A critical review of common models in hydraulic-fracturing simulation: A practical guide for practitioners

Abstract

Abstract In laminated, anisotropic unconventional rocks (ultra-tight shale, tight gas, coal seam, and so forth), the true hydraulic-fracturing mode is a complex, mixed-mode process. The fracture-modeling practitioner often encounters a wide range of fracture models. The selection of an appropriate fracture model has become a daunting task with significant implications for the hydraulic-fracturing pre-treatment design or post-treatment diagnostic evaluation. Therefore, it is crucial to properly identify the assumptions and limitations of each model, and how they affect the modeling output as well as the treatment design. This study provides a critical review of the assumptions, advantages, and disadvantages of the more-common fracture models in hydraulic-fracturing simulation. The models covered range from the simplified 2D, pseudo 3D (P3D), and lumped parameter (LP) models to 3D planar-fracture models (elastically coupled and shear-decoupled models), and to more complex 3D models including the displacement discontinuity model (DDM), discrete fracture network (DFN) model, distinct element model (DEM), bonded particle model (BPM), synthetic rock mass (SRM) using BPM and smooth joint model (SJM), lattice model (including the dual-lattice implementation of SRM), continuum and extended-continuum models, and the meshless models. The advantages and disadvantages of the more advanced 3D models are analyzed in the context of simulating special cases, including alternate fracturing and hydraulic fracturing in naturally fractured reservoirs. In the context of these special cases, the capabilities and limitations of these models in incorporating the complex interactions between the hydraulic fractures, and between hydraulic and pre-existing fractures and joints are discussed. This study outlines how a fracture model should be carefully selected, considering the geological and geomechanical conditions, availability of data, and the scope of investigation.