en
×

分享给微信好友或者朋友圈

使用微信“扫一扫”功能。
作者简介:

孙龙祥(2000—),男,安徽亳州人,在读硕士研究生,从事人工智能与地学交叉学科研究。E-mail:wa21201015@stu.ahu.edu.cn。

通讯作者:

吕文君(1989—),男,安徽合肥人,副研究员,博士。E-mail:wlv@ustc.edu.cn。

中图分类号:TE319

文献标识码:A

文章编号:1009-9603(2023)03-0049-10

DOI:10.13673/j.pgre.202203047

参考文献 1
董庆民,胡忠贵,蔡家兰,等.层序地层与同位素地球化学响应特征及礁滩发育模式——以川东城口—鄂西海槽边缘带长兴组为例[J].石油实验地质,2019,41(2):243-250.DONG Qingmin,HU Zhonggui,CAI Jialan,et al.Sequence stra‐ tigraphy and isotope geochemical response and development pat‐ tern of reef and shoal:a case study of Changxing Formation in the marginal zone of eastern Sichuan Chengkou-western Hubei oceanic trough[J].Petroleum Geology and Experiment,2019,41(2):243-250.
参考文献 2
刘艳红,罗兴旺,张武,等.致密砂岩油藏水淹层测井识别方法及影响因素研究[J].特种油气藏,2022,29(3):57-63.LIU Yanhong,LUO Xingwang,ZHANG Wu,et al.Study on log‐ ging identification method and influencing factors of waterflooded zone in tight sandstone reservoir[J].Special Oil & Gas Reservoirs,2022,29(3):57-63.
参考文献 3
李宁,闫伟林,武宏亮,等.松辽盆地古龙页岩油测井评价技术现状、问题及对策[J].大庆石油地质与开发,2020,39(3):117-128.LI Ning,YAN Weilin,WU Hongliang,et al.Current situation,problems and countermeasures of the well-logging evaluation technology for Gulong shale oil[J].Petroleum Geology & Oil‐ field Development in Daqing,2020,39(3):117-128.
参考文献 4
崔宝文,张顺,付秀丽,等.松辽盆地古龙页岩有机层序地层划分及影响因素[J].大庆石油地质与开发,2021,40(5):13-28.CUI Baowen,ZHANG Shun,FU Xiuli,et al.Organic sequence stratigraphic division and its influencing factors’analyses for Gulong shale in Songliao Basin[J].Petroleum Geology & Oil‐ field Development in Daqing,2021,40(5):13-28.
参考文献 5
陈欢庆,石成方,胡海燕,等.高含水油田精细油藏描述研究进展[J].石油与天然气地质,2018,39(6):1 311-1 322.CHEN Huanqing,SHI Chengfang,HU Haiyan,et al.Advances in fine description of reservoir in high water-cut oilfield[J].Oil & Gas Geology,2018,39(6):1 311-1 322.
参考文献 6
陈维涛,徐少华,孙珍,等.层序地层学标准化理论在陆架坡折-陆坡区的应用——以珠江口盆地中中新世沉积层序为例 [J].石油与天然气地质,2021,42(6):1 414-1 422.CHEN Weitao,XU Shaohua,SUN Zhen,et al.Application of standardized sequence stratigraphy theory in the shelf break-toslope area:a case study of the Middle Miocene sedimentary se‐ quence in the PRMB[J].Oil & Gas Geology,2021,42(6):1 414-1 422.
参考文献 7
刘国强.非常规油气时代的测井采集技术挑战与对策[J].中国石油勘探,2021,26(5):24-37.LIU Guoqiang.Challenges and countermeasures of well logging data acquisition technology in unconventional petroleum explo‐ ration and development[J].China Petroleum Exploration,2021,26(5):24-37.
参考文献 8
王龙,张金川,唐玄.鄂尔多斯盆地下寺湾—云岩地区长7段页岩气测井评价与分布规律研究[J].中国石油勘探,2019,24(1):129-136.WANG Long,ZHANG Jinchuan,TANG Xuan.Logging evalua‐ tion and distribution law of Chang7 shale gas in Xiasiwan-Yun‐ yan area,Ordos Basin[J].China Petroleum Exploration,2019,24(1):129-136.
参考文献 9
MUKHERJEE B,SRIVARDHAN V,ROY P N S.Identification of formation interfaces by using wavelet and Fourier transforms [J].Journal of Applied Geophysics,2016,128:140-149.
参考文献 10
肖波,韩学辉,周开金,等.测井曲线自动分层方法回顾与展望 [J].地球物理学进展,2010,25(5):1 802-1 810.XIAO Bo,HAN Xuehui,ZHOU Kaijin,et al.A review and out‐ look of automatic zonation methods of well log[J].Progress in Geophysics,2010,25(5):1 802-1 810.
参考文献 11
SARG J F,MARKELLO J R,WEBER L J,et al.Carbonate se‐ quence stratigraphy-a summary and perspective with case histo‐ ry,Neogene,Papua New Guinea[C].Jakarta:Proceedings of the International Symposium on Sequence Stratigraphy in SE Asia,1996:137-179.
参考文献 12
ZHANG Benshu,ZHAO Zhui,HAN Chaoyang,et al.Research summary of sequence stratigraphy in salt-lake basin[J].North‐ western Geology,2005,38(1):94-99.
参考文献 13
张文龙,李畅,王帅,等.地层智能划分与对比研究进展及发展趋势探讨[J].信息系统工程,2021,(5):132-135,138.ZHANG Wenlong,LI Chang,WANG Shuai,et al.Research prog‐ ress and development trend of intelligent strata division and cor‐ relation[J].China CIO News,2021,(5):132-135,138.
参考文献 14
李峰峰,郭睿,余义常.层序地层划分方法进展及展望[J].地质科技通报,2019,38(4):215-224.LI Fengfeng,GUO Rui,YU Yichang.Progress and prospect of the division sequence stratigraphy[J].Bulletin of Geological Sci‐ ence and Technology,2019,38(4):215-224.
参考文献 15
刘腾,王军,张京思,等.地震Wheeler域变换结合时频分析技术用于渤海油田岩性油气藏描述[J].岩性油气藏,2018,30(3):124-132.LIU Teng,WANG Jun,ZHANG Jingsi,et al.Application of Wheeler transform combined with time-frequency analysis tech‐ nology to lithologic reservoir characterization of Bohai Oilfield [J].Lithologic Reservoirs,2018,30(3):124-132.
参考文献 16
HAWKINS D M,MERRIAM D F.Optimal zonation of digitized sequential data[J].Journal of the International Association for Mathematical Geology,1973,5(4):389-395.
参考文献 17
李广场,李江林.有序聚类分析在声波测井自动分层中的应用 [J].工程勘察,2007,217(8):73-75.LI Guangchang,LI Jianglin.Application of ordered clustering analysis in automatic stratification of sonic logging[J].Geotech‐ nical Investigation & Surveying,2007,217(8):73-75.
参考文献 18
谢忠怀,李保利.测井地层自动划分的边缘检测最优分割方法 [J].新疆石油地质,2006,27(2):173-175.XIE Zhonghuai,LI Baoli.Edge detection optimum partitioning method for automatic division of log strata[J].Xinjiang Petro‐ leum Geology,2006,27(2):173-175.
参考文献 19
罗德江,汪兴旺,刘伟.基于局部投影和边缘检测的自动分层模型[J].物探与化探,2009,33(1):99-101.LUO Dejiang,WANG Xingwang,LIU Wei.Automatic layer sep‐ aration model based on the local projection and edge detection methods[J].Geophysical and Geochemical Exploration,2009,33(1):99-101.
参考文献 20
罗德江,汪兴旺,刘伟.基于相空间重构和改进最优分割法的自动分层模型[J].地球物理学进展,2009,24(2):675-679.LUO Dejiang,WANG Xingwang,LIU Wei.Automatic separate layer model based on the phase-space reconstruction and im‐ proved optimum division method[J].Progress in Geophysics,2009,24(2):675-679.
参考文献 21
PARTOVI S M A,SADEGHNEJAD S.Geological boundary de‐ tection from well-logs:an efficient approach based on pattern recognition[J].Journal of Petroleum Science and Engineering,2019,176:444-455.
参考文献 22
张明玉.极值方差聚类法在测井分层取值中的应用——以莫北油田莫005井区为例[J].新疆石油地质,2002,23(5):429-431.ZHANG Mingyu.Application of minimax variance cluster analy‐ sis method in well log layering data record-an example of well block Mo-005 in Mobei Oilfield[J].Xinjiang Petroleum Geolo‐ gy,2002,23(5):429-431.
参考文献 23
程玉群,李秀荣,刘铁桩.测井曲线自动分层技术及在杏北地区小层划分中的应用[J].断块油气田,2000,7(2):23-25.CHENG Yuqun,LI Xiurong,LIU Tiezhuang.Technique of autozonation with log and its application to reservoir zonation at Xingbei area[J].Fault-Block Oil and Gas Field,2000,7(2):23-25.
参考文献 24
初颖,吕堂红.基于极值法和聚类分析法的测井曲线自动分层模型——以山东省胜利油井为例[J].长春理工大学学报:自然科学版,2017,40(6):105-110.CHU Ying,LÜ Tanghong.Logging-curve automatic layered model based on the extreme value method and clustering analy‐ sis-taking Shandong Province,Shengli Oilfield as an example[J].Journal of Changchun University of Science and Technology:Natural Science Edition,2017,40(6):105-110.
参考文献 25
郝建华,雷颖,王佳音,等.测井曲线活度法在地层划分中的应用[J].中国石油和化工标准与质量,2012,32(6):170-171.HAO Jianhua,LEI Ying,WANG Jiayin,et al.Application of ac‐ tivity method of logging curve in stratigraphic division[J].China Petroleum and Chemical Standard and Quality,2012,32(6):170-171.
参考文献 26
朱剑兵.基于最大熵谱估计的沉积旋回划分方法探讨[J].岩性油气藏,2012,24(2):67-71.ZHU Jianbing.Division of sedimentary cycle sequences based on maximum entropy spectrum estimation[J].Lithologic Reser‐ voirs,2012,24(2):67-71.
参考文献 27
袁野,王黎,谢锐杰.INPEFA 技术在层序地层划分中的应用:以苏北盆地溱潼凹陷南华区块阜三段为例[J].石油实验地质,2018,40(6):871-876.YUAN Ye,WANG Li,XIE Ruijie.Application of INPEFA tech‐ nology to sequence stratigraphy of the third member of Funing Formation,Nanhua block,Qintong Sag,North Jiangsu Basin[J].Petroleum Geology and Experiment,2018,40(6):871-876.
参考文献 28
BRACEWELL R N.The Fourier Transform and its Applications [M].New York:McGraw-Hill,1986.
参考文献 29
DAUBECHIES I.The wavelet transform,time-frequency local‐ ization and signal analysis[J].IEEE transactions on Information Theory,1990,36(5):961-1 005.
参考文献 30
MCDOUGAL W S,WEIN A J,KAVOUSSI L R,et al.Campbellwalsh urology 11th edition review e-Book[M].New York:Elsevi‐ er Health Sciences,2015.
参考文献 31
阎辉,李鲲鹏,张学工,等.测井曲线的小波变换特性在自动分层中的应用[J].地球物理学报,2000,43(4):568-573.YAN Hui,LI Kunpeng,ZHANG Xuegong,et al.Wavelet trans‐ form properties of well log and their application in automatic segmentation[J].Chinese Journal of Geophysics,2000,43(4):568-573.
参考文献 32
薛波,杨青,张超虹.基于形态学滤波与小波变换的测井曲线自动分层方法[J].地球物理学进展,2020,35(1):203-210.XUE Bo,YANG Qing,ZHANG Chaohong.Automatic stratifica‐ tion method of logging curve based on morphological filtering and wavelet transform[J].Progress in Geophysics,2020,35(1):203-210.
参考文献 33
LI Xia,FAN Yiren,DENG Shaogui,et al.Automatic demarca‐ tion of sequence stratigraphy using the method of well logging multiscale data fusion[J].Petroleum Exploration and Develop‐ ment,2009,36(2):221-227.
参考文献 34
杨建民,黄保纲,刘卫林,等.基于小波变换的浅水三角洲高分辨率层序地层定量划分[J].长江大学学报:自然科学版,2019,16(12):20-24.YANG Jianmin,HUANG Baogang,LIU Weilin,et al.Quantita‐ tive division of high-resolution sequence stratigraphy in shallow water delta based on wavelet transform[J].Journal of Yangtze University:Natural Science Edition,2019,16(12):20-24.
参考文献 35
LIU Hui,JIANG Zaixing.Wavelet transform of geophysical well logging signal and its application to sequence division[C].Yan‐ tai:2010 3rd International Congress on Image and Signal Pro‐ cessing,2010,7:3 470-3 473.
参考文献 36
ZHANG Jinliang,SONG Aixue.Application of wavelet analysis in sequence stratigraphic division of glutenite sediments[C].Washington:2010 International Conference on Challenges in En‐ vironmental Science and Computer Engineering,2010,2:204-207.
参考文献 37
KADKHODAIE A,REZAEE R.Intelligent sequence stratigra‐ phy through a wavelet-based decomposition of well log data[J].Journal of Natural Gas Science and Engineering,2017,40:38-50.
参考文献 38
PAN S Y,HSIEH B Z,LU M T,et al.Identification of strati‐ graphic formation interfaces using wavelet and Fourier trans‐ forms[J].Computers & Geosciences,2008,34(1):77-92.
参考文献 39
MUKHERJEE B,SRIVARDHAN V,ROY P N S.Identification of formation interfaces by using wavelet and Fourier transforms [J].Journal of Applied Geophysics,2016,128:140-149.
参考文献 40
周志华.机器学习[M].北京:清华大学出版社,2016.ZHOU Zhihua.Machine learning[M].Beijing:Tsinghua Universi‐ ty Press,2016.
参考文献 41
CHEN P H,LIN C J,SCHÖLKOPF B,et al.A tutorial on v-sup‐ port vector machines[J].Applied Stochastic Models in Business and Industry,2005,21(2):111-136.
参考文献 42
IQBAL M R A,RAHMAN S,NABIL S I,et al.Knowledge based decision tree construction with feature importance domain knowledge[C].Yantai:2012 7th International Conference on Electrical and Computer Engineering,2012:659-662.
参考文献 43
BELGIU M,DRĂGUŢL.Random forest in remote sensing:a re‐ view of applications and future directions[J].ISPRS Journal of Photogrammetry and Remote Sensing,2016,114:24-31.
参考文献 44
赖强,魏伯阳,吴煜宇,等.基于随机森林的K-近邻算法划分火成岩岩性[J].特种油气藏,2021,28(6):62-69.LAI Qiang,WEI Boyang,WU Yuyu,et al.Classification of igne‐ ous rock lithology with K-nearest neighbor algorithm based on random forest(RF-KNN)[J].Special Oil & Gas Reservoirs,2021,28(6):62-69.
参考文献 45
ABIODUN O I,JANTAN A,OMOLARA A E,et al.State-of-theart in artificial neural network applications:a survey[J].Heliyon,2018,4(11):e00938.
参考文献 46
牟丹,王祝文,黄玉龙,等.基于最小二乘支持向量机测井识别火山岩类型:以辽河盆地中基性火山岩为例[J].吉林大学学报:地球科学版,2015,45(2):639-648.MOU Dan,WANG Zhuwen,HUANG Yulong,et al.Application of least squares support vector machine to lithology identifica‐ tion:taking intermediate/basaltic rocks of Liaohe Basin as an ex‐ ample[J].Journal of Jilin University:Earth Science Edition,2015,45(2):639-648.
参考文献 47
沈凯浪.基于测井曲线的地层划分和测井相分析[D].西安:西安科技大学,2015.SHEN Kailang.Stratigraphic division and sedimentary facies analysis based on logging curves[D].Xi'an:Xi'an University of Science and Technology,2015.
参考文献 48
尚福华,李金成,原野,等.基于改进BP神经网络的地层划分方法[J].计算机技术与发展,2020,30(9):148-153.SHANG Fuhua,LI Jincheng,YUAN Ye,et al.Stratigraphic divi‐ sion method based on improved BP neural network[J].Computer Technology and Development,2020,30(9):148-153.
参考文献 49
李金成.基于涡流优化算法的测井曲线地层对比方法研究[D].大庆:东北石油大学,2020.LI Jincheng.Research on the stratigraphic correlation method of logging curve based on vortex search algorithm[D].Daqing:Northeast Petroleum University,2020.
参考文献 50
ELKATATNY S,AL-ABDULJABBAR A,MAHMOUD A A.New robust model to estimate formation tops in real time using artificial neural networks(ANN)[J].Petrophysics:The SPWLA Journal of Formation Evaluation and Reservoir Description,2019,60(6):825-837.
参考文献 51
徐朝晖,刘钰铭,周新茂,等.基于卷积神经网络算法的自动地层对比实验[J].石油科学通报,2019,4(1):1-10.XU Zhaohui,LIU Yuming,ZHOU Xinmao,et al.An experiment in automatic stratigraphic correlation using convolutional neural networks[J].Petroleum Science Bulletin,2019,4(1):1-10.
参考文献 52
ZHANG Jingya,LI Guojun,ZHANG Juan,et al.A convolutional neural network approach for stratigraphic interface detection[C].Malmo:International Conference on Internet of Things as a Ser‐ vice,2020,346:529-542.
参考文献 53
MIKOLOV T,KARAFIÁT M,BURGET L,et al.Recurrent neu‐ ral network-based language model[C].Chiba,Japan:11th Annual Conference of the International Speech Communication Associa‐ tion,2010:1 045-148.
参考文献 54
FREUND Y,SCHAPIRE R E.Experiments with a new boosting algorithm[C].San Francisco:Proceedings of the Thirteenth Inter‐ national Conference on Machine Learning,1996,96:148-156.
参考文献 55
ZHOU Cuiying,OUYANG Jinwu,MING Weihua,et al.A strati‐graphic prediction method based on machine learning[J].Applied Sciences,2019,9(17):3 553.
参考文献 56
马陇飞,萧汉敏,陶敬伟,等.基于梯度提升决策树算法的岩性智能分类方法[J].油气地质与采收率,2022,29(1):21-29.MA Longfei,XIAO Hanmin,TAO Jingwei,et al.Intelligent li‐ thology classification method based on GBDT algorithm[J].Pe‐ troleum Geology and Recovery Efficiency,2022,29(1):21-29.
参考文献 57
徐鹏晔.基于数据驱动和循环滑动时窗的小层智能划分方法 [J].油气地质与采收率,2022,29(1):113-120.XU Pengye.Small layer intelligent division method based on da‐ ta-driven and cyclic sliding time window[J].Petroleum Geology and Recovery Efficiency,2022,29(1):113-120.
参考文献 58
MEREMBAYEV T,YUNUSSOV R,YEDILKHAN A.Machine learning algorithms for classification geology data from well log‐ ging[C].Kaskelen:2018 14th International Conference on Elec‐ tronics Computer and Computation,2018:206-212.
参考文献 59
GONG Ke,YE Zhihui,CHEN Dong,et al.Investigation on auto‐ matic recognition of stratigraphic lithology based on well log‐ ging data using ensemble learning algorithm[C].Brisbane:SPE Asia Pacific Oil and Gas Conference and Exhibition,2018.
参考文献 60
HARTIGAN J A,WONG M A.Algorithm AS136:a k-means clustering algorithm[J].Journal of the Rroyal Statistical Society:Series C,1979,28(1):100-108.
参考文献 61
HAHSLER M,PIEKENBROCK M,DORAN D.Dbscan:fast density-based clustering with R[J].Journal of Statistical Soft‐ ware,2019,91(1):1-30.
参考文献 62
RINGNÉR M.What is principal component analysis?[J].Nature Biotechnology,2008,26(3):303-304.
参考文献 63
KOHONEN T.Self-organizing maps:ophmization approaches//Artificial Neural Networks[M].UK:North-Holland,1991:981-990.
参考文献 64
阎辉,张学工,李衍达.应用自组织神经网络进行自动地层划分的设计与实现[J].模式识别与人工智能,2000,13(3):309-313.YAN Hui,ZHANG Xuegong,LI Yanda.The application of selforganizing neural network in automatic stratum segmentation[J].Pattern Recognition and Artificial Intelligence,2000,13(3):309-313.
参考文献 65
冯向东,张玉琴,韩红伟,等.基于自组织神经网络在油气分层中的研究[J].计算机技术与发展,2021,31(2):44-48.FENG Xiangdong,ZHANG Yuqin,HAN Hongwei,et al.Re‐ search on oil and gas stratification based on self-organizing neu‐ ral network[J].Computer Technology and Development,2021,31(2):44-48.
参考文献 66
张多,韩逢庆.基于支持向量机和有序聚类的岩层识别[J].智能系统学报,2014,9(1):98-103.ZHANG Duo,HAN Fengqing.Stratum identification based on the SVM and ordered cluster[J].CAAI Transactions on Intelli‐ gent Systems,2014,9(1):98-103.
参考文献 67
HONG Y,KANG C.Automatic well correlation by aligning mul‐ tiple well using deep neural networks[C].Abu Dhabi:Abu Dhabi International Petroleum Exhibition & Conference,2020,SPE202908-MS.
参考文献 68
KARIMI A M,SADEGHNEJAD S,REZGHI M.Well-to-well correlation and identifying lithological boundaries by principal component analysis of well-logs[J].Computers & Geosciences,2021,157:104-942.
参考文献 69
SHEN Yating,WANG Yunyun,MA Zhiguo.Label-expanded manifold regularization for semi-supervised classification[C].Nanjing:2017 12th International Conference on Intelligent Sys‐ tems and Knowledge Engineering,2017:1-4.
参考文献 70
ZHAO Huimin,ZHENG Jianjie,DENG Wu,et al.Semi-super‐ vised broad learning system based on manifold regularization and broad network[J].IEEE Transactions on Circuits and Sys‐ tems I:Regular Papers,2020,67(3):983-994.
参考文献 71
GHORBAN F,HASAN N,VELTEN J,et al.Improving FMGAN through mixup manifold regularization[C].Daegu:2021 IEEE International Symposium on Circuits and Systems,2021:1-5.
参考文献 72
YIN Xucheng,LIU Qian,HAO Hongwei,et al.FMI image based rock structure classification using classifier combination[J].Neu‐ ral Computing and Applications,2011,20(7):955-963.
参考文献 73
ZHANG Wei,WU Tong,LI Zhipeng,et al.Automatic detection of fractures based on optimal path search in well logging images [J].Journal of Sensors,2021,(4):1-10.
参考文献 74
CHANG Ji,LI Jing,KANG Yu,et al.SegLog:geophysical log‐ ging segmentation network for lithofacies identification[J].IEEE Transactions on Industrial Informatics,2021,18(9):6 089-6 099.
参考文献 75
FAN Pengpeng,DENG Rui,QIU Jinquan,et al.Well logging curve reconstruction based on kernel ridge regression[J].Arabian Journal of Geosciences,2021,14(16):1 559.
参考文献 76
MO Xuwen,ZHANG Qiang,LI Xiao.Well logging curve recon‐ struction based on genetic neural networks[C].Zhangjiajie:2015 12th International Conference on Fuzzy Systems and Knowledge Discovery,2015:1 015-1 021.
参考文献 77
SHU Jun,XIE Qi,YI Lixuan,et al.Meta-weight-net:learning an explicit mapping for sample weighting[C].Vancouver:33rd Con‐ ference on Neural Information Processing Systems,2019,172:1 919-1 930.
参考文献 78
ZHANG Zizhao,PFISTER T.Learning fast sample re-weighting without reward data[C].Montreal:Proceedings of the IEEE/CVF International Conference on Computer Vision,2021:725-734.
参考文献 79
CAI Qi,PAN Yingwei,WANG Yu,et al.Learning a unified sam‐ ple weighting network for object detection[C].Seattle:Proceed‐ ings of the IEEE/CVF Conference on Computer Vision and Pat‐ tern Recognition,2020:14 173-14 182.
参考文献 80
MENG Wei,TIA Meng.Unmanned aerial vehicle classification and detection based on deep transfer learning[C].Sanya:2020 In‐ ternational Conference on Intelligent Computing and HumanComputer Interaction,2020:280-285.
参考文献 81
LI Yunsheng,YUAN Lu,CHEN Yinpeng,et al.Dynamic transfer for multi-source domain adaptation[C].New Orleans:Proceed‐ ings of the IEEE/CVF Conference on Computer Vision and Pat‐ tern Recognition,2021:10 998-11 007.
目录contents

    摘要

    基于地球物理测井地层划分相关概念及分类,将测井曲线自动分层方法分为传统方法和人工智能方法,从有监督学习方法和无监督学习方法2个方面分析人工智能方法的应用情况,并综合比较各类地层自动划分方法的优缺点。通过探索相关领域的发展情况,从不同角度思考测井地层划分方法进一步发展所存在的挑战及其解决方法。一是引入半监督学习方法,解决人工标签稀缺问题;二是从分割模型的角度,打破对测井数据的固有认识;三是采用测井曲线重构等方法,解决井段失真或缺失所导致的数据异构问题;四是通过样本加权,解决人工标签错误导致的数据偏差问题;五是采用迁移学习方法,解决不同地区数据分布差异问题。人工智能方法是解决地层划分、岩性识别、储层识别、生产运行中现有难题以及推进测井相关任务数字化转型的重要支撑。

    Abstract

    Given the related concepts and classification of geophysical logging stratigraphic division,this paper divided au‐ tomatic stratification methods of logging curves into traditional methods and artificial intelligence methods and analyzed the application of artificial intelligence technology in logging stratigraphic division from the aspects of supervised and un‐ supervised learning. Then,it comprehensively compared the advantages and disadvantages of various automatic strati‐ graphic division methods. Finally,by exploring the development of related fields,this study considered the challenges and solutions in the future development of logging stratigraphic division from different perspectives. The specific solutions are as follows:①Semi-supervised learning can be introduced to solve the problem of scarce manual labels. ②A new under‐ standing of logging data can be obtained from the perspective of the segmentation model. ③Methods such as logging curve reconstruction can be employed to solve the problem of data heterogeneity caused by the distortion or missing of well sec‐ tions. ④The problem of data deviation caused by manual label errors can be resolved through sample weighting. ⑤Trans‐ fer learning can be used to solve the problem of data distribution differences in different regions. Artificial intelligence tech‐ nology can provide vital support for solving existing problems in stratigraphic division,lithology identification,reservoir identification,as well as operation and production,and promoting the digital transformation of tasks related with logging.

  • 在油气勘探中,地层界面的识别是地球物理和地质解释中必要且繁琐的一部分。在一系列岩石单元中,识别沉积层序边界有不同的信息来源,通过薄片显微照片、测井数据和地震剖面的研究可以确定层序界面,通常将岩心样本、测井曲线以及地震剖面数据进行整合,再由专家详细分析,从而更准确地识别层序边界[1-6]。然而,这类研究面临着核心数据匮乏、测井数据不完整、地震剖面缺乏或质量差等问题[7]。虽然岩心数据往往能代表真实的岩性,但并不适用于在所有钻井或取心井中对整个井段的岩性进行判断。一般油气田的所有油井都有测井记录,因此测井数据作为当下分辨率较高、连续性最好的地质数据之一,可以作为层序地层学研究的主要信息来源[8]。在解释过程中,解释人员可以利用自己的经验和地质知识标记地层中的界面,这可能会导致不一致的分层结果[9]

  • 目前,基于测井曲线进行地层划分的相关综述的时间跨度较长且缺少对最新方法的分析和调研[10-14]。例如,虽然对测井地层划分方法进行了详细介绍分析,但是时间跨度较长,因此对现今方法发展的指导意义较低[11-12],且近年来发表的相关综述文献并未充分展望测井分层方法的发展走向[13-14]。目前人工智能方法迅速发展,行业技术的发展也是日新月异,因此近期相关的综述文献不足以起到对现有方法分析总结及对行业发展方向的启示作用。

  • 为此,笔者对测井地层划分的传统方法进行总结,并引出基于人工智能方法的测井地层划分方法。其中,将传统方法划分为数理统计方法和特征分析方法,将人工智能方法划分为有监督学习方法和无监督学习方法,分别介绍这些方法的基本概念、流程、优缺点以及实际应用情况。进而结合人工智能的最新方法及其在其他领域的应用特点,展望先进人工智能方法在测井地层划分方面的潜在应用前景。

  • 1 测井地层划分原理

  • 地质研究的首要任务之一就是地层划分。近年来,有关地层划分的概念和理论不断补充完善,总的来说依据以下原理:①地层单位在时间、空间上具有一定规模。②划分后的地层内部应具有相当的统一性,或根据实用性划分地层。③地层单位在纵向上的界限必须易于识别且有一定的稳定性。

  • 地层划分的目的是将一个地区的地层进行纵向划分。一般来说,可以直接根据现有的特征实现地层划分,也可以先根据多种特征进行综合分析,再根据分析得出的属性进行地层划分。在技术和方法的不断更新下,地层划分单元趋于精细化,而划分结果的准确性则越难以得到验证。

  • 测井地层划分通常是基于自然电位(SP)、自然伽马(GR)、中子等测井曲线进行分析。自然电位曲线相对幅度、形态特征反映出沉积岩的粒度、分选和泥质含量变化。测井数据在地层划分中的应用主要依据突变(或奇异)点或突变区域的信息。测井数据频率的突变点通常代表粒度、沉积特征以及地层岩性发生变化[15]。在沉积层序划分中,其主要边界为最大海泛面(MFS)、海退面(RS)、海侵面 (TS)和层序边界(SB)。通常,测井响应与层序界面之间存在良好的相关性。根据定义,最大海泛面是海岸线处于最大陆地位置时的沉积表面,当海岸线向陆地移动最大时,会形成许多细粒沉积物(如黏土和页岩)。由于存在放射性且通常富含有机物的页岩和黏土,最大海泛面由更高的伽马射线测井数值表示,包括校正后的无铀伽马射线或总伽马射线,且页岩处测井数值更大。黏土矿物含量越高,中子孔隙度和声波传输时间数值越高,而密度测井数值降低,原因是与石英、长石、方解石和白云石等砂岩和碳酸盐岩的其他成分相比,黏土矿物密度较低。因此,在储层单元中,最大海泛面对应于较高的自然伽马、较高的补偿中子(NPHI)、较高的声波时差(DT)和较低的地层密度补偿(RHOB)。页岩显示出较高的孔隙度和较低的密度。另外,其他层序边界在测井数据中表现出的某些特点,也可以使其作为奇异点进行地层划分。

  • 2 传统测井地层划分方法

  • 2.1 数理统计方法

  • 数理统计方法主要指对地质旋回建立模型,进而确立与数学最优结果的对应关系。有序聚类分析是针对有序样品的一种统计分类方法,也叫最优分割法[16]。李广场等应用有序聚类分析进行声波测井曲线自动分层[17]。谢忠怀等提出一种将边缘检测与最优分割相结合的测井地层自动划分方法[18]。罗德江等将局部投影算法和边缘检测运用于测井曲线的自动分层,通过最优分割方法选出分界点[19];随后,又提出对最优分割方法的改进[20],改进后的自动分层模型更加可靠。

  • 方差分析方法是利用测井数据的层间与层内方差作为统计量来划分地层。PARTOVI 等采用方差分析工具和Tukey多重比较方法评价边界深度确定过程中各参数的有效性,并引入偏差和相对偏差 2 个参数作为测量误差的方法[21]。研究结果表明,该方法具有较好的自动搜索地质边界的能力。张明玉采用极值方差聚类法,实现了对数字化测井曲线的自动分层取值[22]。程玉群等采用层内差异法和聚类分析法相结合实现分层[23]

  • 中值滤波方法的最大特点是从整体上识别测井曲线的地层界面。其全过程分为中值滤波、识别曲线性质以及从分层曲线整体上确定岩性。初颖等通过中值滤波法对测井数据进行分析处理,然后应用极值法和聚类分析法建立测井曲线自动分层模型[24]。活度函数在测井地层划分中最开始被用于微电极测井的倾角计算,其通过寻找曲线的拐点作为分层点[10]。郝建华等指出活度曲线极大值可用于分层,将测井曲线的不同表现性质通过测井曲线的活度能量有机地结合起来[25]

  • 最大熵谱估计是现代功率谱估计方法之一,通过最大熵准则获得功率谱密度。在测井地层划分任务中,朱剑兵利用最大熵谱估计进行沉积旋回划分,可以较好地识别沉积旋回界面及旋回内部的半旋回[26]。袁野等利用最大熵谱分析技术对测井曲线进行合成预测,并对处理后的数据进行误差滤波分析,得到频率趋势线。最后运用不同幅度的趋势拐点来识别不同级次层序界面[27]

  • 2.2 特征分析方法

  • 对曲线特征进行提取分析的方法包括傅里叶变换[28]、小波变换[29] 以及沃尔什变换[30],其中,小波变换能够实现信号分析的时频局部化。阎辉等发现小波变换过零点(极值点)与地层界面存在一定联系,并以此提出以小波变换过零点来区分地层界面的分界点[31]。薛波等提出将测井曲线进行形态学滤波,再通过小波变换模极大值法找出拟地层界面,最后综合不同的测井曲线分层结果确定最终分层结果[32]。LI 等基于小波变换模极大值重构的思想,提出了基于多尺度边缘检测原理的测井数据融合方法[33]。杨建民等采用基于小波变换的时频分析技术,结合测井数据、地震资料定量识别和划分高精度层序地层界面[34]。LIU等基于从小波系数色谱图中得出测井信号的时频特性,根据小波系数色谱图和小波系数曲线的振荡特性,识别出不同的层序边界[35]。ZHANG等提出一种基于 DB5连续小波变换的小波分析方法进行层序地层划分与对比[36]。研究表明,小波分析方法对砂砾岩沉积物的分界是非常适用和有效的。KADKHODAIE 等利用小波变换,成功地从信号的最高频率和低频内容中识别出最大海泛面[37]

  • 近年来,将各种处理方法结合的方法也不断被提出,并且达到了较好的实际应用效果。其中,PAN 等使用 WT 和 FT 方法来分析自然电位和自然伽马测井数据,获得了清晰的地层界面信号[38]。 MUKHERJEE 等利用傅立叶变换和小波变换对测井数据进行去噪,以获得地层界面[39],该方法可用于在工业上的地层界面解释。

  • 2.3 传统方法分析

  • 传统方法中的数理统计方法属于时域分析方法,需进行复杂的求解运算,可以在一定程度上保持岩层内部的均一性和岩层之间的差异性,但是存在精度不足和计算量较大的问题。特征分析方法属于变换域分析方法,其中小波变换能够对测井数据进行时频分析,可以有效提取信号的特征点,但是无法实现对地层的局部和整体信息的有效分辨,如何选取合适的小波函数也是实际应用方面的难点问题。

  • 另外,传统方法往往比较依赖于人工分析,因此其处理结果与专家的工作状态、知识经验等方面紧密相关,精度的稳定性和一致性难以得到保证。大量的测井数据必然导致消耗大量的人力和时间成本。

  • 3 基于人工智能的测井地层划分方法

  • 人工智能是指机器在经过训练之后能够获得人的部分行为能力,从而可以执行人类特征任务,如识别物体、声音以及对环境中的变化做出特定变化等行为,并能从中不断地学习以更有效地解决问题。随着机器学习方法性能的不断提升,促进了人工智能领域相关方法的发展脚步[40]。机器学习方法根据学习任务的特点,可分为有监督学习、无监督学习和半监督学习方法。目前,半监督学习方法还未在该领域中得到应用,且半监督学习方法可以看做有监督学习和无监督学习方法的综合,因此主要对有监督学习和无监督学习方法进行阐述。

  • 3.1 有监督学习方法

  • 训练样本是成对的数据,即每个输入样本存在对应标签,这种任务在机器学习方法中被称为监督学习方法。诸如支持向量机[41]、决策树[42]、随机森林[43-44]、人工神经网络[45] 等监督学习方法,正在逐步被应用于基于测井曲线的地层划分中。支持向量机方法是定义在特征空间的间隔最大的线性分类器。牟丹等以辽河盆地中基性火山岩为例,建立了最小二乘支持向量机岩性识别模型,利用该方法进行全井段的地层自动识别,识别结果的平均正确率达到75%[46]。沈凯浪将同一区域的地层划分看成一个有序多分类问题,建立多个支持向量机分类模型,将专家得出的结果作为训练数据,该方法的地层划分正确率达到90%[47]

  • 神经网络由多层神经元连接组成,每层又可以包含多个神经元,神经网络可以从环境中进行学习,并根据一定的度量规则来改善其行为。尚福华等提出一种基于改进 BP 神经网络的地层划分方法,选取可以较好地反映不同类型地层的特征来构建参数样本,并在实验中采用 6 种测井曲线作为特征参数,建立神经网络模型;利用同一地区中多个标准井的小层进行样本训练,训练后的网络即可用于同类地区的地层划分[48]。李金成利用改进的 BP 神经网络建立地层划分模型,以标准井的 6 条测井曲线及其岩性数据作为训练数据对模型进行训练,进而利用所得模型对验证井进行划分,其结果与人工划分结果基本吻合[49]。ELKATATNY等基于反向传播神经网络模式识别功能的应用,利用钻速、泥浆密度和渗透率建立一种预测地层界面的新模型,以泥浆流速、钻杆旋转、立管压力、扭矩和钻头重量作为输入参数;该研究使用单隐层人工神经网络模型,并采用 Levenberg-Marquardt反向传播函数对模型进行训练,实验结果表明所开发的人工神经网络模型在估算A井和B井测试数据集的地层顶部时具有较高的精度[50]

  • 地层划分与对比任务与图像分割任务具有很大的相似性,因此,用于图像分割任务的神经网络也逐渐被应用于地层划分。图1为一种可应用于地层划分的卷积神经网络结构,以多条测井曲线作为网络的输入,输出为划分出的准层序。徐朝晖等将地层划分看作一种图像分割任务,采用深度学习中广泛应用的卷积神经网络方法,以大庆油田某区块密井网数据为研究对象,选取已知井中的 3 条测井曲线以及地层标签作为数据输入,训练SegNet网络从而实现地层自动划分,实验表明卷积神经网络方法在对测井曲线图像自动对比任务中具有良好的应用效果[51]。ZHANG 等提出一种从地球物理测井数据中检测地层界面的卷积神经网络方法,将数据点转换为二维片段,并将这些片段作为单频道图片输入卷积神经网络进行训练,利用训练好的模型对地层界面进行预测;基于训练好的卷积神经网络模型对 28 口井进行预测,最终的平均预测精度达 89.69%;实验结果表明,地层边界点识别的正确率和召回率均超过 85%,预测的边界点与地面真实值的相对误差仅为1%,满足实际应用要求[52]

  • 此外,其他诸如递归神经网络[53]、Adaboost 算法[54] 等方法也被使用于地层划分中。ZHOU等提出一种用于地质统计序列模拟的机器学习方法,在递归神经网络的基础上,依次建立地层类型和地层厚度的序列模型,通过引入专家驱动学习,在一定程度上提高模型的预测能力,结果表明可以将机器学习方法应用于一系列的地质统计序列模拟[55]。马陇飞等选择 8 个测井参数,构建基于梯度提升决策树算法的岩性识别模型,该方法在岩性识别任务中准确率达到92%[56]。徐鹏晔选择对地质分层敏感的测井曲线作为特征参数,并采用循环滑动时窗方法多次进行样本数据采集,优选不同机器学习方法训练模型,并使用该模型对小层智能划分结果进行预测,实验结果优于常规一次性建模预测方法[57]。 MEREMBAYEV等使用5种机器学习方法来处理基于 KS,PS 数据和梯度(一阶和二阶导数)等附加数据的铀矿床地层分类,现场成功地使用该模型,有助于改进铀矿开采的整个业务流程[58]。GONG等通过引入机器学习方法来提高岩性识别的效率和准确性,以中国龙潜地区大安剖面的3口井为研究对象,采用典型的集成学习方法 Adaboost算法,利用决策树、支持向量机和贝叶斯等典型的单分类机器学习方法进行岩性识别,进而可以在钻井过程中识别地下复杂沉积环境条件下的不同地层[59]

  • 图1 应用于地层划分的卷积神经网络结构

  • Fig.1 Architecture of convolutional neural network applied to stratigraphic division

  • 3.2 无监督学习方法

  • 训练数据不成对出现,即只由 1 组输入向量组成、不含对应的标签的机器学习任务称为无监督学习方法。常用的无监督学习方法包括 k-Means 聚类[60]、DBSCAN聚类[61]、主成分分析[62] 和自组织映射神经网络[63] 等。其中,自组织映射神经网络结构包括输入层和计算层,依靠神经元之间的互相竞争逐步优化网络(图2)。阎辉等通过自组织映射神经网络法处理测井数据,得到自组织原理聚类的不同特征,将分类结果放入一个模式分类器(或观察规方律),使输出节点直接与某特定类别的特征相联系,实验结果与实际人工划分结果的符合率达到 90%[64]。冯向东等通过自组织映射神经网络对测井曲线进行聚类自动分层,首先采用插值的方法去除噪声,然后选取自然伽马、声波时差等 19 个指标作为输入数据训练自组织神经网络,实现对水层、油层等 4 种地层的划分,准确率达 82.17%[65]。在地层划分、岩性识别等任务中有时缺少足够的训练样本,而支持向量机方法需要一定的样本来进行训练,才能达到较好的分类效果,针对该问题张多等利用有序聚类方法基于测井数据进行初步分层,然后以初步分层结果作为训练样本,最后采用训练后的支持向量机分类器对测井数据进行第 2 次分层[66]

  • 图2 自组织映射神经网络结构

  • Fig.2 Self-organizing mapping neural network

  • HONG 等提出一种简单且无监督学习的方法,利用先进的机器学习方法实现测井曲线的自动关联[67]。假设附近的测井曲线在相邻深度点具有类似的特征,引入时间序列对齐概念,可以处理不同长度的多个测井曲线。KARIMI等通过对碳酸盐岩储层多条测井曲线进行主成分分析,研究应用主成分分析的 3 种不同情景,并确定了井间相关性的最佳方法[68]

  • 3.3 人工智能方法分析

  • 人工智能方法能够实现自动化分层,并摆脱人工划分的主观不一致性等问题。利用专家提供的少量标签,有监督方法主要包括以SVM为代表的传统机器学习模型和以深度神经网络为代表的深度学习模型,可以快速建立地层划分模型,并实现无标签井段的快速划分,在运行效率和划分精度上具有较大的优势,并且大大减少了人工和实践消耗。如果不提供标签,也可以使用无监督学习方法实现测井地层划分,主要包括自组织映射神经网络、主成分分析方法等,其不需要成对的数据进行训练,且对小地层的划分具有较好的效果,因此无监督学习方法可以作为测井曲线的预先划分,人工只需对划分结果进行后处理及修正即可。

  • 综上所述,人工智能方法能够快速准确地处理大型测井数据集,避免传统方法中出现的划分结果不一致、精度不稳定等问题。同时,从数据中建立复杂的非线性映射模型,具有较高的泛化能力,可以向其他井进行推广应用,保证划分性能的同时提升工作效率。

  • 4 测井地层划分方法发展趋势

  • 虽然人工智能方法在测井地层划分中得到了一定程度的应用,但仍然面临以下几点挑战。

  • 第一,模型的训练需要人工标签,相对于海量测井数据,人工标签是昂贵且稀缺的。对于该问题,可以采用半监督学习方法进行研究。半监督学习方法是从少量标记数据和大量未标记数据的组合中学习,有标记的样本有助于最终模型的识别能力,而未标记的样本则保持了数据的几何结构。为利用未标记的样本信息,在半监督学习分类中一般采用一些分布的假设。其中,流形正则化是一类被深入研究和常用的半监督分类方法。SHEN等提出一种新的标签扩展流形正则化框架(简称 LE_MR) 用于半监督学习分类,试图扩展标记实例集,以减少“误导性”标记实例的影响,实验结果证明所提出的 LE_MR方法在半监督学习分类中取得了较好的效果[69]。ZHAO等提出一种基于流形正则化框架和广义学习系统的半监督广义学习系统,该方法继承了广义学习系统和流形正则化框架的特点,扩展了广义学习系统应用化框架的特点及应用领域[70]。 GHORBAN等提出一种混合流形正则化,用于执行基于 GAN 的半监督学习分类;与其他基于 GAN 的方法相比,混合流形正则化取得了最好的效果[71]。目前,测井地层划分方法往往依赖于海量的数据,但是在人工成本昂贵的前提下,只能得到少量的标签数据。在少量的标签数据和大量的无标签数据基础上,若只采用现有的有监督学习和无监督学习方法进行测井地层划分,未能充分利用现有的测井数据。因此,在测井数据标签有限、无法精确反映地层信息的情况下,半监督学习方法的引入可以有效解决测井地层划分任务中标签不足的问题。

  • 第二,现有研究大多采用深度点模型,即从某个深度抽取测井样本并对其进行分类,而测井地层划分更适合采用机器学习方法中的分割模型进行研究。YIN 等使用 K-最近邻分类器对岩石结构进行分类,并提取超声波测井图像中的预期特征;该方法将图像分割成大小均匀、致密度适中的超像素块,为后续的裂缝识别提供了依据[72]。ZHANG 等提出一种有效的裂缝识别和提取方法,该方法能提取裂缝区域对应的像素点,并给出高质量的裂缝形态图像[73]。CHANG 等尝试通过输入井段对应的测井数据来预测岩相,并首次将测井岩相识别任务建模为一维语义分割任务,基于真实数据集的大量实验验证了该模型的有效性以及在岩相识别问题上的优越性[74]。研究人员有望将基于上述图像分割、语义分割等方法应用于地层划分任务,从整体和形态上进一步提取测井曲线的特征。

  • 第三,由于井眼条件不佳以及成本控制等导致某些深度段的测井曲线存在失真与缺失,进而导致数据异构。FAN 等采用核岭回归方法进行声学测井曲线重构并取得了一定的效果[75]。MO等提出一种基于遗传神经网络优化算法的电缆测井曲线重构方法,以声学测井曲线的重建为例,确定优化后的遗传神经网络的结构和相关参数,并验证了该方法的有效性[76]。对于不同地区的钻井,测井曲线重构的精度可能会发生变化,说明测井曲线重构的精度不仅与重建方法有关,还与所选地区的地层和测井曲线的质量有关。因此,环境变化导致的测井数据分布偏差会影响重构曲线质量,导致重构的测井曲线不能完全符合地层的真实情况,与真实曲线之间仍存在一定的偏差。因此,跨井测井曲线重构精度的提升仍是重要的挑战,可以采用迁移学习方法来进行研究。

  • 第四,在人工标签存在错误的情况下,测井地层划分模型的训练也会产生极大的不确定性。训练样本重加权是解决不平衡和标签损坏的标签数据偏差的有效方法。SHU 等提出一种可以自适应地直接从数据中学习显式加权函数的方法,在少量无偏元数据的指导下,加权函数的参数更新与分类器的学习同步进行[77]。ZHANG 等提出一种新的基于学习的快速样本重加权方法,该方法在标签噪声鲁棒性方面取得了良好效果,同时显著提高了训练效率[78]。CAI等提出统一的样本加权网络预测样本的任务权重,通过分类损失、回归损失、交并比和概率评分预测样本权重,该方法能够提高目标检测精度[79]。人工标签的获取不仅成本昂贵,而且由于与专家的工作状态、知识经验等紧密相关,导致精度的稳定性和一致性难以得到保证。因此,通过诸如样本加权等方法引入测井地层划分解决人工标签存在错误的问题,以期可以得到较好的效果。

  • 第五,不同地区的地质特征存在一定差异,甚至同一地区不同井段的地质特征也不尽相同,这就可能导致同一种测井地层划分方法在不同井间难以直接应用。近几年,迁移学习方法的发展为这种数据分布存在偏差的情况提供了解决手段。迁移学习方法如图3 所示,通常在源域数据多于目标域数据时采用该方法。MENG 等采用迁移学习方法对无人机图像进行双重分类和检测,通过对 3 个模型进行比较得到适用于迁移的最优模型[80]。LI 等提出动态转移解决领域冲突,其中模型参数适应于样本,将多源域转换为单源域,简化了源域与目标域之间的对齐,与现有的多源域自适应方法相比,该动态迁移学习方法具有更好的自适应性能[81]。面对未知的、多变的应用领域,训练数据总是有限的,导致有限训练数据与模型泛化能力的矛盾。迁移学习方法通过调整预先训练的深度神经网络的权值,仅需使用少量目标域样本数据便可获得良好的效果。因此,迁移学习方法作为解决模型泛化能力问题的有效手段,将成为解决现有测井数据分布存在偏差问题的有效途径。

  • 图3 迁移学习方法

  • Fig.3 Transfer learning method

  • 5 结论

  • 基于人工智能方法的测井地层划分是测井解释的重要发展方向,人工智能方法是解决地层划分、岩性识别、储层识别以及生产运行中现有难题的重要手段,是推进测井数字化转型的重要支撑。笔者对近年来人工智能方法的发展与现阶段测井地层划分方法综合分析并提出以下展望:①引入半监督学习方法,解决人工标签稀缺问题。②从分割模型的角度,打破对测井数据的固有认识。③采用测井曲线重构等方法,解决井段失真或缺失所导致的数据异构问题。④通过样本加权,解决人工标签错误导致的数据偏差问题。⑤采用迁移学习方法,解决不同地区数据分布差异问题。将人工智能、大数据以及物联网等技术深度融入测井关键业务,可以支撑测井相关任务的开发研究、分析、优化等智能化转型,变革传统的测井分析模式,进而保障国家能源开发的稳步推进。

  • 参考文献

    • [1] 董庆民,胡忠贵,蔡家兰,等.层序地层与同位素地球化学响应特征及礁滩发育模式——以川东城口—鄂西海槽边缘带长兴组为例[J].石油实验地质,2019,41(2):243-250.DONG Qingmin,HU Zhonggui,CAI Jialan,et al.Sequence stra‐ tigraphy and isotope geochemical response and development pat‐ tern of reef and shoal:a case study of Changxing Formation in the marginal zone of eastern Sichuan Chengkou-western Hubei oceanic trough[J].Petroleum Geology and Experiment,2019,41(2):243-250.

    • [2] 刘艳红,罗兴旺,张武,等.致密砂岩油藏水淹层测井识别方法及影响因素研究[J].特种油气藏,2022,29(3):57-63.LIU Yanhong,LUO Xingwang,ZHANG Wu,et al.Study on log‐ ging identification method and influencing factors of waterflooded zone in tight sandstone reservoir[J].Special Oil & Gas Reservoirs,2022,29(3):57-63.

    • [3] 李宁,闫伟林,武宏亮,等.松辽盆地古龙页岩油测井评价技术现状、问题及对策[J].大庆石油地质与开发,2020,39(3):117-128.LI Ning,YAN Weilin,WU Hongliang,et al.Current situation,problems and countermeasures of the well-logging evaluation technology for Gulong shale oil[J].Petroleum Geology & Oil‐ field Development in Daqing,2020,39(3):117-128.

    • [4] 崔宝文,张顺,付秀丽,等.松辽盆地古龙页岩有机层序地层划分及影响因素[J].大庆石油地质与开发,2021,40(5):13-28.CUI Baowen,ZHANG Shun,FU Xiuli,et al.Organic sequence stratigraphic division and its influencing factors’analyses for Gulong shale in Songliao Basin[J].Petroleum Geology & Oil‐ field Development in Daqing,2021,40(5):13-28.

    • [5] 陈欢庆,石成方,胡海燕,等.高含水油田精细油藏描述研究进展[J].石油与天然气地质,2018,39(6):1 311-1 322.CHEN Huanqing,SHI Chengfang,HU Haiyan,et al.Advances in fine description of reservoir in high water-cut oilfield[J].Oil & Gas Geology,2018,39(6):1 311-1 322.

    • [6] 陈维涛,徐少华,孙珍,等.层序地层学标准化理论在陆架坡折-陆坡区的应用——以珠江口盆地中中新世沉积层序为例 [J].石油与天然气地质,2021,42(6):1 414-1 422.CHEN Weitao,XU Shaohua,SUN Zhen,et al.Application of standardized sequence stratigraphy theory in the shelf break-toslope area:a case study of the Middle Miocene sedimentary se‐ quence in the PRMB[J].Oil & Gas Geology,2021,42(6):1 414-1 422.

    • [7] 刘国强.非常规油气时代的测井采集技术挑战与对策[J].中国石油勘探,2021,26(5):24-37.LIU Guoqiang.Challenges and countermeasures of well logging data acquisition technology in unconventional petroleum explo‐ ration and development[J].China Petroleum Exploration,2021,26(5):24-37.

    • [8] 王龙,张金川,唐玄.鄂尔多斯盆地下寺湾—云岩地区长7段页岩气测井评价与分布规律研究[J].中国石油勘探,2019,24(1):129-136.WANG Long,ZHANG Jinchuan,TANG Xuan.Logging evalua‐ tion and distribution law of Chang7 shale gas in Xiasiwan-Yun‐ yan area,Ordos Basin[J].China Petroleum Exploration,2019,24(1):129-136.

    • [9] MUKHERJEE B,SRIVARDHAN V,ROY P N S.Identification of formation interfaces by using wavelet and Fourier transforms [J].Journal of Applied Geophysics,2016,128:140-149.

    • [10] 肖波,韩学辉,周开金,等.测井曲线自动分层方法回顾与展望 [J].地球物理学进展,2010,25(5):1 802-1 810.XIAO Bo,HAN Xuehui,ZHOU Kaijin,et al.A review and out‐ look of automatic zonation methods of well log[J].Progress in Geophysics,2010,25(5):1 802-1 810.

    • [11] SARG J F,MARKELLO J R,WEBER L J,et al.Carbonate se‐ quence stratigraphy-a summary and perspective with case histo‐ ry,Neogene,Papua New Guinea[C].Jakarta:Proceedings of the International Symposium on Sequence Stratigraphy in SE Asia,1996:137-179.

    • [12] ZHANG Benshu,ZHAO Zhui,HAN Chaoyang,et al.Research summary of sequence stratigraphy in salt-lake basin[J].North‐ western Geology,2005,38(1):94-99.

    • [13] 张文龙,李畅,王帅,等.地层智能划分与对比研究进展及发展趋势探讨[J].信息系统工程,2021,(5):132-135,138.ZHANG Wenlong,LI Chang,WANG Shuai,et al.Research prog‐ ress and development trend of intelligent strata division and cor‐ relation[J].China CIO News,2021,(5):132-135,138.

    • [14] 李峰峰,郭睿,余义常.层序地层划分方法进展及展望[J].地质科技通报,2019,38(4):215-224.LI Fengfeng,GUO Rui,YU Yichang.Progress and prospect of the division sequence stratigraphy[J].Bulletin of Geological Sci‐ ence and Technology,2019,38(4):215-224.

    • [15] 刘腾,王军,张京思,等.地震Wheeler域变换结合时频分析技术用于渤海油田岩性油气藏描述[J].岩性油气藏,2018,30(3):124-132.LIU Teng,WANG Jun,ZHANG Jingsi,et al.Application of Wheeler transform combined with time-frequency analysis tech‐ nology to lithologic reservoir characterization of Bohai Oilfield [J].Lithologic Reservoirs,2018,30(3):124-132.

    • [16] HAWKINS D M,MERRIAM D F.Optimal zonation of digitized sequential data[J].Journal of the International Association for Mathematical Geology,1973,5(4):389-395.

    • [17] 李广场,李江林.有序聚类分析在声波测井自动分层中的应用 [J].工程勘察,2007,217(8):73-75.LI Guangchang,LI Jianglin.Application of ordered clustering analysis in automatic stratification of sonic logging[J].Geotech‐ nical Investigation & Surveying,2007,217(8):73-75.

    • [18] 谢忠怀,李保利.测井地层自动划分的边缘检测最优分割方法 [J].新疆石油地质,2006,27(2):173-175.XIE Zhonghuai,LI Baoli.Edge detection optimum partitioning method for automatic division of log strata[J].Xinjiang Petro‐ leum Geology,2006,27(2):173-175.

    • [19] 罗德江,汪兴旺,刘伟.基于局部投影和边缘检测的自动分层模型[J].物探与化探,2009,33(1):99-101.LUO Dejiang,WANG Xingwang,LIU Wei.Automatic layer sep‐ aration model based on the local projection and edge detection methods[J].Geophysical and Geochemical Exploration,2009,33(1):99-101.

    • [20] 罗德江,汪兴旺,刘伟.基于相空间重构和改进最优分割法的自动分层模型[J].地球物理学进展,2009,24(2):675-679.LUO Dejiang,WANG Xingwang,LIU Wei.Automatic separate layer model based on the phase-space reconstruction and im‐ proved optimum division method[J].Progress in Geophysics,2009,24(2):675-679.

    • [21] PARTOVI S M A,SADEGHNEJAD S.Geological boundary de‐ tection from well-logs:an efficient approach based on pattern recognition[J].Journal of Petroleum Science and Engineering,2019,176:444-455.

    • [22] 张明玉.极值方差聚类法在测井分层取值中的应用——以莫北油田莫005井区为例[J].新疆石油地质,2002,23(5):429-431.ZHANG Mingyu.Application of minimax variance cluster analy‐ sis method in well log layering data record-an example of well block Mo-005 in Mobei Oilfield[J].Xinjiang Petroleum Geolo‐ gy,2002,23(5):429-431.

    • [23] 程玉群,李秀荣,刘铁桩.测井曲线自动分层技术及在杏北地区小层划分中的应用[J].断块油气田,2000,7(2):23-25.CHENG Yuqun,LI Xiurong,LIU Tiezhuang.Technique of autozonation with log and its application to reservoir zonation at Xingbei area[J].Fault-Block Oil and Gas Field,2000,7(2):23-25.

    • [24] 初颖,吕堂红.基于极值法和聚类分析法的测井曲线自动分层模型——以山东省胜利油井为例[J].长春理工大学学报:自然科学版,2017,40(6):105-110.CHU Ying,LÜ Tanghong.Logging-curve automatic layered model based on the extreme value method and clustering analy‐ sis-taking Shandong Province,Shengli Oilfield as an example[J].Journal of Changchun University of Science and Technology:Natural Science Edition,2017,40(6):105-110.

    • [25] 郝建华,雷颖,王佳音,等.测井曲线活度法在地层划分中的应用[J].中国石油和化工标准与质量,2012,32(6):170-171.HAO Jianhua,LEI Ying,WANG Jiayin,et al.Application of ac‐ tivity method of logging curve in stratigraphic division[J].China Petroleum and Chemical Standard and Quality,2012,32(6):170-171.

    • [26] 朱剑兵.基于最大熵谱估计的沉积旋回划分方法探讨[J].岩性油气藏,2012,24(2):67-71.ZHU Jianbing.Division of sedimentary cycle sequences based on maximum entropy spectrum estimation[J].Lithologic Reser‐ voirs,2012,24(2):67-71.

    • [27] 袁野,王黎,谢锐杰.INPEFA 技术在层序地层划分中的应用:以苏北盆地溱潼凹陷南华区块阜三段为例[J].石油实验地质,2018,40(6):871-876.YUAN Ye,WANG Li,XIE Ruijie.Application of INPEFA tech‐ nology to sequence stratigraphy of the third member of Funing Formation,Nanhua block,Qintong Sag,North Jiangsu Basin[J].Petroleum Geology and Experiment,2018,40(6):871-876.

    • [28] BRACEWELL R N.The Fourier Transform and its Applications [M].New York:McGraw-Hill,1986.

    • [29] DAUBECHIES I.The wavelet transform,time-frequency local‐ ization and signal analysis[J].IEEE transactions on Information Theory,1990,36(5):961-1 005.

    • [30] MCDOUGAL W S,WEIN A J,KAVOUSSI L R,et al.Campbellwalsh urology 11th edition review e-Book[M].New York:Elsevi‐ er Health Sciences,2015.

    • [31] 阎辉,李鲲鹏,张学工,等.测井曲线的小波变换特性在自动分层中的应用[J].地球物理学报,2000,43(4):568-573.YAN Hui,LI Kunpeng,ZHANG Xuegong,et al.Wavelet trans‐ form properties of well log and their application in automatic segmentation[J].Chinese Journal of Geophysics,2000,43(4):568-573.

    • [32] 薛波,杨青,张超虹.基于形态学滤波与小波变换的测井曲线自动分层方法[J].地球物理学进展,2020,35(1):203-210.XUE Bo,YANG Qing,ZHANG Chaohong.Automatic stratifica‐ tion method of logging curve based on morphological filtering and wavelet transform[J].Progress in Geophysics,2020,35(1):203-210.

    • [33] LI Xia,FAN Yiren,DENG Shaogui,et al.Automatic demarca‐ tion of sequence stratigraphy using the method of well logging multiscale data fusion[J].Petroleum Exploration and Develop‐ ment,2009,36(2):221-227.

    • [34] 杨建民,黄保纲,刘卫林,等.基于小波变换的浅水三角洲高分辨率层序地层定量划分[J].长江大学学报:自然科学版,2019,16(12):20-24.YANG Jianmin,HUANG Baogang,LIU Weilin,et al.Quantita‐ tive division of high-resolution sequence stratigraphy in shallow water delta based on wavelet transform[J].Journal of Yangtze University:Natural Science Edition,2019,16(12):20-24.

    • [35] LIU Hui,JIANG Zaixing.Wavelet transform of geophysical well logging signal and its application to sequence division[C].Yan‐ tai:2010 3rd International Congress on Image and Signal Pro‐ cessing,2010,7:3 470-3 473.

    • [36] ZHANG Jinliang,SONG Aixue.Application of wavelet analysis in sequence stratigraphic division of glutenite sediments[C].Washington:2010 International Conference on Challenges in En‐ vironmental Science and Computer Engineering,2010,2:204-207.

    • [37] KADKHODAIE A,REZAEE R.Intelligent sequence stratigra‐ phy through a wavelet-based decomposition of well log data[J].Journal of Natural Gas Science and Engineering,2017,40:38-50.

    • [38] PAN S Y,HSIEH B Z,LU M T,et al.Identification of strati‐ graphic formation interfaces using wavelet and Fourier trans‐ forms[J].Computers & Geosciences,2008,34(1):77-92.

    • [39] MUKHERJEE B,SRIVARDHAN V,ROY P N S.Identification of formation interfaces by using wavelet and Fourier transforms [J].Journal of Applied Geophysics,2016,128:140-149.

    • [40] 周志华.机器学习[M].北京:清华大学出版社,2016.ZHOU Zhihua.Machine learning[M].Beijing:Tsinghua Universi‐ ty Press,2016.

    • [41] CHEN P H,LIN C J,SCHÖLKOPF B,et al.A tutorial on v-sup‐ port vector machines[J].Applied Stochastic Models in Business and Industry,2005,21(2):111-136.

    • [42] IQBAL M R A,RAHMAN S,NABIL S I,et al.Knowledge based decision tree construction with feature importance domain knowledge[C].Yantai:2012 7th International Conference on Electrical and Computer Engineering,2012:659-662.

    • [43] BELGIU M,DRĂGUŢL.Random forest in remote sensing:a re‐ view of applications and future directions[J].ISPRS Journal of Photogrammetry and Remote Sensing,2016,114:24-31.

    • [44] 赖强,魏伯阳,吴煜宇,等.基于随机森林的K-近邻算法划分火成岩岩性[J].特种油气藏,2021,28(6):62-69.LAI Qiang,WEI Boyang,WU Yuyu,et al.Classification of igne‐ ous rock lithology with K-nearest neighbor algorithm based on random forest(RF-KNN)[J].Special Oil & Gas Reservoirs,2021,28(6):62-69.

    • [45] ABIODUN O I,JANTAN A,OMOLARA A E,et al.State-of-theart in artificial neural network applications:a survey[J].Heliyon,2018,4(11):e00938.

    • [46] 牟丹,王祝文,黄玉龙,等.基于最小二乘支持向量机测井识别火山岩类型:以辽河盆地中基性火山岩为例[J].吉林大学学报:地球科学版,2015,45(2):639-648.MOU Dan,WANG Zhuwen,HUANG Yulong,et al.Application of least squares support vector machine to lithology identifica‐ tion:taking intermediate/basaltic rocks of Liaohe Basin as an ex‐ ample[J].Journal of Jilin University:Earth Science Edition,2015,45(2):639-648.

    • [47] 沈凯浪.基于测井曲线的地层划分和测井相分析[D].西安:西安科技大学,2015.SHEN Kailang.Stratigraphic division and sedimentary facies analysis based on logging curves[D].Xi'an:Xi'an University of Science and Technology,2015.

    • [48] 尚福华,李金成,原野,等.基于改进BP神经网络的地层划分方法[J].计算机技术与发展,2020,30(9):148-153.SHANG Fuhua,LI Jincheng,YUAN Ye,et al.Stratigraphic divi‐ sion method based on improved BP neural network[J].Computer Technology and Development,2020,30(9):148-153.

    • [49] 李金成.基于涡流优化算法的测井曲线地层对比方法研究[D].大庆:东北石油大学,2020.LI Jincheng.Research on the stratigraphic correlation method of logging curve based on vortex search algorithm[D].Daqing:Northeast Petroleum University,2020.

    • [50] ELKATATNY S,AL-ABDULJABBAR A,MAHMOUD A A.New robust model to estimate formation tops in real time using artificial neural networks(ANN)[J].Petrophysics:The SPWLA Journal of Formation Evaluation and Reservoir Description,2019,60(6):825-837.

    • [51] 徐朝晖,刘钰铭,周新茂,等.基于卷积神经网络算法的自动地层对比实验[J].石油科学通报,2019,4(1):1-10.XU Zhaohui,LIU Yuming,ZHOU Xinmao,et al.An experiment in automatic stratigraphic correlation using convolutional neural networks[J].Petroleum Science Bulletin,2019,4(1):1-10.

    • [52] ZHANG Jingya,LI Guojun,ZHANG Juan,et al.A convolutional neural network approach for stratigraphic interface detection[C].Malmo:International Conference on Internet of Things as a Ser‐ vice,2020,346:529-542.

    • [53] MIKOLOV T,KARAFIÁT M,BURGET L,et al.Recurrent neu‐ ral network-based language model[C].Chiba,Japan:11th Annual Conference of the International Speech Communication Associa‐ tion,2010:1 045-148.

    • [54] FREUND Y,SCHAPIRE R E.Experiments with a new boosting algorithm[C].San Francisco:Proceedings of the Thirteenth Inter‐ national Conference on Machine Learning,1996,96:148-156.

    • [55] ZHOU Cuiying,OUYANG Jinwu,MING Weihua,et al.A strati‐graphic prediction method based on machine learning[J].Applied Sciences,2019,9(17):3 553.

    • [56] 马陇飞,萧汉敏,陶敬伟,等.基于梯度提升决策树算法的岩性智能分类方法[J].油气地质与采收率,2022,29(1):21-29.MA Longfei,XIAO Hanmin,TAO Jingwei,et al.Intelligent li‐ thology classification method based on GBDT algorithm[J].Pe‐ troleum Geology and Recovery Efficiency,2022,29(1):21-29.

    • [57] 徐鹏晔.基于数据驱动和循环滑动时窗的小层智能划分方法 [J].油气地质与采收率,2022,29(1):113-120.XU Pengye.Small layer intelligent division method based on da‐ ta-driven and cyclic sliding time window[J].Petroleum Geology and Recovery Efficiency,2022,29(1):113-120.

    • [58] MEREMBAYEV T,YUNUSSOV R,YEDILKHAN A.Machine learning algorithms for classification geology data from well log‐ ging[C].Kaskelen:2018 14th International Conference on Elec‐ tronics Computer and Computation,2018:206-212.

    • [59] GONG Ke,YE Zhihui,CHEN Dong,et al.Investigation on auto‐ matic recognition of stratigraphic lithology based on well log‐ ging data using ensemble learning algorithm[C].Brisbane:SPE Asia Pacific Oil and Gas Conference and Exhibition,2018.

    • [60] HARTIGAN J A,WONG M A.Algorithm AS136:a k-means clustering algorithm[J].Journal of the Rroyal Statistical Society:Series C,1979,28(1):100-108.

    • [61] HAHSLER M,PIEKENBROCK M,DORAN D.Dbscan:fast density-based clustering with R[J].Journal of Statistical Soft‐ ware,2019,91(1):1-30.

    • [62] RINGNÉR M.What is principal component analysis?[J].Nature Biotechnology,2008,26(3):303-304.

    • [63] KOHONEN T.Self-organizing maps:ophmization approaches//Artificial Neural Networks[M].UK:North-Holland,1991:981-990.

    • [64] 阎辉,张学工,李衍达.应用自组织神经网络进行自动地层划分的设计与实现[J].模式识别与人工智能,2000,13(3):309-313.YAN Hui,ZHANG Xuegong,LI Yanda.The application of selforganizing neural network in automatic stratum segmentation[J].Pattern Recognition and Artificial Intelligence,2000,13(3):309-313.

    • [65] 冯向东,张玉琴,韩红伟,等.基于自组织神经网络在油气分层中的研究[J].计算机技术与发展,2021,31(2):44-48.FENG Xiangdong,ZHANG Yuqin,HAN Hongwei,et al.Re‐ search on oil and gas stratification based on self-organizing neu‐ ral network[J].Computer Technology and Development,2021,31(2):44-48.

    • [66] 张多,韩逢庆.基于支持向量机和有序聚类的岩层识别[J].智能系统学报,2014,9(1):98-103.ZHANG Duo,HAN Fengqing.Stratum identification based on the SVM and ordered cluster[J].CAAI Transactions on Intelli‐ gent Systems,2014,9(1):98-103.

    • [67] HONG Y,KANG C.Automatic well correlation by aligning mul‐ tiple well using deep neural networks[C].Abu Dhabi:Abu Dhabi International Petroleum Exhibition & Conference,2020,SPE202908-MS.

    • [68] KARIMI A M,SADEGHNEJAD S,REZGHI M.Well-to-well correlation and identifying lithological boundaries by principal component analysis of well-logs[J].Computers & Geosciences,2021,157:104-942.

    • [69] SHEN Yating,WANG Yunyun,MA Zhiguo.Label-expanded manifold regularization for semi-supervised classification[C].Nanjing:2017 12th International Conference on Intelligent Sys‐ tems and Knowledge Engineering,2017:1-4.

    • [70] ZHAO Huimin,ZHENG Jianjie,DENG Wu,et al.Semi-super‐ vised broad learning system based on manifold regularization and broad network[J].IEEE Transactions on Circuits and Sys‐ tems I:Regular Papers,2020,67(3):983-994.

    • [71] GHORBAN F,HASAN N,VELTEN J,et al.Improving FMGAN through mixup manifold regularization[C].Daegu:2021 IEEE International Symposium on Circuits and Systems,2021:1-5.

    • [72] YIN Xucheng,LIU Qian,HAO Hongwei,et al.FMI image based rock structure classification using classifier combination[J].Neu‐ ral Computing and Applications,2011,20(7):955-963.

    • [73] ZHANG Wei,WU Tong,LI Zhipeng,et al.Automatic detection of fractures based on optimal path search in well logging images [J].Journal of Sensors,2021,(4):1-10.

    • [74] CHANG Ji,LI Jing,KANG Yu,et al.SegLog:geophysical log‐ ging segmentation network for lithofacies identification[J].IEEE Transactions on Industrial Informatics,2021,18(9):6 089-6 099.

    • [75] FAN Pengpeng,DENG Rui,QIU Jinquan,et al.Well logging curve reconstruction based on kernel ridge regression[J].Arabian Journal of Geosciences,2021,14(16):1 559.

    • [76] MO Xuwen,ZHANG Qiang,LI Xiao.Well logging curve recon‐ struction based on genetic neural networks[C].Zhangjiajie:2015 12th International Conference on Fuzzy Systems and Knowledge Discovery,2015:1 015-1 021.

    • [77] SHU Jun,XIE Qi,YI Lixuan,et al.Meta-weight-net:learning an explicit mapping for sample weighting[C].Vancouver:33rd Con‐ ference on Neural Information Processing Systems,2019,172:1 919-1 930.

    • [78] ZHANG Zizhao,PFISTER T.Learning fast sample re-weighting without reward data[C].Montreal:Proceedings of the IEEE/CVF International Conference on Computer Vision,2021:725-734.

    • [79] CAI Qi,PAN Yingwei,WANG Yu,et al.Learning a unified sam‐ ple weighting network for object detection[C].Seattle:Proceed‐ ings of the IEEE/CVF Conference on Computer Vision and Pat‐ tern Recognition,2020:14 173-14 182.

    • [80] MENG Wei,TIA Meng.Unmanned aerial vehicle classification and detection based on deep transfer learning[C].Sanya:2020 In‐ ternational Conference on Intelligent Computing and HumanComputer Interaction,2020:280-285.

    • [81] LI Yunsheng,YUAN Lu,CHEN Yinpeng,et al.Dynamic transfer for multi-source domain adaptation[C].New Orleans:Proceed‐ ings of the IEEE/CVF Conference on Computer Vision and Pat‐ tern Recognition,2021:10 998-11 007.

  • 参考文献

    • [1] 董庆民,胡忠贵,蔡家兰,等.层序地层与同位素地球化学响应特征及礁滩发育模式——以川东城口—鄂西海槽边缘带长兴组为例[J].石油实验地质,2019,41(2):243-250.DONG Qingmin,HU Zhonggui,CAI Jialan,et al.Sequence stra‐ tigraphy and isotope geochemical response and development pat‐ tern of reef and shoal:a case study of Changxing Formation in the marginal zone of eastern Sichuan Chengkou-western Hubei oceanic trough[J].Petroleum Geology and Experiment,2019,41(2):243-250.

    • [2] 刘艳红,罗兴旺,张武,等.致密砂岩油藏水淹层测井识别方法及影响因素研究[J].特种油气藏,2022,29(3):57-63.LIU Yanhong,LUO Xingwang,ZHANG Wu,et al.Study on log‐ ging identification method and influencing factors of waterflooded zone in tight sandstone reservoir[J].Special Oil & Gas Reservoirs,2022,29(3):57-63.

    • [3] 李宁,闫伟林,武宏亮,等.松辽盆地古龙页岩油测井评价技术现状、问题及对策[J].大庆石油地质与开发,2020,39(3):117-128.LI Ning,YAN Weilin,WU Hongliang,et al.Current situation,problems and countermeasures of the well-logging evaluation technology for Gulong shale oil[J].Petroleum Geology & Oil‐ field Development in Daqing,2020,39(3):117-128.

    • [4] 崔宝文,张顺,付秀丽,等.松辽盆地古龙页岩有机层序地层划分及影响因素[J].大庆石油地质与开发,2021,40(5):13-28.CUI Baowen,ZHANG Shun,FU Xiuli,et al.Organic sequence stratigraphic division and its influencing factors’analyses for Gulong shale in Songliao Basin[J].Petroleum Geology & Oil‐ field Development in Daqing,2021,40(5):13-28.

    • [5] 陈欢庆,石成方,胡海燕,等.高含水油田精细油藏描述研究进展[J].石油与天然气地质,2018,39(6):1 311-1 322.CHEN Huanqing,SHI Chengfang,HU Haiyan,et al.Advances in fine description of reservoir in high water-cut oilfield[J].Oil & Gas Geology,2018,39(6):1 311-1 322.

    • [6] 陈维涛,徐少华,孙珍,等.层序地层学标准化理论在陆架坡折-陆坡区的应用——以珠江口盆地中中新世沉积层序为例 [J].石油与天然气地质,2021,42(6):1 414-1 422.CHEN Weitao,XU Shaohua,SUN Zhen,et al.Application of standardized sequence stratigraphy theory in the shelf break-toslope area:a case study of the Middle Miocene sedimentary se‐ quence in the PRMB[J].Oil & Gas Geology,2021,42(6):1 414-1 422.

    • [7] 刘国强.非常规油气时代的测井采集技术挑战与对策[J].中国石油勘探,2021,26(5):24-37.LIU Guoqiang.Challenges and countermeasures of well logging data acquisition technology in unconventional petroleum explo‐ ration and development[J].China Petroleum Exploration,2021,26(5):24-37.

    • [8] 王龙,张金川,唐玄.鄂尔多斯盆地下寺湾—云岩地区长7段页岩气测井评价与分布规律研究[J].中国石油勘探,2019,24(1):129-136.WANG Long,ZHANG Jinchuan,TANG Xuan.Logging evalua‐ tion and distribution law of Chang7 shale gas in Xiasiwan-Yun‐ yan area,Ordos Basin[J].China Petroleum Exploration,2019,24(1):129-136.

    • [9] MUKHERJEE B,SRIVARDHAN V,ROY P N S.Identification of formation interfaces by using wavelet and Fourier transforms [J].Journal of Applied Geophysics,2016,128:140-149.

    • [10] 肖波,韩学辉,周开金,等.测井曲线自动分层方法回顾与展望 [J].地球物理学进展,2010,25(5):1 802-1 810.XIAO Bo,HAN Xuehui,ZHOU Kaijin,et al.A review and out‐ look of automatic zonation methods of well log[J].Progress in Geophysics,2010,25(5):1 802-1 810.

    • [11] SARG J F,MARKELLO J R,WEBER L J,et al.Carbonate se‐ quence stratigraphy-a summary and perspective with case histo‐ ry,Neogene,Papua New Guinea[C].Jakarta:Proceedings of the International Symposium on Sequence Stratigraphy in SE Asia,1996:137-179.

    • [12] ZHANG Benshu,ZHAO Zhui,HAN Chaoyang,et al.Research summary of sequence stratigraphy in salt-lake basin[J].North‐ western Geology,2005,38(1):94-99.

    • [13] 张文龙,李畅,王帅,等.地层智能划分与对比研究进展及发展趋势探讨[J].信息系统工程,2021,(5):132-135,138.ZHANG Wenlong,LI Chang,WANG Shuai,et al.Research prog‐ ress and development trend of intelligent strata division and cor‐ relation[J].China CIO News,2021,(5):132-135,138.

    • [14] 李峰峰,郭睿,余义常.层序地层划分方法进展及展望[J].地质科技通报,2019,38(4):215-224.LI Fengfeng,GUO Rui,YU Yichang.Progress and prospect of the division sequence stratigraphy[J].Bulletin of Geological Sci‐ ence and Technology,2019,38(4):215-224.

    • [15] 刘腾,王军,张京思,等.地震Wheeler域变换结合时频分析技术用于渤海油田岩性油气藏描述[J].岩性油气藏,2018,30(3):124-132.LIU Teng,WANG Jun,ZHANG Jingsi,et al.Application of Wheeler transform combined with time-frequency analysis tech‐ nology to lithologic reservoir characterization of Bohai Oilfield [J].Lithologic Reservoirs,2018,30(3):124-132.

    • [16] HAWKINS D M,MERRIAM D F.Optimal zonation of digitized sequential data[J].Journal of the International Association for Mathematical Geology,1973,5(4):389-395.

    • [17] 李广场,李江林.有序聚类分析在声波测井自动分层中的应用 [J].工程勘察,2007,217(8):73-75.LI Guangchang,LI Jianglin.Application of ordered clustering analysis in automatic stratification of sonic logging[J].Geotech‐ nical Investigation & Surveying,2007,217(8):73-75.

    • [18] 谢忠怀,李保利.测井地层自动划分的边缘检测最优分割方法 [J].新疆石油地质,2006,27(2):173-175.XIE Zhonghuai,LI Baoli.Edge detection optimum partitioning method for automatic division of log strata[J].Xinjiang Petro‐ leum Geology,2006,27(2):173-175.

    • [19] 罗德江,汪兴旺,刘伟.基于局部投影和边缘检测的自动分层模型[J].物探与化探,2009,33(1):99-101.LUO Dejiang,WANG Xingwang,LIU Wei.Automatic layer sep‐ aration model based on the local projection and edge detection methods[J].Geophysical and Geochemical Exploration,2009,33(1):99-101.

    • [20] 罗德江,汪兴旺,刘伟.基于相空间重构和改进最优分割法的自动分层模型[J].地球物理学进展,2009,24(2):675-679.LUO Dejiang,WANG Xingwang,LIU Wei.Automatic separate layer model based on the phase-space reconstruction and im‐ proved optimum division method[J].Progress in Geophysics,2009,24(2):675-679.

    • [21] PARTOVI S M A,SADEGHNEJAD S.Geological boundary de‐ tection from well-logs:an efficient approach based on pattern recognition[J].Journal of Petroleum Science and Engineering,2019,176:444-455.

    • [22] 张明玉.极值方差聚类法在测井分层取值中的应用——以莫北油田莫005井区为例[J].新疆石油地质,2002,23(5):429-431.ZHANG Mingyu.Application of minimax variance cluster analy‐ sis method in well log layering data record-an example of well block Mo-005 in Mobei Oilfield[J].Xinjiang Petroleum Geolo‐ gy,2002,23(5):429-431.

    • [23] 程玉群,李秀荣,刘铁桩.测井曲线自动分层技术及在杏北地区小层划分中的应用[J].断块油气田,2000,7(2):23-25.CHENG Yuqun,LI Xiurong,LIU Tiezhuang.Technique of autozonation with log and its application to reservoir zonation at Xingbei area[J].Fault-Block Oil and Gas Field,2000,7(2):23-25.

    • [24] 初颖,吕堂红.基于极值法和聚类分析法的测井曲线自动分层模型——以山东省胜利油井为例[J].长春理工大学学报:自然科学版,2017,40(6):105-110.CHU Ying,LÜ Tanghong.Logging-curve automatic layered model based on the extreme value method and clustering analy‐ sis-taking Shandong Province,Shengli Oilfield as an example[J].Journal of Changchun University of Science and Technology:Natural Science Edition,2017,40(6):105-110.

    • [25] 郝建华,雷颖,王佳音,等.测井曲线活度法在地层划分中的应用[J].中国石油和化工标准与质量,2012,32(6):170-171.HAO Jianhua,LEI Ying,WANG Jiayin,et al.Application of ac‐ tivity method of logging curve in stratigraphic division[J].China Petroleum and Chemical Standard and Quality,2012,32(6):170-171.

    • [26] 朱剑兵.基于最大熵谱估计的沉积旋回划分方法探讨[J].岩性油气藏,2012,24(2):67-71.ZHU Jianbing.Division of sedimentary cycle sequences based on maximum entropy spectrum estimation[J].Lithologic Reser‐ voirs,2012,24(2):67-71.

    • [27] 袁野,王黎,谢锐杰.INPEFA 技术在层序地层划分中的应用:以苏北盆地溱潼凹陷南华区块阜三段为例[J].石油实验地质,2018,40(6):871-876.YUAN Ye,WANG Li,XIE Ruijie.Application of INPEFA tech‐ nology to sequence stratigraphy of the third member of Funing Formation,Nanhua block,Qintong Sag,North Jiangsu Basin[J].Petroleum Geology and Experiment,2018,40(6):871-876.

    • [28] BRACEWELL R N.The Fourier Transform and its Applications [M].New York:McGraw-Hill,1986.

    • [29] DAUBECHIES I.The wavelet transform,time-frequency local‐ ization and signal analysis[J].IEEE transactions on Information Theory,1990,36(5):961-1 005.

    • [30] MCDOUGAL W S,WEIN A J,KAVOUSSI L R,et al.Campbellwalsh urology 11th edition review e-Book[M].New York:Elsevi‐ er Health Sciences,2015.

    • [31] 阎辉,李鲲鹏,张学工,等.测井曲线的小波变换特性在自动分层中的应用[J].地球物理学报,2000,43(4):568-573.YAN Hui,LI Kunpeng,ZHANG Xuegong,et al.Wavelet trans‐ form properties of well log and their application in automatic segmentation[J].Chinese Journal of Geophysics,2000,43(4):568-573.

    • [32] 薛波,杨青,张超虹.基于形态学滤波与小波变换的测井曲线自动分层方法[J].地球物理学进展,2020,35(1):203-210.XUE Bo,YANG Qing,ZHANG Chaohong.Automatic stratifica‐ tion method of logging curve based on morphological filtering and wavelet transform[J].Progress in Geophysics,2020,35(1):203-210.

    • [33] LI Xia,FAN Yiren,DENG Shaogui,et al.Automatic demarca‐ tion of sequence stratigraphy using the method of well logging multiscale data fusion[J].Petroleum Exploration and Develop‐ ment,2009,36(2):221-227.

    • [34] 杨建民,黄保纲,刘卫林,等.基于小波变换的浅水三角洲高分辨率层序地层定量划分[J].长江大学学报:自然科学版,2019,16(12):20-24.YANG Jianmin,HUANG Baogang,LIU Weilin,et al.Quantita‐ tive division of high-resolution sequence stratigraphy in shallow water delta based on wavelet transform[J].Journal of Yangtze University:Natural Science Edition,2019,16(12):20-24.

    • [35] LIU Hui,JIANG Zaixing.Wavelet transform of geophysical well logging signal and its application to sequence division[C].Yan‐ tai:2010 3rd International Congress on Image and Signal Pro‐ cessing,2010,7:3 470-3 473.

    • [36] ZHANG Jinliang,SONG Aixue.Application of wavelet analysis in sequence stratigraphic division of glutenite sediments[C].Washington:2010 International Conference on Challenges in En‐ vironmental Science and Computer Engineering,2010,2:204-207.

    • [37] KADKHODAIE A,REZAEE R.Intelligent sequence stratigra‐ phy through a wavelet-based decomposition of well log data[J].Journal of Natural Gas Science and Engineering,2017,40:38-50.

    • [38] PAN S Y,HSIEH B Z,LU M T,et al.Identification of strati‐ graphic formation interfaces using wavelet and Fourier trans‐ forms[J].Computers & Geosciences,2008,34(1):77-92.

    • [39] MUKHERJEE B,SRIVARDHAN V,ROY P N S.Identification of formation interfaces by using wavelet and Fourier transforms [J].Journal of Applied Geophysics,2016,128:140-149.

    • [40] 周志华.机器学习[M].北京:清华大学出版社,2016.ZHOU Zhihua.Machine learning[M].Beijing:Tsinghua Universi‐ ty Press,2016.

    • [41] CHEN P H,LIN C J,SCHÖLKOPF B,et al.A tutorial on v-sup‐ port vector machines[J].Applied Stochastic Models in Business and Industry,2005,21(2):111-136.

    • [42] IQBAL M R A,RAHMAN S,NABIL S I,et al.Knowledge based decision tree construction with feature importance domain knowledge[C].Yantai:2012 7th International Conference on Electrical and Computer Engineering,2012:659-662.

    • [43] BELGIU M,DRĂGUŢL.Random forest in remote sensing:a re‐ view of applications and future directions[J].ISPRS Journal of Photogrammetry and Remote Sensing,2016,114:24-31.

    • [44] 赖强,魏伯阳,吴煜宇,等.基于随机森林的K-近邻算法划分火成岩岩性[J].特种油气藏,2021,28(6):62-69.LAI Qiang,WEI Boyang,WU Yuyu,et al.Classification of igne‐ ous rock lithology with K-nearest neighbor algorithm based on random forest(RF-KNN)[J].Special Oil & Gas Reservoirs,2021,28(6):62-69.

    • [45] ABIODUN O I,JANTAN A,OMOLARA A E,et al.State-of-theart in artificial neural network applications:a survey[J].Heliyon,2018,4(11):e00938.

    • [46] 牟丹,王祝文,黄玉龙,等.基于最小二乘支持向量机测井识别火山岩类型:以辽河盆地中基性火山岩为例[J].吉林大学学报:地球科学版,2015,45(2):639-648.MOU Dan,WANG Zhuwen,HUANG Yulong,et al.Application of least squares support vector machine to lithology identifica‐ tion:taking intermediate/basaltic rocks of Liaohe Basin as an ex‐ ample[J].Journal of Jilin University:Earth Science Edition,2015,45(2):639-648.

    • [47] 沈凯浪.基于测井曲线的地层划分和测井相分析[D].西安:西安科技大学,2015.SHEN Kailang.Stratigraphic division and sedimentary facies analysis based on logging curves[D].Xi'an:Xi'an University of Science and Technology,2015.

    • [48] 尚福华,李金成,原野,等.基于改进BP神经网络的地层划分方法[J].计算机技术与发展,2020,30(9):148-153.SHANG Fuhua,LI Jincheng,YUAN Ye,et al.Stratigraphic divi‐ sion method based on improved BP neural network[J].Computer Technology and Development,2020,30(9):148-153.

    • [49] 李金成.基于涡流优化算法的测井曲线地层对比方法研究[D].大庆:东北石油大学,2020.LI Jincheng.Research on the stratigraphic correlation method of logging curve based on vortex search algorithm[D].Daqing:Northeast Petroleum University,2020.

    • [50] ELKATATNY S,AL-ABDULJABBAR A,MAHMOUD A A.New robust model to estimate formation tops in real time using artificial neural networks(ANN)[J].Petrophysics:The SPWLA Journal of Formation Evaluation and Reservoir Description,2019,60(6):825-837.

    • [51] 徐朝晖,刘钰铭,周新茂,等.基于卷积神经网络算法的自动地层对比实验[J].石油科学通报,2019,4(1):1-10.XU Zhaohui,LIU Yuming,ZHOU Xinmao,et al.An experiment in automatic stratigraphic correlation using convolutional neural networks[J].Petroleum Science Bulletin,2019,4(1):1-10.

    • [52] ZHANG Jingya,LI Guojun,ZHANG Juan,et al.A convolutional neural network approach for stratigraphic interface detection[C].Malmo:International Conference on Internet of Things as a Ser‐ vice,2020,346:529-542.

    • [53] MIKOLOV T,KARAFIÁT M,BURGET L,et al.Recurrent neu‐ ral network-based language model[C].Chiba,Japan:11th Annual Conference of the International Speech Communication Associa‐ tion,2010:1 045-148.

    • [54] FREUND Y,SCHAPIRE R E.Experiments with a new boosting algorithm[C].San Francisco:Proceedings of the Thirteenth Inter‐ national Conference on Machine Learning,1996,96:148-156.

    • [55] ZHOU Cuiying,OUYANG Jinwu,MING Weihua,et al.A strati‐graphic prediction method based on machine learning[J].Applied Sciences,2019,9(17):3 553.

    • [56] 马陇飞,萧汉敏,陶敬伟,等.基于梯度提升决策树算法的岩性智能分类方法[J].油气地质与采收率,2022,29(1):21-29.MA Longfei,XIAO Hanmin,TAO Jingwei,et al.Intelligent li‐ thology classification method based on GBDT algorithm[J].Pe‐ troleum Geology and Recovery Efficiency,2022,29(1):21-29.

    • [57] 徐鹏晔.基于数据驱动和循环滑动时窗的小层智能划分方法 [J].油气地质与采收率,2022,29(1):113-120.XU Pengye.Small layer intelligent division method based on da‐ ta-driven and cyclic sliding time window[J].Petroleum Geology and Recovery Efficiency,2022,29(1):113-120.

    • [58] MEREMBAYEV T,YUNUSSOV R,YEDILKHAN A.Machine learning algorithms for classification geology data from well log‐ ging[C].Kaskelen:2018 14th International Conference on Elec‐ tronics Computer and Computation,2018:206-212.

    • [59] GONG Ke,YE Zhihui,CHEN Dong,et al.Investigation on auto‐ matic recognition of stratigraphic lithology based on well log‐ ging data using ensemble learning algorithm[C].Brisbane:SPE Asia Pacific Oil and Gas Conference and Exhibition,2018.

    • [60] HARTIGAN J A,WONG M A.Algorithm AS136:a k-means clustering algorithm[J].Journal of the Rroyal Statistical Society:Series C,1979,28(1):100-108.

    • [61] HAHSLER M,PIEKENBROCK M,DORAN D.Dbscan:fast density-based clustering with R[J].Journal of Statistical Soft‐ ware,2019,91(1):1-30.

    • [62] RINGNÉR M.What is principal component analysis?[J].Nature Biotechnology,2008,26(3):303-304.

    • [63] KOHONEN T.Self-organizing maps:ophmization approaches//Artificial Neural Networks[M].UK:North-Holland,1991:981-990.

    • [64] 阎辉,张学工,李衍达.应用自组织神经网络进行自动地层划分的设计与实现[J].模式识别与人工智能,2000,13(3):309-313.YAN Hui,ZHANG Xuegong,LI Yanda.The application of selforganizing neural network in automatic stratum segmentation[J].Pattern Recognition and Artificial Intelligence,2000,13(3):309-313.

    • [65] 冯向东,张玉琴,韩红伟,等.基于自组织神经网络在油气分层中的研究[J].计算机技术与发展,2021,31(2):44-48.FENG Xiangdong,ZHANG Yuqin,HAN Hongwei,et al.Re‐ search on oil and gas stratification based on self-organizing neu‐ ral network[J].Computer Technology and Development,2021,31(2):44-48.

    • [66] 张多,韩逢庆.基于支持向量机和有序聚类的岩层识别[J].智能系统学报,2014,9(1):98-103.ZHANG Duo,HAN Fengqing.Stratum identification based on the SVM and ordered cluster[J].CAAI Transactions on Intelli‐ gent Systems,2014,9(1):98-103.

    • [67] HONG Y,KANG C.Automatic well correlation by aligning mul‐ tiple well using deep neural networks[C].Abu Dhabi:Abu Dhabi International Petroleum Exhibition & Conference,2020,SPE202908-MS.

    • [68] KARIMI A M,SADEGHNEJAD S,REZGHI M.Well-to-well correlation and identifying lithological boundaries by principal component analysis of well-logs[J].Computers & Geosciences,2021,157:104-942.

    • [69] SHEN Yating,WANG Yunyun,MA Zhiguo.Label-expanded manifold regularization for semi-supervised classification[C].Nanjing:2017 12th International Conference on Intelligent Sys‐ tems and Knowledge Engineering,2017:1-4.

    • [70] ZHAO Huimin,ZHENG Jianjie,DENG Wu,et al.Semi-super‐ vised broad learning system based on manifold regularization and broad network[J].IEEE Transactions on Circuits and Sys‐ tems I:Regular Papers,2020,67(3):983-994.

    • [71] GHORBAN F,HASAN N,VELTEN J,et al.Improving FMGAN through mixup manifold regularization[C].Daegu:2021 IEEE International Symposium on Circuits and Systems,2021:1-5.

    • [72] YIN Xucheng,LIU Qian,HAO Hongwei,et al.FMI image based rock structure classification using classifier combination[J].Neu‐ ral Computing and Applications,2011,20(7):955-963.

    • [73] ZHANG Wei,WU Tong,LI Zhipeng,et al.Automatic detection of fractures based on optimal path search in well logging images [J].Journal of Sensors,2021,(4):1-10.

    • [74] CHANG Ji,LI Jing,KANG Yu,et al.SegLog:geophysical log‐ ging segmentation network for lithofacies identification[J].IEEE Transactions on Industrial Informatics,2021,18(9):6 089-6 099.

    • [75] FAN Pengpeng,DENG Rui,QIU Jinquan,et al.Well logging curve reconstruction based on kernel ridge regression[J].Arabian Journal of Geosciences,2021,14(16):1 559.

    • [76] MO Xuwen,ZHANG Qiang,LI Xiao.Well logging curve recon‐ struction based on genetic neural networks[C].Zhangjiajie:2015 12th International Conference on Fuzzy Systems and Knowledge Discovery,2015:1 015-1 021.

    • [77] SHU Jun,XIE Qi,YI Lixuan,et al.Meta-weight-net:learning an explicit mapping for sample weighting[C].Vancouver:33rd Con‐ ference on Neural Information Processing Systems,2019,172:1 919-1 930.

    • [78] ZHANG Zizhao,PFISTER T.Learning fast sample re-weighting without reward data[C].Montreal:Proceedings of the IEEE/CVF International Conference on Computer Vision,2021:725-734.

    • [79] CAI Qi,PAN Yingwei,WANG Yu,et al.Learning a unified sam‐ ple weighting network for object detection[C].Seattle:Proceed‐ ings of the IEEE/CVF Conference on Computer Vision and Pat‐ tern Recognition,2020:14 173-14 182.

    • [80] MENG Wei,TIA Meng.Unmanned aerial vehicle classification and detection based on deep transfer learning[C].Sanya:2020 In‐ ternational Conference on Intelligent Computing and HumanComputer Interaction,2020:280-285.

    • [81] LI Yunsheng,YUAN Lu,CHEN Yinpeng,et al.Dynamic transfer for multi-source domain adaptation[C].New Orleans:Proceed‐ ings of the IEEE/CVF Conference on Computer Vision and Pat‐ tern Recognition,2021:10 998-11 007.