Este trabalho apresenta uma técnica de diferenças finitas para resolver a equação constitutiva Fluido de Segunda Ordem para escoamentos tridimensionais com superfície livre. As equações governantes são resolvidas pelo método de diferenças finitas em uma malha deslocada 3D. A superfície livre é modelada por células marcadoras (Marker-and-Cell) e as condições de contorno a superfície livre são empregadas. O método numérico apresentado neste trabalho foi validado pela comparação entre as soluções numéricas obtidas para o escoamento em um tubo com a solução analítica correspondente para Fluidos de Segunda Ordem. Ao fazer refinamento de malha, a convergência do método numérico foi verificada. Resultados numéricos da simulação do problema do inchamento do extrudado para números de Deborah De \'< OU =\' 0:3 são apresentados / This work presents a finite difference method to simulate three-dimensional viscoelastic flow with free surfaces governed by the constitutive equation Second Order Fluid. The governing equations are solved by the finite difference method in a three-dimensional shifted mesh. The free surface of fluid is modeled by the Marker-and-Cell method which allows for the visualization and the location of the free surface of fluid. The full free surface stress conditions are employed. The numerical method developed in this work is validated by comparing the numerical and analytic solutions for the steady state flow of a Second Order Fluid in a pipe. By using mesh refinement convergence results are given. Numerical results of the simulation of the transient extrudate swell of a Second Order Fluid of the Deborah number De \'< OR =\' 0:3 are presented

Escoamentos viscoelásticos não estacionários com superfícies livres são comuns em muitos processos industriais e diversas técnicas numéricas têm sido empregadas para reproduzir computacionalmente estes processos. A maioria dos modelos empregados utiliza equações diferenciais na definição do tensor de tensões. Porém, para alguns grupos de fluidos complexos, por exemplo, fluidos de Boger, os modelos integrais mostram-se mais capacitados em fornecer uma boa aproximação para os comportamentos não lineares desses fluidos. Este trabalho trata da solução numérica do modelo constitutivo integral KBKZ-PSM para escoamentos transientes bidimensionais com superfícies livres. O método numérico proposto é uma técnica numérica que utiliza diferenças finitas para simular escoamentos com superfícies livres na presença de paredes sólidas. As principais características do método numérico proposto são: solução das equações de conservação de quantidade de movimento e massa utilizando um método semi-implícito; a condição de contorno na superfície livre é acoplada à equação de Poisson, o que garante conservação de massa; a discretização do tempo t é realizada por uma nova técnica numérica; o tensor de Finger é calculado pelo método dos campos de deformação e avançado no tempo pelo método de Euler modificado. Essa nova técnica é verificada em escoamentos cisalhantes e elongacionais. Adicionalmente, uma solução analítica desenvolvida para escoamentos em canais bidimensionais é empregada para verificar e analisar a convergência do método proposto. Com relação a escoamentos com superfícies livres, a convergência é verificada por meio de refinamento de malha nas simulações de um jato incidindo sobre placa rígida e no problema do inchamento do extrudado. Finalmente, o método é aplicado para investigar os problemas jet buckling e inchamento do extrudado de fluidos KBKZ-PSM. / Unsteady viscoelastic free surface flows are common in many industrial processes and a variety of numerical techniques have been employed to simulate these flows. The majority of constitutive models employed are based on differential equations to define the extra stress tensor. However, for some complex fluids, for instance, Boger fluids, integral models are more adequate to approximate the nonlinear behaviour of these fluids. This work deals with the numerical solution of the integral constitutive model KBKZ-PSM for two-dimensional unsteady free surface flows. The proposed numerical method is a numerical technique that employs finite differences to simulate moving free surface flows that interact with solid walls. The main features of the method are: numerical solution of the momentum and mass equations by an implicit method; the pressure condition on the free surface is implicitly coupled with the Poisson equation for obtaining the pressure field from mass conservation; a novel scheme for defining the past times t is employed; the Finger tensor is calculated by the deformation fields method and is advanced in time by the modified Euler method. This new technique is verified by solving shear and uniaxial elongational flows. Moreover, an analytic solution for channel flows is obtained that is used in the verification and convergence analysis of the proposed methodology. For free surface flows, the assessment of convergence lies on the mesh refinement on the simulation of a jet impinging on a flat surface and the extrudade swell problem. Finally, the new method is applied to investigate the jet buckling phenomenon and extrudate swell of KBKZ-PSM fluids.

Heavy industries, such as mining, generate dust in quantities that present an occupational health hazard. Prolonged exposure to the respirable dust has been found to result in many irreversible occupational ailments in thousands of miners. In underground mining applications, a variety of scrubbing systems are used to remove dust near the zones of generation. However, the wire-mesh type fibrous screens in the flooded-bed dust scrubbers used on continuous miners, are prone to clogging due to the accumulation of dust particles. This clogging results in a reduced capture efficiency and a higher exposure to the personnel. This research establishes the Vortecone, an inertial wet scrubber system, as a suitable alternative to the existing filters. The Vortecone accelerates its inlet fluids into a rapid circulatory motion into a vortex chamber, preferentially moving the heavier particles towards the impermeable surface to be trapped by the circulating water film. Vortecones are used on automobile painting lines and capture over-sprayed paint particles with cleaning efficacies exceeding 99 % while requiring only infrequent maintenance. The existing design of the Vortecone could also be altered to control the flow patterns. This dissertation presents detailed computational fluid dynamics (CFD) models to describe air flow patterns in the Vortecone in steady and transient states. Multi-phase spray models were generated to simulate injection of water into the Vortecone. The volume of fraction (VOF) approach was adopted to mimic the air-water interface. The Lagrangian particle tracking method was used to model particle capture on the interface described by the VOF. The CFD models indicate excellent cleaning efficacies, especially of larger particles. Laboratory experiments with optical measurements of aerosols in a reduced scale model of the Vortecone validate the computer models. These experiments which were performed on dust samples with particle sizes 0.3 μm and above, show that the Vortecone captures 90 % particles by mass exceeding about 5.20 and 3.20 μm at air flows of 0.28 m3/s (600 cfm) and 0.38 m3/s (800 cfm), respectively. The development of detailed large eddy simulations (LES) of air flow in the Vortecone provides a novel contribution to research by better resolving the flow patterns. An impactor-type, self-cleaning, non-clogging impingement screen system was designed as a substitute for conventional screens used in continuous miners. The screen could further be used as an efficient dust capturing mechanism with a demister in general mining applications. CFD models and laboratory experiments are presented to establish the cleaning efficacies of the system. Laboratory experiments to investigate the cleaning efficiency of a fibrous-type conventional screen is also discussed. The parameter, filter selection factor, is proposed to compare the performance of the three systems (Vortecone, fibrous screen, and impingement screen) under similar flows. The Vortecone has been found to be the most efficient dust-cleansing system, although it is the most power intensive fillter. The impingement screen shows a similar cleaning efficiency and a much higher availability compared to the conventional fibrous screen. Because of its minimal maintenance requirement, the impingement screen shows significant promise in dust-control applications in mining.

Micro-assembly processes suffer from some breaches due to the continuing trend towards an increase in the production capabilities as well as in the size reduction of the components manipulated. Usual manipulating schemes have reached their limit and capillary forces constitute a valuable alternative strategy. The goal of this work is to describe the dynamics of liquid bridges in the application of micro-assembly processes. The description is obtained using the Kelvin-Voigt model, with a spring, a damper, and a mass connected in parallel, supported by numerical simulations, analytical approximations and experiments. The works is divided into three parts. First we present important aspects of microfluidics, as well as the constitutive equations and an overview of numerical approaches used to describe fluid flow problems with moving interfaces. The second part is devoted to the capillary rise case, intended to validate and to compare the numerical approaches to analytical laws and experimental results. The implementation of the slipping and the dynamic contact angles is discussed. The last part focuses on the dynamics of the liquid bridge. The liquid bridge is confined between two circular and parallel plates and presents an axial symmetry. The description reveals that the stiffness depends on the surface tension and on the shape of the air/liquid interface, the damping coefficient depends on the viscosity and the volume of liquid and the equivalent mass depends on the density and the volume.

Process industries are in general depending, in one way or the other, on fluid mechanics.Specifically, paper manufacturing, which probably is the dominant processindustry in Sweden, is depending on the flow of cellulose fibres suspended in water.As a part of the process the suspension, consisting of fibres in water, is spread out onor between two moving permeable weaves, i.e. wires. The speed of this is usually 10–30 m/s and the suspension is spread out by a plane jet issuing from a headbox nozzle.It has been show that the conditions in the headbox and jet have a large influence onthe quality of the final paper sheet. Primarily, streaks in the paper sheet are believedto be the result of streamwise streaks in the headbox jet.The thesis is aimed at the flow phenomena which occur in the headbox jet. Theinvestigations have been made with numerical calculations, stability theory and modelexperiments using water, as well as experiments with a real paper machine headboxand fibre suspension. In the thesis an introduction to the hydrodynamics of planeliquid jets is presented together with a description of the paper forming process andthe fluid mechanics of headbox flow.The basic flow and stability of a two-dimensional plane liquid jet has been investigatedby numerical calculations, stability theory and experiments. The calculationsof the laminar basic flow is successfully compared to pitot-tube measurements of thestreamwise velocity profile. By visualisations of the flow it is found that wave disturbanceson the jet has a severe effect on the flow. These waves can be predicted bylinear stability theory, which shows the presence of five convectively unstable modes.These can be divided into three types and by comparison with the experiments thetype of the visible waves is determined. These waves seem to initiate a break-up ofthe jet, which leads to strong streamwise streaks inside the jet.By flow visualisation of headbox flow of an experimental paper machine, togetherwith analysis of the resulting paper structure using the wavelet method the correspondencebetween flow disturbances and paper quality was investigated. It was shownthat the wave instability, which is present on the low Reynolds number water jet, alsocan be found in the real the headbox jet. It is shown that these waves play an importantrole in the dynamics of the headbox jet and also have an influence on the final papersheet. / QC 20100825

A two-dimensional Boundary Element Method (BEM) is developed to study the strongly nonlinear interaction between a surface-piercing body and the free-surface. The scheme is applied to problems with and without the possibility of ventilation resulting from the motion and geometric configuration of the surface-piercing body. The main emphasis of this research work is on the development of numerical methods to improve the performance prediction of surface-piercing propellers by including the whole range of free-surface nonlinearities. The scheme is applied to predict the ventilated cavity shapes resulting from the vertical and rotational motion of a blade-section with fully nonlinear free-surface boundary conditions. The current method is able to predict the ventilated cavity shapes for a wide range of angles of attack and Froude numbers, and is in good agreement with existing experimental results. Through a comparison with a linearized free-surface method, the current method highlights the shortcomings of the negative image approach used commonly in two-dimensional and three-dimensional numerical methods for surface-piercing hydrofoils or propellers. The current method with all its capabilities makes it a unique contribution to improving numerical tools for the performance prediction of surface-piercing propellers. The scheme is also applied to predict the roll and heave dynamics of two-dimensional Floating Production Storage and Offloading (FPSO) vessel hull sections within a potential flow framework. The development of the potential flow model is aimed at validating the free-surface dynamics of an independently developed Navier Stokes Solver for predicting the roll characteristics of two-dimensional hull sections with bilge keels. / text

Le modèle de couche mince intégré sur l'épaisseur, Saint-Venant, utilisé classiquement pour simuler la propagation de laves torrentielles et coulées boueuses, repose sur plusieurs approximations concernant la forme des profils de vitesse en zones non-uniformes. Il est pourtant nécessaire d'utiliser ce type de modélisation, comme outil d'aide à la gestion des risques liés aux laves torrentielles. Nous proposons d'éprouver ses hypothèses, en observant une zone fortement non-uniforme, le front de coulées à surface libre et le champ de vitesse à l'intérieur de cette zone.En améliorant notre connaissance de l'évolution de la forme des profils de vitesse (de la dynamique interne) au front de coulées, nous cherchons à améliorer les modèles de couche mince. Cette thèse porte donc sur l'étude de la dynamique interne au front d'écoulements à surface libre de fluides newtoniens et viscoplastiques.Nous avons utilisé le dispositif du canal à fond mobile qui permet de générer des coulées stationnaires dans le référentiel de l'observateur au moyen d'un fond mobile remontant vers l'amont. Nous avons réalisé un travail technique sur ce canal et sur l'analyse des images pour pouvoir mesurer les champs de vitesse à haute résolution spatiale aux fronts de coulées à surface libre de fluides viscoplastiques. L'étude des fluides newtoniens a aussi été réalisée afin de valider les modèles et éprouver le dispositif expérimental.Nous avons comparé les résultats expérimentaux aux solutions théoriques de deux modèles de couche mince adaptés à la rhéologie de Herschel-Bulkley : le modèle classique de la lubrification, à la base du modèle de Saint-Venant et un modèle consistant à l'ordre 1 développé dans cette thèse. Le modèle consistant d'ordre 1 est la somme du modèle à l'ordre 0 (la lubrification) et de termes correctifs qui proviennent des contraintes normales et des termes d'inertie. Dans le cadre de notre configuration du fond mobile remontant vers l'amont, il est possible de déduire la forme du front en cherchant une solution de type onde progressive, sans passer par un modèle intégré dans l'épaisseur.Pour les fluides viscoplastiques, la structure classique du profil de vitesse, avec une zone cisaillée surmontée d'un plug non cisaillé est bien reconnaissable sur nos profils de vitesse en zone uniforme, et en zone faiblement variée. Mais à l'approche du front, cependant, la vitesse de surface augmente, les profils de vitesse expérimentaux deviennent cisaillés sur toute l'épaisseur, conduisant à la disparition du plug à proximité de la ligne de front.Le modèle de lubrification prédit l’existence d'un plug dans le front jusqu'à la ligne de contact, ce qui n'est pas observé expérimentalement. La vitesse de surface du modèle de lubrification augmente à l'approche du front, mais est largement sous-estimée par rapport à la vitesse de surface mesurée. Les vitesses de surface prédites par le modèle d'ordre 1 augmentent plus drastiquement au front, en meilleur accord avec les mesures que le modèle de lubrification. Pour certaines configurations expérimentales l'accord est même très bon. Remarquablement, le cisaillement des profils de vitesse à l'approche du front, observé expérimentalement, est aussi prédit par le modèle d'ordre 1.Les profils de vitesse présentent donc une évolution au front de coulées viscoplastiques en contradiction avec les hypothèses du modèle de Saint-Venant. Le modèle consistant d'ordre 1 permet d'améliorer les prédictions. Un modèle intégré dans l'épaisseur de type Saint-Venant basé sur les développements consistants d'ordre 1 est alors calculé, car il constitue l'étape nécessaire avant d'être intégré dans un outil de simulation opérationnel. / A depth-averaged model based on the thin-layer assumption, called Saint-Venant (Shallow-Water), is classically used to simulate the propagation and the spreading of debris and mud flows. It is based on several approximations concerning the shape of the velocity profile in non-uniform zones. We propose to test these hypotheses, examining a strongly non-uniform zone, the front of free-surface viscoplastic flows and the velocity field within this zone. By improving our knowledge about the internal dynamics in the front zone, we seek to improve the thin-layer models. This thesis therefore focuses on the study of the internal dynamics within the front of viscoplatic free-surface flows.We used the moving conveyor belt to generate stationary flows. We carried out a technical work on this set-up, and specific analysis of images obtained from the high-speed camera, in order to be able to measure velocity fields with a high resolution. The study of a Newtonian fluid was also carried out in order to validate the lubrication model and the experimental device.We compared experimental results to theoretical solutions of two thin-layer models taking into account the Herschel-Bulkley rheology: the classical model of lubrication, which is at the base of Saint-Venant model, and a consistent first-order model specifically developed in this thesis.The first-order model is equal to the zero-order model (lubrication), plus corrective terms derived from the normal stresses and inertia terms.In this study, for the purpose of comparison with our experimental results, we are interested in travelling-wave solutions. We are able to solve the shape of the front without using a depth-averaged model.Far from the front, experimental velocity profiles clearly display the characteristic 2-layer structure predicted by the lubrication solution, with constant values close to the free-surface (plug) and a sheared layer underneath. Closer to surge tip, the shape of experimental longitudinal velocity profilesthen begins to differ from the theoretical prediction. The 2-layer structure tends to disappear, and the profiles display shear across the whole depth ofthe flow. In this tip region, surface velocity also appears to increase faster than its theoretical counterpart. Surface velocity predicted by the first-order model increase more drastically in the tip region, in better agreement with the measurements than the lubrication model. The first-order model predicts a sheared velocity profile when approaching the front, as observed experimentally.The consistent first-order model then provides better predictions about internal dynamics than lubrication model. A depth-integrated model like Saint-Venant, based on consistent first-order developments is then calculated, as a first step before being integrated into an operational simulation tool.

To quench a thermal runaway reaction in a chemical rector, an efficient approach is the introduction of a small quantity of a liquid inhibiting agent, named a “killer”, into the mixing vessel. In this thesis, an experimental approach has been coupled tightly with numerical modelling using Computational Fluid Dynamics (CFD). The first part of this thesis is devoted to a study of the hydrodynamics of partially-baffled mixing vessels, including the free-surface deformation caused by the central vortex. The use of an inhomogeneous, multiphase approach allowed simulation of the free-surface deformation. The capability of this novel method was demonstrated by very good agreement between the numerical predictions and experimental data. In the second part, liquid jet injection at the free-surface was coupled with the vessel hydrodynamics. Numerical results, obtained using an Eulerian-Lagrangian approach, have again shown good agreement with experimental data. These results allowed the jet trajectory to be modelled and its penetration into the agitated vessel was quantified. New mixing criteria were introduced that are specific to this application. Finally, the numerical methods validated at the pilot scale were applied at the industrial scale and allowed the proposal of practical improvements to the safety of the synthesis reactors studied

O presente trabalho tem por objetivo desenvolver uma formulação numérica baseada em Análise Isogeométrica para o estudo de problemas de interação fluido-estrutura (IFE) em aplicações envolvendo corpos rígidos submersos, onde escoamentos incompressíveis de fluidos Newtonianos com superfície livre são considerados. Propõe-se o emprego da Análise Isogeométrica por permitir a unificação entre os procedimentos de pré-processamento e análise, melhorando assim as condições de continuidade das funções de base empregadas tanto na discretização espacial do problema como na aproximação das variáveis do sistema de equações. O sistema de equações fundamentais do escoamento é formado pelas equações de Navier-Stokes e pela equação da conservação de massa, descrita segundo a hipótese de pseudo-compressibilidade, em uma formulação cinemática ALE (Arbitrary Lagrangean- Eulerian). A consideração da superfície livre no escoamento se dá tratando o fluido como um meio bifásico, através do método Level Set. O corpo rígido apresenta não linearidade na rotação e restrições representadas por vínculos elásticos e amortecedores viscosos, sendo a equação de equilíbrio dinâmico resolvida através do método de Newmark. O esquema de acoplamento sólido-fluido adotado é o particionado convencional, que impõe condições de compatibilidade cinemáticas e de equilíbrio sobre a interface sólido-fluido, analisando ambos os meios de maneira sequencial. A discretização das equações governantes é realizada através do esquema explícito de dois passos de Taylor-Galerkin, aplicado no contexto da Análise Isogeométrica. Por fim, são analisados alguns problemas da Dinâmica de Fluidos Computacional, de onde se concluiu que os resultados obtidos são bastante consistentes com os fenômenos envolvidos, com as ferramentas exclusivas da Análise Isogeométrica, como o refinamento k, melhorando a convergência dos resultados. Para escoamentos bifásicos, verificou-se que o método Level Set obteve resultados bastante promissores apresentando, entretanto, uma dissipação numérica excessiva. Propõe-se, para estudos futuros, a elaboração de esquemas numéricos que conservem melhor o volume da fase líquida do escoamento. / The present work aims to development of a numerical formulation based on Isogeometric Analysis for the study of Fluid-Structure Interaction problems in applications involving rigid bodies submerged, considering incompressible Newtonian flows with free surface. The use of the Isogeometric Analysis allows unification between the preprocessing and analysis steps, improving then the continuity of the base functions employed, both in the spatial discretization and approximation of the variables in the system of equations. The fundamental flow equations are formed by the Navier-Stokes and the mass conservation, described by de pseudo-compressibility hypothesis, in an ALE (Arbitrary Lagrangean-Eulerian) kinematic formulation. The free surface consideration of the flow is handled treating the fluid like a two- phase medium, using the Level Set method. The rigid body considers nonlinearity in rotation, and restrictions represented by elastic springs and viscous dampers, with the dynamic equilibrium equation being resolved using the Newmark’s method. The solid-fluid coupling scheme is the conventional partitioned, which imposes kinematics and equilibrium compatibility conditions on the solid-fluid interface, analyzing both mediums in a sequential manner. The governing equations are discretized using the explicit two step Taylor-Galerkin method, applied in an Isogeometric Analisys context. Finally, some Computational Fluid Dinamics problems are analysed, from which it was concluded that the results obtained are quite consistent with phenomena involved, with the unique tools of Isogeometric Analysis, such as k-refinement, improving the convergence of the results. For biphasic flows, it was verified that the Level Set method obtained very promising results, presenting, however, an excessive numerical dissipation. For future studies, it is proposed the elaboration of numerical schemes that better preserve the volume of the liquid phase of the flow.

Submitted by Debora de Oliveira Medeiros null (deboraomedeiros@gmail.com) on 2017-06-13T19:08:30Z No. of bitstreams: 1 dissertacao_arquivar.pdf: 4524135 bytes, checksum: 33a027db23ff40bebc026b9090443d53 (MD5) / Approved for entry into archive by Luiz Galeffi (luizgaleffi@gmail.com) on 2017-06-19T12:58:15Z (GMT) No. of bitstreams: 1 medeiros_do_me_prud.pdf: 4524135 bytes, checksum: 33a027db23ff40bebc026b9090443d53 (MD5) / Made available in DSpace on 2017-06-19T12:58:15Z (GMT). No. of bitstreams: 1 medeiros_do_me_prud.pdf: 4524135 bytes, checksum: 33a027db23ff40bebc026b9090443d53 (MD5) Previous issue date: 2017-04-24 / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / Neste trabalho é apresentado um estudo das equações de Navier Stokes incompressível com superfície livre e métodos de projeção com uma formulação recente denominada laplaciano superficial. Esta formulação tem como finalidade uma melhor descrição da força de tensão superficial e grandezas, como curvatura e vetores tangencial e normal, descritas sobre a interface. Assim, uma condição de tensão normal alternativa é definida, e esta quando discretizada implicitamente e combinada com o método de projeção, descrevem a formulação laplaciano superficial, cuja sua solução é utilizada como condições de contorno para resolver o sistema de equações que descrevem o escoamento. A nova formulação destina-se a resolver um sistema tridiagonal de equações gerado sobre a interface, e usar a solução deste sistema de equações como uma condição de contorno na superfície livre para o sistema linear da correção da pressão no interior do domínio, que é resultante da aplicação do método de projeção. A nova equação que define a condição de tensão normal conta com grandezas definidas na malha euleriana, no contexto Marker-And-Cell (MAC), que devem ser projetadas sobre a malha lagrangeana e também considera a curvatura e os vetores tangente e normal na sua descrição, sendo importante um estudo detalhado de geometria diferencial. Finalmente, variações da formulação laplaciano superficial com tensão superficial para diferenças finitas são aplicadas para resolver os testes numéricos da oscilação da gota e da gota apoiada que possuem solução de referência, além da simulação de um problema com movimento de interface (fountain flow). Nestes testes, concluímos que a variação mais precisa e estável é aquela que aplica uma discretização da equação da tensão normal utilizando médias de valores alocados na malha MAC. / This work presents a study of the Navier-Stokes equations incompressible with free surface and a projection methods with a recent formulation defined as surface laplacian. The purpose of this formulation is to improve the description of the force of stress tension and quantities, as curvature and tangent and normal vectors, present at the interface. Thus, an alternative normal stress condition is defined, and when this is discretized implicitly and combined with the employ of projection method, describe the surface laplacian formulation, whose solution is used as boundary condition to solve the system of equations describing the flow. The new formulation is intended to solve the tridiagonal system of equations generated at the interface, and to use this solution as a boundary condition at free surface for the linear system of the pressure correction inside of domain, which results of the application of the projection method. The new equation used to define the normal stress condition considers quantities defined in the Eulerian mesh, in the Marker-And-Cell context (MAC), that should be projected on the Lagrangian mesh. In addition, in the new equation, it is also considered the influence of the curvature and normal and tangential vectors in your description, so that a detailed study of differential geometry for this computation is important. Finally, variations of the surface laplacian formulation for finite differences are applied for solving numerical tests of the drop oscillation and sessile drop which have reference solutions, beyond of the simulation of a problem with free surface moving (fountain flow). In these tests, we concluded that the most accurate and stable variation is the one that applies a discretization of the normal stress equation using the mean of values in the MAC mesh. / FAPESP: 2015/01243-0