Greening enhanced oil recovery: A solar tower and PV-assisted approach to post-combustion carbon capture with machine learning insights

Abstract

Carbon Capture Utilization and Storage (CCUS) has become a cornerstone in reducing industrial emissions, mainly through Enhanced Oil Recovery (EOR) in underground reservoirs. Conventional post-combustion carbon capture (PCC) systems, however, face significant energy penalty challenges. This study introduces an innovative solar-assisted approach to optimize the EOR factor while reducing the energy penalty. The proposed system uniquely integrates solar tower heliostats and photovoltaic (PV) systems with up to 7 h of energy storage, marking a dual solar energy integration as the core innovation. This hybrid configuration reduces the energy penalty factor from 21.2 % to 7.4 %. To further enhance operational efficiency, the study incorporates a novel compression stream configuration with process integration into the PCC system. Machine learning models, including linear regression, random forest, decision tree, and XGBoost, were employed to model and predict EOR factors using CO₂ streams from a large-scale carbon capture unit at the Abadan power plant in Iran. The decision tree model achieved superior performance with an R² of 0.98 and accurately forecasted an increase in EOR factor from 19 % to 43.16 %. By combining solar-driven energy systems with advanced CO₂ capture and predictive modeling, this study establishes a sustainable and energy-efficient framework for EOR enhancement. The dual integration of solar towers and PV systems represents a significant leap in reducing fossil fuel dependence and carbon emissions while demonstrating practical applicability in high-emission regions like Abadan.