Dents are a common anomaly type found in many pipelines, regardless of diameter, pipe type, pipe grade, environment, etc. It is not feasible or necessary to repair every dent found in a pipeline. In order to determine which dents do require mitigation, standards such as CSA Z662:19, provide a number of screening criteria for the identification of deformations that require further investigation, assessment or remediation. The screening criteria outlined in Section 10.10.4 of the 2019 version of CSA Z662 includes both a depth and a sharpness criteria (i.e. length to depth ratio). For the dents reported by in-line inspection (ILI) that exceed either the dent depth or sharpness criteria, a pipe wall curvature strain-based assessment can be completed to determine their acceptability based on the likelihood of cracking that initiated with the formation of the dent. These screening criteria are intended to be conservative and were designed to identify the dents that are most likely to have unacceptable curvature strain values. However, in November 2018, the American Society of Mechanical Engineers (ASME) corrected the equations given in ASME B31.8 Gas Transmission and Distribution Piping Systems in the non-mandatory Appendix R which is used to calculate equivalent strain on the internal and external pipe surfaces. A previous study involved a comparison of the static curvature strain results using the equivalent strain equations from both the 2016 version of ASME B31.8 and the latest 2020 version (which utilizes the corrected strain equations introduced in the 2018 edition). Over 2,500 dents, in a variety of pipelines were included in the comparison. The study identified that there is an increase in the number of dents with equivalent strains considered unacceptable based on the strain limits stipulated in CSA Z662:19. In addition, the study identified that the current length to depth ratio limit of 20 (the dent sharpness criteria) may not be sufficiently conservative; dents with equivalent strains that exceeded the allowable strain limits were considered acceptable based on the depth and sharpness screening criteria. It was noted that these dents, however, had complex dent geometry as they were either multi-apex or dents interacting with other dents. The scope of the work reported in this paper was to expand on the previous study by increasing the sample size and considering the added strain limits for plain dents outlined in ASME B31.8-2020 (e.g. half the minimum elongation), in addition to the strain limits stated in CSA Z662:19. The study also explored the equivalent strain results for multiple peak dents and dents interacting with other dents, to determine whether the dent screening criteria within CSA Z662:19 are appropriate for dents with complex geometry. The aim of the work was to determine whether the criteria in CSA Z662:19 are still conservative following the correction to the ASME B31.8 strain equations, while further considering additional strain limits, as well as, complex dent geometries. It was concluded that changes to the CSA Z662:19 dent depth and sharpness screening criteria would be prudent to ensure conservatism.

A significant amount of effort has been expended in the area of advancing pipeline dent remaining life assessment methods beginning in the late 1980s and extending to the current day. Initial research efforts were primarily empirical in nature while more recent research efforts have incorporated finite element modelling. Coupled with advancements in assessment techniques, the capabilities of advanced in-line inspection (ILI) tools have increased to a point where they can provide consistent, reliable information that is suitable for dent assessments. As a result of these advancements in assessment models and ILI tools, operators can now perform remaining life assessments using ILI data, and a multitude of remaining life assessment models are available, including solutions from the European Pipeline Research Group (EPRG), Pipeline Research Council International (PRCI), American Petroleum Institute (API), and finite-element based approaches. In addition to these remaining life assessments, many operators routinely perform strain-based assessments based on guidance from ASME B31.8. To date, there have been few studies comparing the various assessment methods on large numbers of dents, and as a result, significant questions persist as to the conservatism inherent in each method. In addition, the EPRG and PRCI methods are largely based on full-scale testing and finite-element models performed with idealized indenter shapes while actual pipeline dents typically exhibit complex shapes and interactions between multiple dents. Each model also has limitations and advantages that are discussed in this paper, such as ease of use and how pipeline geometry and weld association are considered. This paper provides a robust comparison of selected dent assessment methodologies on 220 actual dents from a 24-inch pipeline with depths ranging from 0.6–4.5% OD, and 32 dents from a 30-inch line with depths ranging from 1–2.5% OD. The assessment includes both top and bottom of line dents and investigates the influence of restraint on remaining life. The results presented in the paper are based on high-resolution ILI caliper data collected during two in-line inspections. Furthermore, the paper provides statistical comparisons between strain and remaining life methodologies and also between the various remaining life assessments. The paper also provides a comparison of the restraint parameter from the PRCI model with calculated stress concentration factors from finite-element models. The paper provides a first of its kind comparison of the various methods and discusses how the work may be extended to other pipe diameters and wall thicknesses.

Following a loss of containment incident in July 2016 on a 16-inch diameter pipeline on the south slope of the North Saskatchewan River located in Saskatchewan, Canada, Husky completed extensive studies to understand and learn from the failure. The cause of the incident was ground movement resulting from a landslide complex on the slope involving two deep-seated compound basal shear slides as well as a near surface translational slide in heavily over consolidated marine clays of the Upper Cretaceous Lea Park Formation. One aspect of the studies has been to undertake structural analysis of the pipeline response to the loading imposed from the ground movement to minimize the potential for a similar occurrence from happening in the future and determine the integrity of the pipeline at the time of the assessment. Given the scale and complexity of the landslide, slope stabilization measures were not practical to implement, so repeat ILI using caliper and inertial measurement technology (IMU), in addition to a robust monitoring program was implemented. Realtime monitoring of ground movements, pipe strain and precipitation levels provided a monitoring and early-warning system combined with documented risk thresholds that identified when to proactively shut-in the pipeline. The methodology and findings of the slope monitoring and structural analysis that was undertaken to examine the robustness of the pipeline to withstand future landslide movement are presented herein. The work involved modelling of the pipeline history on the slope including loads that had accumulated in the original pipeline sections based on historical ILI results and slope monitoring. The pipeline orientation was parallel with the ground movement in the landslide complex, so the development of axial strain in the pipeline was the dominant load component, which are particularly damaging in the compression zone. The work provided recommendations and technical basis to continue safe operation of the pipeline with consideration of continuing ground movement and assisted the operator with decisions over the long-term strategy for the pipeline.

Improvements in in-line inspection (ILI) technology have led to an increase in the probability of detection and ability to characterize geometric features such as wrinkles, the assessment of which was introduced into CSA Z662, “Oil & Gas Pipeline Systems”, in the 2015 version. The CSA wrinkle acceptance limits are based predominantly on fatigue assessment criteria; part of the assessment procedure is confirmation that wrinkles are free from associated cracking. In practice, this often restricts the assessment to wrinkles that have already been investigated in-field and where the absence of cracking has been confirmed by non-destructive examination (NDE). This paper describes the assessment of a series of wrinkles that exceeded the CSA height criteria, reported by ILI within field bends in an insulated liquid pipeline. Strain-based assessment, supported by in-field investigations, was used to investigate the likelihood of associated cracking. Utilizing the high resolution caliper ILI tool data, three-dimensional profiles of the wrinkles were generated. Previous work that compared “tool-measured” with “field-measured” profiles identified that caliper tool measurements can underestimate the true depth and profile of wrinkles, this effect is more pronounced for particularly sharp wrinkles. The wrinkle profiles were therefore adjusted based on the historical field-tool correlation. Strain profiles were then calculated using the guidance within ASME B31.8 Appendix R. It was identified that the majority of the wrinkles exceeded the 6% strain limit commonly applied to dents. One field bend containing multiple wrinkles was subsequently excavated in order to gather detailed profile information and to inspect for cracking. Upon excavation, the wrinkles were not visually apparent, but their presence was confirmed following removal of the insulating coating. Profile information was subsequently recorded using laser scanning technology. In addition, NDE confirmed the absence of cracking, despite the fact that the majority of wrinkles were associated with strain levels that exceeded the CSA limiting value, 6%. The laser scan data were then compared with the adjusted “tool-measured” profiles. It was observed that the adjusted measurements based on the ILI tool data were conservative, and in some cases excessively so. The caliper measurements were optimized by identifying a factor that could be systematically applied to the “tool-measured” wrinkle profiles, which provided consistency with the profiles measured by the laser scan, thereby improving the accuracy of the dimensions and strain estimation of the remaining (non-excavated) wrinkles. Finally, a S-N based fatigue assessment was performed using operational cyclic pressure data and estimates of the stress concentration factors associated with the wrinkles. The calculated fatigue lives exceeded the expected operational life of the pipeline.

The assessment of the remaining life of dents in pipelines generally relies on characterizing the structural response to pressure loading and combining a known pressure history with S-N curves to determine a fatigue life. A robust method for determining the structural response of a dent to pressure loading is through the determination of a Stress-Concentration Factor (SCF) derived from the modelling of the dent using Finite-Element Analysis (FEA). For simplicity, most SCF assessments rely on the use of unrestrained models derived directly from deflection data recorded by ILI or laser scan; however, this application can lead to overly high predictions of SCF values when evaluating restrained dents. Explicit modelling of restraint using bespoke indenter profiles and elastic-plastic material models can be used to derive more appropriate SCF values for restrained dents; however, this requires significantly more analytical effort and can sacrifice the fidelity of the shape for complex geometry. An approach that utilizes the efficient modeling and high fidelity of unrestrained elastic models would provide the industry with a reliable and repeatable process for evaluating the fatigue response of restrained dents. The methodology presented within this paper will seek to validate reasonable bounds for unrestrained elastic models that can be applied to cases where restrained dents are indicated. This paper will investigate the feasibility of a plasticity-restraint correction factor that could be applied to elastic SCFs and discuss the implications for dent fatigue assessments. The response of restrained elastic-plastic models will be compared to the response to elastic models for a range of indenter shapes to show the feasibility of this correction factor.