基于LHS-Kriging-DW的页岩气集输管道内腐蚀预测模型研究.docxVIP

2024年第53卷第16期Vol.53No.162024

热加工工艺

HotWorkingTechnology

ISSN1001-3814CN61-1133/TG

DOI:10.14158/j.cnki.1001-3814.rjggy@

基于LHS-Kriging-DW的页岩气集输管道内腐蚀预测模型研究

周逸轩1,彭星煜2,耿月华3

(1.西南石油大学工程学院,四川南充637000;2.西南石油大学石油与天然气工程学院,四川成都610500;3.西南石油大学信息学院,四川南充637000)

摘要院为了准确预测页岩气集输管道的内腐蚀速率袁对DW渊DeWarrd冤模型进行修正袁运用LHS渊拉丁超立方抽样冤优化的Kriging渊克里金方法冤对修正DW进行训练以建立模型并验证遥通过高清漏磁检测器检测某页岩气集输管道内腐蚀速率袁基于PIPESIM模拟该管道现场工况袁运用多元非线性回归修正DW模型遥运用LHS-Kriging对修正DW模型进行训练并优化袁运用建立的模型对另一相邻管道进行了腐蚀速率的预测和现场开挖验证遥现场内检测的结果显示院管线在几处低洼积液处发生了严重的腐蚀袁因此引入了持液率作为修正因子修正DW模型遥选择样本数为500用于LHS-Kriging-DW模型时误差较小遥在实例运用中袁10个开挖检测点有9个都在预测结果的95%置信区间内袁这说明建立的腐蚀速率模型可用于指导现场内腐蚀直接评价等相关工作遥

关键词院页岩气集输管道曰内腐蚀速率曰多元非线性回归曰克里金方法渊Kriging冤曰拉丁超立方抽样渊LHS冤中图分类号院TG174 文献标识码院A 文章编号院1001-3814(2024)16-0113-05

ResearchonInternalCorrosionPredictionModelofShaleGasGatheringPipelineBasedonLHS-Kriging-DWZHOUYixuan1,PENGXingyu2,GENGYuehua3

(1.SchoolofEngineering,SouthwestPetroleumUniversity,Nanchong637000,China;2.SchoolofPetroleumandNatural

GasEngineering,SouthwestPetroleumUniversity,Chengdu610500,China;3.SchoolofInformation,SouthwestPetroleum

University,Nanchong637000,China)

Abstract院Inordertoaccuratelypredicttheinternalcorrosionrateofshalegasgatheringpipeline,theDW(DeWarrd)

modelwasmodified,andthemodifiedDWwastrainedbyKriging(Krigingmethod)optimizedbyLaLHS(latinhypercube

samplinghypercubesampling)toestablishthemodelandverifyit.Theinternalcorrosionrateofashalegasgatheringpipeline

wasdetectedbyhigh-definitionmagneticfluxleakagedetector.BasedonPIPESIM,thefieldconditionofthepipelinewas

simulated,andtheNinfitfunctionwasusedtomodifytheDWmodelbymultivariatenonlinearregression.ThemodifiedDW

modelwastrainedandoptimizedbyLHS-Kriging,andtheintervalestimationofcorrosionrateofanotheradjacentpipeline

waspredictedandverifiedbyfieldexcavation.Theresultsoffield