Estimation of corrosion failure likelihood of oil and gas pipeline based on fuzzy logic approach

Oil and gas pipeline is likely to leak and rupture due to corrosion problems, which seriously impacts the safety operation of pipe systems. Corrosion phenomenon has the nature of uncertainty, which makes the estimation of corrosion failure likelihood (CFL) very difficult. In order to overcome this difficulty, an analytic model for assessing CFL of oil and gas pipelines has been developed by means of fuzzy logic approach. In this model, corrosion failure modes are divided into two types: corrosion thinning and corrosion cracking. Corrosion thinning factor, corrosion cracking factor, inspection effectiveness, and inspection times are introduced as the key influencing factors that affect the CFL of pipelines. The fuzzy rules between the factors and CFL were determined according to the influence degrees of the factors. The model was applied to a natural gas pipeline to assess the CFL. The assessment results can be used as effective reference for working out pipeline inspection and maintenance plans. (C) 2016 Elsevier Ltd. All rights reserved.