This case study describes the crack management program for a pipeline that is NPS26, 7.1 mm wall thickness, Grade X52, flash welded and was constructed from 1954–1957. It transports light crude oil, and experiences pressure cycling though start-stop operations. Excavations have revealed that the pipeline’s flash welded seam contains a variety of manufacturing anomalies. While the majority of these anomalies are benign and stable, some exhibit the potential to grow due to the pressure cycling of the line. Furthermore, in 2010, the pipeline experienced a rupture and leak that were caused by cracks in the longitudinal seam weld. Correspondingly, the cracking threat has been actively managed using in-line inspection (ILI), excavation, and repair programs. The most recent ILI runs were conducted with ultrasonic crack detection (USCD / UTCD) and ultrasonic phased array crack detection (DUO) tools in 2009, 2012 and 2013.
As a part of these ILIs, a comprehensive excavation program comprised of over 300 excavations was conducted to validate the ILI data and mitigate the cracking threat. Unity plots comparing the measurements from each excavated ILI feature with the corresponding field Non-Destructive Examination (NDE) measurements were evaluated to support quantification of the Probabilities of Identification, Detection, and Sizing for these inspections. The results revealed that the ILI results were not repeatable when comparing the data from the three inspections, and not meeting target specifications. Furthermore, advanced analysis was completed to combine the data and evaluate the tool performance reliability for the pipeline. The results showed that the crack ILI tools were not achieving the required reliability targets for the pipeline.
Considering that ILI is often successfully used to support the crack management of pipelines for the vast majority of conditions, the 2010 failures were investigated to determine the causes for the unacceptable ILI performance. The investigation revealed that the distinct peaking, misalignments, and pipe mill grinding associated with the pipelines flash weld, caused challenges for the tool’s detection and sizing capabilities. Therefore in order to ensure the safety of this pipeline, mitigation in addition to crack ILI programs was deemed to be required.
Some options included operating at a reduced pressure, hydrostatic testing, or pipeline replacement. Hydrostatic testing was selected as the preferred option for implementation. This was successfully completed in October 2015 and there were no leaks or ruptures that occurred during the hydrostatic testing. This demonstrated that the pre-test excavation programs, which targeted features exhibiting burst pressures below that of the hydrotest pressure, had mitigated the cracking threat on the pipeline.
The results of the reliability analysis showed that the uncertainties associated with the ILI were higher than acceptable. However as there were no failures during the hydrotest the reliability analysis was conservative for this case in consideration that the pre-test excavation program was able to mitigate the cracking threat. Nonetheless, the process of reviewing and assessing the ILI-field comparisons and evaluating the ILI tool performance remains a critical component of crack ILI management. Conducting alternative mitigation to ILI for crack management, if required, also remains a critical component of crack ILI management. For this pipeline the cracking threat will be re-assessed within 5 years of the hydrostatic test to support continued safe and reliable operation.
NEW DIRECTIONS IN DATA ANALYSIS