Deep Learning Framework for Accurate Static and Dynamic Prediction of CO2 Enhanced Oil Recovery and Storage Capacity

As global warming intensifies, carbon capture, utilization, and storage (CCUS) technology is widely used to reduce greenhouse gas emissions. CO2-enhanced oil recovery (CO2-EOR) technology has, once again, received attention, which can achieve the dual benefits of oil recovery and CO2 storage. However, flexibly and effectively predicting the CO2 flooding and storage capacity of potential reservoirs is a major problem. Traditional prediction methods often lack the ability to comprehensively integrate static and dynamic predictions and, thus, cannot fully understand CO2-EOR and storage capacity. This study proposes a comprehensive deep learning framework, named LightTrans, based on a lightweight gradient boosting machine (LightGBM) and Temporal Fusion Transformers, for dynamic and static prediction of CO2-EOR and storage capacity. The model predicts cumulative oil production, CO2 storage amount, and Net Present Value on a test set with an average R-square (R2) of 0.9482 and an average mean absolute percentage error (MAPE) of 0.0143. It shows great static prediction performance. In addition, its average R2 of dynamic prediction is 0.9998, and MAPE is 0.0025. It shows excellent dynamic prediction ability. The proposed model successfully captures the time-varying characteristics of CO2-EOR and storage systems. It is worth noting that our model is 105–106 times faster than traditional numerical simulators, which once again demonstrates the high-efficiency value of the LightTrans model. Our framework provides an efficient, reliable, and intelligent solution for the development and optimization of CO2 flooding and storage.