Impact of dependent stochastic defect growth on system reliability of corroding pipelines

This paper presents a methodology to evaluate the time-dependent system reliability of corroding pipeline segment containing multiple spatially-correlated active corrosion defects. The pipeline segment is modeled as a series system considering three distinctive failure modes at a given corrosion defect. The defect depth growth is modeled using gamma distributed time-independent growth model and gamma process-based growth model; the spatial correlation between the depth growths of different defects is characterized using the copula function. Numerical examples are used to investigate the impact of the spatially dependent (or correlated) defect depth growth characteristics, the uncertainties in the initial defect sizes (i.e. depth and length) and the model error associated with the burst pressure model on the system failure probability.

► System reliability of pipelines containing dependent corrosion defects is evaluated.
► Growth of individual defects is modeled by the gamma distribution or gamma process.
► Copula functions are used to characterize the spatial dependency of individual defects.
► Impact of uncertainties in the initial defects sizes and the model error is investigated.