1.中国石油西南油气田分公司重庆气矿;2.重庆大学机械与运载工程学院
TE985.8
国家自然科学基金(No.52275518)。
1.Chongqing Gas Field, PetroChina Southwest Oil and Gas Field Company;2.School of Mechanical and Vehicle Engineering, Chongqing University;3.School of Mechanical and Vehicle Engineering,Chongqing University
the National Natural Science Foundation of China (No.52275518)
油气集输管道腐蚀演化行为复杂，实际运行中难以获得充分的数据，且传统经验模型在长期预测中误差较大。为更全面地表征管道腐蚀过程的记忆效应和测量随机误差动态特性，精确预测管道内壁腐蚀深度，提出了一种综合考虑测量误差和记忆效应双重影响下的非马尔可夫维纳过程（Wiener Process）预测模型。通过极大似然估计和贝叶斯推理对模型的未知参数进行估计和更新；基于弱收敛理论和首达失效时间的定义，推导出管道腐蚀深度分布的近似解析式，实现管道腐蚀深度的预测。最后，以重庆气矿某天然气管道内壁的腐蚀监测数据为例验证了该方法的有效性。
The corrosion evolution behavior of oil and gastransportation pipeline is complicated, and sufficient data on corrosion influencing factors is difficult to obtain during actual operation. Additionally, the traditional empirical model exhibit significant errors in the long-term predictions. In order to more comprehensively characterize the dynamic characteristics of memory effect and measurement random error of pipeline corrosion Process, and accurately predict the corrosion depth of pipeline inner wall, a non-Markov Wiener Process prediction model is proposed considering the dual influence of measurement error and memory effect. The unknown parameters of the model are estimated and updated by maximum likelihood estimation and Bayesian inference. Based on the theory of weak convergence and the definition of first reach failure time, the approximate analytical formula of pipeline corrosion depth distribution is derived to achieve the prediction of pipeline corrosion depth. Finally, the corrosion monitoring data of the inner wall of Tiangao Line B section in Chongqing Gas Mine is taken as an example to verify theeffectiveness of the method.
