井下测量工具云端协同平台开发与实践

打开文本图片集

DevelopmentandPracticeofCloud CollaborativePlatform forDownholeMeasurementTools

CheYang1.2Yuan Guangjie1，2Qian Hongyu³Du Weiqiang1，²Wang Chenlong1.2Ding Jiping1，2 （1.CNPCEngineringTechnologyR&DCompanyLimited；2.NationalEngineringResearchCenterofOil&GasDrillnganc CompletionTechnology；3.SchoolofElectrical InformationandEngineeringoftheNortheastPetroleum University）

Abstract： The digital transformation and intelligentdevelopment of the petroleum industry have become a consensus.Thedigital solutionforsurface equipmentis relatively mature，whilethe digitalupgradeof downhole tools is difficult.The measurement tools are mainly single machine version，which is difcult to meet the current requirements for improving driling quality and efficiency.In the paper，based on the remote operation requirements of coreless magnetic steering tool，a platformarchitectureoffive modules，including simulation rehearsal，virtual training，remoteoperation，smart tooland intellgent decision，was designed indetail，achieving a whole process digitization from predrilling risk assessment and in-drilling acquisition and processing topost-drilling feedback and optimization，anda visual interfacewas developed.Moreover，tosolve the problemof low far-field ranging accuracy，multiple magnetic steering data mining algorithms such as support vector machine （SVM），decision tree （DT），multilayer perceptron（MLP）and convolutional neural network（CNN）were built and compared，indicating thatthe robustnessand generalization of the multilayer perceptron algorithm is the best.The field application in 5 wells shows that the efficiency is improved by 30% and the remote measurement accuracy is increased by 20% . The research results provide areference for the development of cloud collaborative digital platform of similar tols.

Keywords： downhole measurementtool; cloud collaboration; platform;data mining;coreless magnetic steering

0 引言

随着信息技术的发展和普及，石油行业的数字化转型和智能化发展已成为共识。（剩余10903字）