- Groundwater flow and contamination studies
- Hydraulic Fracturing and Reservoir Analysis
- Rock Mechanics and Modeling
- Seismic Imaging and Inversion Techniques
- Enhanced Oil Recovery Techniques
- Hydrocarbon exploration and reservoir analysis
- Soil and Unsaturated Flow
- Advanced Mathematical Modeling in Engineering
- Heat and Mass Transfer in Porous Media
- Geophysical and Geoelectrical Methods
- Fire dynamics and safety research
- NMR spectroscopy and applications
- Lattice Boltzmann Simulation Studies
- Pickering emulsions and particle stabilization
- Radiative Heat Transfer Studies
- Combustion and flame dynamics
- Drilling and Well Engineering
- Landslides and related hazards
- Composite Material Mechanics
- Theoretical and Computational Physics
- CO2 Sequestration and Geologic Interactions
- Grouting, Rheology, and Soil Mechanics
- Dam Engineering and Safety
- earthquake and tectonic studies
- Electrostatics and Colloid Interactions
Institut Pprime
2013-2024
Centre National de la Recherche Scientifique
2009-2024
Combustion Institute
2009-2021
École Nationale Supérieure de Mécanique et d'Aérotechnique
2005-2020
Université de Poitiers
2010-2020
Département Mathématiques et Informatique Appliquées
2007-2013
SPX Corporation (United States)
2009
Laboratoire du Futur
1997-2008
Institut de physique du globe de Paris
1996-2000
Wolters Kluwer (Netherlands)
2000
Computed microtomography is applied to a piece of Fontainebleau sandstone in order determine the geometrical structure pores. The topology void space then derived from tomographic image volume. Permeability and conductivity are computed found good agreement with experimental data. Perspectives offered by this new nondestructive method potential resolution one micrometer or less analyzed.
Random packings of grains arbitrary shape are built with an algorithm that is mostly applied to spheres, ellipsoids, cylinders, and parallelepipeds. A systematic account the main geometrical properties such as porosity, reduced specific area, etc. given. The conductivity, permeability, dispersion also systematically determined they shown not depend upon their mode construction.
In homogeneous porous media, the analytical expression of dispersion tensor D* can be calculated by method moments and a multiple scale expansion; symmetric component this is identical in both cases. Numerically, computed two methods, namely B equation random walks. The media are modeled as being spatially periodic; determined function Péclet number for four types unit cells: deterministic, fractal, random, reconstructed. A systematic comparison made with existing numerical experimental...
Flow in fractured porous media was first investigated by Barenblatt and Zheltov [1960] et al. means of the double‐porosity model. A direct, exact, complete numerical solution flow such is given this paper for arbitrary distributions permeabilities matrix fracture network. The network are automatically meshed; equations discretized finite volume method. This code has been so far applied to incompressible fluids statistically homogeneous which schematized as spatially periodic media. Some...
The porosity and the autocorrelation function of pore space are measured on thin sections Fontainebleau sandstones. This information is used to generate artificial porous media which share these statistical properties. Laplace equation numerically solved determine formation factor, or equivalently electrical conductivity. With no adjustable constant, predicted factors were found be in acceptable agreement with experimental ones.
The random geometry of real porous media is analyzed with the objective reproducing it numerically; adequate algorithms are proposed for consolidated materials which may be statistically homogeneous or not, and possess more than one solid phase; packings star-shape grains built to mimic non-consolidated usually obtained by settling processes. macroscopic properties all these can deduced solving local partial differential equations govern phenomena; finite difference schemes used most time. A...
The permeability of a three-dimensional network polygonal fractures is determined by triangulating the and solving two-dimensional Darcy equation in each fracture. general triangulation methodology numerical solution are presented. Networks regular hexagonal detailed; finite-size scaling used to analyze data relative percolation threshold, but conduction exponent $t$ found close its classical value three dimensions; for large fracture densities, shown tend towards mean-field model Snow...
Single fractures are generated by statistical methods and described a probability density of the profile heights spatial covariance function which is either Gaussian or self-afline.The results numerical simulations bised on both 2D Reynolds 3D Stokes equations presented.Total fluxes predicted these for same given discussed.The difference between predictions models analysed range correlation lengths apertures real fractures, it found that total may differ more than two times.When local...
The fundamental issue of reconstructing a porous medium is examined anew in this paper, thanks to sample low-porosity Fontainebleau sandstone that has been analyzed by computed microtomography. Various geometric properties are determined on the experimental sample. A statistical property, namely, probability density covering radius, determined. This used order reconstruct means Poissonian generation polydisperse spheres. In second part, real and reconstructed one compared. most important...
SUMMARY Real or reconstructed porous media can be discretized as a series of elementary cubes, which are filled either with solid fluid phase. Various algorithms, based on pseudo‐diffusion processes, proposed to determine the connected and percolating components pore space. The graph space is obtained by two different methods; most efficient upon homotopic thinning. Topological characteristics, such number loops, derived. Systematic applications these algorithms illustrated computer...
The Laplace equation can be solved in any two- and three-dimensional porous medium by means of a vectorized numerical code. It is applied to several structures such as random media derived from site percolation; close the percolation threshold, critical exponents are found very ones corresponding networks; results usefully compared previous variational upper bounds prediction an approximate space renormalization. Media with double porosity catalyst pellets also addressed. Finally...
The asymptotic behaviors of the permeability isotropic fracture networks at small and large densities are characterized, a general heuristic formula is obtained which complies with limiting accurately predicts these over whole density range. Theses developments based on extensive numerical calculations theoretical arguments inspired by examination flow distribution in fractures densities. Then, results extended to anisotropic Fisher orientations, polydisperse networks, fractured porous...
The two-dimensional displacement joint probability density ${P}_{\ensuremath{\Delta}}(X,Z)$ for water flowing through a bed of glass beads has been measured by means pulsed field-gradient nuclear magnetic resonance. simultaneous particle displacements $X$ and $Z$ perpendicular parallel to the pressure gradient, respectively, at given encoding time \ensuremath{\Delta}, are obtained from an experiment employing orthogonal field gradients. resulting distribution is compared numerical...
The influence of various parameters such as the domain size, exponent power law, smallest radius, and fracture shape on percolation threshold networks has been numerically studied. For large domains, adequate parameter is dimensionless density normalized by product third moment radii distribution factor; for regular polygons, critical depends only slightly fractures; a model proposed fractures with elongated shapes. In small analyzed in terms sum two reduced moments distribution; this...
Two-phase flow in fractured porous media is investigated by means of a direct and complete numerical solution the generalized Darcy equations three-dimensional discrete fracture description. The model applies to arbitrary network geometry, distributions permeabilities matrix fractures. It used here order obtain steady-state macroscopic relative random media. Results are presented as functions mean saturation discussed comparison with simple models.
The percolation threshold of fracture networks is investigated by extensive direct numerical simulations. fractures are randomly located and oriented in three-dimensional space. A very wide range regular, irregular, random shapes considered, monodisperse or polydisperse containing with different and/or sizes. results rationalized terms a dimensionless density. simple model involving new shape factor proposed, which accounts efficiently for the influence shape. It applies good accuracy...