西南石油大学学报(自然科学版) ›› 2019, Vol. 41 ›› Issue (2): 45-52.DOI: 10.11885/j.issn.1674-5086.2017.12.14.04

• 地质勘探 • 上一篇    下一篇

基于模糊数学的油藏干层识别研究

许晓明, 李彦兰, 孙景民   

  1. 中国石油勘探开发研究院, 北京 海淀 100083
  • 收稿日期:2017-12-14 出版日期:2019-04-10 发布日期:2019-04-10
  • 通讯作者: 许晓明,E-mail:xxiaom@petrochina.com.cn
  • 作者简介:许晓明,1984年生,女,汉族,辽宁北票人,工程师,主要从事油气田开发方面的研究工作。E-mail:xxiaom@petrochina.com.cn;李彦兰,1963年生,女,汉族,四川成都人,高级工程师,主要从事油气田开发方面的研究工作。Email:lily@petrochina.com.cn;孙景民,1965年生,男,汉族,河北河间人,高级工程师,主要从事油气田开发地质方面的研究工作。E-mail:sunjingmin@petrochina.com.cn
  • 基金资助:
    国家科技重大专项(2016ZX05010-005)

Study on Identification of Dry Layers Based on Fuzzy Mathematics

XU Xiaoming, LI Yanlan, SUN Jingmin   

  1. PetroChina Research Institute of Petroleum Exploration & Development, Beijing, Haidian 100083, China
  • Received:2017-12-14 Online:2019-04-10 Published:2019-04-10

摘要: 在精细油藏地质认识的基础上进行注采精细调整是高含水老油田挖潜工作的重要出路。通过对渤海湾中深层油藏储层的精细认识及实际生产动态资料分析,认为部分与油井相通的水井干层具有一定的吸水能力,因此有效利用干层能够进一步完善注采关系、提高水驱动用程度以最终提高水驱采收率。决定与油层相通的干层是否可以利用的因素多而复杂,将主成分分析方法与模糊贴近度理论结合,利用各判别变量的综合信息建立一种模糊隶属度判别函数的识别方法,既降低了初始变量的维数,又保证了判别函数的全面性及准确性。实例分析表明,模糊贴近度识别方法对干层的有效性识别准确率超过85%以上,用数值模拟方法进行分析对比,干层的有效利用可使注采关系得到了进一步完善,开采效果明显提高。因此,该综合判别法结果可靠,实用性较强,能够指导油田生产实际。

关键词: 有效干层性识别, 有效干层利用, 模糊贴近度, 主成分分析, 注采关系

Abstract: It is important to make fine adjustments to the injection and production of oil fields based on detailed geological understanding of reservoirs for the development of high-water-cut wells in mature oil fields. This work has studied a detailed description of the middle and deep layers of the oil/oil-water/water reservoir in Bohai Bay and a data analysis of its actual oil production. It is assumed that some dry layers of water wells connected with oil wells demonstrate a certain degree of water absorption. Therefore, the effective use of a dry layer can further increase the injection-production ratio, improve the water drive degree, and ultimately enhance the water absorption ratio. The connection of a dry layer with an oil reservoir depends on many complex factors. This study has employed the principal component analysis and theory of fuzzy closeness to propose an identification method for the membership function of a fuzzy set using the integrated information of each discriminant variable. This method has reduced the dimension of initial variables while ensuring comprehensiveness and accuracy of the discriminant function.Research results show that the proposed method based on fuzzy closeness has achieved an accuracy of over 85% in terms of the effective identification of dry layers. A numerical simulation has been adopted for data analysis and comparison. The effective use of the dry layer has further enhanced the injection-production ratio and significantly improved the oil development. Considering the aforementioned factors, it is established that the comprehensive identification method is reliable and fairly practical for the future development of oilfields.

Key words: effective identification of dry layer, efficient use of dry layer, fuzzy closeness, principal component analysis, injection-production ratio

中图分类号: