Operators need to keep their pipelines fit for purpose, maximise life and control costs. Effective Integrity Management is essential to achieve these aims, and a clear understanding of the relevant threats is key.
External corrosion is one of the main threats faced by operators, costing millions annually in identification, mitigation, and repair. ILI is widely used to identify and size external corrosion. Methods exist to model the growth of corrosion features to predict when they may reach critical dimensions and to define re-inspection schedules. For “unpiggable” pipelines, the situation is more complicated and knowledge-based models, based on data and assumptions for multiple variables that are believed to contribute to corrosion, combined with above ground surveys, are used to identify corrosion “hotspots” for in-field investigation. The process is known as ECDA. These models and surveys can present significant uncertainties and often multiple iterations of costly excavation and model updating are needed to obtain reasonable confidence in pipeline condition.
Data collected by ILI over many years, combined with relevant data such as rainfall, soil type and coating, contains information on corrosion trends across thousands of pipeline segments. Machine learning algorithms trained on this historic data have the potential to substantially reduce uncertainty.
To gain confidence in these new approaches and get the benefit of improved integrity management, requires a collective industry effort involving trials under a variety of conditions. The potential is demonstrated by the good results obtained in recent studies using the concept of “Virtual ILI”, to predict the location, extent, severity, and growth rate of external corrosion.
This paper explores the potential benefits of using “Virtual ILI” in combination with ECDA in pipeline integrity management, and how it can improve decisions.