Trans-Scale Flow Model and Transport Mechanism of Coal Matrix-Cleat System

Summary Matrix pores and the cleat system are widely developed in coal reservoirs. To simultaneously characterize the permeability of the matrix region and the cleat system, we have established a trans-scale flow model of coal reservoirs based on the revised gray lattice Boltzmann method (GLBM) and Shan-Chen model. By associating the partial rebound coefficient ns with the specific spatial position of matrix pores, we consider the permeability of the matrix region in the overall flow. Meanwhile, by adding virtual potential between particles to the evolution equation directly through the force format, we improve the stability of the model and expand the range of density and viscosity ratios for multiphase and multiple-component flow simulation. Finally, based on the flow simulation results, we propose the trans-scale interporosity flow coefficient and revise the macroscopic reservoir numerical model. The results show that for multiscale digital cores with homogeneous characteristics of matrix pores, as the permeability of the matrix region increases, the overall flow velocity at the outlet increases, especially in the face cleat region. Compared with the case without considering the permeability of the matrix region, when the permeability of the matrix region is 0.08 md, the average flow rate increase can reach 10.9%. For multiscale digital cores with heterogeneity characteristics of matrix pores, when the matrix pores’ connectivity direction is perpendicular to the flow direction, the average flow velocity at the outlet is significantly increased, with an average increase of 6.0%. Besides, the trans-scale flow model established in this study successfully characterized the fluid-solid interaction force. Compared with the case without considering the fluid-solid interaction force, the average flow velocity at the outlet increases when the fluid-solid interaction force is attraction and decreases when the fluid-solid interaction force is repulsion. In addition, for reservoir-scale numerical simulation, when considering matrix permeability, the production prediction results are higher than those without considering matrix permeability, especially when the matrix heterogeneity is considered and the matrix pores’ connectivity direction is perpendicular to the flow direction. For reservoir parameter inversion, this will lead to an overestimation of fracture permeability. When the matrix permeability is 0.08 md and the matrix pores’ connectivity direction is perpendicular to the flow direction, the increase in average velocity is 18.4% and the maximum overestimation can be 11.8%.