L'objectif industriel de cette thèse est de proposer un modèle de transfert radiatif dans un réacteur de reformage de méthane. Dans ce procédé, des gaz réactifs circulent dans le réacteur tubulaire rempli de pastilles catalytiques.L'empilement de pastilles constitue un milieu poreux où le rapport de la taille caractéristique des pores sur la dimension radiale du réacteur est grand (1=10 à 1=5). De plus, les très forts gradients de porosité dus à l'organisation des pastilles au voisinage des parois ont un impact important sur les transferts thermiques et en particulier les transferts radiatifs.L'objectif scientifique est de développer et valider un modèle de transfert radiatif applicable à des milieux poreux fortement hétérogènes et anisotropes ne suivant pas la loi de Beer. Dans un premier temps, les propriétés radiatives du milieu homogénéisé équivalent au milieu poreux réel sont complètement déterminées par la fonction de distribution cumulée d'extinction Gext, la fonction de phase p et la porosité Π. Ces fonctions, précédemment introduites pour des milieux homogènes éventuellement anisotropes, sont calculées avec une grande précision par une méthode de Monte Carlo. Elles ont été généralisées ici à des milieux hétérogènes. Il a été montré à partir d'un nouveau critère de validité adapté aux milieux hétérogènes que le milieu homogénéisé équivalent ne suit pas la loi de Beer, en particulier au voisinage des parois.De ce fait, l'équation de transfert radiatif généralisée (GRTE) doit prendre en compte l'émission par un milieu non Beerien fortement hétérogène même à la limite optiquement mince : un coefficient d'absorption n'y a pas de sens physique et des corrélations entre émission et transmission apparaissent dues au caractère non Beerien. Le principe de réciprocité et les propriétés des fonctions d'extinction Gext ont permis d'exprimer rigoureusement les termes sources d'émission dans ce type de milieux fortement hétérogènes non Beeriens. Un facteur de corrélation émission-transmission a été introduit. La GRTE, sous forme intégrale, a été résolue par une méthode de transfert de Monte Carlo. Le modèle complet a été appliqué après validation aux réacteurs de reformage de méthane de Air Liquide. / The industrial goal of this work is to propose a radiative transfer model in a tubular reactor of steam methane reforming. During the reforming process, reactive gases are injected in the tubular reactor filled with catalytic pellets. The packed bed of pellets forms a porous medium, and a particular feature of it is that the characteristic pore size is large compared to the reactor inner dimension. In addition, the organization of the pellets in the near-wall region results in important porosity gradients which have a significant effect on the heat transfer, and more specifically on the radiative transfer.The scientific goal is to develop and validate a radiative transfer model applicable to strongly nonhomogeneous, anisotropic and non Beerian porous media.First, the radiative properties of the homogenised phase equivalent to the real porous medium are completely determined by the cumulated distribution function of extinction Gext, the phase function p, and the local porosity Π. These functions, previously introduced for statistically homogeneous and anisotropic porous media, are calculated very accurately by a Monte Carlo method. They have been extended to statistically non-homogeneous porous media. Similarly, the expression of the validity criterion of the Beer law is extended to statistically anisotropic and non-homogeneous porous media : it is proven that for the considered porous media the Beer law is not valid in the homogenised phase, in particular in the vicinity of the walls. As a result, the Generalized Radiative Transfer Equation (GRTE) is needed and the emission source terms must be determined in a strongly nonhomogeneous non Beerian even at the optically thin limit : an absorption coefficient doesn't have any physical meaning and correlations between emission and transmission appear due to the non-Beerian behavior.The reciprocity principle and the properties of the extinction functions Gext allow the emission source terms in this kind of strongly non-homogeneous and non-Beerian media to be accurately determined. A correlation factor emission-transmission has been introduced. The GRTE has been solved by a Monte Carlo method.The complete model is applied, after validation, to the steam methane reformers in use by Air Liquide.

The applicability and adequacy of three modeling approaches to describe and predict breakthough curves (BTCs) for non-reactive solutes in statistically homogeneous porous media were numerically and experimentally investigated. Modeling approaches were: the convection-dispersion equation (CDE) with scale-dependent dispersivity, mobile-immobile model (MIM), and the fractional convection-dispersion equation (FCDE). In order to test these modeling approaches, a prototype laboratory column system was designed for conducting miscible displacement experiments with a free-inlet boundary. Its performance and operating conditions were rigorously evaluated. When the CDE with scale-dependent dispersivity is solved numerically for generating a BTC at a given location, the scale-dependent dispersivity can be specified in several ways namely, local time-dependent dispersivity, average time-dependent dispersivity, apparent time-dependent dispersivity, apparent distance-dependent dispersivity, and local distance-dependent dispersivity. Theoretical analysis showed that, when dispersion was assumed to be a diffusion-like process, the scale-dependent dispersivity was locally time-dependent. In this case, definitions of the other dispersivities and relationships between them were directly or indirectly derived from local time-dependent dispersivity. Making choice between these dispersivities and relationships depended on the solute transport problem, solute transport conditions, level of accuracy of the calculated BTC, and computational efficiency The distribution of these scale-dependent dispersivities over scales could be described as either as a power-law function, hyperbolic function, log-power function, or as a new scale-dependent dispersivity function (termed as the LIC). The hyperbolic function and the LIC were two potentially applicable functions to adequately describe the scale dependent dispersivity distribution in statistically homogeneous porous media. All of the three modeling approaches described observed BTCs very well. The MIM was the only model that could explain the tailing phenomenon in the experimental BTCs. However, all of them could not accurately predict BTCs at other scales using parameters determined at one observed scale. For the MIM and the FCDE, the predictions might be acceptable only when the scale for prediction was very close to the observed scale. When the distribution of the dispersivity over a range of scales could be reasonably well-defined by observations, the CDE might be the best choice for predicting non-reactive solute transport in statistically homogeneous porous media. / Ph. D.