A Three-Dimensional Stochastic Fracture Network Model for Geothermal Reservoir Stimulation

Abstract

An important issue in geothermal resource development is engineering a fractured reservoir and predicting its future performance. For reaching this objective, numerical modeling of flow and deformation of fractured rock is necessary. This can be achieved by developing deterministic fracture models or models that utilize discrete fracture networks. This study focuses on utilizing stochastic fracture networks to simulate flow in fractured rock and to assess the mechanical rock mass response to stress variations caused by injection/production. A finite element numerical model of fracture network with stochastic description on fracture distribution is presented in this study. In this model, the reservoir is simulated using a system of blocks some of which contain fracture zones and fracture-free matrix zones. The fracture distribution is controlled by the stochastic descriptions of fracture density, size, and orientation, which can be obtained from field data. The model is used to simulate an injection operation. Fluid flow, hydraulic potential, and flow rate are calculated, and the stress response is obtained to determine the fracture aperture change for next injection step. Results show that, as expected, the permeability and flow rate increase in response to injection-induced shear slip in the formation. This is also reflected in the wellbore variation with respect to time.