A Quantum Gravity AI Framework for CO2 Storage Monitoring and Optimization

Abstract

Abstract Gravimetry is a physical method with a large depth of investigation. Traditional applications include surface gravity observations for mining and oil exploration and borehole gravity logging for investigating formation bulk density. Quantum gravity sensors have recently been developed allowing to achieve considerably higher accuracy and signal to noise ratios as compared to conventional gravimetric approaches. Borehole gravity data have some advantages over the surface data, because the sensors are closer to the reservoir better spatial resolution is obtained; and because the deep borehole gravity data are less affected than surface data by near surface changes. We have developed a new AI driven framework for the interpretation and monitoring of CO2 migration for CO2 storage applications. The framework utilize an integrated LSTM -Bayesian inference framework approach that to determine the gravity gradient within the reservoir and infer from this the possible movement in the reservoir. The LSTM framework evaluates the time lapse gravity gradient changes to infer from it the migration of the CO2 movement. We evaluated the framework on a public benchmark dataset of the Pohokura field in New Zealand. The Pohokura field in New Zealand has been investigated as a reservoir for CO2 storage given its acceptable reservoir quality and seal rock structure. The framework was evaluated on simulated CO2 storage migration patterns with multiple scenarios, taking into account the uncertainties that may arise with respect to various potential CO2 migration scenarios. The study outlines the enhanced accuracy and tracking of CO2 front movement within the reservoir based on quantum gravity sensors integrated with an AI framework. The deep learning framework represents an important step at utilizing quantum borehole gravity sensing for CO2 movement monitoring and the optimization of CO2 storage. The AI framework outlined the considerable potential of quantum gravity sensing for CO2 storage monitoring and optimization.