|
|
Analýza stiff soustav diferenciálních rovnic
Šátek, Václav ; Dalík, Josef (oponent) ; Horová, Ivana (oponent) ; Kunovský, Jiří (vedoucí práce)
Řešení tuhých ("stiff") soustav diferenciálních rovnic patří i v současné době stále mezi komplikované úlohy. Základním problémem je přesná definice tuhých systémů. Jednoznačná definice tuhých systémů stále neexistuje. Problémem je dále i detekce tuhých systémů diferenciálních rovnic. K řešení se v praxi využívají implicitní numerické metody nižších řádů, jejichž oblasti stability jsou relativně velké. Ve své práci se zabývám numerickým řešením obyčejných diferenciálních rovnic především numerickým výpočtem využívajícím metody Taylorovy řady. Analyzuji vznik tuhosti jednotlivých systémů diferenciálních rovnic, ukazuji možnou náhradu některých tuhých systémů ekvivalentními systémy bez tuhosti. Dále zkoumám možnost detekce tuhých systémů pomocí členů explicitní Taylorovy řady, zaměřuji se na stabilitu explicitní a implicitní Taylorovy řady. V závěru práce experimentálně ověřuji možnosti řešení tuhých systémů s využitím implicitní Taylorovy řady, zkoumám vhodnost použití víceslovní aritmetiky a navrhuji vhodný paralelizovatelný algoritmus implicitní Taylorovy řady s rekurentním výpočtem členů a Newtonovou iterační metodou (ITMRN).
|
|
|
Stability and convergence of numerical computations
Sehnalová, Pavla ; Dalík, Josef (oponent) ; Horová, Ivana (oponent) ; Kunovský, Jiří (vedoucí práce)
The aim of this thesis is to analyze the stability and convergence of fundamental numerical methods for solving ordinary differential equations. These include one-step methods such as the classical Euler method, Runge-Kutta methods and the less well known but fast and accurate Taylor series method. We also consider the generalization to multistep methods such as Adams methods and their implementation as predictor-corrector pairs. Furthermore we consider the generalization to multiderivative methods such as Obreshkov method. There is always a choice in predictor-corrector pairs of the so-called mode of the method and in this thesis both PEC and PECE modes are considered. The main goal and the new contribution of the thesis is the use of a special fourth order method consisting of a two-step predictor followed by an one-step corrector, each using second derivative formulae. The mathematical background of historical developments of Nordsieck representation, the algorithm of choosing a variable stepsize or an error estimation are discussed. The current approach adapts well to the multiderivative situation in variable stepsize formulations. Experiments for linear and non-linear problems and the comparison with classical methods are presented.
|
|
|
Application of Adaptive Filters in Processing of Solar Corona Images
Druckmüllerová, Hana ; Horová, Ivana (oponent) ; Rušin, Vojtech (oponent) ; Martišek, Dalibor (vedoucí práce)
Solar corona photography counts among the most complicated tasks in astrophotography. It also plays a key role for research of the solar corona. This thesis brings an a complete overview of methods for imaging the solar corona. The thesis contains necessary methematical background, the sequence of steps for image processing, an overview of adaptive filters used for visualization of corona structures in digital images, and new methods are proposed, especially for images which contain more noise than it is typical for images of the white corona taken during total solar eclipses, e.g. images taken with narrow-band filters. The Fourier normalizing-radial-graded filter method that I proposed during my PhD study are based on approximation of pixel values and their variability with trigonometric polynomials using other properties of the image.
|
|
|
Adaptive Filters for 2-D and 3-D Digital Images Processing
Martišek, Karel ; Horová, Ivana (oponent) ; Karpíšek, Zdeněk (oponent) ; Druckmüller, Miloslav (vedoucí práce)
The thesis is concerned with filters for visualization of high dynamic range images. In the theoretical part, the principle of confocal microscopy is described and the term digital image is defined in a mathematically correct way. Both frequency approach (using 2-D and 3-D discrete Fourier transform and frequency filters) and digital geometry approach (using adaptive histogram equalization with adaptive neighbourhood) are chosen for the processing of images. Necessary adjustments when working with non-ideal images containing additive and impulse noise are described as well. The last part of the thesis is interested in 3-D reconstruction from optical cuts of an object. All the procedures and algorithms are also implemented in the software developed as a part of this thesis.
|
|
| |
|
|
Approximations in Stochastic Optimization and Their Applications
Mrázková, Eva ; Horová, Ivana (oponent) ; Štěpánek, Petr (oponent) ; Karpíšek, Zdeněk (vedoucí práce)
Many optimum design problems in engineering areas lead to optimization models constrained by ordinary (ODE) or partial (PDE) differential equations, and furthermore, several elements of the problems may be uncertain in practice. Three engineering problems concerning the optimization of vibrations and an optimal design of beam dimensions are considered. The uncertainty in the form of random load or random Young's modulus is involved. It is shown that two-stage stochastic programming offers a promising approach in solving such problems. Corresponding mathematical models involving ODE or PDE type constraints, uncertain parameters and multiple criteria are formulated and lead to (multi-objective) stochastic nonlinear optimization models. It is also proved for which type of problems stochastic programming approach (EO reformulation) should be used and when it is sufficient to solve simpler deterministic problem (EV reformulation). This fact has the big importance in practice in term of computational intensity of large scale problems. Computational schemes for this type of problems are proposed, including discretization methods for random elements and ODE or PDE constraints. By means of derived approximations the mathematical models are implemented and solved in GAMS. The solution quality is determined by an interval estimate of the optimality gap computed via Monte Carlo bounding technique. Parametric analysis of multi-criteria model results in efficient frontier computation. The alternatives of approximations of the model with reliability-related probabilistic terms including mixed-integer nonlinear programming and penalty reformulations are discussed. Furthermore, the progressive hedging algorithm is implemented and tested for the selected problems with respect to future possibilities of parallel computing of large engineering problems. The results show that it can be used even when the mathematical conditions for convergence are not fulfilled. Finite difference method and finite element method are compared for deterministic version of ODE constrained problem by using GAMS and ANSYS with quite comparable results.
|
host ::
RSS.