Orientadores: Sérgio Tonini Button, José Divo Bressan / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica / Made available in DSpace on 2018-08-21T04:08:51Z (GMT). No. of bitstreams: 1 Martins_MarceloMatos_D.pdf: 11233728 bytes, checksum: 163e9bff5c80ef1c025e8c1083bed472 (MD5) Previous issue date: 2012 / Resumo: A simulação numérica computacional é nos dia de hoje frequentemente aplicada na elaboração de projetos ou análise dos processos de conformação plástica dos metais. A extrusão de metais é um dos principais processos de conformação plástica e largamente aplicado na fabricação de produtos e peças na indústria metal-mecânica. Tradicionalmente, essas análises são feitas utilizando o Método dos Elementos Finitos. Entretanto, há um aumento no interesse dos pesquisadores na utilização do Método dos Volumes Finitos para este fim. A literatura sugere que o escoamento na extrusão de metais pode ser analisado pela formulação do escoamento plástico (flow Formulation). No qual, pode-se assumir como o escoamento de um fluido incompressível e viscoso. Essa hipótese pode ser assumida já que o processo de extrusão é um processo isocórico. O método MacCormack é geralmente aplicado para simular os escoamentos de fluidos compressíveis pelo Método do Volumes Finitos. No escoamento de um fluido incompressível ou no escoamento de metal não existe uma equação para a evolução da variável pressão, sendo necessário a utilização de um método de acoplamento entre a pressão e a velocidade. Este trabalho trata da apresentação de um novo esquema numérico para a determinação de informações sobre o escoamento de um fluido incompressível e viscoso e sobre o escoamento de metal em um processo de extrusão direta, ambos em regime permanente. As equações governantes foram discretizadas pelo Método dos Volume Finitos através do Método de MacCormack explícito para uma malha estruturada e co-localizada. O acoplamento entre a pressão e a velocidade foi feita pelo método SIMPLE. O novo esquema numérico foi aplicado em escoamentos incompressíveis e viscosos para a glicerina e em escoamento de metais em processos de extrusão direta para o chumbo e uma liga de alumínio. O escoamento da glicerina foi avaliado para o caso entre placas paralelas e em dutos circulares sob condição axissimétrica e obtiveram boa concordância em relação ao resultados analíticos. Os campos de velocidades obtidos para a extrusão de metal alcançaram rápida convergência, em torno de 20000 iterações, essa quantidade de iterações foi inferior a quantidade que a glicerina necessitou. Para todos os materiais analisados os resultados numéricos tiveram boa concordância em comparação com resultados analíticos e experimentais obtidas da literatura. O método MacCormack produziu resultados coerentes para o escoamento da glicerina e dos metais sem a necessidade da adição de viscosidade artificial, como sugere a sua definição. Portanto, os resultados numéricos sugerem que o método MacCormack com o SIMPLE pode ser aplicado na resolução de escoamentos de fluidos incompressíveis e na conformação de metais além da sua tradicional aplicação na resolução de escoamentos compressíveis / Abstract: Computational numerical simulation is nowadays largely applied in the design and analysis of metal forming process. Extrusion of metals is one main forming process largely applied in the manufacturing of metallic products or parts. Historically, the Finite Element Method has been applied for decades in extrusion analysis. However, recently in the academy, there is a trend to use Finite Volume Method: literature suggests that metal flow by extrusion can be analysed by the flow formulation. Thus, metal flow can be modelled such us an incompressible viscous fluid. This hypothesis can be assumed because extrusion process is an isochoric process. The MacCormack Method is commonly used to simulate compressible fluid flow by the finite volume method. However, metal extrusion and incompressible fluid flow do not present state equations for the evolution of pressure, and therefore, a velocity-pressure coupling method is necessary to obtain a consistent velocity and pressure fields. Present work proposes a new numerical scheme to obtain information about both incompressible viscous fluid flow and metal flow in the extrusion process, in steady state. The governing equations were discretized by Finite Volume Method, using the Explicit MacCormack Method to structured and collocated mesh. The SIMPLE Method was applied to attain pressure-velocity coupling. These new numerical scheme was applied to incompressible viscous fluid flow of glycerine and forward extrusion process of lead and an aluminium alloy. The numerical results for glicerine fluid flow for parallel plates and axisymmetric flow in circular tube cases had quite good agreement in relation to the analytical solutions. The incompressible metal extrusion velocity fields achieved faster convergence than for liquid glycerine after 20.000 iterations and a good agreement with analytical and experimental results obtained from literature. The MacCormack Method applied for both glycerine and metals produced consistent results without the need of artificial viscosity as employed by the compressible flow simulation approaches. Hence, the present numerical results also suggest that MacCormack Method and SIMPLE can be applied in the solution of incompressible fluid flow and metal forming processes in adition to the traditional application for compressible fluid flow / Doutorado / Materiais e Processos de Fabricação / Doutor em Engenharia Mecânica

Multiphase flows are of great interest to a large variety of industries because flows of two or more immiscible liquids are encountered in a diverse range of processes and equipment. However, the advent of high viscosity oil requires more investigations to enhance good design of transportation system and forestall its inherent production difficulties. Experimental and numerical studies were conducted on water-sand, oil-water and oilwater- sand respectively in 1-in ID 5m long horizontal pipe. The densities of CYL680 and CYL1000 oils employed are 917 and 916.2kg/m3 while their viscosities are 1.830 and 3.149Pa.s @ 25oC respectively. The solid-phase concentration ranged from 2.15e-04 to 10%v/v with mean diameter of 150micron and material density of 2650kg/m3. Experimentally, the observed flow patterns are Water Assist Annular (WA-ANN), Dispersed Oil in Water (DOW/OF), Oil Plug in Water (OPW/OF) with oil film on the wall and Water Plug in Oil (WPO). These configurations were obtained through visualisation, trend and the probability density function (PDF) of pressure signals along with the statistical moments. Injection of water to assist high viscosity oil transport reduced the pressure gradient by an order of magnitude. No significant differences were found between the gradients of oil-water and oil-water-sand, however, increase in sand concentration led to increase in the pressure losses in oil-water-sand flow. Numerically, Water Assist Annular (WA-ANN), Dispersed Oil in Water (DOW/OF), Oil Plug in Water (OPW/OF) with oil film on the wall, and Water Plug in Oil (WPO) flow pattern were successfully obtained by imposing a concentric inlet condition at the inlet of the horizontal pipe coupled with a newly developed turbulent kinetic energy budget equation coded as user defined function which was hooked up to the turbulence models. These modifications aided satisfactory predictions.

The principle of controlled auto-ignition (CAI) is to mix fuel and air homogeneously before compressing the mixture to the point of auto-ignition. As ignition occurs simultaneously, CAI engines operate with lean mixtures preventing high cylinder pressures. CAI engines produce small amounts of nitrogen oxides (NOx) due to low combustion temperatures while maintaining high compression ratios and engine efficiencies. Due to simultaneous combustion and lean mixtures, CAI engines are restricted between low and mid load operations. Various strategies have been studied to improve the load limit of CAI engines. The scope of the project is to investigate the consequences of varying valve timing, as a method to control the mixture temperature within the combustion chamber and therefore, controlling the mixture auto-ignition point. This study presents computational fluid dynamics (CFD) modelling results of transient flow, inside a 0.45 litre Lotus single cylinder engine. After a validation process, a chemical kinetics model is combined with the CFD code, in order to study in-cylinder temperatures, the mixture distribution during compression and to predict the auto-ignition timing. The first part of the study focuses on validating the calculated in-cylinder velocities. A mesh sensitivity study is performed as well as a comparison of different turbulence models. A method to reduce computational time of the calculations is presented. The effects of engine speed on charge delay and charge amount inside the cylinder, the development of the in-cylinder flow field and the variation of turbulence parameters during the intake and compression stroke, are studied. The second part of the study focuses on the gasoline mixture and the variation of the valve timing, to retain different ratios of residual gases within the cylinder. After validation of the model, a final set of CFD calculations is performed, to investigate the effects of valve timing on flow and the engine parameters. The results are then compared to a fully homogeneous mixture model to study the benefits of varying valve duration. New key findings and contributions to CAI knowledge were found in this investigation. Reducing the intake and exhaust valve durations created a mixture temperature stratification and a fuel concentration distribution, prior to auto-ignition. It resulted in extending the heat release rate duration, improving combustion. However, shorter valve timing durations also showed an increase in heat transfer, pumping work and friction power, with a decrease of cylinder indicated efficiency. Valve timing, as a method to control auto-ignition, should only be used when the load limit of CAI engines, is to be improved.

Dans cette dissertation, nous considérons l’écoulement des liquides conducteurs d’électricité dans un champ magnétique externe. De tels écoulements sont décrits par les équations de la magnétohydrodynamique (MHD) quasi-statique, et sont fréquemment rencontrés dans des applications pratiques. Il suit qu’il y a un intérêt fort pour des outils numérques qui peuvent simuler ces écoulements dans des géometries complexes.<p>La première partie de cette thèse (chapitres 2 et 3) est dédiée à la présentation de la machinerie numérique qui a été utilisée et implémentée afin de résoudre les équations de la MHD quasi-statistique (incompressible). Plus précisément, nous avons contribué au développement d’un solveur volumes finis non-structuré parallèle. La discussion sur ces méthodes est accompagnée d’une analyse numérique qui est aussi valable pour des mailles non-structurées. Dans le chapitre 3, nous vérifions notre implémentation par la simulation d’un certain nombre de cas tests avec un accent sur des écoulements dans un champ magnétique intense.<p>Dans la deuxième partie de cette thèse (chapitres 4-6), nous avons utilsé ce solveur pour étudier des écoulements MHD de proche paroi .La première géometrie considérée (chapitre 4) est celle d’une conduite circulaire infini d’axe à haut nombre de Hartmann. Nous avons investitgué la sensitivité des résultats numériques au schéma de discrétisation et à la topologie de la maille. Nos résultats permettent de caractériser in extenso l’écoulement MHD dans une conduite avec des bords bien conducteurs par moyen des lois d’échelle.<p>Le sujet du cinquième chapitre est l’écoulement dans une conduite toroïdale à section carée. Une étude du régime laminaire confirme une analyse asymptotique pour ce qui concerne les couches de cisaillement. Nous avons aussi effectué des simulations des écoulements turbulents afin d’évaluer l’effet d’un champ magnétique externe sur l’état des couches limites limites.<p>Finalement, dans le chapitre 6, nous investiguons l’écoulement MHD et dans un U-bend et dans un coude arrière. Nous expliquons comment générer une maille qui permet de toutes les couches de cisaillement à un coût computationelle acceptable. Nous comparons nos résultats aux solutions asymptotiques. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

No description available.

Vortex Tube (VT) is a simple device having no moving mechanical parts, in which compressed gas at high pressure is injected through one or more tangential nozzles into a vortex chamber resulting in the separation of the inlet flow into two low pressure streams. One of the streams is the peripheral flow that is warmer than the inlet stream while the other is the central (core) flow that is colder than the inlet stream. This separation of the inlet flow into high and low temperature streams is known as temperature or energy separation. It is suggested by many investigators that compressed air of few atmospheres pressure and at room temperature can produce temperatures as high as +200ºC at the hot end (peripheral flow exit) and as low as -50ºC at the cold end (core flow exit) of the VT. Though VTs have large potential for simple heating and cooling applications, the mechanism of energy separation is not clear so far. Based on their studies, many investigators have suggested various theories, different from each other, but having specific lacunas and is an unresolved issue. Also, till date, experimental and industrial designs of the VTs are based purely on empirical correlations. Apart from heating and cooling applications, VTs can also be used for separation of binary gas mixtures and separation of oxygen from two-phase precooled air stream. The conceptual futuristic cryogenic launch vehicle designs are being attempted with in-flight liquid oxygen (LOX) collection system that significantly improves the pay load fraction. Vortex tube technology is one of the few promising technologies for futuristic in-flight LOX separation based launch vehicles. This technology has significant advantages over its counterparts as it is a simple, compact and light weight, and most importantly have no moving parts and unaffected by gravity and orientation. In order that VTs become an acceptable technology for in-flight LOX separation system, it is necessary to achieve minimum oxygen purity of 90% with more than 60% yield (separation efficiency) for the oxygen enriched stream in the VT. A survey of the available open literature has shown very little reported details, in particular, on achieving the required specifications for in-flight LOX separation systems. Till date, the highest LOX purity of 60% with 40% separation efficiency has been reported with VT technology. In view of the above mentioned facts, the work carried out has been focused on to: • Optimize the critical parameters of the VT to achieve maximum energy separation by CFD and experimental studies. • Understand the flow behaviour in the VT by estimating the velocity, temperature and pressure profiles at various locations in the VT and validation of secondary circulation flow and its effect on the performance of energy separation in VT. • Estimation of the energy transfer between the core and the peripheral layers of fluid flow in VT by analytical and CFD methods to propose the most appropriate mechanism of energy separation in VT. • Design and development of a dedicated experimental setup for both energy separation and LOX separation studies in VTs. • Design and fabrication of straight and conical VTs and experimental programme on energy separation and LOX separation. • Development of the VT air separation technology to achieve the required specifications of in-flight LOX separation system for futuristic launch vehicles. With these specific objectives and motivations, the total work was carried out with the following planned and sequential steps: • The first step was the CFD modeling of the VT with the available CFD software (Star-CD) and obtain the energy separation phenomena for a 12mm diameter VT. After gaining sufficient confidence level, optimization of the critical parameters like the air injection nozzle profile, number of nozzles, cold end orifice diameter dc, length to diameter (L/D) ratio, hot gas fraction etc of the VT was carried out through CFD and experimental studies. • The studies show that 6 convergent nozzles perform better in comparison to other configurations like circular helical, rectangular helical, 2 convergent and 6 straight nozzles. The studies also show that cold end orifice diameter (dc) plays an important role on energy separation and bring out the existence of secondary circulation flow with improper design of cold end orifice diameter. Through our studies, the effect of cold end diameter on the secondary circulation flow has been evaluated for the first time. Also, the mechanism of energy transfer in VT based on heat pump mechanism enabled by secondary circulation flow as suggested by some investigators has been evaluated in our studies. The studies show that cold end orifice diameter dc = 7mm is optimum for 12mm diameter VT, which matches fairly with the correlations given by other investigators. The studies confirms that CFD modeling carried out in this work is capable of selecting the correct dc value for a VT, without resorting to the empirical correlations as a design guide or a laborious experimental programme. • Through the CFD and experimental studies on different length to diameter (L/D) ratios and hot gas fractions, maximum hot gas temperature of 391K was obtained for L/D = 30 with hot gas fraction of 12-15 % and minimum cold gas temperature of 267K for L/D = 35 was obtained for cold gas fraction ≈ 60% (lowest cold gas fraction possible with the present experimental system). • CFD analysis has been carried out to investigate the variation of static and total temperatures, static and total pressures as well as the velocity components of the particles as it progresses in the flow field, starting from the entry through the nozzles to the exit of the VT by tracking the particles to understand the flow phenomenon and energy transfer mechanism inside the VT. The studies indicate that the mechanism of energy transfer from the core flow to the peripheral flow in VT is predominantly occurs by the tangential shear work. Thus the investigations reported in the thesis have given a clear understanding of the contributing mechanism for energy separation in VT, which has been an unresolved issue for long time. The net energy transfer between the core and the peripheral fluid has been calculated analytically and compared with the values obtained by CFD model for VTs of L/D ratios equal to 10 and 30. The net energy transfer by analytical and CFD model for VT with L/D = 10 is 159.87W and 154.2W respectively whereas the net energy transfer by analytical and CFD model for VT with L/D = 30 is 199.87W and 192.3W respectively. The results show that CFD results are in very good agreement with the analytical results and CFD can be used as a tool for optimization of the critical parameters and to analyze the flow parameters and heat transfer analysis for VTs. Also, the net energy transfer between the core and peripheral fluids calculated analytically matches very well with that of the net energy transfer by CFD analysis, without considering the effect of acoustic streaming. Thus acoustic streaming may not be the mechanism of energy separation in VT as suggested by some investigators. • By optimizing the critical parameters of the 12mm diameter straight VT through CFD and experimental studies, LOX separation studies have been carried out using both straight and conical VTs of dc = 7mm and of different L/D ratios for high LOX purity and separation efficiency. It is observed that conical (3º divergence) VTs perform better as compared to straight VTs for LOX separation whereas straight VTs perform better for energy separation. The better performance of conical VT as compared to straight VTs can be attributed to its increased surface area for condensation-evaporation phenomenon of oxygen and nitrogen molecules. Experimental studies have been conducted to evaluate the influence of the inlet pressure and the inlet temperature (liquid fraction) on LOX purity. Studies indicate that for achieving high LOX purity for the studied experimental system, the inlet pressure is to be in the range of 6-6.5bar and there exists a very narrow band of inlet temperature zone in which high LOX purity can be achieved. Experimental studies on VTs show that VT can be optimized suitably either for high LOX purity with low separation efficiency or low LOX purity with high separation efficiency by adjusting the hot end mass fraction accordingly. It is also observed that it is not possible to obtain both high purity and high separation efficiency simultaneously with the single VT. Staging approach has to be adapted to achieve higher LOX purity with higher separation efficiency. By staging the VTs, the enriched air stream (hot end outlet flow) from the first stage of VTs is introduced to the inlet of the second stage of VTs. Experimental studies have been conducted to evaluate the design parameters on staging of VTs. LOX purity of 48% with 89% separation efficiency has been achieved for conical first stage VT of L/D = 25. LOX purity of about 94% with separation efficiency of 84% has been achieved for 50% oxygen content at the inlet of the second stage VT. Similarly, LOX purity of 96% with separation efficiency of 73.5% has been achieved for 60% oxygen content at the inlet of the VT. This is the highest LOX purity and separation efficiency reported so far indicating that, conical VT of optimized diameter, L/D ratio and orifice diameter can yield the hot end flow very close to the target value of futuristic in-flight LOX separation based launch vehicles. The present investigation has focused the optimization of the critical parameters of VTs through CFD and experimental studies. It has also given an insight to the mechanism of energy transfer between the core and peripheral flow in VT by evaluating two of the existing theories on mechanism of energy transfer in VT. The studies also highlighted the fact that custom designed and precision fabricated VTs can be very useful for obtaining maximum / minimum temperatures of fluid flow as well as LOX separation with high purity and high separation efficiency needed for futuristic in-flight LOX separation based space launch vehicles.

Orientadores: José Roberto Nunhez, Nicolas Spogis / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Química / Made available in DSpace on 2018-08-22T07:47:05Z (GMT). No. of bitstreams: 1 Vergel_JoseLuisGomez_M.pdf: 8156840 bytes, checksum: f0b5d7b4b0875d4716285f8ec3ce5fa2 (MD5) Previous issue date: 2013 / Resumo: O uso de malhas não estruturadas recentemente tem recebido uma considerável atenção para discretização do domínio computacional em simulações CFD (Vakili & Esfahany, 2009; Cabrales, et al., 2011). A malha não estruturada (comumente tetraédricas) permite uma adequada adaptação em geometrias irregulares e uma geração automática na maioria dos softwares comerciais em CFD. Na literatura constantemente tem sido sugerido que as malhas hexaédricas são preferidas as tetraédricas, porém esta última pode ser usada, desde que alguns cuidados sejam tomados para proporcionar uma boa representação do fenômeno físico do problema. Alguns trabalhos recentemente publicados mostram que as malhas tetraédricas quando utilizadas com cuidado, podem obter resultados satisfatórios (Spogis & Nunhez, 2009). A maioria dos trabalhos desenvolvidos com uso da simulação CFD para processos de mistura utilizam malha hexaédrica, enquanto os resultados obtidos com malhas tetraédricas não possuem a mesma aprovação (Joaquim Junior et al., 2007), parte deste trabalho esta focado em mostrar que elas podem fornecer bons resultados em simulações de processos de mistura. Outro objetivo desse trabalho é mostrar o efeito do refinamento em malhas tetraédricas. Investigou-se a influência dos elementos prismáticos perto da parede sobre parâmetros globais, tais como o número de potência e número de bombeamento. Foi também realizada uma análise inicial da sensibilidade de alguns modelos de turbulência e esquemas de discretização sobre o campo de fluxo produzido. Além disso, simulações com diferentes refinamentos de malha foram feitas usando a abordagem de múltiplos sistemas de referência (MFR); enquanto em outros casos foi proporcionada uma abordagem de malhas deslizantes (SG) no impelidor. Os esquemas de discretização foram limitados aos esquemas Upwind, High resolution e blend fator, uma vez que esquemas diferentes não funcionam adequadamente com elementos tetraédricos. O software comercial de CFD CFX 14.0 foi utilizado para simular os resultados. Um impelidor de pás inclinadas (PBT 45°, com bombeamento para baixo) foi simulado em regime turbulento. Os resultados obtidos pelas simulações CFD para o número de potência e bombeamento, e perfil da componente axial da velocidade são discutidos e comparados com dados experimentais (Machado et al. et al., 2011). A velocidade foi medida usando a técnica PIV (Particle Image Velocimetry). Uma significativa influencia foi observada em alguns parâmetros na determinação de importantes variáveis nos processos de mistura. De forma geral, a predição do perfil da velocidade é influenciada, de alguma forma, pela resolução da malha, pelo modelo de turbulência e pelo esquema de discretização / Abstract: The use of non-structured meshes has received recently considerable attention for discretization of the computational domain in CFD simulations (Vakili & Esfahany, 2009; Cabrales et al., 2011). The non-structured meshes (the most common are tetrahedral) allow adequate adaptation of irregular geometries and easy use of automated algorithms for mesh generation in most commercial CFD software. It has been consistently suggested in the literature that hexahedral meshes are preferred over tetrahedral. However, they can be used, provided that some care is taken to provide a good representation of the physical phenomena of the problem. Recently published works have been proving that if they are used with care, satisfactory results are obtained (Spogis & Nunhez, 2009). Most of works developed with the use of CFD simulation for mixing processes use hexahedral and to date tetrahedral do not have the same acceptance of hexahedral (Joaquim Junior et al., 2007) and part of this work is aimed at showing that they can provide good results when simulating mixing processes. Another objective of this work is to show the effect of mesh refinement in tetrahedral meshes. It is investigated the influence of the prismatic elements near wall on global parameters such as the Power and Flow numbers. An initial analysis of the sensitivity of some turbulence models (Standard k-? model, the shear stress transport (SST) and BSL-Reynolds stress model) and the discretization scheme on the flow field produced were also carried out in this work. In addition, simulations with different mesh resolutions were made using the multiple reference frame (MFR) approach; whereas other cases a sliding mesh modeling was provided for the impeller. The discretization schemes were limited to upwind, high resolution and blend factor, since other schemes do not work well with tetrahedral. The commercial CFD software CFX 14.0 was used to simulate the results. A pitched blade turbine (PBT 45°, down-pumping), was simulated in turbulent flow. The results obtained for the Power and Flow numbers, the axial velocity profile component of the model in CFD are discussed and compared with experimental data (Machado et al. et al., 2011). The velocity was measured using the PIV technique (Particle Image Velocimetry). A significant influence of some parameters is observed on the determination of some important variable mixing. In general, the predictions of the velocity profile are influenced in some way by the resolution of the mesh, turbulence model and the discretization scheme / Mestrado / Desenvolvimento de Processos Químicos / Mestre em Engenharia Química

Orientador: Luiz Otávio Saraiva Ferreira / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica / Made available in DSpace on 2018-08-27T02:13:59Z (GMT). No. of bitstreams: 1 Oliveira_FabiolaMartinsCamposde_M.pdf: 13513306 bytes, checksum: fd54d4a28df68e3217dbd8316b6aeadd (MD5) Previous issue date: 2015 / Resumo: Recentes avanços na tecnologia de processadores multinúcleos e vários-núcleos popularizaram a computação paralela, acelerando a execução de programas e possibilitando a simulação de domínios maiores. Dentre os problemas complexos que requerem alta velocidade de processamento, os problemas de fluidodinâmica computacional se destacam, já que suas simulações tendem a ter um alto custo computacional e exigem grandes domínios de simulação. O método baseado nas Redes de Boltzmann é um método de fluidodinâmica computacional apropriado para o uso de paralelismo que vem ganhando destaque na comunidade científica. Embora haja trabalhos que explorem o paralelismo em GPU nesse método, um simulador eficiente na execução e visualização interativa ainda não foi explorado adequadamente. Assim, a proposta deste trabalho é implementar em GPU um simulador interativo de fluidodinâmica com o método das Redes de Boltzmann. Inicialmente, o simulador foi desenvolvido em linguagem C e foi paralelizado em CPU usando MPI. Em seguida, foi paralelizado em GPU usando CUDA e convertido para linguagem orientada a objetos em C++. Depois, a visualização interativa foi acrescentada utilizando técnicas como interoperabilidade entre CUDA e OpenGL, texturização 3D, fluxo programável da GPU, além de funções de interação com o usuário. O simulador foi validado para casos 2D e 3D em fluxos monocomponentes monofásicos. Além disso, para demonstrar o ganho de desempenho em velocidade de processamento de problemas paralelizados em relação a execuções sequenciais, um conjunto de testes com tamanhos crescentes de domínio foi desenvolvido. O resultado dessa comparação indicou que o simulador implementado em GPU com visualização interativa teve desempenho 71.3 vezes maior em relação à versão sequencial em CPU sem visualização interativa. Dessa forma, a abordagem de paralelização em GPU com visualização interativa mostrou-se muito adequada à execução de simulações fluidodinâmicas, sendo uma ferramenta útil no estudo de escoamentos fluídicos, capaz de executar inúmeros cálculos e lidar com a grande quantidade de memória exigida por simulações fluídicas / Abstract: Recent advances on multicore and many-core processor technology have popularized the parallel computing, accelerating program execution and enabling the simulation of larger domains. Within the complex problems that require a high processing speed, the computational fluid dynamics problems stand out, since their simulations tend to have high computational cost and demand large simulation domains. The method based on the Lattice Boltzmann is an appropriate computational fluid dynamics algorithm to explore parallelism that has been noteworthy in scientific community. Although there are several works that approach GPU parallelism in this method, an efficient simulator implementation and interactive visualization have not been explored adequately. Thus, the purpose of this work is to implement in GPU an interactive fluid dynamics simulator based on the Lattice Boltzmann method. First, the simulator was developed in C language and was parallelized in CPU using MPI. Next, it was parallelized in GPU using CUDA and converted into C++ object-oriented language. Then, the interactive visualization was added using techniques such as CUDA-OpenGL interoperability, 3D texturing, GPU programmable pipeline, and interaction features. The simulator was validated for 2D and 3D cases in single component, single phase flows. Besides that, to show the performance gain in processing velocity of parallelized problems in relation to sequential executions, a test set with increasing domain sizes was developed. This comparison result indicated the GPU-implemented interactive simulator was 71.3 times faster in relation to sequential CPU version without interactive visualization. Thereby, the GPU parallelization approach with interactive visualization showed to be very adequate to fluid dynamics simulations, being a useful tool in fluid flow study, capable of simulating numerous calculations and dealing with the large amount of memory required in fluidic simulations / Mestrado / Mecanica dos Sólidos e Projeto Mecanico / Mestra em Engenharia Mecânica

Die Veranstaltung widmete sich mit der Borverdünnung in Druckwasserreaktoren bzw. mit der Verstopfung der Sumpfansaugsiebe durch freigesetztes Isolationsmaterial schwerpunktmäßig zwei Themen der Reaktorsicherheit, die auch in aktuellen Aufsichtsverfahren eine Rolle spielen. Eingebettet in den internationalen Kontext wollten die Veranstalter die sicherheitstechnische Bedeutung dieser Themen für die deutschen Anlagen beleuchten und die Auswirkungen auf die zu erbringenden Sicherheitsnachweise und den Anlagenbetrieb darstellen. Dabei kamen Gutachter, Vertreter der Forschung, Hersteller und Betreiber gleichermaßen zu Wort. Der Fachtag sollte den Teilnehmern aber insbesondere vermitteln, welche Beiträge die privat und öffentlich finanzierte Reaktorsicherheitsforschung zur Aufklärung der jeweiligen Ereignisabläufe und ihrer sicherheitstechnischen Bedeutung geleistet hat. In diesem Forschungskontext spielen, auch international, die Methoden der so genannten Computational Fluid Dynamics (CFD) eine zunehmende Rolle. Deshalb widmete sich eine Sitzung den Grundlagen, Möglichkeiten und Grenzen von CFD-Methoden. Dabei wurden u.a. Anwendungen zur Borvermischung und zum Verhalten von Mineralwolle im Sumpf präsentiert.

Le monoxyde de carbone déshydrogénase (CODH) catalyse la réduction réversible du dioxyde de carbone par son site actif. En utilisant une méthode électrochimique appelée voltammétrie de film protéique, nous étudions le mécanisme enzymatique de CODH en immobilisant l'enzyme à une surface d'électrode de graphite où le transfert direct d'électrons est possible. Traditionnellement, pour empêcher la déplétion du substrat à l'électrode, les électrochimistes utilisent des électrodes tournantes (RDE). Cependant, comme la CODH est très active, même la RDE ne peut pas empêcher l'épuisement, ce qui masque les caractéristiques cinétiques importantes de l’enzyme. Nous ne pouvons pas résoudre le problème avec RDE, puisque nous l’utilisons déjà à la vitesse maximum. Par conséquent, nous devons concevoir une nouvelle cellule électrochimique. Pour cela, nous avons utilisé des simulations de dynamique des fluides computationnelles pour explorer diverses géométries afin d'en trouver une appropriée. Nous avons commencé par valider notre méthode numérique avec la solution théorique bien définie de la cellule réelle de RDE. Après la bonne validation, nous avons déterminé les vitesses de transport de masse au sein de plusieurs géométries et à basé sur l'optimisation des paramètres géométriques, nous avons atteint notre conception appropriée. Ce nouveau prototype a une électrode graphite uniformément accessible avec un taux de transport trois fois plus rapide que le RDE à des vitesses de solution acceptables. Nous avons construit, mis en place avec succès le système pour caractériser ses performances de transport, et trouvé un excellent accord entre les résultats numériques et expérimentaux / Carbon monoxide dehydrogenase (CODH) catalyzes the reversible reduction of carbon dioxide by its active site. Thus, CODH participates in the first step of fuel production. Using an electrochemical method called protein film voltammetry, we study the enzymatic mechanism of CODH by immobilizing the enzyme at a graphite electrode surface where direct electron transfer is possible. Traditionally, to prevent depletion of the substrate at the electrode, electrochemists use rotating electrodes (RDE). However, since CODH is very active, even RDE cannot prevent depletion, which masks the important kinetic characteristics of the enzyme and complicates the analysis of the enzymatic response.We cannot solve the problem with RDE, since we already use it at maximum speed. Therefore, we must completely change our approach and design a new electrochemical cell. For this, we used computational fluid dynamics (CFD) simulations to explore various geometries to find a suitable one. We began by validating our numerical method with the well-defined theoretical solution of the real cell of RDE. After good validation, we determined the mass transport velocities within several proposed geometries of the flow cell of hydrodynamic channel and jet electrodes. Based on the optimization of geometric parameters, we have achieved our proper design of jet electrode. This new prototype has a uniformly accessible graphite electrode with a transport rate three times faster than the RDE at acceptable solution speeds. We have successfully built and implemented the system to characterize its transport performance. We found an excellent agreement between the numerical and experimental results