Thermo-poroelastic reservoir response of multi-well and multi-fracture geothermal systems

Shallow and deep geothermal reservoirs are potential renewable energy sources. Heat recovery from geothermal reservoirs disrupts the natural temperature distribution and induces thermo-poroelastic responses in the geothermal reservoir. Single-well or double-well systems generally affect a small reservoir volume. Shallow geothermal projects often present a drained response and are unlikely to disturb deep faults. However, an upscaling of geothermal energy requires massive geothermal systems at large depths where the reservoir coupled thermo-poroelastic response to heat drainage is more pronounced and may approach undrained conditions. This work provides novel numerical simulations of geothermal reservoirs completed with (1) multi-stage hydraulic fracturing and (2) closed-loop multilateral-wells to investigate the drained and undrained thermo-poroelastic response when subjected to heat drainage. The numerical solutions are based on the theory of thermo-poroelasticity and solved through the Finite Element Method in two and three-dimensions. A rigorous mechanical contact model is employed to solve for the internal contact tractions and the fracture opening displacements in response to rock cooling. Results show that late-stage shear reactivation is caused mostly by thermo-elastic destressing. Stress changes towards shear failure and potential fault/fracture reactivation are driven by reduction in effective horizontal stresses and therefore are most likely in tectonically passive or extensional environments. The simulations predict thermal recovery factors in the order of 5-50% of the original heat in place for the given initial pressure and stress conditions, mostly limited by a rigorous limit on fracture reactivation of 10% of the thermal reservoir volume. Larger thermal recovery factors could be achieved in (1) tectonically compressive environments, (2) locations where the initial stress anisotropy is low, and (3) deep locations where in-situ effective stresses are large. The inclusion of a limit to fracture reactivation serves as a proxy for maximum allowed induced seismicity, limiting the technically recoverable heat, and provides more realistic estimations of geothermal energy reserves.