Simulation analysis of hydraulic fracture initiation and propagation mechanisms under mixed-mode and reservoir fracturing

With the increasing demand for the development of unconventional oil and gas resources, hydraulic fracturing has become a key technology for enhancing reservoir permeability. However, achieving controlled propagation of fracture networks remains a significant challenge under complex geological conditions. This study integrates theoretical analysis and finite-discrete element method (FDEM) simulations to investigate mixed-mode mechanisms, plastic zone evolution at fracture tips, and anisotropic mechanical responses of shale. Modified fracture criteria-including a T-stress-integrated Mohr-Coulomb criterion and maximum circumferential tensile stress criterion are derived and validated through uniaxial compression and Brazilian splitting tests on Longmaxi Formation shale. Results demonstrate that the modified Mohr-Coulomb criterion effectively predicts anisotropic fracture propagation by characterizing tensile-compressive strength differences, while the plastic zone evolution under maximum circumferential tensile stress is significantly influenced by T-stress: positive T-stress (45°–90°) expands the plastic zone, whereas negative T-stress (0°–45°) contracts it. Lower tensile-to-compressive strength ratios lead to larger plastic zones. An FDEM-based horizontal well fracturing model reveals vertical fracture propagation dominated by bedding plane and interbed fracture extension, forming complex networks, while horizontal fractures initially grow independently before deflecting and interconnecting under maximum principal stress. Sensitivity analysis of perforation spacing identifies 62.5 mm (16 holes/m) as the optimal configuration, achieving Mode II-dominated fracture networks with superior connectivity and stimulation efficiency. Larger spacings (71.4–83.3 mm) result in reduced efficiency or isolated fractures. By coupling stress interference and fluid pressure field dynamics, this study establishes a methodology to balance fracture network complexity and reservoir stimulation efficacy. The findings provide theoretical insights and engineering guidelines for optimizing hydraulic fracturing designs in anisotropic shale gas reservoirs through advanced fracture criteria and FDEM-based multi-physics simulations.