A Novel Physics-Based Subsidiary Fracture Networks Grading and Permeability Equivalence for Efficient Shale Reservoir Simulation in History Matching

Summary Unconventional reservoirs with complex stimulated fracture networks pose formidable challenges for efficient simulation and history matching. Conventional approaches that geometrically resolve all fractures from primary hydraulics to multiscale branches suffer severe computational bottlenecks through unstructured grids and permeability discontinuities, compounded by the sheer number of fracture parameters that require calibration during history matching, which leads to underdetermined inversions. In this study, we introduce a hybrid framework that prioritizes computational efficiency without sacrificing predictive accuracy: (1) explicit modeling of primary hydraulic fractures using an embedded discrete fracture model (EDFM); (2) subsidiary-induced fracture networks rupture-scale-based grading and density estimation via microseismic data; (3) stochastic Oda upscaling converting subsidiary fractures into anisotropic equivalent permeability fields, with uncertainty bounds absorbing subseismic microfracture contributions. By replacing explicit geometric modeling of subsidiary-induced fractures with a parameterized permeability field, this approach eliminates unstructured grid generation for small-scale fractures, enhances numerical stability, and reduces the number of uncertain parameters by greater than 99.75% compared with conventional EDFM workflows. A field application in China’s Gulong Shale Oil No. 1 Pilot Area demonstrates that the framework, when integrated with the ensemble smoother with multiple data assimilation (ES-MDA) to calibrate primary fracture properties (permeability KF and width WF) and subsidiary-induced fracture networks equivalence parameters (partitioned rupture-scale distribution Ef and quantified uncertainty permeability bounds Kmax⁡, Kmin⁡), achieves efficient history matching with eight iterations. This work establishes a scalable and interpretable modeling paradigm for practical history matching and development optimization in shale reservoirs.