Scholarlink

<p>Natural fractures in the carbonate reservoirs of the lower Permian Maokou Formation in the southeastern Sichuan Basin are important controlling factors for oil and gas enrichment in this area. Predicting the natural fracture linear density (FD) is critical across all oil and gas reservoir exploration and development stages. Conventional logging, cross multipole array acoustic (XMAC) logging, and formation micro-imaging (FMI) logging data are used to predict the FD. Rock mechanical parameters are calculated on the basis of conventional and XMAC logging data, and the FD is calculated using FMI logging data. A conditional probability binning ensemble (CPBE) model is established through the quantitative analysis of the correlation between logging data and FD. The FD is predicted with the CPBE model. The natural fracture development degree at the middle of the study area is highest, followed by those in the west and southeast. The proposed method overcomes the limitations of existing methods, such as high cost, the need for third-party software, and the need to perform high-precision numerical simulations. It provides the advantages of simple operation and high reliability and is important for effective FD prediction.</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

Prediction of the natural fracture linear density in Block B of the Sichuan Basin: insights from conditional probability binning ensemble methods

  • 期刊论文

  • 发表日期：

    2025年11月11日

查看全文
  • Mengchu Xiao,
  • Yuntao Li,
  • Mengyang Zhang,
  • Xiaoyun Cheng,
  • Xiang He,
  • Hao Jian

摘要

Natural fractures in the carbonate reservoirs of the lower Permian Maokou Formation in the southeastern Sichuan Basin are important controlling factors for oil and gas enrichment in this area. Predicting the natural fracture linear density (FD) is critical across all oil and gas reservoir exploration and development stages. Conventional logging, cross multipole array acoustic (XMAC) logging, and formation micro-imaging (FMI) logging data are used to predict the FD. Rock mechanical parameters are calculated on the basis of conventional and XMAC logging data, and the FD is calculated using FMI logging data. A conditional probability binning ensemble (CPBE) model is established through the quantitative analysis of the correlation between logging data and FD. The FD is predicted with the CPBE model. The natural fracture development degree at the middle of the study area is highest, followed by those in the west and southeast. The proposed method overcomes the limitations of existing methods, such as high cost, the need for third-party software, and the need to perform high-precision numerical simulations. It provides the advantages of simple operation and high reliability and is important for effective FD prediction.