The contribution of natural fractures to well performance in hydraulically fractured horizontal wells in shale reservoirs is not fully understood. To better understand the influence of natural fractures on well performance, numerical discrete element modelling software was used to simulate the hydraulic fracturing process in shale. Two cases were analyzed: one with natural fractures and the other without natural fractures. For the first case, the modelling results show that critically stressed natural fractures undergo shear failure caused by the injected water from the hydraulic fracturing process, as well as tensile failures on the extreme limits of natural fracture planes. In the second case without natural fractures, there are only tensile failures associated with the hydraulic fracturing process. The microseismic energy of these shear and tensile failure events are simulated by the modelling software and recorded as a function of time or amount of injected fluid. Analyzing the results from both cases reveals a significant difference in the number and magnitude of microseismic events. A variation in the coefficient b of the Gutenberg-Richter’s empirical formula for the magnitude-frequency relation is observed with and without natural fractures. Similar variations in b value measurements are observed using downhole microseismic measurements from hydraulic fracturing in horizontal shale wells. For the downhole microseismic magnitude-frequency relations, it is recommended to use the coefficient b of the Gutenberg- Richter’s empirical formula to indicate the presence or the lack of natural fractures in shale reservoirs.
Downhole, buried array or surface microseimic measurements are used to infer the size of the stimulated rock volume during hydraulic fracturing in vertical and horizontal wells. It is, however, often difficult to establish a significant correlation between microseismic events (dots in the box) and well performance. To better understanding the hydraulic fracturing process in complex rocks various numerical models are being developed. One such approach — discrete element modeling (DEM) — simulates the hydraulic fracturing process in a rock matrix with or without natural fractures. An output of this modeling method is the microseismic energy associated with either the microcrack of the host material and/or microslip along natural fracture/joint surfaces. The modeling of this microseismic energy is complex and needs to be validated.