Géraldine Pichot
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Géraldine Pichot has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Daniel Zegarra Vasquez
Inria 48 rue Barrault 75013 Paris France
3 shared publications - Michel Kern
Inria 48 rue Barrault 75013 Paris France
3 shared publications - Antoine Lejay
Université de Lorraine CNRS IECL Inria F-54000 Nancy France
2 shared publications Alexandre Ern
Ecole des Ponts
1 shared publication- Alexandre Ern
CERMICS, CNRS, ENPC, Institut Polytechnique de Paris, Paris, France
1 shared publication - Baptiste Poirriez
IRISA, Campus de Beaulieu
1 shared publication
Latest Publications
- Open Access
Mathematical and numerical analysis of the mixed formulation of single phase flow in three-dimensional fractured porous media
Three-dimensional fractured-porous media are typically modeled using a reduced approach known as the Discrete Fracture Matrix (DFM) model. In this model, the rock matrix is represented in three dimensions, while the fracture network is represented as a two-dimensional (co-dimension one) structure. The system is governed by a set of partial differential equations (PDEs) that couple the flow in the 3D rock matrix with the flow in the 2D fracture network. Mixed finite element methods are particularly well-suited for discretizing this system because they ensure local mass conservation. This paper presents an existence and uniqueness result for the PDEs in mixed form, assuming that the pressure is continuous across fractures (i.e., the fractures are conductive). Additionally, error estimates for the mixed finite element discretization are derived under a regularity assumption, and a mixed-hybrid approximation for the coupled problem is developed. The theoretical findings are validated through comparison with an analytical solution.
Open AccessMathematical and numerical analysis of the mixed formulation of single phase flow in three-dimensional fractured porous media
Three-dimensional fractured-porous media are typically modeled using a reduced approach known as the Discrete Fracture Matrix (DFM) model. In this model, the rock matrix is represented in three dimensions, while the fracture network is represented as a two-dimensional (co-dimension one) structure. The system is governed by a set of partial differential equations (PDEs) that couple the flow in the 3D rock matrix with the flow in the 2D fracture network. Mixed finite element methods are particularly well-suited for discretizing this system because they ensure local mass conservation. This paper presents an existence and uniqueness result for the PDEs in mixed form, assuming that the pressure is continuous across fractures (i.e., the fractures are conductive). Additionally, error estimates for the mixed finite element discretization are derived under a regularity assumption, and a mixed-hybrid approximation for the coupled problem is developed. The theoretical findings are validated through comparison with an analytical solution.
- Open Access
Domain decomposition preconditioners for efficient parallel simulations of single-phase flow in three-dimensional fractured porous media with a very large number of fractures
Three-dimensional fractured-porous media are classically described by the Discrete Fracture Matrix (DFM) model, where the rock matrix remains three-dimensional, while the fracture network is of co-dimension one. This study deals with efficient solutions of the linear system arising in single-phase flow in large DFMs using a mixed-hybrid finite element method. It demonstrates the effectiveness of two-level domain decomposition methods based on the GenEO framework for preconditioning iterative solvers applied to such flow problems. Due to the mixed-dimensional nature of the geometry, a specific partitioning strategy is required. The coarse spaces for these preconditioners are built by locally solving spectral problems. This paper presents simulations with very large DFMs, with up to hundreds of thousands of fractures, achieving a low iteration count and short total computation time. Furthermore, this work confirms the scalability of these preconditioners.
Open AccessDomain decomposition preconditioners for efficient parallel simulations of single-phase flow in three-dimensional fractured porous media with a very large number of fractures
Three-dimensional fractured-porous media are classically described by the Discrete Fracture Matrix (DFM) model, where the rock matrix remains three-dimensional, while the fracture network is of co-dimension one. This study deals with efficient solutions of the linear system arising in single-phase flow in large DFMs using a mixed-hybrid finite element method. It demonstrates the effectiveness of two-level domain decomposition methods based on the GenEO framework for preconditioning iterative solvers applied to such flow problems. Due to the mixed-dimensional nature of the geometry, a specific partitioning strategy is required. The coarse spaces for these preconditioners are built by locally solving spectral problems. This paper presents simulations with very large DFMs, with up to hundreds of thousands of fractures, achieving a low iteration count and short total computation time. Furthermore, this work confirms the scalability of these preconditioners.
- Open Access
Performance of algebraic preconditioners for large-scale simulations of single-phase flow in three-dimensional fractured porous media
A three dimensional fractured-porous medium is classically described by a reduced model, called the discrete fracture matrix (DFM), where the rock matrix remains three-dimensional, while the fracture network is of co-dimension one. Solving single-phase flow in such a medium is particularly challenging, especially when a large number of fractures is present. This difficulty primarily arises from the intricate complexity introduced by the fracture network. This work demonstrates that solving flow in such systems is feasible, even for discrete fracture networks as large as 90,000 fractures. The mixed-hybrid finite element method is employed to discretize the flow problem, resulting in a large linear system that must be solved. However, the efficiency of direct linear solvers is severely hindered by the three-dimensional nature of the problem, requiring a substantial amount of RAM. Iterative methods, when preconditioned with algebraic multigrid techniques, offer a better alternative in terms of memory usage. However, their effectiveness declines significantly as the fracture network grows and transmissivity within the network becomes highly heterogeneous, as both lead to poor conditioning of the linear system. Even more critically, we show that achieving an adequately low tolerance for iterative solvers is essential for accurate flow simulations. The choice of tolerance strongly influences the quality of the solution, insufficient precision can prevent local mass conservation in some cases, resulting in non-physical values in fully advective transport simulations.
Open AccessPerformance of algebraic preconditioners for large-scale simulations of single-phase flow in three-dimensional fractured porous media
A three dimensional fractured-porous medium is classically described by a reduced model, called the discrete fracture matrix (DFM), where the rock matrix remains three-dimensional, while the fracture network is of co-dimension one. Solving single-phase flow in such a medium is particularly challenging, especially when a large number of fractures is present. This difficulty primarily arises from the intricate complexity introduced by the fracture network. This work demonstrates that solving flow in such systems is feasible, even for discrete fracture networks as large as 90,000 fractures. The mixed-hybrid finite element method is employed to discretize the flow problem, resulting in a large linear system that must be solved. However, the efficiency of direct linear solvers is severely hindered by the three-dimensional nature of the problem, requiring a substantial amount of RAM. Iterative methods, when preconditioned with algebraic multigrid techniques, offer a better alternative in terms of memory usage. However, their effectiveness declines significantly as the fracture network grows and transmissivity within the network becomes highly heterogeneous, as both lead to poor conditioning of the linear system. Even more critically, we show that achieving an adequately low tolerance for iterative solvers is essential for accurate flow simulations. The choice of tolerance strongly influences the quality of the solution, insufficient precision can prevent local mass conservation in some cases, resulting in non-physical values in fully advective transport simulations.
- Open Access
- 4 October 2024
- Published by Springer Nature in Transport in Porous Media
- Vol. 151 (14) , 2625-2645
Random Walk Modeling of Conductive Heat Transport in Discontinuous Media
We consider heat transport within a discontinuous domain by relying on the modeling approach proposed by Baioni et al. Such approach has been specifically designed to address diffusive processes in media with discontinuous physical properties and generalized boundary conditions at the discontinuities. Three algorithms are here applied to estimate the conductive heat transport in a bimaterial medium. The algorithms undergo testing using two test cases that share the same computational domain but differ in terms of their initial conditions. According to the numerical results all the algorithms ensure the conservation of thermal energy and preserve thermal equilibrium under steady state conditions. The Generalized Uffink Method (GUM) and Generalized HYMLA demonstrate sensitivity to the choice of the time step, whereas the Generalized Skew Brownian Motion appears to be unaffected by the value of $$\Delta t$$ . The GUM algorithm presents an optimal trade-off between accuracy and computational time.- 4 October 2024
- Published by Springer Nature in Transport in Porous Media
- Vol. 151 (14) , 2625-2645
Open AccessRandom Walk Modeling of Conductive Heat Transport in Discontinuous Media
We consider heat transport within a discontinuous domain by relying on the modeling approach proposed by Baioni et al. Such approach has been specifically designed to address diffusive processes in media with discontinuous physical properties and generalized boundary conditions at the discontinuities. Three algorithms are here applied to estimate the conductive heat transport in a bimaterial medium. The algorithms undergo testing using two test cases that share the same computational domain but differ in terms of their initial conditions. According to the numerical results all the algorithms ensure the conservation of thermal energy and preserve thermal equilibrium under steady state conditions. The Generalized Uffink Method (GUM) and Generalized HYMLA demonstrate sensitivity to the choice of the time step, whereas the Generalized Skew Brownian Motion appears to be unaffected by the value of $$\Delta t$$ . The GUM algorithm presents an optimal trade-off between accuracy and computational time. - Open Access
Hybrid high-order methods for flow simulations in extremely large discrete fracture networks
We investigate the computational performance of hybrid high-order methods applied to flow simulations in extremely large discrete fracture networks (over one million of fractures). We study the choice of basis functions, the trade-off between increasing the polynomial order and refining the mesh, and how to take advantage of polygonal cells to reduce the number of degrees of freedom.
Open AccessHybrid high-order methods for flow simulations in extremely large discrete fracture networks
We investigate the computational performance of hybrid high-order methods applied to flow simulations in extremely large discrete fracture networks (over one million of fractures). We study the choice of basis functions, the trade-off between increasing the polynomial order and refining the mesh, and how to take advantage of polygonal cells to reduce the number of degrees of freedom.
- 22 August 2020
- Published by Springer Nature
- p. 521-529
A Hybrid High-Order Method for Flow Simulations in Discrete Fracture Networks
- 22 August 2020
- Published by Springer Nature
- p. 521-529
A Hybrid High-Order Method for Flow Simulations in Discrete Fracture Networks
- 1 November 2019
- Published by Elsevier in Journal of Computational Physics
- Vol. 396, 888-904
An exponential timestepping algorithm for diffusion with discontinuous coefficients
- 1 November 2019
- Published by Elsevier in Journal of Computational Physics
- Vol. 396, 888-904
An exponential timestepping algorithm for diffusion with discontinuous coefficients
- 21 April 2014
- Published by Springer Nature
- p. 99-112
A Mortar BDD Method for Solving Flow in Stochastic Discrete Fracture Networks
- 21 April 2014
- Published by Springer Nature
- p. 99-112
A Mortar BDD Method for Solving Flow in Stochastic Discrete Fracture Networks