The aim of this thesis is three-fold: i) to investigate the performance of both the Eulerian-Lagrangian model and the Eulerian-Eulerian model to simulate the turbulent gas-particle flow; ii) to investigate the indoor airflows and contaminant particle flows using the Eulerian-Lagrangian model; iii) to develop and validate particle-wall collision models and a wall roughness model for the Eulerian-Lagrangian model and to utilize these models to investigate the effects of wall roughness on the particle flows. Firstly, the Eulerian-Lagrangian model in the software package FLUENT (FLUENT Inc.) and the Eulerian-Eulerian model in an in-house research code were employed to simulate the gas-particle flows. The validation against the measurement for two-phase flow over backward facing step and in a 90-degree bend revealed that both CFD approaches provide reasonably good prediction for both the gas and particle phases. Then, the Eulerian-Lagrangian model was employed to investigate the indoor airflows and contaminant particle concentration in two geometrically different rooms. For the first room configuration, the performances of three turbulence models for simulating indoor airflow were evaluated and validated against the measured air phase velocity data. All the three turbulence models provided good prediction of the air phase velocity, while the Large Eddy Simulation (LES) model base on the Renormalization Group theory (RNG) provided the best agreement with the measurements. As well, the RNG LES model is able to provide the instantaneous air velocity and turbulence that are required for the evaluation and design of the ventilation system. In the other two-zone ventilated room configuration, contaminant particle concentration decay within the room was simulated and validated against the experimental data using the RNG LES model together with the Lagrangian model. The numerical results revealed that the particle-wall coll ision model has a considerable effect on the particle concentration prediction in the room. This research culminates with the development and implementation of particle-wall collision models and a stochastic wall roughness model in the Eulerian-Lagrangian model. This Eulerian-Lagrangian model was therefore used to simulate the gas-particle flow over an in-line tube bank. The numerical predictions showed that the wall roughness has a considerable effect by altering the rebounding behaviours of the large particles and consequently affecting the particles motion downstream along the in-line tube bank and particle impact frequency on the tubes. Also, the results demonstrated that for the large particles the particle phase velocity fluctuations are not influenced by the gas-phase fluctuations, but are predominantly determined by the particle-wall collision. For small particles, the influence of particle-wall collisions on the particle fluctuations can be neglected. Then, the effects of wall roughness on the gas-particle flow in a two-dimensional 90-degree bend were investigated. It was found that the wa ll roughness considerably altered the rebounding behaviours of particles by significantly reducing the 'particle free zone' and smoothing the particle number density profiles. The particle mean velocities were reduced and the particle fluctuating velocities were increased when taking into consideration the wall roughness, since the wall roughness produced greater randomness in the particle rebound velocities and trajectories.

<p>An Eulerian-Eulerian model for dilute gas-particle turbulent flows is</p><p>developed for engineering applications. The aim is to understand the effect of particles on turbulent flows. The model is implemented in a finite element code which is used to perform numerical simulations. The feedback from the particles on the turbulence and the mean flow of the gas in a vertical channel flow is studied. In particular, the influence of the particle response time and particle volume fraction on the preferential concentration of the particles near the walls, caused by the turbophoretic effect is explored. The study shows that the particle feedback decreases the accumulation of particles on the walls. It is also found that even a low particle volume fraction can have a significant impact on the turbulence and the mean flow of the gas. A model for the particle fluctuating velocity in turbulent gas-particle flow is derived using a set of stochastic differential</p><p>equations. Particle-particle collisions were taken into account. The model shows that the particle fluctuating velocity increases with increasing particle-particle collisions and that increasing particle response times decrease the fluctuating velocity.</p>

An Eulerian-Eulerian model for dilute gas-particle turbulent flows is developed for engineering applications. The aim is to understand the effect of particles on turbulent flows. The model is implemented in a finite element code which is used to perform numerical simulations. The feedback from the particles on the turbulence and the mean flow of the gas in a vertical channel flow is studied. In particular, the influence of the particle response time and particle volume fraction on the preferential concentration of the particles near the walls, caused by the turbophoretic effect is explored. The study shows that the particle feedback decreases the accumulation of particles on the walls. It is also found that even a low particle volume fraction can have a significant impact on the turbulence and the mean flow of the gas. A model for the particle fluctuating velocity in turbulent gas-particle flow is derived using a set of stochastic differential equations. Particle-particle collisions were taken into account. The model shows that the particle fluctuating velocity increases with increasing particle-particle collisions and that increasing particle response times decrease the fluctuating velocity. / QC 20101124

A turbulent two-phase flow model using kinetic theory of granularflows for the particle phase is developed and implmented in afinite element code. The model can be used for engineeringapplications. However, in this thesis it is used to investigateturbulent gas-particle flows through numerical simulations.  The feedback from the particles on the turbulence and the meanflow of the gas in a vertical channel flow is studied. In particular,the influence of the particle response time, particle volumefraction and particle diameter on the preferential concentration ofthe particles near the walls, caused by the turbophoretic effect isexplored. The study shows that when particle feedback is includedthe accumulation of particles near the walls decreases. It is also foundthat even at low volume fractions particles can have a significant impacton the turbulence and the mean flow of the gas. The effect of particles on a developing turbulent vertical upward pipeflow is also studied. The development length is found to substantiallyincrease compared to an unladen flow. To understand what governs thedevelopment length a simple estimation was derived, showing that itincreases with decreasing particle diameters in accordance with themodel simulations. A model for the fluctuating particle velocity in turbulentgas-particle flow is derived using a set of stochastic differentialequations taking into account particle-particle collisions. Themodel shows that the particle fluctuating velocity increases whenparticle-particle collisions become more important and that increasingparticle response times reduces the fluctuating velocity. The modelcan also be used for an expansion of the deterministic model for theparticle kinetic energy. / QC20100726

La protection des opérateurs et la surveillance des ambiances de travail en cas de mise en suspension d’aérosols radioactifs, dans un local ventilé d’une installation nucléaire, requièrent la connaissance de l’évolution spatio-temporelle de la concentration en particules, en tout point du local considéré. L’estimation précise de cette concentration a fait l’objet du développement de modèles spécifiques de transport et de dépôt d’aérosols dans un local ventilé, dans le cadre d’une thèse cofinancée par l’IRSN et EDF, en collaboration avec l’IMFT. Un formalisme eulérien de glissement est utilisé pour modéliser le transport des aérosols. Celui-ci est basé sur une unique équation de transport des concentrations en particules (« Diffusion-Inertia model »). L’étude spécifique du dépôt d’aérosols en parois a permis de développer un modèle de couche limite, qui consiste à déterminer précisément le flux de dépôt de particules en parois, quels que soient le régime de dépôt et l’orientation de la surface considérée. Les modèles de transport et de dépôt finalement retenus ont été implantés dans Code_Saturne, un logiciel de mécanique des fluides. La validation de ces modèles a été effectuée à partir de données de la littérature en géométries simples, puis sur la base de campagnes expérimentales de traçage dans des locaux ventilés d’environ 30 m&#179; et 1500 m&#179;. / When particulate radioactive contamination is likely to become airborne in a ventilated room, assessment of aerosol concentration in every point of this room is important, in order to ensure protection of operators and supervision of workspaces. Thus, a model of aerosol transport and deposition has been developed as part of a project started with IRSN, EDF and IMFT. A simplified eulerian model, called “diffusion-inertia model” is used for particle transport. It contains a single transport equation of aerosol concentration. The specific study of deposition on walls has permitted to develop a boundary condition approach, which determines precisely the particle flux towards the wall in the boundary layer, for any deposition regime and surface orientation.The final transport and deposition models retained have been implemented in a CFD code called Code_Saturne. These models have been validated according to literature data in simple geometries and tracing experiments in ventilated rooms, which have been carried out in 30 m&#179; and 1500 m&#179; laboratory rooms.