Prediction of Steel Pipeline Internal Corrosion Status with Markov Chain Theory

Pipeline plays an important role in transporting energy across the nation. Due to its inherent property, pipe may corrode after years of service, which may bring potential risk to the safety operation of it. This paper proposed a model to predict the corrosion status of pipeline with Markov chain theory. The basic principle of Markov chain theory was introduced and definition of transition probability matrix was established. A case study of predicting steel pipeline corrosion status was conducted. This case includes three corrosion data of a certain pipe corrosion flaw at different inspection periods and criteria for evaluating corrosion status was defined based on ASME B31G. Results show that the pipeline will fail after 33 years of service, though this result may vary according to the selection of initial status, it could somehow indicate the general growth trend of the flaw. Grey theory was also briefly introduced herein to tackle with discontinuous time sequence data. Conclusions drawn from the study could serve as references for optimizing pipeline inspection schedule.