Search

Colloque avec actes et comité de lecture. Internationale.; International audience; Une extension d'un modèle numérique pour la simulation à la microéchelle de la combustion en milieu poreux (Debenest et al., 2005a) est présentée. L'enrichissement porte sur le modèle chimique, qui fait notamment intervenir quatre espèces réactives (C, O2, CO et CO2), et quatre réactions dont trois à la surface des grains et une homogène en phase gazeuse. Un premier ensemble de résultats est discuté qui met en évidence différents régimes possibles.

The aim of the present work is to study the in situ combustion (ISC) process at inter-well scale in a fractured reservoir. Due to the complexity associated with the ISC process, highly heterogeneous nature of the fractured reservoirs and some unsuccessful attempts in the past to put the process into practice, the subject of ISC in fractured systems has been receiving little interest and there are still many essential open questions in this area. It is very challenging to answer the question whether the ISC process could be applied in a heavy oil fractured reservoir or not. And if the answer is positive, what is (are) the dominant oil recovery mechanism(s) and finally, how can we model and simulate this process, at least, at inter-well scale. This work tries to give answers to some of these questions. In this regard, we followed a step by step procedure. In the first step, general literature concerning the combustion process in porous media and particularly that related to the combustion process in an oil reservoir was reviewed. Some other references about the modeling of fracture reservoirs were also reviewed. This led us to distinguish some of the main challenges in this area and define a methodology for the rest of the work. Based on this methodology, the first target was to understand and to characterize the behavior of a combustion front at small (Darcy) scale. The second target was to apply the knowledge of the first part to propose a suitable model for ISC at larger scale. To this end, a commercial thermal reservoir simulator (STARS) was used. The simulator was validated for both simple process for which an analytical solution is available and for a more complex process where the laboratory results are on hand. Then, after the validation part, the numerical tool has been used to widely investigate the conditions where a reaction front can propagate in a fractured core. This allowed us to understand some of the leading mechanisms (oxygen diffusion coefficient for extinction/ propagation of combustion front and matrix permeability for oil production). Some other numerical studies provided us with some understanding about the most important mechanism(s) of oil production. Thereafter, some single block simulations were done to investigate the two-dimensional behavior of the ISC process, based on which the underlying process was found to be diffusion dominated both for heat and mass transfers. These results also helped us to distinguish the characteristic length scale of some important parameters (temperature, coke concentration, combustion front, etc.) which can give useful information about the large scale model. After that, an experimental part has been performed to find propagation conditions of ISC at laboratory scale. This was done by varying both the operational conditions (flowrate, pressure and oxygen concentration) and the characteristics of the fractured system (aperture, surface area, permeability). This permitted us to find that in some suitable conditions there is a possibility to generate a combustion front in a fractured system containing heavy oil. To give an idea about the modeling of the process at larger scale, some fine grid simulations are also performed using a multi-block model. By analyzing the results of this model some guidelines are proposed for the large scale model. At the end, a short discussion about the upscaling of an easy system (solid-gas combustion using an Arrhenius law as a function for the mass sink term in a conductive system) is presented based on an upscaling using the volume averaging method.

In a composting process the monitoring of heat flows is a useful tool in terms of phenomenological comprehension and diagnosis of the process. Indeed, the temperature fields are generic markers of the micro-biological phenomena governing a composting process. Moreover, as heat flows are largely influenced by the air flowing through the porous solid waste, the temperature fields also characterize the influence of the air flow distribution on the composting process. However, at the industrial scale, it may be difficult to measure precisely the temperature distribution. In this paper, an in situ experimental data acquisition method of these thermal fields is described. Obtained data were used to reconstruct the evolution of the temperature in the whole composting reactor along the industrial composting process, in order to understand the influence of the air flow distribution. On the basis of these results, a model for the description of the thermal behaviour and the optimization of the aeration system was proposed.

Solution mining is a well known method for extracting salt by dissolving this mineral with water. In this analysis about cavity dissolution modeling, we consider the case of a binary system, i.e., a chemical solute constitutes the solid that is dissolved by a "solvent" (mainly water). Rock salt dissolution is controlled by thermodynamic equilibrium at the interface, i.e., equality of the chemical potentials. In this paper, a local non-equilibrium Diffuse Interface Model (DIM) and an explicit treatment of the brine-salt interface (using an ALE (Arbitrary Lagrangian-Eulerian) technique) are introduced in order to describe dissolution problems in such binary cases. The control equations are obtained by upscaling micro-scale balance equations for a solid-liquid dissolution problem using a volume averaging theory. This results in a diffuse DIM model for dissolution. After the mathematical formulation, dissolution test cases are presented. We show the main features of the method. An optimum expression for the solid liquid exchange coefficient is obtained by comparison with reference solutions obtained by ALE simulations. Illustrations of the interaction between natural convection and forced convection in dissolution problems are presented. The ability of the method to track accurately the time and space evolution of the rock salt-fluid interface is shown through several examples.