Research Status and Development Trend of Drilling Digital Twin Technology
-
摘要:
在第四次工业革命技术浪潮的推动下，油气钻井行业正朝着信息化、数字化、智能化方向快速发展，钻井数字孪生技术成为行业前沿与热点。钻井数字孪生技术是将真实钻井工程映射到虚拟空间，建立集成多学科、多物理量、多尺度的钻井过程全生命周期虚拟仿真模型，实现钻前演练、钻中优化、钻后分析等功能，保障安全、高效、低成本钻进，提高复杂油气储层钻井效率。在分析数字孪生技术在钻井工程中的应用现状的基础上，将钻井数字孪生分为钻机数字孪生和井筒数字孪生，并提出钻井数字孪生五维系统架构；同时，分析了钻井数字孪生未来发展趋势，包括钻井数据实时高效传输、地质模型精细表征、多领域一体化建模与仿真、仿真模型动态自适应更新、机理与数据融合建模、安全高效的人机交互及云边协同软件系统架构，并提出我国钻井数字孪生技术发展的相关建议。研究结果可为钻井数字孪生技术体系的构建提供参考，对于推动钻井行业智能化革新具有一定的指导意义。
Abstract:Driven by the technological impetus of the Fourth Industrial Revolution, the oil and gas drilling industry is rapidly advancing towards informatization, digitization, and intelligentization, with drilling digital twin technology emerging as a frontier and hotspot in the field. Drilling digital twin technology aims to map real drilling operations into virtual space and establish integrated, multi-disciplinary, multi-physical, and multi-scale virtual simulation models throughout the entire lifecycle of drilling. This enables functions such as pre-drilling rehearsal, in-drilling optimization, and post-drilling analysis, ensuring safe, efficient, and cost-effective drilling while enhancing the drilling efficiency of complex oil and gas formation. The current application status of digital twin technology in drilling engineering was introduced, and drilling digital twins were categorized into rig digital twins and wellbore digital twins. A five-dimensional system architecture for drilling digital twins was proposed. Furthermore, future development trends in drilling digital twins were analyzed, including real-time and efficient transmission of drilling data, refinement and quantification of geological models, multi-domain integrated modeling and simulation, dynamic adaptive updating of simulation models, the integration of mechanistic and data-driven modeling, safe and efficient human-machine interaction, and cloud-edge collaborative software system architecture. Relevant suggestions for the development of drilling digital twin technology in China were also proposed. The research findings could serve as a reference for establishing a drilling digital twin technology system and provide guidance for promoting intelligent innovation in the drilling industry.
-
水平井桥塞分段压裂已成为非常规油气高效开发的核心技术之一[1]。桥塞用于封隔已压裂井段裂缝，通常桥塞中心管中空，需井口投球、落入桥塞球座，起到密封桥塞中心管的作用。当桥塞球座处于完全密封状态，则为有效坐封[2]。一旦球座坐封失效，易发生压裂砂堵、重复压裂已改造井段等问题，无法正常后续作业。目前主要依靠人为观察井口压力变化特征判断坐封的有效性，然而对于非典型压力特征，难以快速准确识别，这已成为制约水平井桥塞分段压裂技术发展的瓶颈之一。
近年来，以大数据、机器学习、超强算力为基础的新一代人工智能技术蓬勃发展[3–4]，基于海量压裂历史数据，通过人工智能算法从大数据中学习数据变化特征[5–7]，形成了压裂工况智能诊断方法，达到实时诊断压裂工况的目的。前人已开展基于大数据分析的压裂起止时刻、暂堵、球座坐封等工况诊断研究。A. Ramirez等人[8]采用分类算法结合泵压曲线和专家经验，实现压裂作业起始与终止时刻的识别；M. M. Awad等人[9]利用小波变换方法将施工泵压蕴含的能量信息与裂缝扩展物理过程相关联，实现了单位时间内裂缝扩展事件数的定量表征；袁彬等人[10]结合长短期记忆神经网络、反向传播神经网络等多种模型，实现了泵球、前置酸降压、暂堵压裂、砂堵等事件的智能识别；盛茂等人[11]利用聚类算法、特征参数阈值法分析压裂施工数据，建立了暂堵有效性评价模型。Shen Yuchang等人[12]利用包含地面泵压以及排量的施工曲线图，基于识别图像的U-Net架构深度学习算法，建立了桥塞球座坐封起止时刻的识别模型，识别准确率达95%。该研究的判别特征是单一的排量下降，而复杂地层压裂作业过程中往往存在大量的排量下降现象，但这些并不都是由桥塞球座坐封造成的，因此仅以排量下降作为识别坐封工况的特征具有一定局限性。
为此，笔者融合专家经验定性判识和坐封数据特征挖掘定量标注，滑动窗口数据切片形成5 792组样本数据，优选井口压力–排量二维输入的长短期记忆神经网络，建立了压裂球座坐封有效性智能诊断模型；并采用欠采样平衡数据集方式提升模型判识精度，实现了每秒输出诊断结果，为桥塞球座坐封有效性实时自动诊断提供了方法。
1. 球座坐封特征参数提取
1.1 压裂球座坐封工况数据分割与标签标注
水平井桥塞分段压裂时，压裂投球坐封阶段，排量先降至0.5～1.0 m3/min，维持压裂球以较低速度坐入桥塞球座；当球座被完全密封，井筒内流体憋压，此时井口压力显著升高[13]；压力达到地层破裂压力使地层破裂后，井口压力骤降；随后逐级提高排量至压裂设计值，井口压力缓慢上升，整个投球坐封阶段井口压力呈现陡升—陡降—平缓上升的显著特征（见图1）。根据该特征，采用滑动窗口方式，对球座坐封工况进行数据分割：在排量开始降低且累计排量小于井筒体积的时段寻找第一次压力突升的时刻，并标记为坐封开始时刻；在滑动时间窗口寻找压力降落结束时刻，向后继续寻找至排量开始增加的时刻，并标记为坐封结束时刻。
工况分割后形成198段坐封数据，结合专家经验对井口压力和排量的变化特征进行判识，对每段坐封数据打标签，分为有效坐封和无效坐封2类标签，分别将其对应的时刻标记为数字1和0，最终得到有效坐封168段，无效坐封30段。
1.2 有效坐封数据特征分析
整合并提取168段有效坐封数据共有的特征点，分别是压力升高前的最小值、压力升高后达到的峰值、压力下降后的最小值及排量开始增加时刻的压力回升值，形成1—2，2—3，3—4明显的三阶段特征（见图2（a））；统计3个阶段对应的持续时间及井口压力变化值，得到各阶段的斜率分布和小提琴图（见图2（b）和（c））。
![]()
图 2 井口压力三阶段变化特征及斜率分布情况Figure 2. Characteristics of wellhead pressure change and slope distribution in three stages从图2可知，阶段一压力上升变化斜率区间[0.14，5.62]；阶段二压力下降变化斜率区间 [−0.01，−7.39]，分布范围较大；阶段三压力上升变化斜率区间[0.02，1.36]，与前2个阶段相比，分布区间更为集中。尽管有效坐封数据均呈显著的井口压力陡升—陡降—平缓上升的三阶段特征，但是数据样本分布范围较大、且不一致，无法形成明确的诊断规则实现准确诊断。
1.3 无效坐封数据特征分析
统计分析30段无效坐封数据，发现存在2类形态的无效坐封曲线（见图3）。其中，A类无效坐封虽然存在阶段一对应的压力升高过程，但阶段二的压力会降至低于压力升高前的起始值；B类无效坐封的井口压力在降低后直至提排量前始终未呈现回升趋势。
![]()
图 3 坐封失效时对应的2类井口压力曲线Figure 3. Two types of wellhead pressure curves corresponding to setting failure2. 基于长短期记忆神经网络模型建立
2.1 神经网络结构设计
压裂泵注曲线具有显著的长时序性和数据关联性，即压裂全过程具有较长的时间跨度，且每个时间步长之间的压力存在依赖关系，为此优选长短期记忆神经网络[14–18]，其特有的记忆门控单元可捕捉序列数据中的长期依赖关系[19–20]。神经网络设计采用二分类问题的设计思路，其中隐含层初始设置为256层，输入维度分别设置为一维和二维，输出设置为代表坐封有效与失效的1或0标签，对每一种输入特征值进行批标准化处理；选择Softmax激活函数对每一维度相同位置的数值进行Softmax运算，每次模型调用时对待训练参数矩阵和待训练偏置项进行初始化处理；选用Adam优化器处理二分类问题，初始学习率设置为0.01；使用交叉熵损失函数表征模型输出的有效坐封标签与实际有效坐封标签的偏差，来衡量该网络在此数据集上对坐封有效性识别效果的好坏。每次输入神经网络的训练集样本数初始设置为128个，迭代200次，每次迭代输出一次损失函数，保留最后一次训练参数，并计算准确率。神经网络结构如图4所示。
![]()
图 4 二维输入的长短期记忆神经网络结构Figure 4. Structure of LSTM neural network with two-dimensional input图4中，t表示时间，p和Q分别表示井口压力与排量的时序值，X和Y表示LSTM网络的输入和输出，σ和tanh分别表示sigmoid激活函数和tanh激活函数，参数C、h、f、i和o分别代表LSTM网络的细胞状态、隐藏状态、遗忘门、输入门和输出门。
2.2 标签数据切片处理
全连接层的神经元网络用分割出的坐封段作为训练数据集时，训练集太过冗长，会大大增加迭代时间，影响训练效果[21]，因此采用滑动窗口切片送入模型的方式来缩短训练的数据长度[22]。
统计每段数据中有效坐封的3个阶段特征的时长，得出坐封所需的时间最长为252 s。为保证特征被全部包含，将时间跨度增加，初步设定切片窗口为300 s。考虑到时间序列前后数据的相关性，设置移动步长为50 s，即窗口每次向后移动50 s，以保证对同样的一段坐封特征，其前后段的数据都能作为有效坐封的样本输入，同时也增加了样本数量。窗口从左到右，依次对每个时刻的标签进行扫描，当该窗口第一秒和最后一秒的标签均为0，且中间有且仅有一段连续为1的标签时，视为此窗口包含了一个完整的坐封段，并将整个窗口标记为标签1，作为一个有效坐封样本；当窗口内数据标签均为0，或者由1开始与结束，即无坐封段或坐封段不完整时，整个窗口标记为标签0，作为一个无效坐封样本。滑动窗口切片标注如图5所示。
198段坐封工况数据按照时间窗口300 s、移动步长50 s切片后形成5 792个样本，其中有效坐封383个，占比仅6.61%。当二分类模型中标签为1的数据量极少时，神经网络被重复传入大量的无效坐封样本，从而无法学习到有效坐封的特征。为此，采用欠采样平衡数据集方式[23]，从切片后的样本中等比例地提取标签为1和0的样本，总计766个，再以8∶2的比例划分为训练集和测试集，最终形成610个样本的训练集，156个样本的测试集。
3. 模型训练与结果分析
为考察压力和排量变化对坐封有效性判识效果的影响，分别建立井口压力一维输入和井口压力–排量二维输入的长短期记忆神经网络模型进行对比训练。首先，调整数据切片时间窗口为300，400和500 s，随着时间窗口增长，准确率由88%降至70%，表明过长的时间窗口导致样本包含更多冗余的数据信息，从而输入的干扰特征增多，因此时间窗口选择300 s；然后，调整批量大小为64，128和256，学习率分别为0.001，0.01和0.1，进行组合训练，训练结果如图6（a）所示。训练结果表明，批量大小为256时，模型准确率整体偏低，仅为50%～70%；批量大小为64、学习率为0.001时，井口压力一维输入模型的准确率最高为91.7%；批量大小为128、学习率为0.01时，井口压力–排量二维输入模型的准确率最高为96.8%，相比井口压力一维输入模型提高5.1百分点。2种模型准确率最高时对应的损失函数变化曲线如图6（b）所示。由图6（b）可以看出，迭代至第25次时，井口压力一维输入模型的损失函数降至0.30，而井口压力–排量二维输入模型的损失函数降至0.15，收敛速度更快，最终趋近于0.10。
验证集选用长庆油田合水区块51段未参与训练的压裂数据。将井口压力和排量数据以时间窗口300 s、移动步长1 s滑动输入模型，模型调用训练准确率最高的权重参数进行判识，实时输出坐封工况判识标签。若井口压力从排量降至送球排量时开始运行到累计液量达到一个井筒体积时仍未呈现三阶段特征，则判识为无效坐封。对比专家经验标签，井口压力一维输入的模型准确率为73.7%，井口压力–排量二维输入的模型准确率为84.3%。将模型识别出的43段有效坐封段绘制成瀑布图（见图7（a）），可以看出，虽然井口压力数据跨度分布较大，但该模型均能正确判识，验证了模型的有效性；将实际有效坐封、但模型误判为无效坐封的8段数据绘制成瀑布图（见图7（b）），发现此类曲线在压力突升至峰值后有一段时间的缓慢爬升，未能被该模型识别，其原因是此类情况下输入样本不足，长短期记忆神经网络未能学习到该类曲线的特征。
4. 结 论
1）针对压裂桥塞球座坐封有效性难以形成有效规则、快速准确判识的问题，提出了人工智能技术辅助诊断方法。融合专家经验定性判识和坐封数据特征挖掘定量标注，建立了基于长短记忆神经网络的压裂球座坐封有效性智能诊断模型，并采用欠采样平衡数据集方式提高模型的预测精度。
2）压裂桥塞球座有效坐封时，井口压力呈现陡升—陡降—平缓上升特征，但各阶段变化值分布离散，持续时间跨度大；无效坐封时，井口压力呈2种形态，一种是陡降幅度超过陡升幅度，一种是缺少平缓上升阶段特征。
3）利用未参与训练的51个样本验证模型，井口压力–排量二维输入模型成功识别出43个有效坐封段，准确率达84.3%。
-
[1] 门相勇,王陆新,王越,等. 新时代我国油气勘探开发战略格局与2035年展望[J]. 中国石油勘探,2021,26(3):1–8. MEN Xiangyong, WANG Luxin, WANG Yue, et al. Strategic pattern of China’s oil and gas exploration and development in the new era and prospects for 2035[J]. China Petroleum Exploration, 2021, 26(3): 1–8.
[2] 邹才能,赵群,王红岩,等. 非常规油气勘探开发理论技术助力我国油气增储上产[J]. 石油科技论坛,2021,40(3):72–79. ZOU Caineng, ZHAO Qun, WANG Hongyan, et al. Theory and technology of unconventional oil and gas exploration and development helps China increase oil and gas reserves and production[J]. Petroleum Science and Technology Forum, 2021, 40(3): 72–79.
[3] 袁光杰,付利,王元,等. 我国非常规油气经济有效开发钻井完井技术现状与发展建议[J]. 石油钻探技术,2022,50(1):1–12. YUAN Guangjie, FU Li, WANG Yuan, et al. The up-to-date drilling and completion technologies for economic and effective development of unconventional oil & gas and suggestions for further improvements[J]. Petroleum Drilling Techniques, 2022, 50(1): 1–12.
[4] 邹才能,杨智,张国生,等. 非常规油气地质学理论技术及实践[J]. 地球科学,2023,48(6):2376–2397. ZOU Caineng, YANG Zhi, ZHANG Guosheng, et al. Theory, technology and practice of unconventional petroleum geology[J]. Earth Science, 2023, 48(6): 2376–2397.
[5] 苏义脑,路保平,刘岩生,等. 中国陆上深井超深井钻完井技术现状及攻关建议[J]. 石油钻采工艺,2020,42(5):527–542. SU Yinao, LU Baoping, LIU Yansheng, et al. Status and research suggestions on the drilling and completion technologies for onshore deep and ultra deep wells in China[J]. Oil Drilling & Production Technology, 2020, 42(5): 527–542.
[6] 汪海阁,黄洪春,毕文欣,等. 深井超深井油气钻井技术进展与展望[J]. 天然气工业,2021,41(8):163–177. WANG Haige, HUANG Hongchun, BI Wenxin, et al. Deep and ultra-deep oil/gas well drilling technologies: progress and prospect[J]. Natural Gas Industry, 2021, 41(8): 163–177.
[7] 李根生,宋先知,田守嶒. 智能钻井技术研究现状及发展趋势[J]. 石油钻探技术,2020,48(1):1–8. LI Gensheng, SONG Xianzhi, TIAN Shouceng. Intelligent drilling technology research status and development trends[J]. Petroleum Drilling Techniques, 2020, 48(1): 1–8.
[8] LI Gensheng, SONG Xianzhi, TIAN Shouceng, et al. Intelligent drilling and completion: a review[J]. Engineering, 2022, 18(11): 33–48.
[9] 李根生,宋先知,祝兆鹏,等. 智能钻完井技术研究进展与前景展望[J]. 石油钻探技术,2023,51(4):35–47. LI Gensheng, SONG Xianzhi, ZHU Zhaopeng, et al. Research progress and the prospect of intelligent drilling and completion technologies[J]. Petroleum Drilling Techniques, 2023, 51(4): 35–47.
[10] 杨传书. 数字孪生技术在钻井领域的应用探索[J]. 石油钻探技术,2022,50(3):10–16. YANG Chuanshu. Exploration for the application of digital twin technology in drilling engineering[J]. Petroleum Drilling Techniques, 2022, 50(3): 10–16.
[11] 张好林,杨传书,李昌盛,等. 钻井数字孪生系统设计与研发实践[J]. 石油钻探技术,2023,51(3):58–65. ZHANG Haolin, YANG Chuanshu, LI Changsheng, et al. Design and research practice of a drilling digital twin system[J]. Petroleum Drilling Techniques, 2023, 51(3): 58–65.
[12] SUGIURA J, SAMUEL R, OPPELT J, et al. Drilling modeling and simulation: current state and future goals[R]. SPE 173045, 2015.
[13] MAYANI M G, ROMMETVEIT R, OEDEGAARD S I, et al. Drilling automated realtime monitoring using digital twin[R]. SPE 192807, 2018.
[14] LAGRANGE E. Developing a digital twin: the roadmap for oil and gas optimization[R]. SPE 195790, 2019.
[15] THORESEN K E, KYLLINGSTAD Å, HOVLAND S, et al. Using an advanced digital twin to improve downhole pressure control[R]. SPE 194088, 2019.
[16] VARADARAJAN P A, ROGUIN G, ABOLINS N, et al. A digital twin for real-time drilling hydraulics simulation using a hybrid approach of physics and machine learning[R]. OTC 31278, 2021.
[17] SHATIRY M S A, HARUN F, HUSNI Z H, et al. A review of the digital twin technology application in energy industry for performance improvement[J]. AIP Conference Proceedings, 2023, 2827(1): 040012.
[18] 张霖. 关于数字孪生的冷思考及其背后的建模和仿真技术[J]. 系统仿真学报,2020,32(4):1–10. ZHANG Lin. Cold thinking on digital twins and the modeling and simulation techniques behind them[J]. Journal of System Simulation, 2020, 32(4): 1–10.
[19] TAO Fei, LIU Ang, HU Tianliang, et al. Digital twin driven smart design[M]. London: Academic Press, 2020.
[20] 庄存波,刘检华,熊辉,等. 产品数字孪生体的内涵、体系结构及其发展趋势[J]. 计算机集成制造系统,2017,23(4):753–768. ZHUANG Cunbo, LIU Jianhua, XIONG Hui, et al. Connotation, architecture and trends of product digital twin[J]. Computer Integrated Manufacturing Systems, 2017, 23(4): 753–768.
[21] 张霖,陆涵. 从建模仿真看数字孪生[J]. 系统仿真学报,2021,33(5):995–1007. ZHANG Lin, LU Han. Discussing digital twin from of modeling and simulation[J]. Journal of System Simulation, 2021, 33(5): 995–1007.
[22] 王巍,刘永生,廖军,等. 数字孪生关键技术及体系架构[J]. 邮电设计技术,2021(8):10–14. WANG Wei, LIU Yongsheng, LIAO Jun, et al. Key technology and architecture of digital twin[J]. Designing Techniques of Posts and Telecommunications, 2021(8): 10–14.
[23] 本刊编辑部. 美欧军工领域发力数字孪生技术应用[J]. 国防科技工业,2019(2):36–37. Editorial Department of This Magazine. The US and Europe are making efforts to apply digital twin technology in the military industry[J]. Defence Science & Technology Industry, 2019(2): 36–37.
[24] 方伟光,聂兆伟,刘宸宁,等. 数字孪生驱动的武器装备智能保障技术研究[J]. 系统工程与电子技术,2023,45(4):1247–1260. FANG Weiguang, NIE Zhaowei, LIU Chenning, et al. Research on digital twin driven intelligent weaponry support technology[J]. Systems Engineering and Electronics, 2023, 45(4): 1247–1260.
[25] 胡权. 数字孪生体:第四次工业革命的通用目的技术[M]. 北京:人民邮电出版社,2021. HU Quan. Digital twin: general-purpose technologies (GPT) of the fourth industrial revolution[M]. Beijing: Posts and Telecommunications Press, 2021.
[26] 王林,乔建梅. 国外数字孪生技术发展及对我国的启示[J]. 上海信息化,2021(11):52–55. WANG Lin, QIAO Jianmei. The development of digital twin technology abroad and its implications for China[J]. Shanghai Informatization, 2021(11): 52–55.
[27] 陶飞,程颖,程江峰,等. 数字孪生车间信息物理融合理论与技术[J]. 计算机集成制造系统,2017,23(8):1603–1611. TAO Fei, CHENG Ying, CHENG Jiangfeng, et al. Theories and technologies for cyber-physical fusion in digital twin shop-floor[J]. Computer Integrated Manufacturing Systems, 2017, 23(8): 1603–1611.
[28] 苏兴华,詹胜,胡刚. 石油钻井数字孪生架构设计[J]. 信息系统工程,2021(11):26–30. SU Xinghua, ZHAN Sheng, HU Gang. Design of digital twin architecture for oil drilling[J]. China CIO News, 2021(11): 26–30.
[29] 杨少龙,孙延浩,向先波,等. 船舶数字孪生及其服务全生命周期研究综述[J]. 舰船科学技术,2020,42(21):1–8. YANG Shaolong, SUN Yanhao, XIANG Xianbo, et al. Ship digital twin and a review of life-cycle service[J]. Ship Science and Technology, 2020, 42(21): 1–8.
[30] 陶飞,刘蔚然,刘检华,等. 数字孪生及其应用探索[J]. 计算机集成制造系统,2018,24(1):1–18. TAO Fei, LIU Weiran, LIU Jianhua, et al. Digital twin and its potential application exploration[J]. Computer Integrated Manufacturing Systems, 2018, 24(1): 1–18.
[31] ROMMETVEIT R, BJØRKEVOLL K S, HALSEY G W, et al. eDrilling: a system for real-time drilling simulation, 3D visualization, and control[R]. SPE 106903, 2007.
[32] MAYANI M G, SVENDSEN M, OEDEGAARD S I. Drilling digital twin success stories the last 10 years[R]. SPE 191336, 2018.
[33] ROMMETVEIT R, GHOLAMI MAYANI M, NABAVI J, et al. Automatic realtime monitoring of drilling using digital twin technologies enhance safety and reduce costs[R]. SPE 197465, 2019.
[34] KEMAJOU V, SAMUEL R. Real-time well monitoring and engineering analysis of drilling activities: Intelligent rig state detection and prediction with uncertainty[C]//ASME 2020 39th International Conference on Ocean, Offshore and Arctic Engineering. New York: ASME, 2020: V011T11A001.
[35] ISLAM N, ROSENER A, SOUZA W, et al. Smarter well engineering concepts aid in reducing planning time and increasing ROP[C]//First EAGE Digitalization Conference and Exhibition. Odijk: European Association of Geoscientists & Engineers, 2020: 1–5.
[36] BIMASTIANTO P, KHAMBETE S, ALSAADI H, et al. Digital twin implementation on current development drilling, benefits and way forward[R]. SPE 202795, 2020.
[37] 井玉霏. 石油钻机提升系统的数字孪生模型构建[D]. 西安:西安石油大学,2023. JING Yufei. Construction of digital twin model for lifting system of oil drilling rig[D]. Xi'an: Xi'an Shiyou University, 2023.
[38] SRIKONDA R, RASTOGI A, OESTENSEN H. Increasing facility uptime using machine learning and physics-based hybrid analytics in a dynamic digital twin[R]. OTC 30723, 2020.
[39] WU Bo, KOU Yufeng, LIU Jun, et al. Research on the application of digital twin technology in the structural safety assessment of deep-water semi-submersible platforms[R]. ISOPE-I 22-296, 2022.
[40] MURRAY P, WATTIS Z, BAIN B, et al. Towards a digital twin supporting risk based decision making for offshore installations[R]. SPE 195717, 2019.
[41] 蒋爱国,王金江,谷明,等. 数字孪生驱动半潜式钻井平台智能技术应用[J]. 船海工程,2019,48(5):49–52. JIANG Aiguo, WANG Jinjiang, GU Ming, et al. Application of intelligent technology of semi-submersible drilling platform driven by digital twin[J]. Ship & Ocean Engineering, 2019, 48(5): 49–52.
[42] STRAW M, MISTRY R, LEWIS S, et al. Predictive digital twin for performance and integrity[R]. OTC 31863, 2022.
[43] 刘文岭,韩大匡. 数字孪生油气藏:智慧油气田建设的新方向[J]. 石油学报,2022,43(10):1450–1461. LIU Wenling, HAN Dakuang. Digital twin system of oil and gas reservoirs: a new direction for smart oil and gas field construction[J]. Acta Petrolei Sinica, 2022, 43(10): 1450–1461.
[44] SHIRANGI M G, FURLONG E, SIMS K S. Digital twins for well planning and bit dull grade prediction[R]. SPE 200740, 2020.
[45] SHI Jibin, DOURTHE L, LI D, et al. Real-time reamer vibration predicting, monitoring, and decision-making using hybrid modeling and a process digital twin[R]. SPE 208795, 2022.
[46] BOUKREDERA F S, HADJADJ A, YOUCEFI M R. Drill string torsional vibrations modeling with dynamic drill pipe properties measurement and field validation[J]. Journal of Energy Resources Technology, 2022, 144(2): 024502. doi: 10.1115/1.4051299
[47] SHIRANGI M G, ETTEHADI R, ARAGALL R, et al. Digital twins for drilling fluids: advances and opportunities[R]. SPE 199681, 2020.
[48] SAINI G S, ASHOK P, VAN OORT E. Predictive action planning for hole cleaning optimization and stuck pipe prevention using digital twinning and reinforcement learning[R]. SPE 199548, 2020.
[49] ANJOS J L, ARANHA P E, MARTINS A L, et al. Digital twin for well integrity with real time surveillance[R]. OTC 30574, 2020.
[50] KARPOV R B, ZUBKOV D Y, MURLAEV A V, et al. Drilling performance and data quality control with live digital twin[R]. SPE 206527, 2021.
[51] 张鑫鑫,梁博文,张晓龙,等. 智能钻井装备与技术研究进展[J]. 煤田地质与勘探,2023,51(9):20–30. doi: 10.12363/issn.1001-1986.23.06.0324 ZHANG Xinxin, LIANG Bowen, ZHANG Xiaolong, et al. Research progress of intelligent drilling equipment and technology[J]. Coal Geology & Exploration, 2023, 51(9): 20–30. doi: 10.12363/issn.1001-1986.23.06.0324
[52] 赵恩普,刘晓晨. 浅析井场数据远程传输系统的设计与实现[J]. 中国管理信息化,2018,21(18):66–67. ZHAO Enpu, LIU Xiaochen. Design and implementation of remote transmission system for well site data[J]. China Management Informationization, 2018, 21(18): 66–67.
[53] 何治亮,赵向原,张文彪,等. 深层–超深层碳酸盐岩储层精细地质建模技术进展与攻关方向[J]. 石油与天然气地质,2023,44(1):16–33. HE Zhiliang, ZHAO Xiangyuan, ZHANG Wenbiao, et al. Progress and direction of geological modeling for deep and ultra-deep carbonate reservoirs[J]. Oil & Gas Geology, 2023, 44(1): 16–33.
[54] 叶新,潘清,董正宏. 多领域建模仿真方法综述[J]. 软件,2014,35(3):233–236. YE Xin, PAN Qing, DONG Zhenghong. Comparison of multi-domain modeling and simulation technology[J]. Software, 2014, 35(3): 233–236.
[55] ZHANG Chengkai, SONG Xianzhi, SU Yinao, et al. Real-time prediction of rate of penetration by combining attention-based gated recurrent unit network and fully connected neural networks[J]. Journal of Petroleum Science and Engineering, 2022, 213: 110396. doi: 10.1016/j.petrol.2022.110396
[56] 祝兆鹏,张瑞,宋先知,等. 数据拓展和增量更新的井底压力实时预测方法[J]. 石油机械,2023,51(6):1–10. ZHU Zhaopeng, ZHANG Rui, SONG Xianzhi, et al. Real-time prediction of bottom hole pressure based on data expansion and incremental update[J]. China Petroleum Machinery, 2023, 51(6): 1–10.
[57] 肖立志. 机器学习数据驱动与机理模型融合及可解释性问题[J]. 石油物探,2022,61(2):205–212. doi: 10.3969/j.issn.1000-1441.2022.02.002 XIAO Lizhi. The fusion of data-driven machine learning with mechanism models and interpretability issues[J]. Geophysical Prospecting for Petroleum, 2022, 61(2): 205–212. doi: 10.3969/j.issn.1000-1441.2022.02.002
[58] KARNIADAKIS G E, KEVREKIDIS I G, LU Lu, et al. Physics-informed machine learning[J]. Nature Reviews Physics, 2021, 3(6): 422–440. doi: 10.1038/s42254-021-00314-5
[59] CURINA F, TALAT QUSHCHI A, ALDANY A. A case study for the development and use of a well control simulator as a digital twin of a real scenario[R]. SPE 206530, 2021.
[60] FERRARA P, RICCI MACCARINI G, POLONI R, et al. Virtual reality: new concepts for virtual drilling environment and well digital twin[R]. IPTC 20267, 2020.
[61] CHATAR C, LOKESH R, KAMBOJ V, et al. Extended reality and gamification for drilling[R]. SPE 212532, 2023.
[62] 王文明,侯春来,武振宇,等. 海洋无隔水管修井的数字孪生框架与可视化交互[J]. 计算机集成制造系统,2021,27(2):423–431. WANG Wenming, HOU Chunlai, WU Zhenyu, et al. Frame and visualization for digital twin of marine riserless well intervention[J]. Computer Integrated Manufacturing Systems, 2021, 27(2): 423–431.
[63] 李浩,刘根,文笑雨,等. 面向人机交互的数字孪生系统工业安全控制体系与关键技术[J]. 计算机集成制造系统,2021,27(2):374–389. LI Hao, LIU Gen, WEN Xiaoyu, et al. Industrial safety control system and key technologies of digital twin system oriented to human-machine interaction[J]. Computer Integrated Manufacturing Systems, 2021, 27(2): 374–389.
[64] 金镭,秦雪伦,江如意,等. 油气行业数字孪生研究述评与展望[J]. 世界石油工业,2022,29(5):17–25. JIN Lei, QIN Xuelun, JIANG Ruyi, et al. Review and prospect of digital twin research in oil and gas industry[J]. World Petroleum Industry, 2022, 29(5): 17–25.
下载:
计量
- 文章访问数: 434
- HTML全文浏览量: 93
- PDF下载量: 189
