Speaker: Nina Fermet-Quinet

Date: Thursday 25th of November 2021, 1:15 pm.

Abstract:

La Guyane est un territoire géo-écologique notamment caractérisée par la présence d’or primaire et secondaire. L’exploitation de cette ressource génère des impacts positifs (emplois directs, développement de l’activité locale…) et négatifs (destruction de la biodiversité, pollution…). Dans une précédente thèse, une méthodologie d’évaluation des risques positifsetnégatifs baséesur différents scénariosdedéveloppement minier a été testée et validée afin de pouvoir comparerdifférents scénariosà l’échelle duterritoire. L'objectif demathèse est de généraliser cette approche en considérant (i) d’autres événements générateurs que la rupture de digue minière ; (ii) des impacts ordinaires causés par l’activité normale des projets (iii) l’activité illégale sur le territoire. Suite à une première mission de terrain en Guyane, les observations et les données récoltées seront exposées afin de discuter des nouvelles perspectives de recherches quisont associées.

 

Speaker: Narges Dashtbesh

Date: Thursday 18th of November 2021, 1:15 pm.

Abstract:

I will present in this seminar a sequential approach to accelerate immiscible multiphase flow modelling in heterogeneous porous media using discontinuous Galerkin methods and dynamic mesh coarsening. This approach involves dynamic domain decomposition and different solution strategies in the different flow regions, using a criterion that can be fastly evaluated. I will present a technique to estimate the position of the saturation front and identify the flow zones that need high-resolution gridding. The accuracy of the proposed approach is evaluated through test cases from the SPE10 model.
In the second part, I will briefly talk about my postdoc project on groundwater contaminations by nonaqueous phase liquids (NAPL), as part of the QUASPER project.

 

Speaker: Enrico Scarpa

Date: Thursday 4th of November 2021, 1:15 pm.

Abstract:

Proper representation of connected features is crucial when modeling the fluid circulation in the subsurface. In that framework, connectivity metrics are essential tools to describe a reservoir and its hydrodynamic behavior. This seminar will briefly review the role of connectivity metrics developed for a channelized deep-water system. The discussed metrics cover static and dynamic points of view. The static connectivity metrics depend only on the parameter fields such as hydraulic conductivity or geological facies. Conversely, dynamic connectivity metrics are related to physical processes such as flow or transport. This talk wants to highlight the influences of the stacking pattern of channels and their impact on simple flow models. The purpose is to discuss the relations between the geological representation of connected components, connectivity measures, and the hydrodynamic behaviors of heterogeneous deposits.

Speaker: Ever-Dennys Coarita-Tintaya

Date: Thursday 28th of October 2021, 1:15 pm.

Abstract:

Context of this work is related to underground hydrogen storage in salt caverns. The operational phase (injecting and extracting, according to daily or seasonal cycles) can affect stability of the cavern and initiate development of fractured zones. This can modify the transport properties and therefore potentially cause hydrogen leakage problems around the cavern.

Based on the mechanical and hydromechanical behaviour of salt rock known from experimental data, as well as on the rheological models available for this material, we propose a model that considers the short-term and long-term behaviour of salt rock. In the short-term behaviour we have a macroscopic elastoplastic and damage constitutive model. In the long-term behaviour, the three creep types are considered: (i) transient (or primary) creep; (ii) steady-state (or stationary) creep; and (iii) acceleration (or tertiary) creep. Thus, this model describes the key mechanisms of salt rock behaviour.

To evaluate and validate the numerical implementation in Comsol Multiphysics®, triaxial tests were simulated and the results obtained correlate well with the theoretical criteria. The application of the proposed model was evaluated on salt caverns where hydromechanical simulations were carried out under saturated conditions. The analysed scenarios were shallow and very depth caverns, as well as seasonal and daily hydrogen operating cycles.

Speaker: Capucine Legentil

Date: Thursday 21st of October 2021, 1:15 pm.

Abstract:

In this work, we present a local updating method to test different contact depth scenarios and assess their impact on wave propagation in the subsurface. We propose to locally modify a 2D geological model and run time-dependent elastic simulations. The input model triangulation is conforming to geological structures. The 2D meshed model is locally updated, which means that only the reservoir compartment is modified. Several model geometries are generated by inserting a new interface, in this paper a gas-water contact that is defined by a scalar field. We quantitatively evaluate the impact of the gas-water contact depth on elastic wave propagation. We run the numerical simulations with Hou10ni2D code, which is based on a Discontinuous Galerkin method. The simulation results are compared to a reference depth by computing the L2-norm at a set of seismic receivers. Results show a consistent behavior: we observe a positive correlation between the depth difference and global L2-norm for all receivers. This approach could therefore be integrated into an inversion loop to determine the position of the fluid contact and reduce uncertainties in the reservoir model from a few seismic sensors.

Speaker: Emilio Abi Aad

Date: Thursday 14th of October 2021, 1:15 pm.

Abstract:

L’objectif principal de cette thèse est de comprendre le comportement mécanique d’un massif rocheux soumis à des sollicitations mécaniques. Notre objectif à terme est de modéliser une exploitation minière creusée dans un massif rocheux discontinu, en utilisant une modélisation physique. Ce type de modélisation permet d’obtenir des résultats quantitatifs en complément de la modélisation numérique. Néanmoins, la transposition des résultats de l’échelle réduite à l’échelle du prototype nécessite de respecter les facteurs d’échelle.

L’originalité de ce travail consiste à développer une méthode innovante, basée sur les techniques d’impression 3D combinant du sable et un liant phénolique, pour introduire explicitement des joints rocheux à propriétés mécaniques et géométriques contrôlées. L’utilisation de cette technique peut permettre la réalisation de modèles réduits discontinus de géométrie très précise et reproductibles.


Speaker: Nathan Amrofel

Date: Thursday 7th of October 2021, 1:15 pm.

Abstract:

Pour la pérennité des roches hôtes utilisées comme barrière dans les dépôts de déchets radioactifs, il est d'une grande importance de comprendre la migration des gaz dans ces formations rocheuses sédimentaires fortement saturées en eau. Compte tenu des fortes pressions générées, la percolation du gaz engendre une dilation des chemins percolants, pouvant conduire à un déplacement localisé de l’eau loin de ces chemins, sous l’effet de la compression de la matrice argileuse. Ces phénomènes entraînent à leur tour un endommagement et une dégradation des propriétés mécaniques de la roche. Si la pression continue d’augmenter, des fractures en tension peuvent se développer. D’autre part, cette percolation du gaz dans un milieu très peu perméable initialement saturé va s’effectuer dans des très faibles gammes de vitesse d’écoulement. Il a été démontré que les effets de séchage associés à ces faibles vitesses peuvent impacter les processus de transfert du gaz. Compte tenu des tailles nanométriques des pores, l’équilibre thermodynamique entre le fluide et le gaz va dépendre de la courbure de l’interface liquide-gaz et sera alors régie par l’équation de Kelvin. Afin de mieux comprendre le transport du gaz, une étude numérique couplée hydro-mécanique à l'échelle des pores est proposé. Le modèle est proposé en formulation Smoothed Particle Hydrodynamics (SPH) et sera appliqué à l’argilite du Callovo-Oxfordien.

Speaker: Jeremie Giraud

Date: Thursday 30th of September 2021, 1:15 pm.

Abstract:

To reduce uncertainties in reconstructed images, geologic information must be introduced in a numerically robust and stable way during the geophysical data inversion procedure. In the context of potential (gravity) data inversion, it is important to bound the physical properties by providing probabilistic information on the number of lithologies and ranges of values of possibly existing related rock properties (densities). For this purpose, we have introduced a generalization of bounding constraints for geophysical inversion based on the alternating direction method of multipliers. The flexibility of the proposed technique enables us to take into account petrophysical information as well as probabilistic geologic modeling, when it is available. The algorithm introduces a priori knowledge in terms of physically acceptable bounds of model parameters based on the nature of the modeled lithofacies in the region under study. Instead of introducing only one interval of geologically acceptable values for each parameter representing a set of rock properties, we define sets of disjoint intervals using the available geologic information. Different sets of intervals are tested, such as quasidiscrete (or narrow) intervals as well as wider intervals provided by geologic information obtained from probabilistic geologic modeling. Narrower intervals can be used as soft constraints encouraging quasidiscrete inversions. The algorithm is first applied to a synthetic 2D case for proof-of-concept validation and then to the 3D inversion of gravity data collected in the Yerrida Basin (Western Australia). Numerical convergence tests show the robustness and stability of the bound constraints that we apply, which is not always trivial for constrained inversions. This technique can be a more reliable uncertainty reduction method as well as an alternative to other petrophysically or geologically constrained inversions based on the more classic “clustering” or Gaussian-mixture approaches.

Speaker: Luyu Wang

Date: Thursday 23rd of September 2021, 1:15 pm.

Abstract:

In this presentation, firstly, discrete fractures and their stochastic generations based on key geometrical parameters are presented. After a short revisit of hierarchical modelling, Discrete Fracture Modeling (DFM) on unstructured grids is introduced. According to the concept of DFM, a hybrid-dimensional scheme for fluid flow simulation is devised with adaptive iteration. The hybrid-dimensional objects are depicted in porous media, with high contrast hydraulic properties. Then, a mixed-finite-element (FE) scheme is presented to handle contact mechanics on fractures using Lagrange multipliers (LMs) in the framework of constrained variational principle. In contrast to standard-FE, LMs are introduced in the expanded system as an additional unknown. Later, FE-cohesive zone model is presented as an alternative method to mechanical simulation. Finally, several numerical tests are performed to study fluid flow as well as geomechanics of fractured media using the developed numerical schemes.