Employing Artificial Intelligence to Enhance Reservoir Property Estimation in Waste Sequestration Wells: Pressure Transient Analysis Through Physics-Informed Machine Learning

Abstract Global efforts towards environmentally safe waste disposal necessitate Waste Slurry Injection (WSI) wells to be monitored for operational compliance, safety, sustainability, and optimization of waste disposal and vaulting capacity. Pressure Transit Analysis (PTA) is a comprehensive and time-consuming analysis conducted on shut-in data collected post-injection to assess reservoir response to the injection and ensure compliance. Our engineering team performs PTA on over 7,000 injection batches annually, which opens a venue for AI optimization. This paper presents a Physics Informed Machine Learning (PIML) method that enhances PTA accuracy, predicting reservoir properties to enhance WSI and waste disposal. Our methodology integrates standard PTA with machine learning informed by physical science and geomechanical principles. We have developed and trained a Physics-Informed Machine Learning (PIML) model using shut-in data, to enable early prediction of the behavior of hydraulic fractures and other reservoir properties prior to further injections. This model is validated against 35,000 injection batches and their corresponding PTA results from 10 wells over the last 5 years and allows for better storage capacity management and operational forecasting. The application of the PIML model demonstrates a significant improvement in estimating the formation response prior to the injection operations, solely based on controllable parameters which can be changed at the surface. Results indicate that our model can predict injectivity, formation stress, fracture closure pressure and time, wellhead pressure at closure, bottom-hole instantaneous shut-in pressure, fracture half-length, skin, and transmissivity with high confidence and accuracy, thus enabling more informed decision-making during the disposal process. Observations from multiple test scenarios confirm that our model reliably estimates critical reservoir characteristics under varying conditions. Conclusions drawn from these experiments suggest that incorporating machine learning with physical science principles substantially reduces the risks associated with hydraulic fracturing during waste disposal. Additionally, the ability to predict reservoir behavior ahead of injections can lead to better compliance with safety standards and reduced adverse environmental impact, or setting a new benchmark for operational practices in the industry. Through the introduction of a novel integration of physics-informed machine learning with conventional PTA techniques, this paper contributes new insights into the preemptive estimation of reservoir properties, potentially impacting operational protocols in underground waste disposal. Our methodology and findings provide actionable intelligence that could influence future designs and operational strategies in the energy industry, enhancing both safety and efficiency.