杨红梅;薛敏;杨泱;马磊磊;冯志强;

• 1:长治市综合检验检测中心
• 2:中国石油长庆油田油气工艺研究院
• 3:西南石油大学石油与天然气工程学院

摘要(Abstract)：

针对页岩油藏焖井开发过程中，合理焖井时间难以确定、影响因素多、计算难度大等问题，开展了基于机器学习的页岩油藏合理焖井时间优化研究。基于Y区块268口压裂水平井物性与施工参数，进行了数值模拟合理焖井时间的循环优选计算，并生成机器学习数据集。利用支持向量回归、多变量线性回归以及多层神经网络方法，分别建立了合理焖井时间预测模型，输入参数包括11个主要物性与施工参数。测试结果表明，新模型计算的合理焖井时间具有较高精度，预测准确率最高可达到94%。通过对比可以得知，在该模型条件下，支持向量回归法的准确率明显高于其他两种方法，具有较高适用性。毛管力大小、入地液量以及原油黏度对合理焖井时间影响较大，相关系数分别为0.202,0.170和0.159,在焖井方案设计中应重点考虑。经过机器学习优化后，Y区块X-1生产井累产油增长了约8.5%。

关键词(KeyWords)： 焖井时间;机器学习;支持向量回归;多层神经网络;页岩油井

Abstract:

Keywords:

基金项目(Foundation): 国家科技重大专项(2017ZX05069)

作者(Authors): 杨红梅;薛敏;杨泱;马磊磊;冯志强;

参考文献(References)：

• [1] 王睿.致密油藏压后闷井蓄能机理与规律的数值模拟研究[D].北京：中国石油大学，2019.WANG Rui.Numerical simulation study on mechanism and law of energy storage in shut-in schedule after fracturing of tight oil[D].Beijing:China University of Petroleum,2017.
• [2] 朱维耀，张启涛，岳明，等.裂缝网络支撑剂非均匀分布对开采动态规律的影响[J].工程科学学报，2020,42(10):1318-1324.ZHU Weiyao,ZHANG Qitao,YUE Ming,et al.Effect of uneven distribution of proppant in fracture network on exploitation dynamic characteristics[J].Chinese Journal of Engineering,2020,42(10):1318-1324.
• [3] 杜洋，雷炜，李莉，等.页岩气井压裂后焖排模式[J].岩性油气藏，2019,31(3):145-151.DU Yang,LEI Wei,LI Li,et al.Shut-in and flow-back pattern of fractured shale gas wells[J].Lithologic Reservoirs,2019,31(3):145-151.
• [4] 刘敦卿.压裂液微观渗吸与“闷井”增产机理研究[D].北京：中国石油大学，2017.LIU Dunqing.Research on microcosmic laws of fracture fluid imbibition and mechanisms of productivity enhancement by “shut-in” in unconventional hydrocarbon reservoir[D].Beijing:China University of Petroleum,2017.
• [5] 马莉，张驰，刘敦卿，等.涪陵页岩气田压裂后闷井工艺适应性初探[J].特种油气藏，2019,26(1):147-151.MA Li,ZHANG Chi,LIU Dunqing,et al.Preliminary study on the well-soaking adaptability after fracturing in fuling shale gasfield[J].Special Oil & Gas Reservoirs,2019,26(1):147-151.
• [6] 刘刚，杨东，梅显旺，等.松辽盆地古龙页岩油大规模压裂后闷井控排方法[J].大庆石油地质与开发，2020,39(3):147-154.LIU Gang,YANG Dong,MEI Xianwang,et al.Method of well-soaking and controlled flowback after large-scale fracturing of Gulong shale oil reservoirs in Songliao Basin[J].Petroleum Geology & Oilfield Development in Daqing,2020,39(3):147-154.
• [7] 张相春，杨亚静，刘军龙，等.页岩油开发合理焖井时间数值模拟[J].西安石油大学学报(自然科学版),2021,36(3):71-76.ZHANG Xiangchun,YANG Yajing,LIU Junlong,et al.Numerical simulation of reasonable soaking time in development of shale oil[J].Journal of Xi’an Shiyou University (Natural Science Edition),2021,36(3):71-76.
• [8] 王金龙，腊丹萍，雷兆丰，等.A区页岩油蓄能压裂后合理焖井时间研究[J].石油化工应用，2020,39(6):88-90.WANG Jinlong,LA Danping,LEI Zhaofeng,et al.Study of reasonable well shut-in time in a shale oil reservoir after fracturing[J].Petrochemical Industry Application,2020,39(6):88-90.
• [9] 李杨.水平井压裂后焖井技术的研究[J].辽宁化工，2020,49(7):794-796.LI Yang.Study of well shut-in technology after fracturing of horizontal well[J].Liaoning Chemical Industry,2020,49(7):794-796.
• [10] 孙健，李琪，陈明强，等.基于机器学习的油气水层随钻识别模型优选[J].西安石油大学学报(自然科学版),2019,34(5):79-85.SUN Jian,LI Qi,CHEN Mingqiang,et al.Optimization of model for identification of oil /gas and water layers while drilling based on machine learning[J].Journal of Xi’an Shiyou University (Natural Science Edition),2019,34(5):79-85.
• [11] 柴明锐，程丹，张昌民，等.机器学习方法对砂砾岩岩屑成分的预测：以西北缘 X723 井百口泉组为例[J].西安石油大学学报(自然科学版),2017,32(5):22-28.CHAI Mingrui,CHENG Dan,ZHANG Changmin,et al.Prediction of debris composition in glutenite by machine learning method:a case study in Baikouquan Formation of well X723 in the NW margin of Junggar Basin[J].Journal of Xi’an Shiyou University (Natural Science Edition),2017,32(5):22-28.
• [12] 黄诚，潘雯晋.基于机器学习的石油多峰模型研究及应用[J].西南石油大学学报(自然科学版),2020,42(6):75-81.HUANG Cheng,PAN Wenjin.Research and Application of Oil Multi-peak Model Based on Machine Learning[J].Journal of Southwest Petroleum University (Science and Technology Edition),2020,42(6):75-81.
• [13] 赵艳红，姜汉桥，李洪奇，等.基于机器学习的单井套损预测方法[J].中国石油大学学报(自然科学版),2020,44(4):57-67.ZHAO Yanhong,JIANG Hanqiao,LI Hongqi,et al.Research on predictions of casing damage based on machine learning[J].Journal of China University of Petroleum (Edition of Natural Science),2020,44(4):57-67.
• [14] 谷建伟，任燕龙，王依科，等.基于机器学习的平面剩余油分布预测方法[J].中国石油大学学报(自然科学版),2020,44(4):39-46.GU Jianwei,REN Yanlong,WANG Yike,et al.Prediction methods of remaining oil plane distribution based on machine learning[J].Journal of China University of Petroleum (Edition of Natural Science),2020,44(4):39-46.
• [15] 卫浪，蒲红宇，向辉，等.基于改进神经网络的丙烷回收流程多目标优化[J].石油与天然气化工，2021,50(1):66-71.WEI Lang,PU Hongyu,XIANG Hui,et al.Multi-objective optimization of propane recovery process based on improved BP neural network[J].Chemical Engineering of Oil & Gas,2021,50(1):66-71.
• [16] 梁生荣，李文君，翁军利，等.基于BP神经网络的天然气采气管线甲醇加注量预测及其分配管网优化[J].石油与天然气化工，2020,49(6):45-52.LIANG Shengrong,LI Wenjun,WENG Junli,et al.Quantitative prediction of the methanol injection for natural gas production pipelines based on BP artificial neural networks and optimization of the distribution network[J].Chemical Engineering of Oil & Gas,2020,49(6):45-52.
• [17] ZHANG QITAO,WEI CHENJI,WANG YUHE,et al.Potential for Prediction of Water Saturation Distribution in Reservoirs Utilizing Machine Learning Methods[J].Energies,2019,12(19),3597.
• [18] ROYA TALEBI,MOHAMMAD GHIASI,HOSSEIN TALEBI,et al.Application of soft computing approaches for modeling saturation pressure of reservoir oils[J].Journal of Natural Gas Science and Engineering,2014,20:8-15.
• [19] KHOSRAVI A,G PABON J J,KOURY R N N,et al.Using machine learning algorithms to predict the pressure drop during evaporation of R407C[J].Applied Thermal Engineering,2018,133:361-370.
• [20] MAHATO J K,GUPTA S K.Exploring applicability of artificial intelligence and multivariate linear regression model for prediction of trihalomethanes in drinking water[J].International Journal of Environmental Science and Technology,2021:1-14.
• [21] 周德胜，师煜涵，李鸣，等.基于核磁共振实验研究致密砂岩渗吸特征[J].西安石油大学学报(自然科学版),2018,33(2):51-57.ZHOU Desheng,SHI Yuhan,LI Ming,et al.Study on spontaneous imbibition feature of tight sandstone based on NMR experiment[J].Journal of Xi’an Shiyou University (Natural Science Edition),2018,33(2):51-57.
• [22] 刘华，王卫红，陈明君，等.页岩储层多尺度渗流实验及数学模型研究[J].西安石油大学学报(自然科学版),2018,33(4):66-71.LIU Hua,WANG Weihong,CHEN Mingjun,et al.Seepage experiment and mathematical model of multi-scale shale reservoir[J].Journal of Xi’an Shiyou University (Natural Science Edition),2018,33(4):66-71.