Numerical investigation of fracture propagation mechanism subjected to cyclic soft fracturing in naturally fractured HDR reservoirs

Cyclic Soft Fracturing (CSF) has been proposed to solve the problems of simple fracture network, high breakdown pressure and seismic risk during conventional hydraulic fracturing in hot dry rock (HDR) reservoirs. However, few studies have investigated its application in naturally fractured reservoirs and the interaction between natural fractures (NFs) and CSF-induced hydraulic fractures. In this paper, a thermal-hydraulic-mechanical (THM) coupling numerical model with random discrete fracture network (DFN) is established based on logging data of Gonghe basin, China to explore the reservoir stimulation characteristics subjected to CSF injection protocol. The results show that there are four interaction mode between NFs and CSF-induced fractures, namely cross, arrest, activate, and offset. There NFs are more likely to be activated by CSF method with dominant shear failure. Compared with monotonous hydraulic fracturing (MHF), CSF can reduce the breakdown pressure and the maximum AE energy by 12% and 50%, respectively. The fracture network formed by CSF are more complex with several branch fractures, which further connect with NFs and broaden the stimulated reservoir volume. Under the same injection volume, CSF produces a 10% longer total fracture length than MHF, but the increase of average fracture width is limited resulting in minor improvement of reservoir permeability. The results are expected to provide an insight into the stimulation mechanisms of CSF method in HDR development.