西南石油大学学报(自然科学版) ›› 2018, Vol. 40 ›› Issue (6): 131-138.DOI: 10.11885/j.issn.1674-5086.2018.03.18.01

• 石油与天然气工程 • 上一篇    下一篇

煤层气水平井随钻地层判识技术研究

张禾, 张弓   

  1. 西南石油大学机电工程学院, 四川 成都 610500
  • 收稿日期:2018-03-18 出版日期:2018-12-01 发布日期:2018-12-01
  • 通讯作者: 张禾,E-mail:zhanghe@swpu.edu.cn
  • 作者简介:张禾,1973年生,男,汉族,四川南充人,副教授,主要从事测控技术方面的研究工作。E-mail:zhanghe@swpu.edu.cn;张弓,1993年生,男,汉族,四川成都人,硕士研究生,主要从事测控技术方面的研究工作。E-mail:81037261@qq.com
  • 基金资助:
    四川省应用基础研究基金(2016JY0049)

Logging-while-drilling Stratigraphic Identification Techniques for Horizontal Wells for Coalbed Methane

ZHANG He, ZHANG Gong   

  1. School of Mechatronic and Electrical Engineering, Southwest Petroleum University, Chengdu, Sichuan 610500, China
  • Received:2018-03-18 Online:2018-12-01 Published:2018-12-01

摘要: 煤层气水平井是提高煤层气单井产量的有效技术措施,为确保煤层气水平井的有效钻遇率,需要进行随钻地层判识,进而对钻井施工过程实施地质导向。现有的煤层气水平井随钻地层判识技术大都以井下随钻数据进行分析,判识准确率有待提高。针对目前中国煤层气水平井钻井过程中普遍采用的仪器组合方式,在满足煤层气低成本开发的客观要求的前提下,选取了自然伽马、声波时差、补偿密度、补偿中子4个参数进行分析,采用支持向量机算法对上述测井参数进行归一化处理,根据随钻地层判识要求选择了符合条件的核函数,对相关参数进行了优化,通过对输出结果的分类实现了煤层气水平井随钻地层判识,判识准确度满足现场应用的要求。

关键词: 煤层气, 水平井, 支持向量机, 测井数据, 随钻地层判识

Abstract: Existing methods of logging-while-drilling stratigraphic identification for horizontal wells for coalbed methane are mainly based on the logging-while-downhole drilling data. However, the accuracy of these methods needs to be improved. This study employs horizontal drilling jig configurations widely adopted for coalbed methane exploration in China, and takes the cost-effectiveness into account. Four parameters are selected and analyzed, namely natural gamma, interval transit time, compensated density, and compensated neutron. The support vector machine algorithm is used to normalize data collected for these parameters, and to optimize the parameters with the appropriate kernel functions. By classifying the final outputs, this study realizes logging-while-drilling stratigraphic identification for horizontal wells for coalbed methane. Such an identification method is accurate enough for field applications.

Key words: coalbed methane, horizontal well, support vector machine, well-logging data, logging-while-drilling stratigraphic identification

中图分类号: