Assessment of reliability for subterranean corroded pipelines in cold regions using Monte Carlo method and BP neural network

Abstract

Due to the complexity and instability of the environment in which the buried oil pipeline is located in cold regions, the corrosion of the outer wall of the pipeline often occurs, and the corrosion poses a serious threat to the safety of the pipeline. In order to ensure the safe operation of buried oil pipelines, it is particularly urgent to study the reliability of operating pipelines. In this paper, the external temperature field and soil effect were taken into account, and the finite element method was used to calculate and analyze the results. The fitted formula which can predict the failure pressure of buried corroded pipeline was obtained. After that, the reliability of buried corroded pipelines was studied by using Monte Carlo method and BP neural network. The results show that the fitted formula for predicting the failure pressure of pipelines is more accurate than traditional code. The corrosion depth has the most significant impact on the reliability of pipelines, being 1.85 times that of the corrosion length. The impact of corrosion width, burial depth, frost-heaving rate, and length of the frost-heaving section on the reliability of corroded pipelines is relatively small. The frost-heaving section within 5000 mm can play a certain positive role in the reliability of corroded pipelines.