Abstract The permeability of coal is the key parameter both in primary and enhanced coalbed methane recovery. The natural cleat system in coal serves as the primary pathway for gas flow in coal seams though mineralisation in cleats and is known to significantly reduce coal permeability. This paper reports on a numerical simulation of the pore network evolution of coal subject to cleat demineralisation. A high-resolution micro-computed tomography scanner was used to characterize the micro-structures of three anthracite coal samples. The mineral phases available in the coal samples were selectively removed to different extents (20%, 40%, 60%, 80% and 100%) and merged into the pore space using image processing techniques. In this way, the coal demineralisation process could be simulated and its impact on porosity and permeability studied. Comprehensive pore structure characterizations, including porosity, connectivity and tortuosity, were then conducted on the reconstructed pore network using Avizo software. Pore network models were also extracted to investigate changes in the pore and throat attributes. The lattice Boltzmann method was adopted to identify the absolute permeability changes with cleat demineralisation. The results reveal that demineralisation can increase coal porosity and permeability up to a percolation threshold. Although porosity was enhanced prior to the percolation threshold, the coal permeability was not enhanced due to poor pore connectivity. The permeability changed rapidly close to the percolation threshold, depending on the degree of demineralisation, and an exponential relation was observed between permeability and the amount of demineralisation. According to the observations, complete removal of the mineral phase can significantly increase the connected porosity while reducing the pore tortuosity, resulting in several orders of magnitude increase in coal permeability. This study shows that cleat demineralisation is an effective permeability enhancement technique for coalbed methane recovery, if very high demineralisation can be achieved.

中文翻译：

---

使用图像处理技术通过脱矿来表征煤的孔隙度和渗透率演变：微计算机断层扫描研究

摘要 煤的渗透率是煤层气一次采收和强化采收的关键参数。煤中的天然割理系统是通过割理中的矿化作用在煤层中流动的主要途径，并且已知会显着降低煤的渗透率。本文报告了受割理脱矿作用的煤的孔隙网络演化的数值模拟。高分辨率微型计算机断层扫描仪用于表征三个无烟煤样品的微观结构。煤样中可用的矿物相被选择性地去除到不同程度（20%、40%、60%、80% 和 100%），并使用图像处理技术合并到孔隙空间中。通过这种方式，可以模拟煤炭脱矿过程并研究其对孔隙度和渗透率的影响。然后使用 Avizo 软件对重建的孔隙网络进行全面的孔隙结构表征，包括孔隙率、连通性和曲折度。还提取了孔隙网络模型以研究孔隙和喉道属性的变化。采用格子 Boltzmann 方法来识别随着割理脱矿作用的绝对渗透率变化。结果表明，脱矿作用可以将煤的孔隙度和渗透率增加到渗透阈值。尽管在渗透阈值之前孔隙度增加，但由于孔隙连通性差，煤渗透率并未增加。渗透率在接近渗透阈值时迅速变化，这取决于脱矿程度，并且在渗透率和脱矿量之间观察到指数关系。据观察，完全去除矿物相可以显着增加连通孔隙度，同时降低孔隙曲折度，使煤的渗透率提高几个数量级。这项研究表明，割理脱矿是一种有效的提高煤层气回收渗透率的技术，前提是可以实现非常高的脱矿。