Dynamic safety assessment of oil and gas pipeline containing internal corrosion defect using probability theory and possibility theory

Abstract Safety assessment of oil and gas pipeline containing corrosion defects is one of the significant techniques for ensuring safe operation of the pipeline. In practice, the parameters involved in conventional assessment procedures have the nature of uncertainty (i.e. randomness and fuzziness). The traditional methods deal with the uncertainty through probability method or fuzzy set theory. However, each of them cannot effectively handle the coexistence of both random parameters and fuzzy parameters. In addition, corrosion is time-dependent, which means only if the aggressive environment exists, corrosion defects could gradually propagate until the pipeline failure occurs. With respect to these problems, this paper proposed a new safety assessment method for oil and gas pipeline containing internal corrosion defects. The method can be used to estimate the failure probability bounds of a pipeline with time by introducing corrosion rate into the limit state function of the pipeline, meanwhile, combining probability theory with possibility theory to deal with random and fuzzy uncertainties. To demonstrate the feasibility of the model, an application example was presented. In the case, the failure probability bounds at different service times were calculated. Furthermore, the effects of uncertainty of parameters (i.e. pipe geometry, corrosion defect size, material mechanical properties, and operating pressure) and their mean values on failure probability bounds were discussed. The results show that the pipe wall thickness has the most important impact on corrosion failure probability of the pipe, following by the yield stress of pipe material, the outer diameter of the pipe, corrosion defect depth, ultimate tensile stress of the material, and the defect length.