Numerical Modeling of Hydraulic Stimulation and Long-Term Fluid Circulation at the Utah FORGE Project

Abstract

In the coming months, the Utah FORGE project plans to connect two vertically offset inclined wells – 16A and 16B – using multistage hydraulic fracturing. Six fracturing stages will be performed, varying: fluid type, cluster spacing, flow rate, total volume, and (possibly) proppant type. In this paper, we perform numerical simulations of these stimulations to predict: fracture geometry, the number of frac hits that will be observed in the DAS fiber at the production well (16B), the flow rate that will be achieved during initial circulation tests, and the thermal drawdown that will occur during long-term circulation tests. The simulation model is calibrated to the fracture geometry observations from the previously performed Stage 3 (consisting of cross-linked gel) in the 16A well. The upcoming stimulations will provide an opportunity to test model predictions and assumptions. In addition to the Base Case model, we run sensitivity analysis simulations varying key model inputs. The results show that the number of frac hits (and the ultimate connectivity between the wells) depends on the effective fracture toughness and the fracture conductivity during propagation. Simulations suggest that thermoelastic stress effects have potential to cause early thermal breakthrough during circulation. A simulation is performed using inflow control at the production well (and including thermoelastic stress reduction), and it shows that inflow control can prevent premature breakthrough and cause thermoelastic stress to have a positive, rather than negative effect, on overall thermal extraction. Finally, a simulation is performed circulating CO2 instead of water. The net effect of using CO2 is a moderate increase in the rate of thermal extraction, caused by the differences in viscosity, density, and heat capacity.