Jet engines often operate under dirty conditions where large amounts of particulate matter can be ingested, especially, sand, ash and dirt. Particulate matter in different engine components can lead to degradation in performance. The objective of this dissertation is to investigate sand transport and deposition in the internal cooling passages of turbine blades. A simplified rectangular geometry is simulated to mimic the flow field, heat transfer and particle transport in a two pass internal cooling geometry. Two major challenges are identified while trying to simulate particle deposition. First, no reliable particle-wall collision model is available to calculate energy losses during a particle wall interaction. Second, available deposition models for particle deposition do not take into consideration all the impact parameters like impact velocity, impact angle, and particle temperature. These challenges led to the development of particle wall collision and deposition models in the current study. First a preliminary simulation is carried out to investigate sand transport and impingement patterns in the two pass geometry by using an idealized elastic collision model with the walls of the duct without any deposition. Wall Modeled Large Eddy Simulations (WMLES) are carried to calculate the flow field and a Lagrangian approach is used for particle transport. The outcome of these simulations was to get a qualitative comparison with experimental visualizations of the impingement patterns in the two pass geometry. The results showed good agreement with experimental distributions and identified surfaces most prone to deposition in the two pass geometry. The initial study is followed by the development of a particle-wall collision model based on elastic-plastic deformation and adhesion forces by building on available theories of deformation and adhesion for a spherical contact with a flat surface. The model calculates deformation losses and adhesion losses from particle-wall material properties and impact parameters and is broadly applicable to spherical particles undergoing oblique impact with a rigid wall. The model is shown to successfully predict the general trends observed in experiments. To address the issue of predicting deposition, an improved physical model based on the critical viscosity approach and energy losses during particle-wall collisions is developed to predict the sand deposition at high temperatures in gas turbine components. The model calculates a sticking or deposition probability based on the energy lost during particle collision and the proximity of the particle temperature to the softening temperature. For validation purposes, the deposition of sand particles is computed for particle laden jet impingement on a coupon and compared with experiments conducted at Virginia Tech. Large Eddy Simulations are used to calculate the flow field and heat transfer and particle dynamics is modeled using a Lagrangian approach. The results showed good agreement with the experiments for the range of jet temperatures investigated. Finally the two pass geometry is revisited with the developed particle-wall collision and deposition model. Sand transport and deposition is investigated in a two pass internal cooling geometry at realistic engine conditions. LES calculations are carried out for bulk Reynolds number of 25,000 to calculate flow and temperature field. Three different wall temperature boundary conditions of 950 oC, 1000 oC and 1050 oC are considered. Particle sizes in the range 5-25 microns are considered, with a mean particle diameter of 6 microns. Calculated impingement and deposition patterns are discussed for different exposed surfaces in the two pass geometry. It is evident from this study that at high temperatures, heavy deposition occurs in the bend region and in the region immediately downstream of the bend. The models and tools developed in this study have a wide range of applicability in assessing erosion and deposition in gas turbine components. / Ph. D.

Ce manuscrit se concentre sur l’étude de la dynamique de drainage de gouttes d’eau dans une phase continue d’huile de paraffine. Les gouttes sont de taille millimétrique, déformables et évoluent dans un domaine de simulation carré de 1 cm de coté. La simulation du comportement de tels systèmes pose le problème général de la description numérique des écoulements multiphasiques non stationnaires. Un modèle simplifié dans une géométrie à deux dimensions est proposé et simulé en volumes finis. Il inclut les propriétés physico-chimiques des interfaces et notamment les phénomènes de coalescence et l’évolution d’un tensioactif soluble dans les gouttes. L’effet des conditions aux limites sur le drainage d’une unique goutte est étudié. Le rôle de la coalescence sur ce drainage est également décrit pour un modèle de deux gouttes. Quelques simulations sont enfin proposées avec des systèmes dispersés plus complexes. / This manuscript focuses on the description of the settling dynamics of water droplets in a continuous phase of paraffin oil. Droplets are of millimetre size, deformable and evolve in a square simulation domain of 1 cm side. The simulations of the behaviour of such systems raise the general problem of the numerical description of the flows occurring in multiphase unsteady systems. A simplified model in a two dimensional geometry is used and integrated with a finite volume numerical technique. It includes the interfacial mechanical and chemical properties and in particular the coalescence phenomena and the evolution of a water soluble surfactant. The effect of the boundary conditions on the drainage of a unique droplet is studied. The role of drop-drop coalescence on this drainage is also described for a model with two droplets. Some simulations are finally proposed with more complex dispersed systems.

A tomografia por sensoriamento elétrico representa uma técnica de grande potencial para a otimização de processos normalmente associados às indústrias do petróleo e química. Entretanto, o emprego de técnicas tomográficas em processos industriais envolvendo fluidos multifásicos ainda carece de métodos robustos e computacionalmente eficientes. Nesse contexto, o principal objetivo deste trabalho é contribuir para o desenvolvimento de métodos para a solução do problema tomográfico com base em algoritmos genéticos específicos para a fenomenologia do problema abordado (interação do campo elétrico com o campo hidrodinâmico), bem como a adaptação do algoritmo para processamento em paralelo. A idéia básica consiste em partir de imagens qualitativas, fornecidas por uma sonda de visualização direta, para formar um modelo da distribuição interna do contraste elétrico e refiná-lo iterativamente até que variáveis de controle resultantes do modelo numérico se igualem às suas homólogas, determinadas experimentalmente. Isso pode ser feito usando um funcional de erro, que quantifique a diferença entre as medidas externas não intrusivas (fluxo de corrente elétrica real) e as medidas calculadas no modelo numérico (fluxo de corrente elétrica aproximado). De acordo com a abordagem funcional adotada, pode-se modelar a reconstrução numérica do contraste elétrico como um problema de minimização global, cuja função objetivo corresponde ao funcional de erro convenientemente definido e o mínimo global representa a imagem procurada. A grande dificuldade está no fato do problema ser não linear e mal-posto, o que reflete na topologia da superfície de minimização, demandando um método especializado de otimização para escapar de mínimos locais, pontos de sela, mínimos de fronteira e regiões praticamente planas. Métodos de otimização poderosos, como os algoritmos genéticos, embora apresentem elevado esforço computacional na obtenção da imagem procurada, são melhor adaptáveis ao problema em questão. Desse modo, optou-se pelo uso de algoritmos genéticos paralelos nas arquiteturas mestre-escravo, ilha, celular e híbrida (combinando ilha e celular). O desempenho computacional dos algoritmos desenvolvidos foi testado em um problema de reconstrução da imagem tomográfica de um escoamento vertical a bolhas. De acordo com os resultados, a arquitetura híbrida é capaz de obter a imagem desejada com um desempenho computacional melhor, quando comparado ao desempenho das arquiteturas mestre-escravo, ilha e celular. Além disso, estratégias para melhorar a eficiência do algoritmo foram propostas, como a introdução de informações a priori, derivadas de conhecimento físico do problema tomográfico (fração de vazio e coeficiente de simetria do escoamento), a inserção de uma tabela hash para evitar o cálculo de soluções já encontradas, o uso de operadores de predação e de busca local. De acordo com os resultados, pode-se concluir que a arquitetura híbrida é um método apropriado para solução do problema de tomografia por impedância elétrica de escoamentos multifásicos. / Tomography by electrical sensing represents a technique of great potential for the optimization of processes usually associated with petroleum and chemical industries. However, the employment of tomographic techniques in industrial processes involving multiphase flows still lacks robust and computationally efficient methods. In this context, the main objective of this thesis is to contribute to the development of solution methods based on specific genetic algorithms for the phenomenology of the tomographic problem (interaction between electric and hydrodynamic fields), as well as the adaptation of the algorithm to parallel processing. From qualitative images provided by a direct imaging probe, the basic idea is to generate a model of electric contrast internal distribution and refine it repeatedly until control variables resulting from the numerical model equalize their counterparts, determined experimentally. It can be performed by using an error functional to quantify the difference between non-intrusive external measurements (actual electric current flow) and measurements calculated in a numerical model (approximate electric current flow). According to the functional approach, the numerical reconstruction of the electrical contrast can be treated as a global minimization problem in which the fitness function is an error functional conveniently defined and the global minimum corresponds to the sought image. The major difficulty lies in the nonlinear and ill-posed nature of the problem, which reflects on the topology of the minimization surface, demanding a specialized optimization method to escape from local minima, saddle points, boundary minima and almost plane regions. Although powerful optimization methods, such as genetic algorithms, require high computational effort to obtain the sought image, they are best adapted to the problem in question, therefore parallel genetic algorithms were employed in master-slave, island, cellular and hybrid models (combining island and cellular). The computational performance of the developed algorithms was tested in a tomographic image reconstruction problem of vertical bubble flow. According to the results, the hybrid model can obtain the sought image with a better computational performance, when compared with the other models. Besides, strategies to improve the algorithm efficiency, such as the introduction of a priori information derived from the physical knowledge of the tomographic problem (void fraction and symmetry coefficient of the flow), the insertion of a hash table to avoid the calculation of solutions already found, the use of predation and local search operators were proposed. According to the results, it is possible to conclude that the hybrid model is an appropriate method for solving the electrical impedance tomography problem of multiphase flows.

Fluid flows mainly driven by capillary forces are presented in this thesis. By means of modeling and simulations, interesting dynamics in capillary-driven flows are revealed such as coalescences, breakups, precursor films, flow instabilities, rapid spreading, rigid body motions, and reactive wetting. Diffuse-interface methods model a fluid interface as having a finite thickness endowed with physical properties such as surface tension. Two diffuse-interface models that are based on the free energy of the system are presented. The binary model, more specifically the coupled Navier-Stokes/Cahn-Hilliard equations, was used to study different two-phase flows including problems related to microfluidics. Numerical issues using this model have been addressed such as the need for mesh adaptivity and time-step restrictions. Moreover, the flexibility of this model to simulate 2D, axisymmetric, and 3D flows has been demonstrated. The factors affecting reproducibility of microdroplet depositions performed under a liquid medium are investigated. In the deposition procedure, sample solution is dispensed from the end of a capillary by the aid of a pressure pulse onto a substrate with pillar-shaped sample anchors. In both the experimental and numerical study it was shown that the deposited volume mainly depends on the capillary-substrate distance and anchor surface wettability. Furthermore, a critical equilibrium contact angle has been identified below which reproducible depositions are facilitated. The ternary model is developed for more complicated flows such as liquid phase sintering. With the introduction of a Gibbs energy functional, the governing equations are derived, consisting of convective concentration and phase-field equations which are coupled to the Navier-Stokes equations with surface tension forces. Arbitrary phase diagrams, surface energies, and typical dimensionless numbers are some input parameters into the model. Detailed analysis of the important capillary phenomena in liquid phase sintering such as reactive and nonreactive wetting and motion of two particles connected by a liquid bridge are presented. The dynamics of the wetting is found to match with a known hydrodynamic theory for spreading liquids. Factors affecting the equilibrium configuration of the particles such as equilibrium contact angles and volume ratios are also investigated. / QC 20100823

Multiphase flow is a familiar phenomenon from daily life and occupies an important role in physics, engineering, and medicine. The understanding of multiphase flows relies largely on the theory of interfaces, which is not well understood in many cases. To date, the Navier-Stokes-Korteweg equations and the Cahn-Hilliard equation have represented two major branches of phase-field modeling. The Navier-Stokes-Korteweg equations describe a single component fluid material with multiple states of matter, e.g., water and water vapor; the Cahn-Hilliard type models describe multi-component materials with immiscible interfaces, e.g., air and water. In this dissertation, a unified multiphase fluid modeling framework is developed based on rigorous mathematical and thermodynamic principles. This framework does not assume any ad hoc modeling procedures and is capable of formulating meaningful new models with an arbitrary number of different types of interfaces. In addition to the modeling, novel numerical technologies are developed in this dissertation focusing on the Navier-Stokes-Korteweg equations. First, the notion of entropy variables is properly generalized to the functional setting, which results in an entropy-dissipative semi-discrete formulation. Second, a family of quadrature rules is developed and applied to generate fully discrete schemes. The resulting schemes are featured with two main properties: they are provably dissipative in entropy and second-order accurate in time. In the presence of complex geometries and high-order differential terms, isogeometric analysis is invoked to provide accurate representations of computational geometries and robust numerical tools. A novel periodic transformation operator technology is also developed within the isogeometric context. It significantly simplifies the procedure of the strong imposition of periodic boundary conditions. These attributes make the proposed technologies an ideal candidate for credible numerical simulation of multiphase flows. A general-purpose parallel computing software, named PERIGEE, is developed in this work to provide an implementation framework for the above numerical methods. A comprehensive set of numerical examples has been studied to corroborate the aforementioned theories. Additionally, a variety of application examples have been investigated, culminating with the boiling simulation. Importantly, the boiling model overcomes several challenges for traditional boiling models, owing to its thermodynamically consistent nature. The numerical results indicate the promising potential of the proposed methodology for a wide range of multiphase flow problems. / text

Cette thèse porte sur l'étude des ballotements dans les réservoirs de satellites à poste, lors des phases de manoeuvre à faible accélération. En effet la bulle de gaz d'hélium servant à pressuriser le réservoir se met en mouvement, générant ainsi des perturbations sur la stabilité globale du satellite. Afin de mener à bien cette étude, des méthodes numériques ont été développées, avec une méthode de frontières immergées pour prendre en compte les parois du réservoir.Le code est utilise la méthode Level-Set pour capturer l'interface, et gère les sauts à l'aide de la méthode Ghost-Fluid. Un solveur BlackBox Multigrid est également développé pour améliorer lesperformances de calcul. Une étude est présentée dans le dernier chapitre pour définir quelques lois de comportements en fonction des vitesses et accélérations générées lors des manoeuvres. / The core study of this PhD thesis is the sloshing in satellite tanks, during low acceleration maneuvers. Indeed the helium bubble used to pressurize the tank moves, thus generating perturbations on the global stability of the satellite. In order to understand this problem, numerical schemes have been developed, such as an immersed boundary method to model the tank wall. The numerical tool uses a Level-Set function coupled to a Ghost Fluid Method to track the interface and to account for the jump conditions.A BlackBox Multigrid Solver have been developed to improve computational cost. Finally a study is presented in the last chapter to predict the behaviour of the fluids with a varying rotational speed generated during some classical maneuvers.

Cette thèse porte sur l’étude expérimentale de l’extraction liquide-liquide au sein d’une colonne de laboratoire reproduisant de manière très simplifiée les appareils utilisés pour retraiter le combustible nucléaire (procédé hydro-métallurgique PUREX). Une technique optique non intrusive, dite de réfractométrie ou diffractométrie arc-en-ciel, a été spécifiquement développée pour caractériser la granulométrie (et donc l’aire interfaciale) et la composition de la phase dispersée. Cette dernière est composée de gouttes millimétriques de mélanges d’alcanes en ascension dans une colonne remplie d’eau au repos. Les spécificités de l’arc-en-ciel produit par ces gouttes à faible indice de réfraction relatif ont été étudiées à l’aide de différents modèles de diffusion asymptotiques de la lumière et la théorie électromagnétique de Lorenz-Mie. Grâce au développement de différentes approches directes et inverses (paramétriques), il a été démontré numériquement et expérimentalement que cette technique optique permettait d’estimer individuellement (ou collectivement) diamètre (moyen) et fraction de mélange de quelques dizaines de milliers de gouttes à quelques pour cents près. Des travaux préliminaires ont été réalisés sur l’extraction, résolue dans le temps de l’acétone, d’un milieu continu aqueux vers une goutte sessile composée initialement de toluène pur. Ils ont montré que l’on pouvait réellement estimer les constantes de transfert (de diffusion et de distribution moléculaire) à partir de l’analyse de l’évolution temporelle des arcs-en-ciel et en utilisant une méthode inverse qui intègre un modèle de diffusion moléculaire à symétrie radiale et un modèle électromagnétique ad hoc. / This thesis deals with the experimental study of liquid-liquid extraction within a laboratory column reproducing in a simple way the apparatuses used in a nuclear fuel reprocessing process (PUREX). A non-intrusive optical technique, called rainbow refractometry or diffractometry, has been specifically developed to characterize the size (and hence the interfacial area) and the composition of the dispersed phase. The latter is first time composed of millimeter droplets of alkanes mixtures free rising in a column filled with water at rest. Specificities of the rainbow produced by these droplets with a low relative refractive index were studied using different asymptotic light scattering models and the Lorenz-Mie electromagnetic theory. Thanks to the development of different direct and reverse (parametric) approaches, it was demonstrated both numerically and experimentally that this optical technique allows estimating individually (or collectively) the diameter (in average) and the mixing fraction of a few tens of thousands of droplets with a few percent of accuracy. Preliminary works have been carried out on the time-resolved extraction of acetone from an aqueous continuous medium to a sessile droplet initially composed of pure toluene. They have revealed that the transfer (diffusion and molecular splitting) constants can be estimated from the analysis of the temporal evolution of the rainbow signals using an inverse method that integrates a molecular diffusion model with a radial symmetry and an electromagnetic light scattering model.

Dans les travaux de recherche développés durant cette thèse, une première approche desécoulements magnétohydrodynamiques (MHD) multiphasiques est proposée. Cette approche seconcentre sur les phénomènes liés à l’interaction entre une dynamique interfaciale et un écoulementde coeur MHD. Le couplage induit entre rhéologie de surface etMHDde sous-phase est particulièrementillustré par le développement d’un viscosimètre annulaire surfacique, dédié à l’étude desmétaux liquides progressivement oxydés. En premier lieu sont introduits les éléments théoriquespropres à laMHDet à la rhéologie de surface. La modélisation de leur couplage fait intervenir deuxparamètres interfaciaux : les viscosités surfaciques dilatationnelle et de cisaillement. L’influence respectivede ces deux paramètres sur l’écoulement MHD de sous-phase est étudiée analytiquementet numériquement dans le cas d’un écoulement (stratifié) annulaire MHD permanent. Dans la configurationretenue, un champ magnétique uniforme vertical est imposé, perpendiculairement à lasurface liquide graduellement oxydée. Le rôle décisif des contraintes visqueuses interfaciales concernantl’(in)activation des couches de Hartmann est démontré, conduisant à des topologies atypiquesd’écoulement MHD. Le viscosimètre annulaire MHD est ensuite proposé en tant que méthodeexpérimentale originale, permettant la mesure sélective des viscosités surfaciques de fluides électroconducteurs.Les premières campagnes expérimentales aboutissent à une estimation de la viscositéde cisaillement interfacial d’un alliage métallique (GaInSn) liquide à température ambiante. Finalement,une ouverture sur l’écoulement MHD à proximité d’une inclusion gazeuse sphérique rigideest discutée en annexe de ce projet, en lien avec des conditions mécaniques variables à l’interfaceliquide/gaz pilotées par la rhéologie de surface, constituant une première approche vers la descriptiondes écoulements MHD dispersés. / In this thesis work, a first approach towards the description of magnetohydrodynamic(MHD) multiphase flows is proposed, based on the investigation of the role of surface rheologyin permanent regime. The study of the coupling between bulk MHD and surface rheology is particularlymotivated by the development of an annular surface viscometer devoted to liquid metalstopped with an oxidation layer. First, theoretical foundations of MHD and of surface rheologyare separately introduced. The modelling brings out a strong coupling between bulk and surfacevelocities, the latter being dependent on two interfacial parameters: the surface shear and dilatationalviscosities. Their respective influence is analytically and numerically investigated in the caseof a permanent annular (stratified) MHD end-driven flow. In the considered geometry, a vertical(uniform) magnetic field is imposed, perpendicular to the gradually oxidising liquid surface. Thecontribution of planar surface viscous stresses to the possible electrical activation of Hartmannlayers is demonstrated, leading to a variety of atypical MHD flow patterns. The annular MHDviscometer is then developed as a first device able to perform selective measurement of the surfaceviscosities of electroconductive fluids. First experiments lead to an estimated value of the surfaceshear viscosity for a liquid alloy (GaInSn) at room temperature. Some hints are finally given to investigatethe study case of aMHDflow past a rigid fluid sphere, with varying interfacial conditionsgoverned by surface rheology, as an additionalwork constituting a first step towards the descriptionof dispersed MHD flows.

A tomografia por sensoriamento elétrico representa uma técnica de grande potencial para a otimização de processos normalmente associados às indústrias do petróleo e química. Entretanto, o emprego de técnicas tomográficas em processos industriais envolvendo fluidos multifásicos ainda carece de métodos robustos e computacionalmente eficientes. Nesse contexto, o principal objetivo deste trabalho é contribuir para o desenvolvimento de métodos para a solução do problema tomográfico com base em algoritmos genéticos específicos para a fenomenologia do problema abordado (interação do campo elétrico com o campo hidrodinâmico), bem como a adaptação do algoritmo para processamento em paralelo. A idéia básica consiste em partir de imagens qualitativas, fornecidas por uma sonda de visualização direta, para formar um modelo da distribuição interna do contraste elétrico e refiná-lo iterativamente até que variáveis de controle resultantes do modelo numérico se igualem às suas homólogas, determinadas experimentalmente. Isso pode ser feito usando um funcional de erro, que quantifique a diferença entre as medidas externas não intrusivas (fluxo de corrente elétrica real) e as medidas calculadas no modelo numérico (fluxo de corrente elétrica aproximado). De acordo com a abordagem funcional adotada, pode-se modelar a reconstrução numérica do contraste elétrico como um problema de minimização global, cuja função objetivo corresponde ao funcional de erro convenientemente definido e o mínimo global representa a imagem procurada. A grande dificuldade está no fato do problema ser não linear e mal-posto, o que reflete na topologia da superfície de minimização, demandando um método especializado de otimização para escapar de mínimos locais, pontos de sela, mínimos de fronteira e regiões praticamente planas. Métodos de otimização poderosos, como os algoritmos genéticos, embora apresentem elevado esforço computacional na obtenção da imagem procurada, são melhor adaptáveis ao problema em questão. Desse modo, optou-se pelo uso de algoritmos genéticos paralelos nas arquiteturas mestre-escravo, ilha, celular e híbrida (combinando ilha e celular). O desempenho computacional dos algoritmos desenvolvidos foi testado em um problema de reconstrução da imagem tomográfica de um escoamento vertical a bolhas. De acordo com os resultados, a arquitetura híbrida é capaz de obter a imagem desejada com um desempenho computacional melhor, quando comparado ao desempenho das arquiteturas mestre-escravo, ilha e celular. Além disso, estratégias para melhorar a eficiência do algoritmo foram propostas, como a introdução de informações a priori, derivadas de conhecimento físico do problema tomográfico (fração de vazio e coeficiente de simetria do escoamento), a inserção de uma tabela hash para evitar o cálculo de soluções já encontradas, o uso de operadores de predação e de busca local. De acordo com os resultados, pode-se concluir que a arquitetura híbrida é um método apropriado para solução do problema de tomografia por impedância elétrica de escoamentos multifásicos. / Tomography by electrical sensing represents a technique of great potential for the optimization of processes usually associated with petroleum and chemical industries. However, the employment of tomographic techniques in industrial processes involving multiphase flows still lacks robust and computationally efficient methods. In this context, the main objective of this thesis is to contribute to the development of solution methods based on specific genetic algorithms for the phenomenology of the tomographic problem (interaction between electric and hydrodynamic fields), as well as the adaptation of the algorithm to parallel processing. From qualitative images provided by a direct imaging probe, the basic idea is to generate a model of electric contrast internal distribution and refine it repeatedly until control variables resulting from the numerical model equalize their counterparts, determined experimentally. It can be performed by using an error functional to quantify the difference between non-intrusive external measurements (actual electric current flow) and measurements calculated in a numerical model (approximate electric current flow). According to the functional approach, the numerical reconstruction of the electrical contrast can be treated as a global minimization problem in which the fitness function is an error functional conveniently defined and the global minimum corresponds to the sought image. The major difficulty lies in the nonlinear and ill-posed nature of the problem, which reflects on the topology of the minimization surface, demanding a specialized optimization method to escape from local minima, saddle points, boundary minima and almost plane regions. Although powerful optimization methods, such as genetic algorithms, require high computational effort to obtain the sought image, they are best adapted to the problem in question, therefore parallel genetic algorithms were employed in master-slave, island, cellular and hybrid models (combining island and cellular). The computational performance of the developed algorithms was tested in a tomographic image reconstruction problem of vertical bubble flow. According to the results, the hybrid model can obtain the sought image with a better computational performance, when compared with the other models. Besides, strategies to improve the algorithm efficiency, such as the introduction of a priori information derived from the physical knowledge of the tomographic problem (void fraction and symmetry coefficient of the flow), the insertion of a hash table to avoid the calculation of solutions already found, the use of predation and local search operators were proposed. According to the results, it is possible to conclude that the hybrid model is an appropriate method for solving the electrical impedance tomography problem of multiphase flows.

La démolition de bâtiments par foudroyage est une technique devenue classique. Elle génère des nuisances qui doivent être prises en compte, sous peine de rendre caduc ce type d'activité. Les principales nuisances générées par le foudroyage sont : - les projections générées par l’explosion et la dislocation du bâtiment ; - l’onde de surpression générée par l’explosif ; - la vibration du sol induite par l’effondrement du bâtiment ; - la création et la propagation d’un nuage de poussière. Les deux premiers points sont maitrisés. En ce qui concerne les vibrations, aucune méthode n’existe pour les limiter. Seuls, des fossés creusés permettent de bénéficier d’un effet d’ombre vis-à-vis de certains bâtiments. Les nuisances générées par les poussières et les manières de les limiter commencent à être une question fréquemment posée. Dans le cadre de la destruction d’installation classée pour la protection de l’environnement, cette question devient critique, car ce nuage peut déplacer des particules nocives. Actuellement, aucune information sur les mécanismes de production ou sur les mécanismes de propagation des poussières n’a été élaborée. Cette thèse propose un protocole de mesure pour l'étude des caractéristiques du nuage de particule. Elle présente les premières informations sur la production et la propagation des particules. Et pour finir, elle présente les premières simulations numériques du phénomène. / The demolition of buildings caving is a technique which has become classic. The demolition generates nuisance that must be taken into account, under penalty of rendering this type of demolition. The main nuisance generated by the caving are: - The projections generated by the explosion and the disintegration of the building; - The pressure wave generated by the explosive; - The ground vibration induced by the collapse of the building; - The creation and spread of a cloud of dust. The first two points are controlled. With regard to vibration, no method exists to limit them. Only the ditches can provide access to a shadow effect vis-à-vis some buildings. Nuisances generated by dust and ways to limit them begin to be a common question. As part of the destruction of classified installations for environmental protection, this issue becomes critical because the cloud can move harmful particles. Currently, no information on the mechanisms of production or on the mechanisms of spread of dust has been developed. This thesis proposes a measurement protocol for the study of the characteristics of cloud particles. It prensetes the first information on the production and propagation of the particles. And finally presents the first numerical simulation of the phenomenon.