A hybrid algorithm for inspection planning of subsea pipelines subject to corrosion-fatigue degradation

The corrosion-fatigue degradation is a significant concern causing subsea pipeline failure. This paper presents an inspection planning methodology for subsea pipelines subject to corrosion-fatigue degradation using dynamic Bayesian network (DBN) and improved Adaptive Genetic Algorithm (IAGA). The classical decision tree is used to represent the inspection process of pipeline. The corrosion-fatigue degradation is simulated using DBN to estimate the failure probability of subsea pipeline. IAGA is used to automatically present the inspection schemes and determine the optimal solution due to its strong optimization ability. A case study is utilized to illustrate the feasibility of the methodology. The results show that the cost is the lowest when the pipeline is inspected three times (11, 19 and 25 years) in its service, which is 58.0235 million dollars. The methodology can be used to guide the daily inspection planning of subsea pipeline subject to corrosion-fatigue degradation.