A multiscale hydro‐geochemical‐mechanical approach to analyze faulted CO₂ reservoirs

Abstract

This paper applies a multiscale hydro‐geochemical‐mechanical approach to analyze faulted CO2 reservoirs using the STOMP‐CO2‐R code that is coupled to the ABAQUS® finite element package. STOMP‐CO2‐R models the reactive transport of CO2 causing mineral composition changes that are captured by an Eshelby‐Mori‐Tanka model implemented in ABAQUS®. A three‐dimensional (3D) STOMP‐CO2‐R model for a reservoir containing an inclined fault was built to analyze a formation containing a reaction network with five minerals: albite, anorthite, calcite, kaolinite, and quartz. A 3D finite element mesh that exactly maps the STOMP‐CO2‐R grid was developed for coupled analyses. The model contains alternating sandstone and shale layers. The impact of reactive transport of CO2 on the geomechanical properties of reservoir rocks are studied in terms of mineral composition changes that affect their geomechanical responses. Simulations assuming extensional and compressional stress regimes with and without coupled geochemistry are performed to study the stress regime effect on the risk of hydraulic fracture. The fault slip is examined as functions of stress regime, geomechanical, geochemical‐mechanical effects, fault inclination, and position. The results show that mineralogical changes due to CO2 injection reduce the permeability and elastic modulus of the reservoir, leading to increased fluid pressure and risk of fracture in the injection location and the caprock seal. Shear failure in the fault leading to fault reactivation was not predicted to occur. However, stress regime, fault inclination, and fault position in light of the coupled hydro‐geochemical‐mechanical analysis have an important impact on the slip tendency factor and elastic fault slip. © 2016 Society of Chemical Industry and John Wiley & Sons, Ltd.

Statistical based hydromechanical models to estimate poroelastic effects of CO2 injection into a closed reservoir

“…Coupled hydromechanical modeling is typically used for evaluation of the poroelastic effect of injection as well as the resulting geomechanical outcomes such as the potential for fracturing in reservoirs, analysis of slippage along faults, surface uplift, and associated seismicity . Different numerical approaches including finite element method, finite difference method, discrete element method, and boundary element method have been used to address in situ stress changes and rock deformation . Combinations of different methods have also been developed to address the poroelastic response of injection.…”

“…8,[16][17][18][19][20][21][22] Different numerical approaches including finite element method, finite difference method, discrete element method, and boundary element method have been used to address in situ stress changes and rock deformation. [23][24][25][26][27][28] Combinations of different methods have also been developed to address the poroelastic response of injection. One commonly used tool is TOUGH-FLAC modeling, which is based on linking the finite-volume code for the simulation of multiphase fluid flow (TOUGH2) with the finite-difference code for the simulation of geomechanics (FLAC), and has been extensively used to predict stress changes that might activate the faults and induce seismicity.…”