Thesis (Ph. D.)--University of Texas at Austin, 2006. / Vita. Includes bibliographical references.

Thesis (Ph. D.)--University of Washington, 2005. / Vita. Includes bibliographical references (p. 214-234).

Currently standard first order single-point upstream weighting methods are employed in reservoir simulation for integrating the essentially hyperbolic system components. These methods introduce both coordinate-line numerical diffusion (even in 1-D) and cross-wind diffusion into the solution that is grid and geometry dependent. These effects are particularly important when steep fronts and shocks are present and for cases where flow is across grid coordinate lines. In this thesis, families of novel edge-based and cell-based truly multidimensional upwind formulations that upwind in the direction of the wave paths in order to minimise crosswind diffusion are presented for hyperbolic conservation laws on structured and unstructured triangular and quadrilateral grids in two dimensions. Higher resolution as well as higher order multidimensional formulations are also developed for general structured and unstructured grids. The schemes are coupled with existing consistent and efficient continuous CVD (MPFA) Darcy flux approximations. They are formulated using an IMPES (Implicit in Pressure Explicit in Saturation) strategy for solving the coupled elliptic (pressure) and hyperbolic (saturation) system of equations governing the multi-phase multi-component flow in porous media. The new methods are compared with single point upstream weighting for two-phase and three-component two-phase flow problems. The tests arc conducted on both structured and unstructured grids and involve full-tensor coefficient velocity fields in homogeneous and heterogeneous domains. The comparisons demonstrate the benefits of multidimensional and higher order multidimensional schemes in terms of improved front resolution together with significant reduction in cross-wind diffusion.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / à corrente o uso de inspeÃÃo nÃo-destrutiva ultrassÃnica na detecÃÃo de descontinuidades nos mais diversos materiais utilizados na indÃstria Adicionalmente informaÃÃes sobre a microestrutura do material inspecionado podem ser obtidas a partir do processamento da sÃrie temporal produzida durante a inspeÃÃo A simulaÃÃo do ensaio ultrassÃnico representa uma importante ferramenta para o entendimento e previsÃo da interaÃÃo da onda mecÃnica com o meio No entanto faz-se necessÃrio primeiramente modelar o meio que reproduza as caracterÃsticas de uma amostra objeto de estudo por onde a onda propaga Cinco meios unidimensionais compostos por domÃnios (representando grÃos) com tamanho mÃdio distinto e a mesma densidade mÃdia foram definidos neste trabalho SimulaÃÃes de propagaÃÃo de ondas ultrassÃnicas nos meios modelados foram executadas para quatro diferentes frequÃncias de ondas Concomitantemente foram capturados sinais ultrassÃnicos sobre cinco amostras de aÃo contendo diferentes tamanhos mÃdios de grÃo utilizando transdutores de 2.25 5.0 10.0 e 20.0 MHz Todos os sinais obtidos foram submetidos à detrended fluctuation analysis DFA e rescaled range analysis R/S duas tÃcnicas de anÃlise de flutuaÃÃes em sÃries temporais com vista a filtrar informaÃÃes espÃrias e avaliar influÃncia das variÃveis selecionadas (tamanho de grÃo e frequÃncia do sinal) sobre os sinais ultrassÃnicos obtidos Por fim à feita uma comparaÃÃo entre os dados simulados e experimentais e avaliaÃÃo da qualidade da simulaÃÃo / It is usual the application of ultrasonic non-destructive evaluation to detect discontinuities in different materials applied in industry Furthermore information about the microstructure of the inspected material can be obtained by signal processing of time series produced during the inspection Simulation of ultrasonic testing can be an important tool to help the understanding and predicting the interaction of mechanical wave in materials First of all computational materials modeling to reproduce the characteristics of specimens through which the wave propagates is necessary Five different one-dimensional materials composed of domains (grains) with different average size and same density were designed in this work Simulations of ultrasonic wave propagation in the modeled materials were performed at four different wave frequencies Concomitantly ultrasonic signals were acquired from five steel samples containing different average grain sizes using probes with central frequency of 225 5 10 and 20 MHz Detrended fluctuation analysis DFA and rescaled range analysis R/S two techniques for analyzing fluctuations in time series were used to process the signals in order to filter out spurious information and evaluate the influence of selected variables (grain size and frequency of the signal) on the ultrasonic signals obtained Finally simulated and experimental data and compared in order to evaluate the simulation performance

No description available.

This work details the development of a computational tool that can accurately model strongly-coupled fluid-structure-interaction (FSI) problems, with a particular focus on thin-walled structures undergoing large, geometrically non-linear deformations, which has a major interest in, amongst others, the aerospace and biomedical industries. The first part of this work investigates improving the efficiency with which a stable and robust in-house code, Elemental, models thin structures undergoing dynamic fluid-induced bending deformations. Variations of the existing finite volume formulation as well as linear and higher-order finite element formulations are implemented. The governing equations for the solid domain are formulated in a total Lagrangian or undeformed conguration and large geometrically non-linear deformations are accounted for. The set of equations is solved via a single-step Jacobi iterative scheme which is implemented such as to ensure a matrix-free and robust solution. Second-order accurate temporal discretisation is achieved via dual-timestepping, with both consistent and lumped mass matrices and with a Jacobi pseudo-time iteration method employed for solution purposes. The matrix-free approach makes the scheme particularly well-suited for distributed memory parallel hardware architectures. Three key outcomes, not well documented in literature, are highlighted: the issue of shear locking or sensitivity to element aspect ratio, which is a common problem with the linear Q4 finite element formulation when subjected to bending, is evaluated on the finite volume formulations; a rigorous comparison of finite element vs. finite volume methods on geometrically non-linear structures is done; a higher-order finite volume solid mechanics procedure is developed and evaluated. The second part of this work is concerned with fluid-structure interaction (FSI) modelling. It considers the implementation and coupling of a higher order finite element structural solver with the existing finite volume fluid-flow solver in Elemental. To the author’s knowledge, this is the first instance in which a strongly-coupled hybrid finite element–finite volume FSI formulation is developed. The coupling between the fluid and structural components with non-matching nodes is rigorously assessed. A new partitioned fluid-solid interface coupling methodology is also developed, which ensures stable partitioned solution for strongly-coupled problems without any additional computational overhead. The solver is parallelised for distributed memory parallel hardware architectures. The developed technology is successfully validated through rigorous temporal and mesh independent studies of representative two-dimensional strongly-coupled large-displacement FSI test problems for which analytical or benchmark solutions exist. / Dissertation (MEng)--University of Pretoria, 2012. / Mechanical and Aeronautical Engineering / unrestricted

A new implicit and compact optimization-based method is presented for high order derivative calculation for finite-volume numerical method on unstructured meshes. Highorder approaches to gradient calculation are often based on variants of the Least-Squares (L-S) method, an explicit method that requires a stencil large enough to accommodate the necessary variable information to calculate the derivatives. The new scheme proposed here is applicable for an arbitrary order of accuracy (demonstrated here up to 3rd order), and uses just the first level of face neighbors to compute all derivatives, thus reducing stencil size and avoiding stiffness in the calculation matrix. Preliminary results for a static variable field example and solution of a simple scalar transport (advection) equation show that the proposed method is able to deliver numerical accuracy equivalent to (or better than) the nominal order of accuracy for both 2nd and 3rd order schemes in the presence of a smoothly distributed variable field (i.e., in the absence of discontinuities). This new Optimization-based Gradient REconstruction (herein denoted OGRE) scheme produces, for the simple scalar transport test case, lower error and demands less computational time (for a given level of required precision) for a 3rd order scheme when compared to an equivalent L-S approach on a two-dimensional framework. For three-dimensional simulations, where the L-S scheme fails to obtain convergence without the help of limiters, the new scheme obtains stable convergence and also produces lower error solution when compared to a third order MUSCL scheme. Furthermore, spectral analysis of results from the advection equation shows that the new scheme is better able to accurately resolve high wave number modes, which demonstrates its potential to better solve problems presenting a wide spectrum of wavelengths, for example unsteady turbulent flow simulations.

Computational thermo-fluid models of a free surface flow under the dominant radiative transfer have been developed for the design and control of a modern optic fiber drawing process. Although modeling of the fiber drawing process has been of interest for the past three decades, most of the previous studies were limited to low draw speeds and small preforms. Large preforms drawn at high speeds have been used in modern fiber drawing systems to improve production efficiency and reduce cost. Several assumptions commonly made in previous studies have to be relaxed to address the new challenges. In this study, instead of using the Rosseland approximation, the radiative transfer equation (RTE) was solved directly for the radiation fluxes using the finite volume method (FVM). The complete two-dimensional free surface flow was simulated along with the coupling of the radiative transfer. Unlike most of the previous studies that only considered the furnace domain and that assumed the glass velocity at the exit, we included the post-chamber in the computation domain and predicted the fiber solidification location. Furthermore, the mixed convection of the air in the post-chamber was also considered, and was shown to have significant effects on the fiber solidification. On the basis of the computational model, a reduced order model (ROM) was developed for a mixed HŁ /LQG controller designed to regulate the fiber diameter under the effects of disturbances. The ROM was derived on the basis of the computational model. Optimal numerical eigenfunctions were obtained through the Karhunen-Loeve expansion using the computational model. The GalerkinŁ s method was then applied to obtain the state space ROM. The numerical model was shown to be efficient and was verified experimentally. The ROM characterizes the dynamics of the system accurately as compared with the computational model. The simulations using the full computational model showed that the closed-loop system is robust and superior to the open-loop system in the regulation of fiber diameter. The modeling and control methods can be applied to the design optimization and parameter regulation of the high-speed large-preform draw processes as well as other manufacturing processes that involve similar thermal-fluid transports.

A suspension flow model based on the "suspension balance" approach has been developed. This work modifies the model to allow the solution of suspension flows under general flow conditions. This requires the development of a frame-invariant constitutive model for the particle stress which can take into account the spatially-varying local kinematic conditions. The mass and momentum balances for the bulk suspension and particle phase are solved numerically using a finite volume method. The particle stress is based upon the computed rate of strain and the local kinematic conditions. A nonlocal stress contribution corrects the continuum approximation of the particle phase for finite particle size effects. Local kinematic conditions are accounted through the local ratio of rotation to extension in the flow field. The coordinates for the stress definition are the local principal axes of the rate of strain field. The developed model is applied to a range of problems. (i) Axially-developing conduit flows are computed using both the full two-dimensional solution and the more computationally efficient "marching" method. The model predictions are compared to experimental results for cross-stream particle concentration profiles and axial development lengths. (ii) Model predictions are compared to experiments for wide-gap circular Couette flow of a concentrated suspension in a shear-thinning liquid. With minor modification, the suspension flow model predicts the major trends and results observed in this flow. (iii) Comparisons are made to experiments for an axisymmetric contraction-expansion. Model predictions for a two-dimensional planar contraction flow test the influence of model formulation. The variation of the magnitude of an isotropic particle normal stress with local kinematic conditions and anisotropy in the in-plane normal stresses are both explored. The formulation of the particle phase stress is found to have significant effects on the solid fraction and velocity. (iv) Finally, for a rectangular piston-driven flow and an obstructed channel flow, a "computational suspension dynamics" study explores the effect of particle migration on the bulk flow field, system pressure drop and particle phase composition.

Dielectric Barrier Discharge (DBD) excimer lamp is a very attractive source for Ultraviolet (UV) radiation. It has a number of advantages compared to the mercury lamp which is the main lamp used in the industry for UV production. Some of these advantages are instant UV radiation (no warm-up period), narrow UV spectrum, longer life times and simple construction. The DBD UV lamp can be used in number of applications like water disinfection, Plasma Display Panels (PDP) and surface treatment in the semiconductor industry. Yet, the full industrial application of this lamp still faces some problems mainly related to finding the optimum electrical driver to maximize the efficiency of such a lamp. This includes the type of the electrical waveform to generate and the power electronic driver to produce it. In this thesis, firstly a physically based circuit model for the DBD lamp using the Finite Volume Method (FVM) is developed. This model provides the electrical and optical characteristics of the lamp. Using this model the sensitivity of the lamp efficiency to the proposed electrical waveform has been determined. Secondly, the order of this FVM model has been reduced to obtain a model which is used in the design procedure of the proposed driver. Since the DBD lamp has a capacitive nature, a current controlled driver is proposed in this thesis as opposed to most of the published drivers which are voltage controlled drivers. The design of this driver is intended to enhance the electrical to optical efficiency of the lamp and therefore enhancing the overall efficiency of the system. The driver topology permits direct control of the peak lamp current and the operating frequency of the supplied current to the DBD lamp. The width of the current pulses is determined by the transformer magnetizing inductance and the lamp capacitance. Experimental results of the proposed driver connected to a XeCl DBD lamp are presented to validate the performance of the driver and to prove the concept of such a current controlled driver. The proposed driver performance is compared to a voltage source driver which was also implemented. The proposed driver produced higher overall system efficiency but at the expense of a reduction in the driver efficiency as compared to the voltage source driver. The complete system, which consists of the developed FVM based model and the equivalent circuit of the proposed driver, was simulated and the results were compared to the experimental results to validate the accuracy of the developed model for the DBD lamp.