Predictive Modeling and Simulation of CO2 Trapping Mechanisms: Insights into Efficiency and Long Term Sequestration Strategies

This study presents a comprehensive analysis of CO₂ trapping mechanisms in reservoirs, integrating reservoir simulations, geochemical modeling, and machine learning techniques to enhance the understanding of carbon capture and storage (CCS). A 2D reservoir model was developed to simulate CO₂ injection dynamics, incorporating various trapping mechanisms: residual, solubility, mineralization, and structural, under realistic geomechanical and geochemical conditions. Advanced machine learning models, including Random Forest, Gradient Boosting, and Decision Trees, were employed to act as Smart Proxy models, significantly reducing computational time while maintaining predictive accuracy. Results highlight the critical role of hysteresis and aquifer dynamics in enhancing trapping efficiency and long-term CO₂ storage stability. The study underscores the potential of combining traditional reservoir engineering methods with machine learning-driven analytics to optimize CCS strategies, ensuring sustainable and secure storage solutions.