Integrated simulation of multi-stage hydraulic fracturing in unconventional reservoirs

In recent years, the multi-stage hydraulic fracturing technology has been widely applied to unconventional reservoirs. In order to optimize the stimulation performance and achieve economic production rates, the underlying physics and the impacts of different designing parameters must be quantitatively understood prior to the operations. Although many numerical models have been developed for this purpose, few are capable of properly taking into account the interplay of different mechanisms, such as fluid flow in the horizontal wells, fracture propagation, and proppant transport. A novel 3D thermal-hydro-mechanical model entitled UFrac is developed in this paper. This model integrates the various mechanisms to capture the interactions among wellbores, fractures and reservoir rocks. The fluid flow inside the wells and the fractures is simulated with the finite volume method (FVM), the elastic deformations of rock mass are calculated with the 3D displacement discontinuity method (DDM), and fracture propagation is simulated with the fixed grid method. The UFrac model is capable of simulating the combined effects of multiple fracturing operation parameters, elastic interaction between fractures, and temperature redistribution induced by flow exchange. By using the UFrac model, we investigate several key problems in hydraulic fracturing such as fracture height growth, perforation spacing optimization, and thermal effect on proppant placement. Important conclusions are drawn from the simulation results. The strength of stress interaction is found to be related to the fracture geometries. Stress shadow effect does not only affect the flow partitioning and fracture size distribution, but also influence the proppant transport. Friction loss in the wellbore can affect the decisions on spacing optimization. Moving inner fractures closer to the heel of the wells would be beneficial for fracture propagation balancing. Fluid viscosity loss due to heating from the surrounding formations results in longer but narrower fractures and faster settling of proppant. The modeling of proppant distribution aids in better characterization of the fracture conductivity, thus provides more reliable prediction of the well productivity. (C) 2016 Elsevier B.V. All rights reserved.National Natural Science Foundation of China [51674010]SCI(E)ARTICLEtanghuiying@pku.edu.cn875-8923