A Data-Driven Model for Rapid Evaluation of Miscible CO₂-WAG Flooding in Heterogeneous Clastic Reservoirs

The compositional flow simulation model was frequently used to evaluate the miscible water alternating CO₂ flooding (CO₂-WAG). The uncertainty and sensitivity analysis have to be conducted to examine the parameters mostly affecting the performance of the process. Accordingly, multiple simulation runs require to be constructed which is a time-consuming procedure and finally increase the computational cost. This paper presents a simplistic approach to assess the miscible CO₂-WAG flooding in an Iraqi oilfield through developing a statistical proxy model. The Central Composite Design (CCD) was employed to build the proxy model to determine the incremental oil recovery (ΔFOE) as a function of seven reservoir and operating parameters (permeability, porosity, ratio of vertical to horizontal permeability, cyclic length, bottom hole pressure, ratio of CO₂ slug size to water slug size, and CO₂ slug size). In total, 81 compositional simulation runs were conducted at field-scale to establish the proxy model. The validity of the model was investigated based on statistical tools; the Root Mean Squared Error (RMSE), R-squared statistic and the adjusted R-squared statistic of 0.0095, 0.9723 and 0.9507 confirmed the reliability of the model. The most influential and the optimum values of the parameters that lead to the higher ΔFOE during miscible CO₂-WAG process were identified through proxy modeling analysis. The developed model was created based on the Nahr Umr reservoir in Subba oilfield and can be applied to roughly estimate the ΔFOE during the miscible CO₂-WAG process at the same geological conditions as Nahr Umr reservoir.