Numerical Simulation of Multi-cluster Fracturing Using the Triaxiality Dependent Cohesive Zone Model in a Shale Reservoir with Mineral Heterogeneity

There are still some major issues in the study of fracture network evolution during multi-cluster fracturing in shale reservoirs, such as the simplistic assumption that reservoir rocks are homogeneous and completely brittle, without considering that real rocks are often aggregates of different minerals and may present ductility and heterogeneity. It is important to study the evolution of the multi-cluster fracture network in quasi-brittle shale reservoirs to improve the stimulated reservoir volume (SRV). For this purpose, a 2D, coupled stress-seepage-damage field multi-cluster fracturing numerical model with global cohesive zone was developed in this paper. We conducted triaxial compression acoustic emission experiments using real shale samples and developed a generic trapezoidal TSL softening model that includes triaxial effects. The globally embedded 0-thickness cohesive elements ensures that hydraulic fractures can propagate randomly and thus reflect stress interference among multiple clusters. At the same time, the X-ray diffraction (XRD) experiment was used to determine the mineral content of rock, and the finite element mesh was then processed using the Weibull distribution probability density function to simulate the mineral heterogeneity of rock. In addition, the dynamic distribution of injection fluid flow during multi-cluster fracturing is implemented based on the Bernoulli equation. The cohesive parameters were validated by Brazilian splitting test, and the model was then used to parametrically study the evolution law of the fracture network and the variation characteristics of the flow rate into each cluster. The results show that using conventional bilinear TSL will result in a larger SRV than trapezoidal TSL, and reservoir heterogeneity may further exaggerate the drainage area when using bilinear TSL model. In addition, compared with a homogeneous isotropic reservoir, a highly heterogeneous reservoir has a more balanced flow rate into each cluster during multi-cluster fracturing, and this can significantly increase the complexity of fracture network. By increasing the number of clusters, it is possible to alleviate the stress interference on some internal clusters, which may also promote frequent fracture merging around the wellbore. Compared with increasing the number of clusters, reducing the perforation diameter can better compensate for the stress interference suffered by internal clusters, and make the flow rate into each cluster more balanced, thus improving the SRV.