Beyond the Spec: Validating An In-Line Inspection Tool for Metal Loss with Low Expected Probability of Detection

Thursday - January 28 | 15:40 · 7.4 ILI Verification

Fraser Gray1, Susannah Turner1, Aaron Lockey1, Tim Turner1

1Highgrade Associates, Newcastle upon Tyne, UK

This paper discusses in detail the methods used to validate an ILI tool beyond its original specification and the improved understanding of integrity that follows. This demonstrates the strength of tool validation, particularly when incomplete ILI datasets are expected (i.e. many anomalies are unreported), and when confidence in sizing certainty is low. A case study is considered for a production pipeline transporting gas from offshore to an onshore terminal where In-line inspection (ILI) identified deep internal corrosion along the entirety of the route. Integrity assessment based on the findings of the ILI resulted in restriction of the maximum allowable operating pressure (MAOP). However, field validation of a small number of accessible anomalies in the onshore section identified that the metal loss was on average 40% shallower than reported by the ILI, indicating that the ILI may have significantly oversized the anomalies. Further information was therefore required to understand whether the condition was as severe as indicated by the ILI. It was identified that ILI using an alternative tool would be the most appropriate method to better understand the internal condition of the pipeline. However, due to operational and practical limitations, the choice of ILI technology was restricted. It was a key requirement that the new ILI should provide information to confirm the pipeline's condition with confidence. Specifically, that the information should be sufficient to reject the conclusion from the original ILI regarding the presence of multiple, deep internal corrosion anomalies. In addition, a significant proportion of the population of anomalies reported by the original ILI had diameters below the thresholds for the stated probability of detection (POD) and sizing specifications of the new ILI. Given this, the POD and sizing performance of the ILI tool needed to be tested, quantified and validated, including for anomalies outside the stated limits. To achieve this, a pull-through test spool was used. This was designed considering the potential nature of anomalies in the pipeline. It was expected that the new ILI tool would have a low POD for small diameter metal loss and a large number of these smaller anomalies were therefore introduced into the test spool. This ensured that sufficient anomalies were detected by the pull through test to allow a statistically confident sizing model to be developed for metal loss, beyond the tool's stated specification. On successful completion of the pipeline ILI run, the findings were analysed accounting for the sizing performance developed from the test spool, review of correlations between the original and new ILIs, and the available field validation data. Integrity assessment considered both the anomalies reported by the new ILI and the potential population of unreported anomalies. Ultimately, it was concluded that the original MAOP could be reinstated.

Read/Download PaperGradeAvailable after the event