Numerical simulation of carbon dioxide migration in a sandstone aquifer considering the fluid-solid coupling

摘要

Very limited investigations have been done on the numerical simulation of carbon dioxide (CO_2) migration in sandstone aquifers taking consideration of the interactions between fluid flow and rock stress. Based on the poroelasticity theory and multiphase flow theory, this study establishes a mathematical model to describe CO_2 migration, coupling the flow and stress fields. Both finite difference method (FDM) and finite element method (FEM) were used to discretize the mathematical model and generate a numerical model. A case study was carried out using the numerical model on the Jiangling sandstone aquifer in the Jianghan basin, China. The rock mechanics parameters of reservoir and overlying strata of Jiangling depression were obtained by triaxial tests. A two-dimensional model was then built to simulate carbon dioxide migration in the sandstone aquifer. The numerical simulation analyzes the carbon dioxide migration distribution rule with and without considering capillary pressure. Time-dependent migration of CO_2 in the sandstone aquifer was analyzed, and the result from the coupled model was compared with that from a traditional non-coupled model. The calculation result indicates a good consistency between the coupled model and the non-coupled model. At the injection point, the CO_2 saturation given by the coupled model is 15.39 % higher than that given by the non-coupled model; while the pore pressure given by the coupled model is 4.8 % lower than that given by the non-coupled model. Therefore, it is necessary to consider the coupling of flow and stress fields while simulating CO_2 migration for CO_2 disposal in sandstone aquifers. The result from the coupled model was also sensitized to several parameters including reservoir permeability, porosity, and CO_2 injection rate. Sensitivity analyses show that CO_2 saturation is increased non-linearly with CO_2 injection rate and decreased non-linearly with reservoir porosity. Pore pressure is decreased non-linearly with reservoir porosity and permeability, and increased nonlinearly with CO_2 injection rate. When the capillary pressure was considered, the computed gas saturation of carbon dioxide was increased by 10.75 % and the pore pressure was reduced by 0.615 %.