Evaluation of the fault reactivation dynamics using fully coupled hydro-mechanical models

Jose Antonio Torres,Michel Boisson,Claude Gapillou
DOI: https://doi.org/10.2139/ssrn.4285679
2022-01-01
SSRN Electronic Journal
Abstract:In this paper we present a fully-coupled hydro-mechanical (HM) model designed to evaluate potential containment risks presented when injecting fluids in a faulted reservoir. The model captures dynamic fracture re-opening and fluid leakage associated to pressure changes. To validate our modelling and simulation approach we investigated two cases. First we validated a stress-dependent fracture permeability relationship using a conceptual benchmark case with analytical solution. Then, we simulated the Mont Terri fault stimulation FS-B experiment. This in-situ experiment involves controlled injection of fluids in the proximity of a fault surrounded by a clay-rich caprock. In our study we employed commercially available software: the geomodelling tasks were carried out using the PetrelTM platform, while the simulation tasks were performed with COMSOL Multiphysics®. With these tools we developed a 3-D physics-based model that estimates the pressure, stress and local displacement changes caused by the fluid injection. The architecture of the fault zone is represented using 3-D structural objects transferred from a more realistic geological model that was built using a comprehensive characterization dataset. The 3-D FEM model, named “bedding-parallel fractured” model, is composed of several geological features such as a fault core damaged zone that is surrounded by a rock matrix, and a set of explicit fractures parallel to the bedding planes. The discrete fractures, modelled using thin structures, act as high permeability pathways that strongly influence the poroelastic response. The mathematical model relies on the classical Biot’s theory of poroelasticity. Results indicate that the local fracture transmissibility can increase up to 6 orders of magnitude as a result of a reduction in the normal effective stress due to a change in pressure up to 9MPa. Simulation results emphasize the critical role of the fracture connectivity on the poroelastic effects. To sum up, we have devised a methodology which takes into account complex hydro-mechanical interactions in fractured media. Our method is a practical option to evaluate potential risks associated with storage integrity and induced seismicity on field operations. Future studies need to be done to establish multiscale approaches for coupling larger-scale effects, such as the CO2 plume evolution, with local-scale effects such as fault slip or damage mechanics.
What problem does this paper attempt to address?