Scholarlink

<p>To address issues such as insufficient productivity release and poor process parameter adaptability in the fracturing development of deep coal seams, this study takes vertical wells and horizontal wells in deep coal seams of the Linxing and Shenfu blocks as research objects. It conducts data analysis on fractured wells from three dimensions: geology, engineering, and production, identifies the key determinants of productivity by combining machine learning models, and optimizes the commingled fracturing process for sand-coal superimposed reservoirs. The results show that the main controlling factors of post-fracturing productivity for horizontal wells, in order of importance, are preflush ratio &gt; fracture pressure gradient &gt; displacement &gt; gas content per ton of coal &gt; sanding intensity; while for vertical wells, the order is reservoir thickness &gt; gas content per ton of coal &gt; closure pressure gradient &gt; displacement &gt; reservoir depth. The intelligent control program constructed based on the Gaussian Kriging surrogate model can output optimal construction parameters, achieving a balance between maximizing the stimulated reservoir volume (SRV) of fracture networks and minimizing construction costs. The research results offer technical support for effective development of deep coal seams.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Study on Deep Coal Seam Fracturing Technology and Productivity-Influencing Factors

  • 期刊论文

  • 发表日期：

    2026年2月20日

查看全文
  • Ling Chen,
  • Bumin Guo,
  • Jinwei Shen,
  • Jian Zhao,
  • Yantao Xu,
  • Shuan Li,
  • Yichang Zhang,
  • Ximo Qu

摘要

To address issues such as insufficient productivity release and poor process parameter adaptability in the fracturing development of deep coal seams, this study takes vertical wells and horizontal wells in deep coal seams of the Linxing and Shenfu blocks as research objects. It conducts data analysis on fractured wells from three dimensions: geology, engineering, and production, identifies the key determinants of productivity by combining machine learning models, and optimizes the commingled fracturing process for sand-coal superimposed reservoirs. The results show that the main controlling factors of post-fracturing productivity for horizontal wells, in order of importance, are preflush ratio > fracture pressure gradient > displacement > gas content per ton of coal > sanding intensity; while for vertical wells, the order is reservoir thickness > gas content per ton of coal > closure pressure gradient > displacement > reservoir depth. The intelligent control program constructed based on the Gaussian Kriging surrogate model can output optimal construction parameters, achieving a balance between maximizing the stimulated reservoir volume (SRV) of fracture networks and minimizing construction costs. The research results offer technical support for effective development of deep coal seams.