Real-Time Prediction of Stress Response of Tubular Joints on the Jacket Platform Using a Reduced-Order Model  

作　　者：LENG Jian-cheng MAO Hou-bin DIAO Kai-xin FENG Hui-yu 

机构地区：[1]School of Mechanical Science and Engineering,Northeast Petroleum University,Daqing 163318,China

出　　处：《China Ocean Engineering》2025年第1期13-26,共14页中国海洋工程(英文版)

基　　金：financially supported by the National Natural Science Foundation of China(Grant No.11472076).

摘　　要：Jacket platforms constitute the foundational infrastructure of offshore oil and gas field exploitation.How to efficiently and accurately monitor the mechanical properties of jacket structures is one of the key problems to be solved to ensure the safe operation of the platform.To address the practical engineering problem that it is difficult to monitor the stress response of the tubular joints of jacket platforms online,a digital twin reduced-order method for real-time prediction of the stress response of tubular joints is proposed.In the offline construction phase,multi-scale modeling and multi-parameter experimental design methods are used to obtain the stress response data set of the jacket structure.Proper orthogonal decomposition is employed to extract the main feature information from the snapshot matrix,resulting in a reduced-order basis.The leave-one-out cross-validation method is used to select the optimal modal order for constructing the reduced-order model(ROM).In the online prediction phase,a digital twin model of the tubular joint is established,and the prediction performance of the ROM is analyzed and verified through using random environmental load and field environmental monitoring data.The results indicate that,compared with traditional numerical simulations of tubular joints,the ROM based on the proposed reduced-order method is more efficient in predicting the stress response of tubular joints while ensuring accuracy and robustness.

关 键 词：offshore jacket platform digital twin tubular joints proper orthogonal decomposition leave-one-out cross validation real-time prediction 

分 类 号：TG1[金属学及工艺—金属学]