Lixin Wei, Zhuo Wang, Qing Miao, Xingyu Li, Feng Yan, and Xin Ouyang. Evaluation of Pipeline Crack Defects Based on Random Reliability. Dynamic Systems and Applications 30 (2021) No.5, 667-682

https://doi.org/10.46719/dsa20213051

ABSTRACT.
Crack defects will exist in long-distance oil and gas pipeline inevitably in the process of manufacturing, construction, or service. Such defects are generally evaluated according to the Level2A—Normal assessment in the ‘Guide to methods for assessing the acceptability of flaws in metallic structures’. But it can’t reflect the uncertainty of defects size and tube peculiarity. And its results are strongly influenced by whether to use the subentry safety factor and only draw a conclusion that the defects are acceptable or not. So proposed the pipe body crack random reliability evaluation method based on Monte Carlo simulation. Firstly, determine the regularities of distribution of the corresponding evaluation parameters, then generate a certain amount of random numbers, and evaluate the parameters for each set of random sampling. When the simulation sampling frequency reaches a certain number, the result of evaluation tends to a stable failure probability value. A pipe with 5 different size cracks has been evaluated. The results of level 2A assessment are as follows: without using the subentry safety factor, the evaluation results for all acceptable, and when using the subentry safety factor, there are 3 sizes of creaks unacceptable. Using a random reliability evaluation method based on Monte Carlo simulation, the evaluation result is a certain failure probability value. Due to considering the randomness of the size of the defect and the tube performance, random reliability evaluation can correctly reflect the possibility of pipeline failure and then provide a quantitative reference basis for the maintenance plan for the management department.

Keywords: Monte Carlo simulation; Pipe crack; Integrity evaluation; Reliability analysis