页岩地层测井岩性识别技术发展现状  

Development Status of Logging⁃Based Lithology Identification Technology for Shale Formations

作　　者：陈秀娟[1] 冯镇涛 曾芙蓉 胡建波 徐松 CHEN Xiujuan;FENG Zhentao;ZENG Furong;HU Jianbo;XU Song(Research Institute of Exploration and Development,Jianghan Oilfield Company,Sinopec,Wuhan,Hubei 430223,China;School of Resources,China University of Geosciences(Wuhan),Wuhan,Hubei 430074,China)

机构地区：[1]中国石化江汉油田分公司勘探开发研究院,武汉430223 [2]中国地质大学(武汉)资源学院,武汉430074

出　　处：《新疆石油地质》2024年第6期742-752,共11页Xinjiang Petroleum Geology

基　　金：中国石化江汉油田科研项目(JKD4623003)。

摘　　要：页岩油气是中国最具发展潜力的非常规油气资源,已成为非常规油气勘探开发的热点。中国页岩多为陆相沉积,岩性变化快,矿物种类多,物性条件差,非均质性强,连续性差,仅利用常规测井资料解释方法无法精细识别岩性,致使页岩储集层特征难以有效表征,制约了油气储量评估与油气开发。为有效识别页岩岩性,系统调研了国内外测井岩性识别技术,梳理了基于测井解释及不同测井手段的岩性识别技术,重点剖析了基于机器学习的测井岩性识别技术,阐述了各技术的方法原理,归纳总结了各技术的优缺点及适用性,对该领域发展趋势进行了展望。Shale reservoirs contribute the most promising unconventional oil and gas resources in China and have become a hotspot in unconventional oil and gas exploration and development.Shale formations in China are mostly continental,with varying lithologies,diverse minerals,poor physical properties,strong heterogeneity,and poor continuity.These characteristics make it difficult to accurately identify lithology only using conventional logging interpretation methods,which in turn hinders the effective characterization of shale reservoirs and severely constrains reserves estimation and oil/gas development activities.In order to effectively identify the lithology of shale formations,the logging⁃based lithology identification technologies at home and abroad were systematically reviewed,and the lithology identification technologies based on logging interpretation and logging techniques were introduced.The logging lithology identification technologies based on machine learning were dissected in respect to their principles,advantages,disadvantages,and applicability.Finally,the prospects of logging⁃based lithology identification technologies for shale formations were proposed.

关 键 词：页岩 测井岩性识别 机器学习 测井响应特征 神经网络 测井解释 

分 类 号：P631.842[天文地球—地质矿产勘探]