Analytical Assessment of the Remaining Strength of Corroded Pipelines and Comparison With Experimental Criteria

Abstract

Recently published analytical solutions for the remaining strength of a pipeline with narrow axial and axisymmetric volumetric flaws are described in this paper, and their experimental and numerical validation are reviewed. Next, the domains of applicability of each solution are studied, some simplifications suitable to steel pipelines are introduced, and an analytical model for the remaining strength of corroded steel pipelines is presented. This analytical solution is compared with the standards most widely used in the industry for assessment of corroded pipelines: ASME B31G, modified ASME, and DNV RP-F101. The empirical and analytical solutions are compared with respect to their most relevant parameters: critical (or flow) stress, flaw geometry parameterization, and Folias or bulging factor formulation. Finally, two common pipeline steels, API 5L grades X42 and X100, are selected to compare the different corrosion assessment methodologies. Corrosion defects of 75%, 50%, and 25% thickness reduction are evaluated. None of the experimental equations take into account the strain-hardening behavior of the pipe material, and therefore, they cannot properly model materials with very dissimilar plastic behavior. The comparison indicates that the empirical methods underestimate the remaining strength of shallow defects, which might lead to unnecessary repair recommendations. Furthermore, it was found that the use of a parameter employed by some of the empirical equations to model the assumed flaw shape leads to excessively optimistic and nonconservative results of remaining strength for long and deep flaws. Finally, the flaw width is not considered in the experimental criteria, and the comparative results suggest that the empirical solutions are somewhat imprecise to model the burst of wide flaws.