Discussion on“Data-driven Methods to Predict the Burst Strength of Corroded Line Pipelines Subjected to Internal Pressure https://doi.org/10.1007/s11804-022-00263-0”  

“采用数据驱动方法对受内压作用的受腐蚀管线断裂强度预报研究”讨论

作　　者：Mojtaba Mokhtari Robert E.Melchers 

机构地区：[1]Centre for Autonomous Marine Operations and Systems,Department of Marine Technology,Norwegian University of Science and Technology(NTNU),NO-7491 Trondheim,Norway [2]Centre for Infrastructure Performance and Reliability,The University of Newcastle,Newcastle,NSW 2300,Australia

出　　处：《哈尔滨工程大学学报(英文版)》2025年第1期249-251,共3页Journal of Marine Science and Application

摘　　要：The authors(Cai et al.,2022)claim that their proposed machine learning(ML)models,which are based on three typical ML algorithms and are trained to predict the burst capacity of pitting corroded pipelines,perform better than the existing semi-empirical formulas recommended by the international engineering code developers,DNV and ASME.The authors’assessments of the semi-empiri‐cal burst capacity formulas in Figure 10(a)and Table 8(Cai et al.,2022)incorrectly indicate that DNVGL-RP F101,ASME B31G,and modified ASME B31G are dan‐gerously unsafe due to significantly overestimating burst pressures in several cases.In contrast to the results and conclusions in Cai et al.

关 键 词：Internal FIGURE ASME 

分 类 号：TE988.2[石油与天然气工程—石油机械设备]