Coal deformation during gas depletion primarily involves the effective stress controlled bulk compression and gas sorption/desorption-induced coal matrix swelling/shrinkage. The two concurrent deformation behaviors can significantly impact coal permeability. Majority coal permeability models are defined as a cubic function of coal porosity, in which the porosity is a linear superposition of bulk strain and sorption strain. Nevertheless, laboratory measurements have observed that besides the sorption-induced swelling the injected gas pressure can also hydrostatically compress coal matrix. The observations reflect the complexity of fracture-matrix interaction on the coal porosity, which brings significant challenges to the reliable prediction of the evolution of coal permeability. In this study, we developed an experimental approach to separating the sorption-induced swelling strain of coal matrices and hydrostatic compression. We also modified a coal-gas interaction model by introducing an internal swelling coefficient and implemented into a coupled numerical model. Our experimental results based on nitrogen injection show that gas pressure rise can increase both the swelling strain and hydrostatic compressive strain of the coal specimen. If the measured coal strain is uncorrected, the corresponding isothermal sorption-strain parameters will be underestimated, resulting in the overestimation of the coal permeability. Numerical modeling results show that when the confining stress remains unchanged, the matrix swelling contribution to narrow the fracture width can attenuate with the increasing pore pressure. When the effective stress remains unchanged, the matrix swelling contribution can increase. Comparison of the published permeability data and these modeling results suggests that the actual contribution of the matrix deformation to fracture is significantly related to geomechanical state of coal reservoir, indicating that the internal swelling coefficient should be considered as a key variable when selecting a coal permeability model.

瓦斯枯竭过程中的煤变形主要涉及有效的应力控制的整体压缩和瓦斯吸附/解吸引起的煤基质膨胀/收缩。两种同时发生的变形行为会显着影响煤的渗透性。多数煤渗透率模型定义为煤孔隙度的三次函数，其中孔隙度是体积应变和吸附应变的线性叠加。然而，实验室测量已经观察到，除了吸附引起的溶胀之外，注入的气压还可以流体静压地压缩煤基质。这些观察结果反映了煤层孔隙度与裂缝-基质相互作用的复杂性，这给可靠地预测煤渗透性演化提出了重大挑战。在这个研究中，我们开发了一种实验方法来分离吸附引起的煤基质溶胀应变和静水压力。我们还通过引入内部溶胀系数修改了煤-气相互作用模型，并将其实现为耦合数值模型。我们基于氮气注入的实验结果表明，气体压力升高可以增加煤样品的膨胀应变和静水压应变。如果测得的煤应变未经校正，则相应的等温吸附应变参数将被低估，从而导致煤渗透率的高估。数值模拟结果表明，当约束应力保持不变时，随着孔隙压力的增加，基体膨胀作用使裂缝宽度变窄的作用会减弱。当有效应力保持不变时，基体溶胀的贡献会增加。对已发表的渗透率数据和这些建模结果的比较表明，基质变形对裂缝的实际贡献与煤储层的地质力学状态显着相关，表明在选择煤渗透率模型时应将内部溶胀系数视为关键变量。 。