Markov chain modelling for time evolution of internal pitting corrosion distribution of oil and gas pipelines

Abstract

A continuous time non-homogenous linear growth pure birth Markov model was used to predict the future pit depth distribution of internally corroded oil and gas pipelines. A negative binomial distribution was used for calculating the transition probability functions of the pit depths whilst pit depth growth was estimated for low, moderate, high and severe pitting corrosion rates using field measured data of pit depths, temperatures, CO2 partial pressures, pH and flow rates. The Markov predicted results agreed well with field measured pit depth data from X52 grade pipeline and L-80 and N-80 grades offshore well tubing.