Structural Integrity recently performed probabilistic fracture mechanics (PFM) analysis of a gas transmission pipeline for a major U.S. operator. The analysis yielded interesting insights in several areas:
Pressure testing has long been considered the gold standard for assuring pipeline integrity. By testing at a factor (e.g., 1.25x or 1.5x) above the Maximum Allowable Operating Pressure (MAOP), any size critical flaws in the line would fail at this pressure level and are thus removed prior to future service. Subcritical flaws that remain after the test will be smaller than the critical flaw sizes during operation, and thus can be assumed to have some margin for growth before they become critical in service. Flaw growth rates can be calculated based on operational and environmental factors to establish a reassessment interval for future testing or inspections.
In-Line Inspection (ILI) technology has improved significantly in the areas of Probability Of Detection (POD) and flaw sizing accuracy, such that a greater level of safety may now be achieved through ILI. With accurate ILI and an associated repair criterion (i.e., repair all flaws greater than a specified size), smaller flaws and a greater numbers of flaws will be identified and repaired. This is particularly important when aggressive crack growth mechanisms are present, such as Stress Corrosion Cracking (SCC).
This point is illustrated in Figure 1 above, which presents the results of a probabilistic analysis of about 100 miles of gas transmission piping that had experienced an SCC failure and was subjected to a 100% ILI using advanced EMAT UT technology. The plots compare probability of failure versus time following the ILI or following a spike Hydrotest at a stress level equal to 100% of specified Minimum Yield Strength (SMYS).
While the results are acceptable with both techniques, they show that EMAT ILI outperforms Hydrotest – i.e. the EMAT ILI probability of failure (red curve) is consistently about an order of magnitude lower than the Hydrotest probability (dashed curve). This is significant finding, as Hydrotesting can cost significantly more than ILI. In many cases, Hydrotesting is not a viable alternative since the pipeline is not looped or back-fed and therefore may not be taken out of service. A cautionary note, however, is that this result is highly dependent on the quality of the ILI technique. The above PFM analysis incorporated vendor-specified sizing error margins and PODs, which were validated via in-ditch and destructive measurements on a substantial number of detected features.
The PFM analysis can also provide guidance in establishing reassessment intervals. In theory, one just specifies an acceptable probability of failure (e.g., the horizontal green or yellow lines in Figure 1), and selects the intersection of that line with the applicable curve. For example, using the green line (10-4 failure probability), a 4-year reassessment interval is predicted for Hydro-test versus 6.7 years for EMAT ILI. Of course, acceptable failure probability is not obvious or easy to establish, so a perhaps more appropriate use of such PFM results is to evaluate existing reassessment guidance and make comparisons. For the subject pipeline, the operator’s internal SCC management plan specifies a 3-year Hydrotest interval when a failure has occurred. Referring to Figure 1, that interval corresponds to a rupture probability between the 10-4 and 10-5 lines. An equivalent integrity level could be achieved by instead performing EMAT ILI on a 5-year interval.
Finally, the PFM analysis can be used to evaluate tradeoffs among specific provisions of an integrity management plan. For example, the Figure 1 ILI result was based on an ILI repair criterion that requires all detected flaws greater than AMSE B31.8S Category 2 (flaws that would fail at 125% of MAOP) to be repaired. In Figure 2, this criterion was enhanced to specify repair of >Category 2 flaws plus any detected flaws with lengths greater than 2 inches, regardless of depth. The results show that this enhanced repair criterion adds about 1 to 2 years to the 10-5 and 10-4 reassessment intervals. Thus, an operator can use these results to compare the economics of enhanced repair criteria versus extended reassessment intervals, while maintaining the same level of pipeline integrity. PFM evaluations such as this could be used to perform many types of integrity management cost-benefit evaluations.