Scholarlink

<p>Tensile fracture propagation is a critical aspect of both coalbed methane exploitation and underground coal mining. A significant gap in the existing knowledge is the quantitative understanding of tensile fractures’ initiation and propagation within the context of the existing pore fracture network. This research addresses this gap using such cutting-edge technologies as deep learning, computed tomography (CT) scanning, scanning electron microscope (SEM), and mechanical loading under Brazilian Test conditions. Coal samples from the Ordos Basin of China underwent step-by-step loading, and CT scanning was conducted before and after each loading step. Then, the tensile fractures’ evolution was investigated. A deep learning technique was employed for accurate image segmentation. The developed models were transformed into a user-friendly Windows-based App, providing a valuable resource for fellow researchers. The analyses revealed insights into the initiation, propagation, extension, and growth of tensile fractures. Two-dimensional (2D) and three-dimensional (3D) analyses considered fracture length, aperture, density, and volume, leading to the formulation of relevant rock mechanical and geomechanical models. This research contributes valuable insights into the intricate interplay between natural fractures and tensile fractures’ dynamics based on the pore space evolution pattern. The findings have significant implications for coalbed methane extraction and coal mining. </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

Tensile Fracture Propagation in Deep Coalbed Methane Layers Under Brazilian Test: A Quantitative Analysis Using Digital Rock Technology and Deep Learning-Based Image Segmentation

  • 期刊论文

  • 发表日期：

    2025年6月16日

查看全文
  • Naser Golsanami,
  • Dingrui Guo,
  • Shanilka G. Fernando,
  • Mustafa Kumral,
  • Lishuai Jiang,
  • Thanuja Raveendrasinghe,
  • Behzad Saberali,
  • Ghasem Saedi,
  • Weihcao Yan,
  • Elham Bakhshi,
  • Qazi Adnan Ahmad,
  • Olga V. Dolbnya,
  • Roman Kozlov,
  • Mahmoud Behnia,
  • Madusanka N. Jayasuriya

摘要

Tensile fracture propagation is a critical aspect of both coalbed methane exploitation and underground coal mining. A significant gap in the existing knowledge is the quantitative understanding of tensile fractures’ initiation and propagation within the context of the existing pore fracture network. This research addresses this gap using such cutting-edge technologies as deep learning, computed tomography (CT) scanning, scanning electron microscope (SEM), and mechanical loading under Brazilian Test conditions. Coal samples from the Ordos Basin of China underwent step-by-step loading, and CT scanning was conducted before and after each loading step. Then, the tensile fractures’ evolution was investigated. A deep learning technique was employed for accurate image segmentation. The developed models were transformed into a user-friendly Windows-based App, providing a valuable resource for fellow researchers. The analyses revealed insights into the initiation, propagation, extension, and growth of tensile fractures. Two-dimensional (2D) and three-dimensional (3D) analyses considered fracture length, aperture, density, and volume, leading to the formulation of relevant rock mechanical and geomechanical models. This research contributes valuable insights into the intricate interplay between natural fractures and tensile fractures’ dynamics based on the pore space evolution pattern. The findings have significant implications for coalbed methane extraction and coal mining.