Abstract In ladle metallurgy, gas stirring and the behaviour of the slag layer are very important for alloying and the homogenization of the steel. When gas is injected through a nozzle located at the bottom of the ladle into the metal bath, the gas jet exiting the nozzle breaks up into gas bubbles. The rising bubbles break the slag layer and create an open-eye. The size of the open-eye is very important as the efficiency of the metal-slag reactions depend on the interaction between the slag and steel created during the stirring process, and information about the position and size of the open-eye is important for effective alloying practice. Moreover, the open-eye has an effect on the energy balance since it increases heat losses. In this study, experimental measurements and numerical simulations were performed to study the effect of different operating parameters on the formation of the open-eye and mixing time in a water model and industrial ladle. Experimental measurements were performed to study the effect of the gas flow rate, slag layer thickness, slag layer densities and number of porous plugs in a 1/5 scale water model and in a 150-ton steelmaking ladle. For numerical modelling, a multi-phase volume of fluid (VOF) model was used to simulate the system including the behaviour of the slag layer. The numerical simulation of the open-eye size and mixing time was found to be in good agreement with the experimental data obtained from the water model and data obtained from the industrial measurements. / Tiivistelmä Senkkametallurgiassa kaasuhuuhtelu ja kuonakerroksen käyttäytyminen ovat tärkeitä teräksen seostamisen ja homogenisoinnin näkökulmasta. Senkan pohjalla sijaitsevasta suuttimesta puhallettava kaasu hajoaa kupliksi, jotka rikkovat kuonakerroksen ja muodostavat avoimen silmäkkeen. Avoimen silmäkkeen koko on yhteydessä voimakkaampaan kuonan emulgoitumiseen, joka tehostaa metallisulan ja kuonan välisiä reaktioita. Tietoa avoimen silmäkkeen paikasta ja koosta tarvitaan myös tehokkaaseen seostuspraktiikkaan. Avoin silmäke vaikuttaa lisäksi prosessin energiataseeseen lisäten sen lämpöhäviöitä. Tässä tutkimuksessa tutkittiin kokeellisesti ja laskennallisesti erilaisten operointiparametrien vaikutusta avoimen silmäkkeen muodostumiseen vesimallissa ja terässenkassateollisessa senkassa. Kokeellisia mittauksia tehtiin kaasuhuuhtelun, kuonakerroksen paksuuden, ja suuttimien määrän vaikutuksen tutkimiseksi 1/5-mittakaavan vesimallissa ja 150 tonnin terässenkassa. Numeerisessa mallinnuksessa systeemin ja siihen lukeutuvan kuonakerroksen käyttäytymisen simuloimiseen käytettiin volume of fluid (VOF) –monifaasimenetelmää. Avoimen silmäkkeen kokoon ja sekoittumisaikaan liittyvien numeeristen simulointien havaittiin vastaavan hyvin vesimallista ja teollisista mittauksista saatua kokeellista aineistoa.

Water management is a serious issue that affects the performance and durability of PEM fuel cells. It is known, from previous experimental investigations, that surface wettability has influence on water behaviour and fuel cell performance. This finding has lead researchers to develop numerical tools for further investigation of the liquid water behaviour in gas channels. The Volume-of-Fluid (VOF) method has been used in a wide range of studies for its advantage of showing the multi-phase interface in a Computational Fluid Dynamics (CFD) simulation to understand liquid water behaviour in gas channels. In this thesis, numerical study has been carried out to examine the behaviour of liquid water in gas channels. The dynamic movement of the liquid water in the channel and the associated pressure drop, water saturation and water coverage of the GDL have been investigated. Firstly, flow diffusion into the GDL was examined to determine its effect on liquid droplet behaviour in a small section of a gas channel. Furthermore, the effects of the percentage of flow diffusion, GDL wettability, pore size, and water inlet velocity were investigated. Fluid diffusion into GDL found to have insignificant impact on liquid water behaviour so further investigations has been carried with a solid GDL surface. Secondly, gas channel geometry effect on liquid water behaviour was studied. Square, semicircle, triangle, trapezoid with a long base and trapezoid with a short base were compared to find suitable cross section geometry to carry wall wettability investigations. Among the examined geometries, the square cross section showed reasonable results for both scenarios of geometry design, fixed Reynolds number and fixed GDL interface. The effect of wall wettability was assessed by comparing nine different wall/GDL wettability combinations for straight and bend channels. Wall wettability found to have an impact on liquid water behaviour but not as much as GDL wettability. It affects liquid water saturation in the channel by a great deal by accumulating water in the channel edges affecting water behaviour. This was also proven in the last test case of a long channel where water accumulation was investigated by running the calculation until the percentage of water saturation is stabilized. It is also concluded that changing wall wettability from hydrophobic to hydrophilic doubles the percentage of channel occupied by liquid water and increases the time to reach steady state.

Made available in DSpace on 2014-06-11T19:23:39Z (GMT). No. of bitstreams: 0 Previous issue date: 2007-12-04Bitstream added on 2014-06-13T20:50:44Z : No. of bitstreams: 1 minussi_rb_me_ilha.pdf: 1918415 bytes, checksum: 5176f1d27a7b361288abf94e48919e04 (MD5) / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / Denominam-se problemas tipo rompimento de barreira os fenômenos nos quais um fluido é liberado de maneira abrupta. Quando o fluido apresenta natureza hiperconcentrada, a relação entre a tensão de cisalhamento e a taxa de deformação pode se tornar não-linear, passando a apresentar reologia não-Newtoniana. Problemas deste tipo podem ser encontrados em muitos fenômenos tanto na natureza quanto em processos industriais. O estudo de tal problema é, geralmente, conduzido usando simplificações, como a aproximação de águas rasas e a separação do escoamento em regimes dominantemente inerciais ou viscosos. O presente trabalho é composto de duas partes, uma experimental e outra, numérica. No campo experimental, duas soluções controladas são usadas: soluções aquosas de açúcar e de Carbopol 940, esta última com várias concentrações volumétricas. O aparato experimental consiste em um canal retangular de acrílico, contendo uma comporta, a montante da qual, o fluido é retido e, pela ruptura (levantamento da comporta), começa a escoar. O escoamento é estudado através de técnicas avançadas de filmagem. No campo numérico, são realizadas simulações usando o programa CFX, no qual é usado um método de rastreamento de interface, o VOF e sem o emprego das simplificações citadas. Os resultados experimentais são comparados com os numéricos e com resultados da literatura que usam tais simplificações. Na comparação a aproximação de águas rasas, apesar de descrever bem a forma da interface, se distancia dos valores reais da posição da frente de onda. / The dam break problem describes a phenomenon in which there is an abrupt release of fluid. When the fluid is hiperconcentrated, the relation between the shear stress and the strain rate can become non-linear, and so present a non-Newtonian rheology. The non-Newtonian dam break problem may be found in many phenomena in nature and industrial process. The study of such a problem is, generally, conducted using simplified hypothesis such as the shallow water approximation and the separation of the flow in inertial and viscous dominated regimes. The present work is composed of two parts, one experimental and other, numerical. In the experimental field, two controlled solutions were used: water solutions of sugar and of Carbopol 940, the last one with a wide range of volume concentrations. These fluids have, respectively, Newtonian and non-Newtonian rheologies. The experimental setup consists of an acrylic rectangular channel, which has a dam and upstream of that the fluid is retained and, by the rupture, it begins to flow. The flow is studied by using advanced filming techniques. In the numerical field, simulations are conducted using the CFX software, which uses an interface tracking method, the VOF, and without the shallow water approximation and the division of the flow. So the experimental, numerical and literature results, that uses such simplifications, are compared and it is showed that the shallow water approximation, however describes very well the shape of the surface, is not accurate in calculate the wave front position.

Tsunamis are one of the most catastrophic natural events impacting coastal regions often generated by undersea earthquakes. Nevertheless, in enclosed basins, i.e., fjords, reservoirs and lakes, subaerial or submarine landslides can initiate devastating tsunamis with similar consequences. Although a subaerial or submarine landslide that impinges into a large water body can generate a tsunami, subaerial landslides are much more efficient tsunami generators than its counterpart. In this study we aim to integrate laboratory scale experiments of tsunami generation by subaerial landslide with numerical models. The work focuses on the numerical validation of two three-dimensional Navier-Stokes (3D-NS) models, FLOW-3D and our developed model TSUNAMI3D. The models are validated based on previous large scale laboratory experiments performed by a tsunami research team lead by Dr. Hermann Fritz, Georgia Institute of Technology. Three large scale landslide scenarios were selected from the set of laboratory experiments, namely, fjord like, headland and far field coastline. These scenarios showed that complex wave fields can be generated by subaerial landslides. The correct definition and evolution of the wave field are key to accurate modeling the ensuing tsunami and its effect in coastal regions. In this study, comparisons are performed between numerical results and laboratory experiments. Methodology and key parameters for soil rheology are defined for model validations. Results of the models are expected to be under the allowable errors indicated by the National Tsunami Hazard Mitigation Program (NTHMP), National Oceanic and Atmospheric Administration (NOAA) guidelines for validation of tsunami numerical models. The ultimate goal of this research is to obtain better tsunami calculation tools for real-world application of 3-D models for landslide tsunamis, which are necessary for the construction of inundation maps in the Gulf of Mexico and the Caribbean regions.

Orientador: Geraldo de Freitas Maciel / Banca: Sérgio Said Mansur / Banca: Jean Paul Vila / Resumo: Denominam-se problemas tipo rompimento de barreira os fenômenos nos quais um fluido é liberado de maneira abrupta. Quando o fluido apresenta natureza hiperconcentrada, a relação entre a tensão de cisalhamento e a taxa de deformação pode se tornar não-linear, passando a apresentar reologia não-Newtoniana. Problemas deste tipo podem ser encontrados em muitos fenômenos tanto na natureza quanto em processos industriais. O estudo de tal problema é, geralmente, conduzido usando simplificações, como a aproximação de águas rasas e a separação do escoamento em regimes dominantemente inerciais ou viscosos. O presente trabalho é composto de duas partes, uma experimental e outra, numérica. No campo experimental, duas soluções controladas são usadas: soluções aquosas de açúcar e de Carbopol 940, esta última com várias concentrações volumétricas. O aparato experimental consiste em um canal retangular de acrílico, contendo uma comporta, a montante da qual, o fluido é retido e, pela ruptura (levantamento da comporta), começa a escoar. O escoamento é estudado através de técnicas avançadas de filmagem. No campo numérico, são realizadas simulações usando o programa CFX, no qual é usado um método de rastreamento de interface, o VOF e sem o emprego das simplificações citadas. Os resultados experimentais são comparados com os numéricos e com resultados da literatura que usam tais simplificações. Na comparação a aproximação de águas rasas, apesar de descrever bem a forma da interface, se distancia dos valores reais da posição da frente de onda. / Abstract: The dam break problem describes a phenomenon in which there is an abrupt release of fluid. When the fluid is hiperconcentrated, the relation between the shear stress and the strain rate can become non-linear, and so present a non-Newtonian rheology. The non-Newtonian dam break problem may be found in many phenomena in nature and industrial process. The study of such a problem is, generally, conducted using simplified hypothesis such as the shallow water approximation and the separation of the flow in inertial and viscous dominated regimes. The present work is composed of two parts, one experimental and other, numerical. In the experimental field, two controlled solutions were used: water solutions of sugar and of Carbopol 940, the last one with a wide range of volume concentrations. These fluids have, respectively, Newtonian and non-Newtonian rheologies. The experimental setup consists of an acrylic rectangular channel, which has a dam and upstream of that the fluid is retained and, by the rupture, it begins to flow. The flow is studied by using advanced filming techniques. In the numerical field, simulations are conducted using the CFX software, which uses an interface tracking method, the VOF, and without the shallow water approximation and the division of the flow. So the experimental, numerical and literature results, that uses such simplifications, are compared and it is showed that the shallow water approximation, however describes very well the shape of the surface, is not accurate in calculate the wave front position. / Mestre

It is well known that surface tension of a liquid has a decisive role in flow dynamics and the eventual equilibrium state, especially in confined flows under low gravity conditions and also in free surface flows. One such instance of a combination of these two cases where surface tension plays an important role is in the microgravity environment of a spacecraft propellant tank. In this specific case both propellant acquisition and residual propellant estimation are critical to the mission objectives particularly in the end-of-life phase. While there have been a few studies pertaining to the equilibrium state in given geometric configurations, the transient flow leading to final state from an initial arbitrary distribution of propellant is rarely described. The present study is aimed at analysing the dynamic behaviour of the liquids under reduced gravity through numerical simulation and also addresses the specific case of propellant flow transient in a cone-in-a-sphere type of tank configuration proposed by Lal and Raghunandan which is likely to result in both improved acquisition and life time estimation of spacecraft. While addressing this specific problem, the present work aims to study the transient nature of such surface tension driven flows in a general form as applicable to other similar problems also. Volume of Fluid (VOF) method for multiphase model in ANSYS FLUENT was adapted with suitable changes for generating numerical solutions to this problem. Simulations were run for three different cone angles of 17o, 21o & 28o with a flat liquid surface for full scale models to measure the rise height and time of rise. Two scaled models of ½ and 1/10th of the original dimensions with the same liquid configuration of the 28o cone angle case were simulated to see if the time scales involved would come down for experimental feasibility. A third simulation of the 1/10th scale model was run with the liquid spread in the tank to imitate the general conditions found in the propellant tank in microgravity. To understand the behaviour of liquids in the microgravity state to changing physical parameters, a set of simulations was run using liquid phases as water and hydrazine with different physical parameters of temperature and surface tension. The theory put forward by Lal and Raghunandan was found to stand firm. In the case of the cone angle of 28o it was observed that in the final equilibrium state the liquid collected towards the apex of the cone with the larger volume fraction of liquid accumulating inside the cone. An addition of a cylindrical section at the bottom of the cone seems to help although not uniformly for all case. The equilibrium settling times for all the three cone angle cases were in the order of 300 to 600 seconds for simulations on a spherical tank of diameter two metres which was close to the actual tank dimension used on spacecraft. Scaled down simulations of 1/10th and ½ the tank geometry with both flat liquid surfaces and spread out liquid volumes showed that the smaller models had equilibrium settling times which were considerably lower (in the order of tens of seconds) than the full scale models. Although smaller, these time scales are larger than the maximum time scales available in drop tower tests which provide a maximum free fall time of around 9 to 10 seconds. Validation of the proposed configuration by flying an aircraft in a parabolic flight path is a possibility that could be explored for the scaled down models since the zero-g duration for these flights is on an average between 15-20 seconds.

L’émergence de bruits auparavant inaudibles dans les réservoirs à carburants automobiles requiert des constructeurs une meilleure compréhension des phénomènes physiques intervenants au sein de leurs produits. Dans cette thèse, différents travaux ont été conduits autour de l’étude du ballottement de fluide dans une cuve rigide rectangulaire partiellement remplie de fluide et soumise à une excitation extérieure. La première partie présente un état de l’art sur le sloshing suivant trois approches complémentaires - approche analytique, approche numérique et approche expérimentale - permettant d’orienter les travaux. Dans une deuxième partie, une étude préliminaire sur le sloshing dans une cuve rectangulaire soumise à une excitation harmonique forcée est réalisée. La confrontation des résultats numériques entre une approche linéaire - basée sur la théorie d’écoulement potentiel tenant compte de la viscosité du fluide [Schotté et Ohayon, 2013] - et une approche non linéaire commerciale – basée sur la résolution des équations de Navier-Stokes - permet de définir un paramètre de linéarité. Ce dernier permet de déterminer les cas de sloshing qui nécessitent une résolution non linéaire et ceux pour lesquels la théorie linéaire suffit pour prédire le phénomène. La troisième partie de ce document présente une étude expérimentale du ballottement de fluide dans une cuve rectangulaire rigide soumise à un freinage automobile. Deux niveaux de remplissage créant deux types d’impacts contre les parois (avec et sans enfermement de poche d’air) ont été analysés. Les essais menés ont permis de mesurer les forces engendrées par le mouvement du fluide, les pressions d’impact en paroi ainsi que le champ de vitesse par méthode Particle Image Velocimetry (PIV). Ce chapitre constitue une importante base de données expérimentales ayant permis d’étudier précisément le phénomène physique. L’étude est complétée par une confrontation des résultats expérimentaux avec des résultats Computational Fluid Dynamics (CFD). Enfin, pour conclure ce mémoire, une étude du sloshing dans un réservoir en tenant compte de la Fluid-Structure Interaction (FSI) est présentée. Le choix du couplage a été porté sur un schéma partitionné itératif faible avec, dans un premier temps, une approche potentielle instationnaire, puis avec une approche Volume Of Fluid (VOF) pour la physique fluide. Les limites d’un tel couplage dans le cas d’étude d’un réservoir partiellement rempli de fluide et attaché de manière flexible en fonction du rapport de masse fluide-réservoir ont été mises en évidence. La correction du schéma de couplage par l’effet de masse ajoutée présentée dans [Song et al., 2013] permet la résolution d’un système couplé quel que soit le rapport de masse en jeu et améliore de manière significative la convergence en réduisant également fortement le temps de calcul. / The present thesis focuses on an investigation of the sloshing phenomenon in a partially filled fuel tank submitted to a harmonic excitation motion. In the first part, the confrontation of numerical results between a linear approach - taking into account viscosity - and a nonlinear approach based on a commercial code leads to define a parameter of linearity. This parameter allows determining cases of sloshing who require non-linear resolution and those who need a linear theory to predict the phenomenon. An experimental study of fluid sloshing in a rectangular tank submitted to an automotive braking is conducted. Tests leaded allow measuring global forces engendered by the motion of the fluid, pressure of fluid impact and velocity field by PIV. This chapter provides an important data base and helps to investigate on the physical phenomenon. This study is completed by CFD results. To conclude, a numerical model for fluid-structure interactions is presented. Limits of this segregated partitioned coupling in case of sloshing in tank flexibly attached are highlighted, depending mostly on the mass ratio between fluid and tank structure. An added-mass term is integrated to the corrected staggered scheme ensuring systematically the convergence of the coupled solution and reducing significantly the iterations required.

High speed planing hulls are currently widely used for example in recreational and emergency vessel applications. However, very little CFD research has been done for planing vessels, especially for those with stepped hulls. A validated CFD method for planing stepped hulls could be a valuable improvement for the design phase of such hulls. In this thesis, a CFD method for stepped hulls, with a primary focus on two-step hulls, is developed using STAR-CCM+. As a secondary objective, porpoising instability of two-step hulls is investigated. The simulations are divided into two parts: In the first part a method is developed and validated with existing experimental and numerical data for a simple model scale planing hull with one step. In the second part the method is applied for two two-step hulls provided with Hydrolift AS. A maximum two degrees of freedom, trim and heave, are used, as well as RANS based k-w SST turbulence model and Volume of Fluid (VOF) as a free surface model. The results for the one-step hull mostly corresponded well with the validation data. For the two-step hulls, validation data did not exists and they were first simulated with a fixed trim and sinkage and compered between each other. In the simulations with free trim and heave both hulls experienced unstable porpoising behavior.

Early in the ship design process, naval architects must often evaluate and compare multiple hull forms for a specific set of requirements. Analytical tools are useful for quick comparisons, but they usually specialize in a specific hull type and are therefore not adequate for comparing dissimilar hull types. Scale model hydrodynamic testing is the traditional evaluation method, and is applicable to most hull forms. Scale model tests are usually performed on the largest model possible in order to achieve the most accurate performance predictions. However, such testing is very resource intensive, and is therefore not a cost effective method of evaluating multiple hull forms. This thesis explores the testing of small scale models. It is hypothesized that although the data acquired by these tests will not be accurate enough for performance predictions, they will be accurate enough to rank the performance of the multiple hull forms being evaluated.

[ES] La creciente preocupación por los problemas medioambientales, la disponibilidad de combustibles fósiles unido a la gran demanda de vehículos, han llevado a los gobiernos a regular las emisiones emitidas a la atmósfera. Existen propuestas de adoptar fuentes de energía renovables. Sin embargo, la sustitución de los combustibles derivados del petróleo no será fácil, rápida o rentable, y el transporte propulsado por motores de combustión interna (ICE) seguirá destacando en los próximos años. La eficiencia de la combustión y el rendimiento del motor están influenciados por el complejo proceso de inyección. La inyección directa de gasolina (GDI) aumenta el ahorro de combustible y cumple los requisitos de emisiones contaminantes, aunque queda potencial por descubrir. Por ello, ha sido objeto de estudio en los últimos años y, en consecuencia, de la presente Tesis. Este trabajo tiene como motivación mejorar el entendimiento en el campo del GDI. La compleja naturaleza transitoria del proceso de inyección hace que el estudio experimental sea un desafío. La Mecánica de Fluidos Computacional (CFD) surge como una potente alternativa a los experimentos y ha sido adoptada para esta investigación. Bajo este contexto, el objetivo de la presente Tesis es desarrollar una metodología predictiva para la caracterización hidráulica del inyector, capaz de ser aplicada a las actuales y futuras generaciones de inyectores GDI, independientemente de las características del inyector y del software de estudio. Una vez validada, el objetivo posterior es utilizar los resultados para analizar el comportamiento del chorro. Este enfoque busca seguir los pasos de la comunidad científica sustituyendo la práctica experimental. La validación de la metodología se lleva a cabo mediante su aplicación en dos inyectores GDI solenoides multi-orificio diferentes. Además, se han utilizado dos códigos CFD comerciales: CONVERGE y StarCCM+. La metodología predictiva se centra en el estudio del flujo interno y el campo cercano para caracterizar hidráulicamente el inyector. El problema a tratar se define como un sistema multifásico en un marco Euleriano y considerando un único fluido. El tratamiento del flujo multifásico se realiza mediante el enfoque Volume-of-Fluid (VOF). Además, se emplea el Homogeneous Relaxation Model (HRM) para considerar el intercambio de masa entre las fases líquida y vapor debido a cavitación y flash boiling. La turbulencia se ha tratado a partir de los enfoques Reynolds-Averaged Navier-Stokes (RANS) y Large Eddy Simulations (LES). Por otro lado, en cuanto al estudio del flujo externo, se ha adoptado el Discrete Droplet Model (DDM). La atomización y el chorro están influenciados por la geometría de la tobera, por lo que la estrategia de acoplamiento del flujo interno y externo complementa los análisis. Se han adoptado enfoques de acoplamiento unidireccional y mapeado, utilizando como parámetros de entrada los datos de flujo interno de la validada metodología. Esta Tesis aporta una nueva y valiosa metodología predictiva con una elevada precisión a la hora de caracterizar el proceso de inyección en comparativa con datos experimentales. Por otro lado, es directamente trasferible a distintos códigos de cálculo así como aplicable a inyectores con características dispares sin perjudicar las exigencias del modelo. La correcta caracterización del flujo interno ha permitido emplear los datos obtenidos para analizar el comportamiento del chorro eliminando la necesidad de usar datos experimentales. Los resultados obtenidos capturan el comportamiento macroscópico del chorro con una precisión comparable a los experimentos. Aunque todavía hay muchos retos que afrontar, la presente Tesis supone un gran avance en el campo del GDI. El remarcable progreso se debe al desarrollo y uso de una metodología totalmente predictiva, que permite prescindir de la mayoría de los experimentos para contribuir a una mayor y más amplia visión de la física del proceso de inyección. / [CA] La creixent preocupació pels problemes ambientals, la limitada disponibilitat de combustibles fòssils, acompanyat a la gran demanda de vehicles, ha portat el govern a regular els nivells d'emissions emesos a l'atmosfera. Existeixen propostes d'adoptar fonts d'energia renovables. Tanmateix, la substitució dels combustibles líquids derivats del petroli no es durà a terme de forma fàcil, ràpida o rentable, i el transport propulsat per motors de combustió interna (ICE) continuarà destacant en els pròxims anys. L'eficiència de la combustió i el rendiment del motor són fortament influenciats pel complex procés d'injecció. La injecció directa de gasolina (GDI) augmenta l'estalvi de combustible i complix amb els requisits d'emissions, encara que queda molt potencial per descobrir. Per això, aquest ha sigut objecte d'investigació en els últims anys i, com a conseqüència, d'aquesta Tesi. Aquest treball té com a motivació millorar l'enteniment en el camp del GDI. La complexa natura transitòria de la injecció fa que l'estudi experimental siga força complex. La Mecànica de Fluids Computacional (CFD) sorgeix com una potent alternativa als experiments, i ha sigut adoptada per aquesta investigació. Baix aquest mateix context, es proposa com a objectiu principal d'aquesta Tesi el desenvolupament d'una metodologia predictiva per a la caracterització hidràulica de l'injector, capaç de ser aplicada a les actuals i futures generacions d'injectors GDI (independentment de les característiques de l'injector i del software d'estudi). Una vegada validada, el posterior objectiu és analitzar el comportament de l'esprai. Aquest enfocament busca seguir els passos de la comunitat científica substituint la pràctica experimental. La validació de la metodologia ha sigut duta a terme mitjançant la seva aplicació en dos injectors GDI solenoides multi-orifici. A més, s'han utilitzat dos software CFD comercials: CONVERGE i StarCCM+. La metodologia predictiva se centra en l'estudi del flux intern i el camp proper per tal de caracteritzar hidràulicament l'injector. El problema a tractar es defineix en base a un sistema multi-fàsic en un marc Eulerià i considerant un únic fluid. El tractament del fluid multi-fàsic es realitza mitjançant l'aproximació Volume-of-Fluid (VOF). A més, s'utilitza el Homogeneous Relaxation Model (HRM) per tal de considerar l'intercambi de massa entre les fases líquida i vapor degut als fenòmens de cavitació i flash boiling. La turbulència s'ha tractac a través dels enfocaments Reynolds-Averaged Navier-Stokes (RANS) i Large Eddy Simulations (LES). Pel que fa a l'estudi del fluix extern, s'ha adoptat el Discrete Droplet Model (DDM). Sent conscients que el comportament l'atomització i l'esprai estan influenciats per la geometria de la tovera, l'estratègia d'acoblament del flux intern i extern complementa les anàlisis. S'han adoptat els enfocaments d'acoblament unidireccional i mapejat, utilitzant com a paràmetres d'entrada les dades del flux intern obtingudes amb la validada metodologia. Aquesta Tesi aporta una nova i valuosa metodologia predictiva amb una elevada precisió a l'hora de caracteritzar el procés d'injecció en comparativa amb dades experimentals. És directament transferible a diversos codis de càlcul així com aplicable a injectors amb característiques dispars sense perjudicar les exigències del model. La correcta caracterització del flux intern ha permès utilitzar les dades obtingudes per tal d'analitzar el comportament de l'esprai, eliminant la necessitat d'emprar dades experimentals. Els resultats obtinguts d'aquest estudi capturen el comportament macroscòpic de l'esprai amb una precisió comparable als experiments. Encara que queden molts reptes per afrontar, aquesta Tesi aporta un important avanç al camp del GDI. La ruptura prové del desenvolupament i ús d'una metodologia completament predictiva, que substitueix els experiments requerits i així contribueix a una millor i més ampla visió de la física del procés d'injecció. / [EN] Concerns about climate change, availability of fuel resources and the high demand for vehicles, have led governments to regulate the level of pollution emitted by engines into the atmosphere. There is a strong desire to adopt renewable and sustainable energy sources. However, the substitution of liquid fuels derived from petroleum will not emerge easily, quickly or economically, and Internal Combustion Engines (ICE) will continue to excel for the next few years. Combustion efficiency and engine performance are strongly influenced by the complex fuel injection process. Gasoline Direct Injection (GDI) strategies increase fuel economy and meet emission requirements, although many challenges remain, which has therefore been one of the main research objectives in recent years and of this Thesis. The present research aims to provide a better understanding in the field of GDI. The transient and complex nature of the injection process makes the experimental study of GDI quite challenging. Therefore, Computational Fluid Dynamics (CFD) emerges as a powerful alternative adopted for this research. In this context, the main objective of the present Thesis is to develop a predictive methodology capable of being applied to current and future generations of GDI injectors, regardless of the injector features and the software employed, for the hydraulic characterization of the injector. Once validated, the subsequent goal is to employ the obtained results to analyze the behavior of the spray downstream of the injector. The approach attempts to follow the footsteps of the research community to avoid experimental practice. The predictive methodology has been validated through its application to two multi-hole solenoid GDI injectors with different features. In addition, the mentioned methodology has been evaluated using diverse commercial software: CONVERGE and StarCCM+. The methodology focuses on the study of the internal and near-field flow to hydraulically characterize the injector. So the problem to be addressed is a multi-phase system, performed in an Eulerian framework, modeled through a single-fluid approach. The multi-phase flow is treated by means of the Volume-of-Fluid (VOF) approach. Homogeneous Relaxation Model (HRM) is employed to consider the mass exchange between liquid and vapor fuel phases, due to cavitation and flash boiling. The turbulence treatment has been performed from both Reynolds-Averaged Navier-Stokes (RANS) and Large Eddy Simulations (LES) approaches. Regarding the external flow study, the Discrete Droplet Model (DDM) has been adopted. In addition, being aware that atomization and spray behavior is greatly influenced by the nozzle geometry, the coupling strategy of the internal and external flow complements the analyses. One-way coupling and mapping approaches have been adopted, using as input parameters the internal flow data obtained from the already validated methodology. Accordingly, this Thesis provides a new and valuable predictive methodology, which has demonstrated a high accuracy in characterizing the flow behavior during the injection process through comparison with experimental data. It has also proven to be directly transferable to different CFD software and applicable to injectors with dissimilar characteristics without compromising the requirements of the model. The correct internal flow characterization has made it possible to employ the obtained data to analyze the spray patterns, which eliminates the need to consider experimental data. The outcomes of this study macroscopically capture the jet behavior with an accuracy comparable to experiments under different operating conditions. Although there are still many challenges to face, the present Thesis brings a breakthrough in the field of GDI. The quantum leap arises from the development and use of a fully predictive methodology, allowing to avoid most experiments to contribute to a greater and broader vision of the injection process physics. / María Martínez García has been founded through a grant from the Government of Generalitat Valenciana with reference ACIF/2018/118 and financial support from the European Union. These same institutions, Government of Generalitat Valenciana and the European Union, supported through a grant for pre-doctoral stays out of the Comunitat Valenciana with reference BEFPI/2020/057 the research carried out during the stay at Aerothermochemistry and Combustion Systems Laboratory, Swiss Federal Institute of Technology, ETH Zurich, Switzerland. Special gratitude from the author to both institutions, Government of Generalitat Valenciana and the European Union, for making this dream possible / Martínez García, M. (2022). Computational Study of the Injection Process in Gasoline Direct Injection (GDI) Engines [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/185180 / TESIS