Reliability assessment of corroded pipeline considering multiple defects interaction based on an artificial neural network method

Because of the corrosivity of the external environment and internal media, oil, and gas pipelines are prone to be corroded. Corrosion defect, as one of the most common and dangerous pipeline damages, could weaken the loading capacity of a pipeline and may result in serious pipeline incidents, such as pipeline leakage and rupture. According to previous in-line inspection records, corrosion defects on the pipeline walls commonly don't exist in isolation. The reliability of corroded pipelines significantly affected by the interaction of multiple corrosion defects. However, hardly any previous research involves the reliability assessment of pipelines with multiple corrosions.In this paper, a simulation-based method is proposed to estimate the reliability of pipelines with multiple corrosions. The reliability assessment of this method is realized by integrating multiple approaches, including finite element analysis, sensitivity analysis, Monte Carlo simulation, and artificial neural networks (ANN). A new interaction rule considering the effect of corrosion depth for multiple corrosions is developed based on finite element analysis. The optimized PCORRC burst model determines the limit state of corroded pipelines. Sensitivity analysis is employed to reduce the number of ANN inputs for performance improvement. Data sets used to train and test the artificial neural network are generated by Monte Carlo Simulation. The proposed method is compared with the traditional reliability analysis method through a case study, and the results show that the new method could achieve accurate reliability prediction for pipelines with multiple corrosions while improving computational efficiency.