Effect of Shearing on Non-Darcian Fluid Flow Characteristics through Rough-Walled Fracture

Abstract

The heterogeneous fracture geometry induced by the presence of roughness and shearing complicates the fracture flow. This paper presents a numerical investigation of the non-Darcian flow characteristics of rough-walled fractures during shear processes. A series of fracture flow simulations were performed on four types of fractures with different joint roughness coefficients (JRCs), and the different shear displacements were imitated by degrees of mismatch on two fracture surfaces. The results show that the disorder of fracture geometries and the increase in flow rate are the main causes for the emergence of an eddy flow region, which can significantly reduce the fracture conductivity and change the fracture flow from linear to nonlinear. The Forchheimer equation provides a good model for the nonlinear relationship between the hydraulic gradient and the flow rate in the fracture flow. When the shear displacement or JRC increased, the linear permeability coefficient kv decreased, while the nonlinear coefficient β increased. A three-parameter equation of β was used to examine the inertial effect induced by the fracture roughness JRC and the variation coefficient of aperture distribution σs/em. The critical Reynolds number was a combined effect of aperture, viscous permeability, and inertial resistance, assuming the flow becomes non-Darcian when the inertial part is greater than 10%.

Assessment of nonlinear flow behaviors through real rough rock fractures during shearing

“…In addition, shear induces expansion of rock fractures and surface mismatch, resulting in changes in contact characteristics and flow channels, thus affecting the flow behaviors in fractures. In recent years, scholars have studied the changes in fluid flow behavior during fracture shearing, mainly using the numerical simulations and experimental studies (Xie et al 2015;Zou et al 2017;Zhou et al 2018;Li et al 2020;Zhang et al 2017Zhang et al , 2021 and experimental studies (Rong et al 2016;Yin et al 2017;Wang et al 2020).…”