Logo Geosciences

Partenaires

CNRS
Logo Rennes1
Logo OSUR
Logo Doc OSUR

Géosciences Rennes
UMR 6118
Université de Rennes1
Campus de Beaulieu
35042 Rennes Cedex

02 23 23 60 76


Rechercher

Sur ce site

Sur le Web du CNRS


Accueil du site > Français > Les annonces de séminaires et thèses > Séminaire de Amélie NEUVILLE (Faculty of Mathematics and Natural Sciences, Department of Physics, Oslo)

Séminaire de Amélie NEUVILLE (Faculty of Mathematics and Natural Sciences, Department of Physics, Oslo)

Date : 21 septembre 2012, 11 h

Titre : Hydro-thermal exchanges in rough fractures : numerical simulations and experiments.

Lieu : Salle de Conférences de l’OSUR (Bât. 14 B, RdC)

Résumé :
Hydraulic and thermal exchanges in nature take often place in fractured bedrocks. The morphology of fractures, which can be characterized using photogrammetry, laser profilometry, or computer tomography, influences the hydro-thermal flow which occurs when a cold fluid is injected into a hot fractured bedrock. This is shown under lubrication approximations – which assume that the fracture morphology varies in a smooth way – by solving the Stokes equation together with a bidimensional (integrated over the thickness) advection-diffusion equation. In most cases, we notice that the roughness reduces the efficiency of the heat transfer between the rock and the fluid.

However some features which are observed on the field, like time-dependent temperature during pumping, cannot be explained with this model. To go beyond these lubrication assumptions we have developed a fully tridimensional (3D) lattice Boltzmann (LB) code. In these LB simulations, two sorts of fictitious particles are used : one to solve the mass transport, and the other one for the energy transport. As a first step, we study the hydrothermal exchanges (at low Reynolds numbers and moderate Peclet numbers) around a single sharp cavity perturbing an otherwise flat fracture, with an invariant third dimension. Beyond some critical slope for the boundaries, 3D effects (like recirculation) become important and require to go beyond the lubrication approximations to be properly modeled. In addition, the effect of rock cooling (neglected in the lubrication approximations) is quantified. Simulations of natural convection will also be shown.

The spatial variability of the hydraulic flow also induces variations in the stress field around the fracture. Simulations with a real 3D rock morphology obtained by computer tomography scan have also been carried out. This is of interest for further mechanical investigation purposes.

Finally we also present an experimental setup that we want to use to further investigate and calibrate our simulations. An infrared camera monitors in space and time the temperature field of cold water injected through a rough fracture.