My thesis investigates innovative approaches to carbon management and hydrogen storage to support the transition to a low-carbon future, particularly in enhanced oil recovery (EOR) contexts. Employing advanced analytics and machine learning, the study focuses on evaluating the effectiveness of flue gas and CO2 in Water Alternating Gas (WAG) injection within low porosity and permeability fractured carbonate reservoirs. A computational model using Eclipse (E300) software simulates reservoir behavior and evaluates hybrid EOR methods, revealing a significant oil recovery factor increase.
The research also explores CO2 geo-sequestration integrated with EOR, demonstrating that flue gas exhibits superior storage capacity and better reservoir pressure maintenance compared to CO2. Sensitivity analyses indicate that increased reservoir porosity, permeability, and injection rates enhance gas storage capacity and recovery rates. Additionally, a data-driven framework for offshore CO2 storage site screening is developed, leveraging machine learning algorithms, particularly Deep Neural Networks (DNN), achieving high predictive accuracy.
Furthermore, the study addresses hydrogen storage potential in geological formations by developing predictive models for water-hydrogen-rock interactions using a comprehensive dataset. The findings underscore the critical role of advanced analytics in refining carbon management practices and optimizing hydrogen storage capabilities. This research provides essential insights for the energy sector's sustainable transition, highlighting the integration of CO2 storage with EOR as a viable strategy for reducing greenhouse gas emissions while enhancing oil recovery metrics.