Thesis (Ph. D.)--Worcester Polytechnic Institute. / UMI no.: 99-37530. Includes bibliographical references (leaves 221-238).

Thesis (M.S. in Mechanical Engineering) Naval Postgraduate School, March 1997. / Thesis advisor, Ashok Gopinath. Includes bibliographical references (p. 47-48). Also available online.

Thesis supervisors: Larry J. Weber, Nathan C. Young. Includes bibliographic references (p. 152-155).

Thesis supervisors: Marcela Politano, Larry J. Weber. Includes bibliographic references (p. 95-97).

The marker and cell method is an efficient co-volume technique suitable for the solution of incompressible flows using a Cartesian mesh. For flows around complex geometries the use of an unstructured mesh is desirable. For geometric flexibility an unstructured mesh implementation is desirable. A co-volume technique requires a dual orthogonal mesh, in the triangular case the Delaunay-Voronoi dual provides the means for determining this dual orthogonal mesh in an unstructured mesh framework. Certain mesh criteria must be placed on the Delaunay-Voronoi to ensure it meets the dual orthogonal requirements. The two dimensional extension of the marker and cell method to an unstructured framework is presented. The requirements of the mesh are defined and methods in their production are discussed. Initially an explicit time stepping scheme is implemented which allows efficient simulation of incompressible fluid flow problems. Limitations of the explicit time stepping scheme that were discovered, mean that high Reynolds number flows that require the use of stretched meshes cannot produce solutions in a reasonable time period. A semi-implicit time stepping routine removes this limitation allowing these types of flows to be successfully modelled. To validate the solvers accuracy and demonstrate its performance, a number of test cases are presented. These include the lid driven cavity, flow over a backward facing step, inviscid flow around a circular cylinder, unsteady flow around a circular cylinder, flow around an SD7003 aerofoil, flow around a NACA0012 aerofoil and flow around a multi element aerofoil. The investigation although revealing a high dependence on the quality of the mesh still demonstrates that accurate results can be obtained efficiently. The efficiency is demonstrated by comparison to the in-house 2D incompressible finite volume solver for flow around a circular cylinder. For this case the unstructured MAC method produced a solution four times faster than the finite volume code.

In this research, the finite element method (FEM) was used to solve the nonlinear, incompressible, transient, three dimensional Navier-Stokes equations in their non-conservative form. Linear tetrahedron elements were employed with the elegant, equal order interpolation for both pressure and velocity. The characteristic based split scheme was formulated in a fully implicit manner to circumvent the time step restrictions of the classical explicit formulations. The monolithic (single step, fully coupled solution procedure for pressures and velocity) form of the CBS scheme was also derived and its suitability was positively demonstrated. Casting the CBS scheme in a monolithic framework, results in the generation of a pressure stabilization term in the mass conservation equation, thereby circumventing the LBB restriction by the elimination of the zero pressure block. An account of all the steps involved in discretizing the Navier-Stokes equations (both in split and monolithic frameworks) was presented in meticulous detail, which included the derivation of the convective and pressure stabilization terms, linearization of the non-linear terms and the consequent derivation of the highly efficient analytical jacobian matrix, along with the temporal and spatial discretizations of the corresponding terms. The monolithic and the split version of the CBS scheme were integrated into a parallel, scalable and extensible Fortran90 software called IFENs. The development of IFENs started during the course of this research and all of its components have been designed and implemented by the author of this thesis. Multi processor parallelism was achieved using the Intel implementation of the most widely used/preferred, Message Passing Interface (MPI) standard. The parallel support needed for the use of a variety of parallel, linear, iterative solvers belonging to the Krylov subspace family (e.g. GMRES and its variants, CG, BiCG, BiCG- stab, etc.), parallel non linear solvers belonging to the Newton-Krylov family (line search newton, trust region newton, nonlinear GMRES, etc.) and parallel preconditioners (incomplete LU, Additive Shwarz Method - ASM, algebraic multigrid, etc.), was provided by the incorporation of PETSc into IFENs. PETSc is a state of the art, non-trivial toolkit, which represents a collection of several parallel libraries useful in high performance scientific computing. Keeping in mind the specific requirements of IFENs, a custom mesh partitioner was implemented. It operated on meshes that were renumbered using bandwidth reducing algorithms like Revere Cuthill Mckee. The possibility of using established domain decomposition libraries like ParMETIS was explored and demonstrated to be counter productive for the demands of this research. After the preliminary testing and validation of the procedures adopted before and during the execution of IFENs, large, high definition domains representative of human arteries (specifically, carotid bifurcations, found in the neck) were considered and the complete incompressible set of Navier-Stokes equations were solved for pressure and velocity fields. During the tenure of this research more than 1000 recorded parallel test cases were executed to test various components of IFENs, as well as various simulations representative of a wide variety of problems. IFENs can easily handle meshes with tens of millions of elements. The largest mesh used for the purpose of this research contained 14.58 million tetrahedrons and 2.489 million nodes, which on average required just 7 minutes per timestep, while executing the classical split framework of the CBS scheme. Results from the simulation of 9 carotid meshes, representative of 4 carotid geometries were presented and found to be in good agreement with the available ultrasound data. The flow fields were analysed and post processed using different techniques for each case. The haemodynamic wall parameters like time averaged wall shear stress and oscillatory shear index were calculated and mapped onto the corresponding boundary nodes. The region in the carotid bifurcation susceptible to the deposition of plaques and consequent stenosis were pointed out and other anomalies were highlighted.

Orientador: José Roberto Nunhez / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Química / Made available in DSpace on 2018-08-25T09:46:44Z (GMT). No. of bitstreams: 1 Villamizar_UrbanoMontanez_M.pdf: 2761421 bytes, checksum: b36bf4d1123da6ad78402c7d206517a3 (MD5) Previous issue date: 2013 / Resumo: Os Erlenmeyers agitados têm sido utilizados como uma ferramenta em pesquisa na área biotecnológica e laboratórios industriais, especialmente em estágios iniciais, onde graças a sua praticidade, muitos experimentos podem ser realizados simultaneamente com um baixo custo e virtualmente sem supervisão. No entanto, esses dispositivos só fornecem informação limitada dos processos fenomenológicos que acontecem dentro do Erlenmeyer, tais como a velocidade rotacional, que dá uma ideia dos requerimentos de mistura, etc. Esta limitação pode dificultar a transição de um novo processo da bancada experimental para a escala piloto ou industrial. Alguns estudos de pesquisa têm sido realizados para determinar importantes variáveis de processo, tais como consumo de potência volumétrico, capacidade de transferência de oxigênio, estresse hidrodinâmico, etc. A fluidodinâmica computacional (CFD) tem ganhado importância recentemente no estudo dos fenômenos de transporte, graças aos avanços no desenvolvimento de software especializado, e poder computacional. O objetivo desse trabalho é simular o processo de agitação desenvolvido em um Erlenmeyer agitado em máquinas orbitais, utilizando o pacote computacional CFX versão 14. Os resultados CFD são comparados com os dados experimentais disponíveis para validar o modelo com o objetivo de estudar a fluidodinâmica desenvolvida nestes dispositivos / Abstract: Erlenmeyers have been used as a tool in many biotechnology research and industrial laboratories, especially in its early stages when many experiments can be performed simultaneously at low cost and nearly without supervision. However, these devices offer only limited information on the phenomenological processes occurring within these shake flasks, as the rotational speed which gives an idea of the mixing requirements, etc. This limitation could be a difficulty when trying to scale up new processes developed in laboratories to a pilot plant scale or an industrial processes. Some experimental research has been carried out to determine important process variables in Erlenmeyer agitation such as volumetric power consumption, oxygen transfer capacity, hydrodynamic stress, etc. Computational fluid dynamics (CFD) has recently gained importance in the study of transport phenomena, thanks to advances in the development of specialized software, and computational power. The objective of this work is to simulate the mixing process in an Erlenmeyer flask. The software used is CFX version 14. The CFD results is compared with the experimental data available to validate this computational model in order to study the fluid dynamics that develops in these devices / Mestrado / Desenvolvimento de Processos Químicos / Mestre em Engenharia Química

Orientadores: Milton Mori, Marcos Akira D'Ávila / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Química / Made available in DSpace on 2018-08-20T21:16:08Z (GMT). No. of bitstreams: 1 MiillerLopes_MariaFernanda_M.pdf: 2798310 bytes, checksum: d508c014dbb2db75fd6d052fdf5a76c7 (MD5) Previous issue date: 2012 / Resumo: Esta pesquisa tem por objetivo avaliar os resultados numéricos por Fluidodinâmica Computacional (CFD) e experimentais pela técnica de Velocimetria por Imagem de Partícula (PIV) e a técnica de Shadow Imaging em uma coluna de bolhas de 1m de altura, 14,5 cm de diâmetro com um distribuidor de gás de 5 furos de 1 mm de diâmetro cada. O sistema estudado foi o sistema bifásico água-ar. O estudo foi realizado no regime homogêneo de operação para as velocidades superficiais de gás de 0,3 cm/s, 0,5 cm/s e 0,7 cm/s. Os experimentos de PIV e shadow imaging foram realizados nas mesmas condições operacionais. Para uma boa representação do escoamento pela técnica de PIV foi necessário o tratamento de 2500 fotos para as velocidades mais baixas e 3500 fotos para a velocidade mais alta. Para a técnica de shadow imaging foi necessário o tratamento de 2000 fotos para a determinação do diâmetro médio das bolhas. Nos testes numéricos foram avaliados dois modelos de arraste, o de Ishii-Zuber (1979) e o de Zhang-Vanderheyden (2002), para a turbulência foi utilizado o modelo k-?, e o diâmetro de bolha utilizado foi o obtido experimentalmente pela técnica de shadow imaging. Perfis de velocidade média axial de líquido obtidos experimentalmente foram comparados com os dados numéricos. Análises dos tensores de Reynolds, energia cinética turbulenta e intensidade turbulenta também foram avaliados. Para a velocidade de 0,3 cm/s e 0,5 cm/s ambos os modelos de arraste representaram bem o escoamento. Para a velocidade de 0,7 cm/s o modelo de arraste de Zhang-Vanderheyden representou melhor o escoamento / Abstract: This research aims to evaluate the numerical results from Computational Fluid Dynamics (CFD) simulations and experimental results obtained using Particle Image Velocimetry (PIV) and Shadow Imaging in a bubble column of 1m height, diameter of 14,5cm with a gas distributor with 5 holes of 1 mm diameter each one. The system studied was water-air. The study was performed in homogeneous operation regime for the superficial gas velocity of 0.3 cm/s, 0.5 cm/s and 0.7 cm/s. The PIV and shadow imaging experiments were performed under the same operation conditions. For a good flow representation of PIV data it was necessary to treat 2500 photos to the low superficial gas velocities and 3500 photos to the highest superficial gas velocity. For the shadow imaging technique it was necessary to treat 2000 photos for determining the bubble diameters. In the numerical tests two drag models were evaluated: Ishii-Zuber (1979) and Zhang-Vanderheyden (2002); for turbulence the k-? model was used and the bubble diameter used in the simulations was obtained experimentally by the shadow imaging technique. Average axial velocity profiles of fluid obtained experimentally were compared with numerical results. Analysis of Reynolds tensor, turbulent kinetic energy and turbulent intensity were also evaluated. For superficial gas velocities of 0,3 cm/s and 0,5 cm/s both drag models show good agreement with experimental data. For superficial gas velocities of 0,7 cm/s Zhang-Vanderheyden drag model showed better agreement with experimental data / Mestrado / Desenvolvimento de Processos Químicos / Mestra em Engenharia Química

Catalytic converters are widely used in the automotive industry to comply with increasingly stringent emissions regulations. The flow distribution across the catalyst substrate significantly aects its conversion eciency. Measuring the flow in a catalyst system is challenging; computational fluid dynamics (CFD) provides an alternative approach for the assessment of different design concepts and is therefore commonly employed to model flow behaviour. This thesis studies the application of CFD to modelling ow in a two-dimensional system consisting of a catalyst monolith downstream of a wide-angled planar diuser, with total included angle 60°. Computational models are developed using the commercial CFD software STAR-CCM+. Flow predictions are compared to experimental data collected by Mat Yamin, (2012) and also as part of this study. Measurements were obtained on a two-dimensional isothermal flow rig using particle image velocimetry (PIV) and hot-wire anemometry (HWA). Steady flow studies compare different methods of modelling the monolith. Models include the common approach of modelling the monolith as a porous medium and the computationally expensive individual channels model. A hybrid model is developed that combines the two approaches, benefiting from the respective merits of each method. Two monolith lengths are considered, with flow at varying Reynolds numbers. The porous model predicts the downstream velocity prole well for the shorter monolith but overpredicts flow maldistribution for the longer monolith. The inclusion of an entrance effect to account for the pressure losses associated with oblique entry into the monolith channels is studied. Best agreement in downstream velocity is observed when the pressure losses are limited using a critical angle approach. The individual channels model is found to be the most consistently accurate across monolith lengths, attributable to the accurate capture of flow behaviour upon entry into the monolith channels. A novel hybrid model is proposed, which combines the computational efficiency of the porous model with the geometrical accuracy of individual channels. The model is evaluated and is found to provide results similar to the individual channels model, with improved predictions of velocity maxima and minima. Pulsating flow studies present three transient flow regimes with similar inlet pulse shapes and varying Reynolds number and frequency. Predicted velocities in the diuser are in good agreement with PIV flow fields, however CFD predicts higher magnitudes at the shear layer. The model predicts large residual vortices present at the end of the cycle where experimental data shows none; it is concluded that CFD underpredicts turbulence diffusion. Evidence of cyclic variation in experimental data highlights the limitation of URANS turbulence models.

M.Ing. / The characterisation of the flow field and thermal performance of supply air windows (airflow windows operating in supply mode) have been a topic of interest for at least two decades. Computational Fluid Dynamics (CFD) as well as other simulation methods have been used to model and characterise the flow field, temperature distributions and thermal performance of the supply air window in recent years. Where experimental validation of the velocity (only outlet velocity) and temperature predictions has been provided the error between experiment and CFD (and other forms of simulation) is in the order of 50 % and 3 ◦C (10-13 %), respectively. Furthermore, a large part of the literature does not have experimental validation of the simulation results. The significant error in many of the studies, that provide experimental val- idation of the velocity field, is attributed to inappropriate turbulence mod- els, unrealistic boundary conditions, neglecting significant three-dimensional effects, solar radiation effects not entirely accounted for, mesh sensitivity studies neglected and material properties of glass and air assumed constant. The aim of this research was to characterise a supply air window in terms of its velocity field, temperature distributions and thermal performance. This was done by mathematically modelling the fluid dynamics and heat trans- fer processes in a supply air window and solving the model in a commer- cial CFD code, namely ANSYS Fluent 12.1. Furthermore, an experimental rig was designed, constructed and used to measure the flow field and tem- peratures with the aim of validating the CFD models. The CFD models incorporated appropriate turbulence models, realistic boundary conditions, three-dimensional effects, solar radiation, temperature dependent material properties and a mesh sensitivity study. The CFD models and experiments were setup for forced and natural flow conditions. Laser Doppler Velocimetry has not been used for velocity field measure- ments in an airflow window to date. The experimental setup made use of Laser Doppler Velocimetry to measure the velocity field and turbulence in- tensities. The Laser Doppler Velocimeter (LDV) probe was positioned using a three axis computer controlled traversing mechanism. Furthermore, flow visualisation experiments were done to qualitatively capture the flow field. The results from the CFD are partially in good agreement with the exper- imental work. Qualitatively the flow field as predicted by CFD is in good agreement with the results from the flow visualisation experiments. Quan- titatively the results from the CFD are in good agreement with the tem- perature measurements, however, there is noticeable error between the LDV readings and the velocities as well as turbulence intensity values predicted by CFD. The error, with regards to velocity and turbulence intensity, may be attributed to the experimental error caused by problems with flow seeding as well as the isotropic turbulence assumption inherent in the turbulence model (SST k − ω) used.