地球物理学进展 ›› 2019, Vol. 34 ›› Issue (3): 1017-1024.doi: 10.6038/pg2019CC0131

• 应用地球物理学Ⅰ(油气及金属矿产地球物理勘探) • 上一篇    下一篇

银额盆地居延海坳陷烃源岩TOC测井预测方法优选及应用

陈浩1,2,刘洛夫1,2,*(),窦文超1,2,肖遥1,2,刘晓翔1,2,陈怡婷1,2,王梦尧1,2,王曦蒙1,2   

  1. 1. 中国石油大学(北京)油气资源与探测国家重点实验室,北京 102249
    2. 中国石油大学(北京)盆地与油藏研究中心,北京 102249;
  • 收稿日期:2018-08-03 修回日期:2019-03-10 出版日期:2019-06-20 发布日期:2019-07-09
  • 通讯作者: 刘洛夫 E-mail:liulf@cup.edu.cn
  • 作者简介:陈浩,男,1994年生,湖北公安人,硕士研究生,主要从事石油地质研究.(E-mail: 583783142@qq.com)
  • 基金资助:
    中国地质调查局地质调查项目(DD20160200);中国地质调查局地质调查项目(DD20160172)

Optimal selection and application of prediction methods for TOC of source rocks based on logging data in Juyanhai depression,Yingen-Ejin Banner basin

CHEN Hao1,2,LIU Luo-fu1,2,*(),DOU Wen-chao1,2,XIAO Yao1,2,LIU Xiao-xiang1,2,CHEN Yi-ting1,2,WANG Meng-yao1,2,WANG Xi-meng1,2   

  1. 1. State Key Laboratory of Petroleum Resources and Prospecting,China University of Petroleum(Beijing),Beijing 102249,China
    2. Basin and Reservoir Research Center,China University of Petroleum(Beijing),Beijing 102249,China;
  • Received:2018-08-03 Revised:2019-03-10 Online:2019-06-20 Published:2019-07-09
  • Contact: Luo-fu LIU E-mail:liulf@cup.edu.cn

摘要:

由于钻井取心成本以及样品来源等原因的限制,利用测井资料来预测烃源岩中的有机碳含量(TOC)成为一种常用手段.本文剖析了前人提出的关于烃源岩TOC测井预测方法,包括自然伽马法、ΔLogR法及多元回归法,结合居延海坳陷的地质特征以及有关测井参数,对不同方法的预测结果进行误差对比分析,优选认为多元回归法能够较好的应用于研究区烃源岩TOC含量预测,并提出了适用于本研究区的以ΔLogR、深侧向电阻率和自然伽马为基本参数的多元回归预测模型.将该模型应用到研究区额探1井和额探2井的烃源岩TOC含量预测,在白垩系和二叠系识别出了多段厚度大于10 m的优质烃源岩,平均TOC含量都大于1.5%.由此表明优选出的预测模型对于居延海坳陷烃源岩TOC含量的预测具备重要意义,具有良好的应用前景.

关键词: 居延海坳陷, 烃源岩, 有机碳含量, 测井预测, 多元回归

Abstract:

Due to the limitation of core drilling cost,core samples are uneasy to obtain.Therefore,using logging data to predict the Total Organic Carbon content(TOC)in source rock has become a common method.Combining the geological characteristics of Juyanhai depression and related logging parameters,this paper analyzes different methods on the TOC logging prediction of hydrocarbon source rocks,including gamma-ray model,ΔLogR model and multiple regression model.According to error analysis of prediction results of different methods,multiple regression model is considered as the best method to predict the content of the TOC in source rock in the research area.Multiple regression prediction model based on ΔLogR,RLLD and GR parameters is established in this paper and is applied to the TOC logging prediction of hydrocarbon source rocks in Etan1 and Etan2 wells in the research area.From the results of the practical application,a series of high quality source rocks with thickness of more than 10 m located in the Cretaceous and Permian are identified and the average content of TOC is greater than 1.5%.It is shown that multiple regression model established in this paper is of great significance to the prediction of the content of TOC in source rocks in Juyanhai depression and also has a good application prospect.

Key words: Juyanhai depression, Hydrocarbon source rocks, Organic carbon content, Prediction by well-logging data, Multiple regression model

中图分类号: