Turbulent flows present structures with a wide range of scales. The computation of the complete physics of a turbulent flow (termed DNS) is very expensive and is, for the time being, limited to low and medium Reynolds number flows. As a way to capture high Reynolds number flows, a part of the physics complexity has to be modeled. Large eddy simulation (LES) is a simulation strategy where the large turbulent eddies present on a given mesh are captured and the influence of the non-resolved scales onto the resolved ones is modeled. The present thesis reports on the development and validation of a methodology in order to apply LES for complex wall-bounded flows. Discretization methods and LES models, termed subgrid scale models (SGS), compatible with such a geometrical complexity are discussed. It is proved that discrete a kinetic energy conserving discretization of the convective term is an attractive solution to perform stable simulations without the use of an artificial dissipation, as upwinding. The dissipative effect of the SGS model is thus unaffected by any additional dissipation process. The methodology is first applied to a developed parallel fourth-order incompressible flow solver for cartesian non-uniform meshes. In order to solve the resulting Poisson equation, an efficient multigrid solver is also developed. The code is first validated using DNS (Taylor-Green vortex, channel flow, four-vortex system) and LES (channel flow), and finally applied to the investigation of an aircraft two-vortex system in ground effect. The methodology is then applied to improve a RANS-based industrial unstructured compressible flow solver, developed at CENAERO, to perform well for LES applications. The proposed modifications are tested successfully on the unsteady flow past a sphere at Reynolds of 300 and 10000, corresponding to the subcritical regime.

Turbulent flows present structures with a wide range of scales. The computation of the complete physics of a turbulent flow (termed DNS) is very expensive and is, for the time being, limited to low and medium Reynolds number flows. As a way to capture high Reynolds number flows, a part of the physics complexity has to be modeled. Large eddy simulation (LES) is a simulation strategy where the large turbulent eddies present on a given mesh are captured and the influence of the non-resolved scales onto the resolved ones is modeled. The present thesis reports on the development and validation of a methodology in order to apply LES for complex wall-bounded flows. Discretization methods and LES models, termed subgrid scale models (SGS), compatible with such a geometrical complexity are discussed. It is proved that discrete a kinetic energy conserving discretization of the convective term is an attractive solution to perform stable simulations without the use of an artificial dissipation, as upwinding. The dissipative effect of the SGS model is thus unaffected by any additional dissipation process. The methodology is first applied to a developed parallel fourth-order incompressible flow solver for cartesian non-uniform meshes. In order to solve the resulting Poisson equation, an efficient multigrid solver is also developed. The code is first validated using DNS (Taylor-Green vortex, channel flow, four-vortex system) and LES (channel flow), and finally applied to the investigation of an aircraft two-vortex system in ground effect. The methodology is then applied to improve a RANS-based industrial unstructured compressible flow solver, developed at CENAERO, to perform well for LES applications. The proposed modifications are tested successfully on the unsteady flow past a sphere at Reynolds of 300 and 10000, corresponding to the subcritical regime.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O presente trabalho propõe analisar metodologias para o cálculo do gradiente em malhas não-estruturadas do tipo Voronoi que são utilizadas no método de Volumes Finitos. Quatro metodologias para o cálculo do gradiente são testadas e comparadas com soluções analíticas. As técnicas utilizadas são: Método do Balanço de Forças, Método do Mínimo Resíduo Quadrático, Método da Média dos Gradientes Projetos e Método da Média dos Gradientes Projetados Corrigidos. Uma análise por série de Taylor também foi feita, e as equações analíticas comparadas com resultados numéricos. Os testes são realizados em malhas cartesianas e malhas triangulares, que em um trabalho anterior apresentaram alguns resultados inconsistentes. A influência do ponto gerador e do ângulo de rotação é analisada. É verificado que a posição do ponto gerador e a metodologia utilizada em cada malha influencia no cálculo do gradiente. Dependendo da malha e da metodologia utilizada, as equações analíticas indicaram que existem erros associados, que prejudicam o cálculo do gradiente. / Presente work propose examine methodologies for calculate gradient using a non-structured mesh of Voronois type, used on finite volume method. Four methodologies for calculate gradient are tested and compared with analytical solutions. The techniques used are: the Forces Balance Method, Minimum Quadratic Residue Method, Average Projected Gradient Method and the Revised Average Projected Method. An analysis using Taylor series was also made, and the analytical equations compared with numerical results. Tests are performed over Cartesian and triangular meshes, second one which in a previous work showed some inconsistent results. The influence of the gerator point displacement and rotation angle is analyzed. It was found that the position of the generator point and the methodology used influences gradient value. Accordingly to the mesh and the methodology used, analytical equations indicates that there are associated errors, which affect gradient value.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O presente trabalho propõe analisar metodologias para o cálculo do gradiente em malhas não-estruturadas do tipo Voronoi que são utilizadas no método de Volumes Finitos. Quatro metodologias para o cálculo do gradiente são testadas e comparadas com soluções analíticas. As técnicas utilizadas são: Método do Balanço de Forças, Método do Mínimo Resíduo Quadrático, Método da Média dos Gradientes Projetos e Método da Média dos Gradientes Projetados Corrigidos. Uma análise por série de Taylor também foi feita, e as equações analíticas comparadas com resultados numéricos. Os testes são realizados em malhas cartesianas e malhas triangulares, que em um trabalho anterior apresentaram alguns resultados inconsistentes. A influência do ponto gerador e do ângulo de rotação é analisada. É verificado que a posição do ponto gerador e a metodologia utilizada em cada malha influencia no cálculo do gradiente. Dependendo da malha e da metodologia utilizada, as equações analíticas indicaram que existem erros associados, que prejudicam o cálculo do gradiente. / Presente work propose examine methodologies for calculate gradient using a non-structured mesh of Voronois type, used on finite volume method. Four methodologies for calculate gradient are tested and compared with analytical solutions. The techniques used are: the Forces Balance Method, Minimum Quadratic Residue Method, Average Projected Gradient Method and the Revised Average Projected Method. An analysis using Taylor series was also made, and the analytical equations compared with numerical results. Tests are performed over Cartesian and triangular meshes, second one which in a previous work showed some inconsistent results. The influence of the gerator point displacement and rotation angle is analyzed. It was found that the position of the generator point and the methodology used influences gradient value. Accordingly to the mesh and the methodology used, analytical equations indicates that there are associated errors, which affect gradient value.

Αντικείμενο της παρούσας εργασίας είναι η πρόβλεψη τόσο της κίνησης, αλληλεπίδρασης και παραμόρφωσης δύο φυσαλίδων λόγω μεταβολής της πίεσης στο περιβάλλον ιξώδες υγρό, όσο και της ανοδικής κίνησης μιας φυσαλίδας λόγω άνωσης σε ένα Νευτωνικό ή ιξωδοπλαστικό ρευστό. Για τη μοντελοποίηση των αλληλεπιδρώντων φυσαλίδων, αναπτύχθηκε μιας νέα ελλειπτική μεθόδος κατασκευής του υπολογιστικού πλέγματος προκειμένου να αντιμετωπιστούν επιτυχώς τα ιδιάζοντα σημεία (πόλοι) των φυσαλίδων και οι μεγάλες παραμορφώσεις των διεπιφανειών τους. Με τη μέθοδο αυτή η πύκνωση του πλέγματος περιορίζεται μόνο στις περιοχές που είναι αναγκαίο, μειώνοντας έτσι το υπολογιστικό κόστος και αυξάνοντας την ακρίβεια των υπολογισμών. Για την επίλυση των παρακάτω προβλημάτων χρησιμοποιήθηκε η μέθοδος των μικτών πεπερασμένων στοιχείων κατά Galerkin. Στην περίπτωση των αλληλεπιδρώντων φυσαλίδων έχει εξετασθεί η επίδραση του σχετικού μεγέθους τους, της συχνότητας και του εύρους μεταβολής της επιβαλλόμενης πίεσης και πότε οδηγούν σε έλξη ή άπωση των φυσαλίδων. Στην περίπτωση ελκτικής δύναμης, ακολουθείται η κίνηση και η παραμόρφωσή τους μέχρι του σημείου που έρχονται σε επαφή, όπου αυτό είναι εφικτό. Για τη μελέτη του προβλήματος της φυσαλίδας που ανέρχεται λόγω άνωσης, υποθέτουμε αξονική συμμετρία και μόνιμη κατάσταση. Σύγκριση των προβλέψεών μας για το σχήμα των φυσαλίδων και το πεδίο ροής γύρω τους με προηγούμενα θεωρητικά και πειραματικά αποτελέσματα για Νευτωνικά ρευστά έδειξε άριστη συμφωνία. Στην περίπτωση του ιξωδοπλαστικού ρευστού εξετάστηκαν λεπτομερώς οι παραμορφώσεις των φυσαλίδων σαν συνάρτηση των αριθμών Bingham, Bond και Αρχιμήδη και υπολογίσθηκαν οι συνθήκες υπό τις οποίες είναι δυνατή η παγίδευση της φυσαλίδας μέσα σε αυτό. / The present study deals with the numerical simulation of the motion, interaction and deformation of two bubbles due to variation of the pressure of the ambient Newtonian fluid, and the buoyancy-driven rise of a bubble in a Newtonian or a viscoplastic fluid. A new elliptic mesh generation method is developed in order to deal with the singular points (poles) of the bubbles and the large deformations of their surface. This method permits us to increase the mesh resolution only in the regions that is necessary, decreasing thus the computational cost and increasing the precision of our calculations. The following problems are solved using the mixed finite element/Galerkin method. In the case of the interacting bubbles the effect of their relative size, the frequency and the width of the imposed pressure is examined as well as the conditions that lead in attraction or repulsion of the bubbles. In the case that attractive forces exist, the motion and the deformation of the bubbles followed up to the point that they come in contact, whenever this is possible. In order to study the problem of the bubble that rises due to buoyancy, axial symmetry and steady flow is assumed. Our results for the shape of the bubbles and the flow around them are in very good agreement with previous theoretical and experimental results for Newtonian fluids. The deformations of the bubbles rising in a viscoplastic material are also examined for various values of the Bingham, Bond and Archimedes numbers and the conditions under which entrapment of a bubble is possible are determined.