A Bayesian network‐based susceptibility assessment model for oil and gas pipelines suffering under‐deposit corrosion

Abstract

Oil and gas pipelines are exposed to harsh operating conditions that facilitate their susceptibility to complex corrosion mechanisms. This affects their integrity and results in failure with associated consequences. Capturing these complex corrosion phenomena requires a robust approach. This study proposes the application of a dynamic probabilistic model to capture the key influential factors that contribute to the complex under‐deposit corrosion (UDC) mechanism in oil and gas pipelines. The Bayesian network model assesses the pipeline's susceptibility (degradation rate) to the UDC, capturing parametric dependencies. The predicted corrosion rates are input data for the corrosion propagation prediction. Three semi‐empirical corrosion propagation models are used for a comparative assessment to establish the degree of susceptibility given the prevalent influential factors and model parameters. The proposed approach is tested on an offshore pipeline, and the degree of impact of the key influential parameters is predicted. The result shows a percentage increase in the degradation rate by 18.7%, 33.2%, 35.8%, and 63.4%, respectively, for the various interaction scenarios. The present approach offers an adaptive and robust technique that would provide an early warning guide on the rate of pipeline degradation to aid integrity management for offshore assets suffering from deposit corrosion.