基于土壤腐蚀主成分的金属管道退化维纳过程研究

目前,对埋地管道的土壤腐蚀过程研究很少考虑各因素之间的相关作用。首先分析土壤腐蚀的因素,采用主成分分析法(PCA)筛选造成管道腐蚀的主要因素;其次应用维纳过程(Wiener Process)结合幂律应力加速方程建立针对埋地管道的非线性退化模型;再根据采集的土壤数据通过马尔科夫链蒙特卡洛(MCMC)法估计模型参数值,最终实现埋地管道剩余寿命和可靠性的预测。以川气东输利川段埋地管道进行实例分析,结果表明模型考虑了土壤腐蚀因素之间的关联性对管道的影响,同时克服了维纳过程的线性随机性的缺陷,比较客观地反映了管道的腐蚀情况,能够有效地预测土壤腐蚀环境下埋地管道的剩余寿命和可靠性,具有较好的准确性和适用性。

Wiener Process of Degradation of Metal Pipelines Based on Principal Components of Soil Corrosion

Firstly,factors of soil corrosion were analyzed and principal component analysis method was used to get main factors that caused corrosion.Secondly,the nonlinear regression model by Wiener process and power-law accelerated equation was built.Then the collected data were used to estimate the unknown parameter values of the model by Markov Chain Monte Carlo (MCMC).Finally,the residual life and reliability of pipelines were predicted.In addition,the pipelines buried in Lichuan of East natural gas transmission project were utilized by as an example.Result showed that the model that considered the relationship between the factors of soil corrosion and overcame the linear stochastic of Wiener process reflected the pipeline corrosion objectively,and predicted the residual life of buried pipelines effectively in the soil corrosion environment.In general,the model had good accuracy and applicability.