摘要
致密岩层中总有机碳(TOC)含量往往直接指示了油气藏的所在,但致密岩层往往岩性配置复杂,利用常规地球物理技术难以识别有效烃源岩。研究提出一种基于最优化估算和贝叶斯统计分类的TOC井-震联合预测技术,即将常规方法估算的TOC作为初始值,利用构建的岩石密度计算模型和最优化理论对TOC初始值进行校正,得到与实验室样点最佳匹配的TOC测井曲线;在TOC敏感参数分析的基础上,采用贝叶斯统计分类方法将反演的TOC敏感参数转换为TOC概率体空间分布。实际应用于湖相致密泥灰岩预测,为高产油井ST3井的部署提供了可靠依据,并验证了该技术的有效性。该项技术可推广应用于具有类似地质背景的有效烃源岩预测。
The total organic carbon(TOC)content in tight rock formations often directly indicates the location of oil and gas reservoirs,but dense rock formations often have complex lithology configurations,so it is difficult to identify effective source rocks by using conventional geophysical techniques.This paper proposed a TOC well-seismic joint prediction technology based on optimal estimation and Bayesian statistical classification.The TOC estimated by the conventional method was taken as the initial value,and the initial value of TOC was corrected by using the constructed rock density calculation model and optimization theory.The TOC log curve with the best matching with the laboratory sample was obtained.Based on the analysis of TOC sensitive parameters,the inver‐sion TOC sensitive parameters were converted into TOC probability volume spatial distribution by Bayesian sta‐tistical classification method.This technology was applied to the prediction of lacustrine dense marl reservoirs,providing a reliable basis for the deployment of high-yield oil wells ST3.The practical application results verified the effectiveness of the technology,and the technology can be applied to the prediction of effective source rocks with similar geological features.
作者
赵万金
高海燕
闫国亮
郭同翠
ZHAO Wanjin;GAO Haiyan;YAN Guoliang;GUO Tongcui(PetroChina Research Institute of Petroleum Exploration&Development-Northwest,Lanzhou 730020,China;Lanzhou University of Finance and Economics,Lanzhou 730020,China;PetroChina Research Institute of Petroleum Exploration&Development,Beijing 100083,China)
出处
《岩性油气藏》
CSCD
北大核心
2020年第1期86-93,共8页
Lithologic Reservoirs
基金
中国石油天然气集团有限公司科学研究与技术开发项目“海外重点战略大区勘探技术研究与应用”（编号：2018A-4305）、“非均质储层流体因子构建新方法研究”（编号：2019A-3310）联合资助
关键词
致密岩性
总有机碳
岩石物理模型
最优化
贝叶斯
井-震联合
tight lithology
total organic carbon
rock physics model
optimization
Bayesian
well-seismic joint
