Statistical Modelling of Pitting Corrosion: Extrapolation of the Maximum Pit Depth-Growth

Abstract

Pitting corrosion is one of the main threats in the pressure vessels integrity and also causes the failure of buried pipelines steels that transport sour gas, crude oil or condensate hydrocarbon, for this reason, a reliability assessment of pressurized vessels and buried pipelines based on probabilistic mathematical modelling to estimate the remaining life-time due to pitting corrosion damage is extensively employed. Herein, a methodology for probabilistic mathematical modelling of the pits initiation process and its depth growth process is developed; both uncertain processes are well represented by stochastic models. In this methodology two stochastic models are applied; Poisson process is used to model pit initiation and Gamma process to model the pit depth-growth. Such methods are validated using data produced by computer modeling procedures. On the other hand, in the oil industry it is common not to inspect the entire vessels surface; instead of this only a small part of the surface is under inspection. According to this, the use of Block Maxima (BM) and Peak-Over-Threshold (POT) models “EXTREME VALUE STATISTICS” to characterize the probability distribution of maximum pit depths is also approached. The results indicate that POT model can evaluate efficiently the maximum pitting corrosion depths.