Development of a hybrid cost-based risk integrity assessment model for burst failure of pipeline systems with interacting corrosion defects

Burst is one of the primary causes for the failures of pipelines with corrosion defects. In this study, a hybrid costbased risk integrity assessment model (CBRIA) was developed to schedule the maintenance plans during the entire service period for the pipeline systems consisting of multiple corroded segments with either the single or interacting defects: the finite element method (FEM) and the probabilistic methodologies were coupled to estimate the burst capacity and the corresponding failure probabilities, motivated on capturing the non-linear behaviors of structural response; Subsequently, they were embedded into the multi-object genetic algorithm (MOGA) to search the optimal solutions with two goals, mitigation of the failure risk and minimization of the life cycle cost. After validated against the burst tests and traditional risk analysis methodologies, the CRBIA model was applied to a case study. The performances of the CBRIA model in different cases were evaluated. For the single pipeline segment, this model yielded reasonable schedules in capturing the failure evolution characteristics. For pipeline systems with multi-segments, an iteration scheme with the constraint-dependent coding pattern of chromosomes was developed into the CBRIA model to supervise the solution searching, where the inner mechanism between the constraint and the model performances was discussed in depth. Furtherly, a parametric study was performed to explore the effects of the structural response on the failure risk. Besides, the growth rate on corrosion depth had more remarkable effects on the failure risk than the wall thickness, mainly attributed to the nature of larger uncertainties.