditional) states and inputs of generators are used which enables one to leverage linear models with the model predictive control (MPC). Nonlinearities are represented outside the control in maps between MPC inputs and the physical inputs. Third, a recently proposed linearized trajectory (LTMPC) is considered. Finally, the performance of the controllers is examined utilizing alternative models of the synchronous machine that have been proposed for power system analysis.

The first objective of this dissertation is to present hybrid numerical-analytical approaches and reduced-order models to simulate microelectromechanical systems (MEMS) in multi-physics fields. These include electric actuation (AC and DC), squeeze-film damping, thermoelastic damping, and structural forces. The second objective is to investigate MEMS phenomena, such as squeeze-film damping and dynamic pull-in, and use the latter to design a novel RF-MEMS switch. In the first part of the dissertation, we introduce a new approach to the modeling and simulation of flexible microstructures under the coupled effects of squeeze-film damping, electrostatic actuation, and mechanical forces. The new approach utilizes the compressible Reynolds equation coupled with the equation governing the plate deflection. The model accounts for the slip condition of the flow at very low pressures. Perturbation methods are used to derive an analytical expression for the pressure distribution in terms of the structural mode shapes. This expression is substituted into the plate equation, which is solved in turn using a finite-element method for the structural mode shapes, the pressure distributions, the natural frequencies, and the quality factors. We apply the new approach to a variety of rectangular and circular plates and present the final expressions for the pressure distributions and quality factors. We extend the approach to microplates actuated by large electrostatic forces. For this case, we present a low-order model, which reduces significantly the cost of simulation. The model utilizes the nonlinear Euler-Bernoulli beam equation, the von K´arm´an plate equations, and the compressible Reynolds equation. The second topic of the dissertation is thermoelastic damping. We present a model and analytical expressions for thermoelastic damping in microplates. We solve the heat equation for the thermal flux across the microplate, in terms of the structural mode shapes, and hence decouple the thermal equation from the plate equation. We utilize a perturbation method to derive an analytical expression for the quality factor of a microplate with general boundary conditions under electrostatic loading and residual stresses in terms of its structural mode shapes. We present results for microplates with various boundary conditions. In the final part of the dissertation, we present a dynamic analysis and simulation of MEMS resonators and novel RF MEMS switches employing resonant microbeams. We first study microbeams excited near their fundamental natural frequencies (primary-resonance excitation). We investigate the dynamic pull-in instability and formulate safety criteria for the design of MEMS sensors and RF filters. We also utilize this phenomenon to design a low-voltage RF MEMS switch actuated with a combined DC and AC loading. Then, we simulate the dynamics of microbeams excited near half their fundamental natural frequencies (superharmonic excitation) and twice their fundamental natural frequencies (subharmonic excitation). For the superharmonic case, we present results showing the effect of varying the DC bias, the damping, and the AC excitation amplitude on the frequency-response curves. For the subharmonic case, we show that if the magnitude of the AC forcing exceeds the threshold activating the subharmonic resonance, all frequency-response curves will reach pull-in. / Ph. D.