A machine learning‐based approach to the multiobjective optimization of CO ₂ injection and water production during CCS in a saline aquifer based on field data

The presence of carbon capture and storage (CCS) projects is important due to the growing production greenhouse gases, especially dioxide (CO2). Our target functions have been chosen because importance CO2 in CCS requirement for producing less water requiring production. As a proxy reservoir simulations, support vector regression, artificial neural network (ANN), multivariate adaptive regression spline (MARS) used. It was determined that MARS had higher accuracy based on examining these three data-driven models with available field data. was, however, very close ANN. gave root mean square error (RMSE), absolute (MAE), R2 values 2.78%, 1.95%, 0.998, respectively, predicting test data, yet 3.73%, 3.53%, 0.995 blind RMSE, MAE, evaluate machine learning (ML) model were 3.58%, 2.81%, 0.997, while results data 3.94%, 2.83%, respectively. By reducing amount computational load time taken reproduce simulation by MARS, it will be beneficial optimization. This study highlights application ML approach coupling genetic algorithms optimizing With proposed frameworks, preprocessing, feature selection, two-stage validation model, optimizer are performed aquifer repository containing series optimal solutions developed used projects.