Fingerprinting Burst and Collapse Events in Well Casings Using Electromagnetic Logs

Abstract Well casings can burst or collapse due to sudden changes in pressure within the well that can put stress on the casings. The likelihood of occurrence of burst or collapse events can be exacerbated by the presence of corrosion on the casings that creates weak spots for these events to happen. Bursting may involve the casing breaking open or developing ruptures, while collapsing involves the casing collapsing inward on itself which may also be associated with breaks in the walls of the casing. Both scenarios can lead to well integrity issues, water contamination, and other safety hazards, so it is essential to prevent and address these issues promptly through proper maintenance and inspection protocols. A key component in the remediation process is the identification of the anomaly type on each one of the casings. Several technologies have been utilized for monitoring the integrity of well casings, among which only electromagnetic (EM) technologies can be used to inspect outer casings beyond the immediate pipe in which the tool is run. In this work, we used multi-frequency, multi-spacing EM logs to detect the presence of burst or collapse events and identify the anomaly type on each pipe. A database of signatures for different combinations of burst and collapse events on nested casings is created using three-dimensional modeling of mockup fixtures. The database serves to fingerprint different failure modes that can be encountered in practice under pressure imbalance sequences within the different annuli. Measured data are matched against the signatures in the database to detect the presence of burst or collapse events and determine the most likely failure mode. Examples using modelled fixtures showed the sensitivity of EM measurements to differentiate different combinations of burst and collapse events on the casings. The signature of burst and collapse events in EM logs are different than those of metal loss due to corrosion, and therefore conventional processing is deficient when it comes to identifying the type of anomaly of the pipes in the presence of burst and collapse events. The proposed workflow allows for fingerprinting burst and collapse events in well casings, and that can be used in combination with pressure testing for planning remediation operations.