A combination of hydraulic fracturing and horizontal wells is now being used to tap geothermal energy from naturally fractured reservoirs. Fully grid-based numerical models are currently used to simulate heat recovery from enhanced geothermal systems (EGS). Such models require a fine unstructured mesh and are computationally expensive. In this paper we present a computationally efficient model that allows us to accurately simulate fracture propagation, fluid flow, and heat transfer in networks of natural fractures that may be created in naturally fractured geothermal reservoirs.

The integrated simulator is developed by combining the displacement discontinuity method (DDM) for fracture propagation in naturally fractured reservoirs with a general Green's function solution for fluid and heat flow from the matrix to the fracture. This eliminates the need to discretize the matrix domain resulting in a very computationally efficient solution. A discrete fracture network (DFN) approach is used to represent the pre-existing natural fractures.

The model is first validated against an analytical solution for fluid flow and heat transfer in a rock matrix with a single fracture. The computation time with and without discretizing the rock matrix shows a 100-fold reduction in computation cost with very little loss in accuracy. Parametric studies are conducted to investigate the effect of the distribution of natural fracture density, length, and orientation. The results show that the efficiency of tapping geothermal energy is impacted by geometrical and topological complexities of the fracture network and in particular the connectivity of backbone fractures. It is, therefore, important to optimize (not maximize) the connectivity and complexity of the backbone fracture network. The computationally efficient model presented here provides a practical tool for optimizing operational parameters for efficient geothermal production.

You can access this article if you purchase or spend a download.