Tectonically stable deep saline reservoirs are considered to be the most abundant carbon dioxide (CO2) sequestration sites. Pressure, temperature, salinity and other characteristics of these geologic formations vary at each injection site. It is essential to understand the roles of geologic and engineering factors to optimize the CO2 injection conditions and to quantify the associated risks. Factors such as the magnitude of injection-induced reservoir pressure, quantity of supercritical phase CO2 that comes in contact with caprock, and the amount of residually trapped CO2 govern the fate of CO2, and provide quantitative assessment of the storage integrity.A streamlined protocol was developed using response surface methodology to quantify the risks related to CO2 injection. The proposed methodology includes the design of simulation scenarios, selection and screening of parameters, multiple-linear regression of outcomes, and the development of probability density functions (PDF) of various potential risk factors. Multiphase numerical simulations were performed to understand the behavior of the injected CO2 and associated parameters in deep saline reservoirs with prescribed geometries and petrophysical properties. Formation thickness, formation depth, porosity, horizontal permeability, brine density, and the end-point residual CO2 saturation were the six critical parameters identified that affected important outcomes. A six-factor Box-Behnken experimental design procedure was used to establish an understanding of the sensitivity of the parameters on the important factors, and for subsequently establishing response surfaces. Closed boundary domains with different operational constraints were employed. A stepwise, sequential regression method was used to determine statistically significant coefficients of a response surface model. Monte Carlo simulations with logical distributions of input parameters were performed using the response surface coefficients. Uncorrelated and correlated porosity-permeability distributions were used to generate two types of probability density functions (PDF). PDFs of CO2 plume extent under the caprock and average reservoir overpressure after injection were generated given all of the variability in the input parameters. These results will allow initial screening of a large number of potential injection sites without detailed simulations of each.
All Science Journal Classification (ASJC) codes
- Industrial and Manufacturing Engineering
- Management, Monitoring, Policy and Law