Interpretation of Real-Time Pressure Measurements to Detect CO₂ Leakage

Carbon Capture and Storage (CCS) has been gaining support and popularity as one of the most viable CO₂ emission mitigation methods. In order to assure underground CO₂ storage safety and reduce leakage risk, different CO₂ Monitoring techniques must be utilized. In-zone reservoir pressure, which is transmitted by Permanent Down-hole Gauges (PDG), is a widely used monitoring parameter that can provide important indications when CO₂ migration/leakage occurs. As part of a monitoring package, a Real-Time Intelligent CO₂ Leakage Detection System (RT-ILDS) was developed for CO₂ storage project at Citronelle Dome, Alabama. This system, which is designed based on Pattern Recognition Technology and Smart Wells, is able to identify the location and amount of the CO₂ leakage at the reservoir level using real-time pressure data from PDGs.

In this work, history matched reservoir simulation model (based on 11 months of actual injection/pressure data) was used for CO₂ leakage modeling study. High frequency real time pressure streams were processed with a novel technique to form a new data driven RT-ILDS which was able to detect leakage characteristics in a short time(less than a day). RT- ILDS also demonstrated high precision in quantifying leakage characteristics subject to complex rate behaviors. Finally the performance of RT- ILDS was examined under different conditions as multiple well leakage, availability of additional monitoring well, uncertainty in the reservoir model, CO₂ leakage through the cap rock and multi-well leakage.

The objective of this study was to proof the concept and feasibility of using real time pressure data from PDGs in order to notice occurrence of leakage and identify its characteristics as location and rate in a real CO₂ storage project by utilizing data mining techniques.