基于同伦分析方法和神经网络的复杂腐蚀缺陷海底管道屈曲压力计算方法  

Calculation Method of Buckling Pressure of Submarine Pipelines with Complex Corrosion Defects Based on Homotopy Analysis Method and Neural Network

作　　者：陈严飞[1,3] 阎宇峰 敖川 贾鲁生 侯富恒 高志浩 刘瑞昊 Chen Yanfei;Yan Yufeng;Ao Chuan;Jia Lusheng;Hou Fuheng;Gao Zhihao;Liu Ruihao(National Engineering Laboratory for Pipeline Safety/Beijing Key Laboratory of Urban Oil and Gas Distribution Technology,China University of Petroleum,Beijing 102246,China;Research Institute of China National Offshore Oil Corporation,Beijing 100028,China;State Key Laboratory of Structural Analysis for Industrial Equipment,Dalian University of Technology,Dalian 116024,China)

机构地区：[1]中国石油大学(北京)油气管道输送安全国家工程实验室/城市油气输配技术北京市重点实验室,北京102249 [2]中国海洋石油集团有限公司研究总院,北京100028 [3]大连理工大学工业装备结构分析国家重点实验室,辽宁大连116024

出　　处：《中国海洋大学学报（自然科学版）》2024年第6期156-164,178,共10页Periodical of Ocean University of China

基　　金：国家自然科学基金项目(52171285);大连理工大学工业装备结构分析国家重点实验室开放基金项目(GZ19119)资助。

摘　　要：针对含复杂形状腐蚀缺陷海底管道屈曲压力计算难的问题,本研究中采用同伦分析方法对含复杂腐蚀缺陷海底管道的屈曲压力进行了计算,并用有限元方法进行了对比验证,进而根据计算结果提出了基于神经网络模型的海底管道屈曲压力的计算方法。结果显示:同伦分析方法可以有效、准确地计算含复杂腐蚀缺陷海底管道的屈曲压力,在相同情况下抛物线形腐蚀管道的临界屈曲压力大于椭圆形管道的临界屈曲压力。基于神经网络模型的腐蚀海底管道屈曲压力计算方法具有良好的泛化性能,可以很好地应用于工程现场的计算。Aiming at the problem that it is difficult to calculate the buckling pressure of submarine pipelines with complex corrosion defects.The buckling pressure of the submarine pipelines with complex corrosion defects was calculated by the homotopy analysis method,and the finite element method was used for comparison and verification.According to the calculation results,a calculation method of buckling pressure of submarine pipelines based on neural network model is proposed.The results show that the homotopy analysis method can effectively and accurately calculate the buckling pressure of submarine pipelines with complex corrosion defects;under the same conditions,the critical buckling pressure of the parabolic corroded submarine pipelines is greater than that of the elliptical submarine pipelines.The proposed calculation method of buckling pressure of corroded submarine pipelines based on neural network model has good generalization performance and can be well applied in practical engineering.

关 键 词：海底管道 腐蚀缺陷 屈曲压力 同伦分析方法 神经网络模型 

分 类 号：TE88[石油与天然气工程—油气储运工程]