Microseismicity-constrained discrete fracture network models for stimulated reservoir simulation

Abstract

The effectiveness of hydraulic fracture stimulation in low-permeability reservoirs was evaluated by mapping microseismic events related to rock fracturing. The geometry of stage by stage event point sets were used to infer fracture orientation, particularly in the case where events line up along an azimuth, or have a planar distribution in three dimensions. Locations of microseismic events may have a higher degree of uncertainty when there is a low signal-to-noise ratio (either due to low magnitude or to propagation effects). Low signal-to-noise events are not as accurately located in the reservoir, or may fall below the detectability limit, so that the extent of fracture stimulated reservoir may be underestimated. In the Bakken Formation of the Williston Basin, we combined geologic analysis with process-based and stochastic fracture modeling to build multiple possible discrete fracture network (DFN) model realizations. We then integrated the geologic model with production data and numerical simulation to evaluate the impact on estimated ultimate recovery (EUR). We tested assumptions used to create the DFN model to determine their impact on dynamic calibration of the simulation model, and their impact on predictions of EUR. Comparison of simulation results, using fracture flow properties generated from two different calibrated DFN scenarios, showed a 16% difference in amount of oil ultimately produced from the well. The amount of produced water was strongly impacted by the geometry of the DFN model. The character of the DFN significantly impacts the relative amounts of fluids produced. Monitoring water cut with production can validate the appropriate DFN scenario, and provide critical information for the optimal method for well production. The results indicated that simulation of enhanced permeability using induced microseismicity to constrain a fracture flow property model is an effective way to evaluate the performance of reservoirs stimulated by hydraulic fracture treatments.

A sensitivity analysis of the effect of pumping parameters on hydraulic fracture networks and local stresses during shale gas operations

“…Previous sensitivity studies have considered the effect on fracture volume, aperture and porosity when changing fracture length [25]; the effect on complexity of the fracture network when altering cohesion, the in-situ stress ratio, internal friction angle and injection rate [26,27] and the effect on the lateral respect distance (the lateral distance from a known fault that hydraulic fracturing should occur in order not to reactivate the fault) for different fracture intensities [19]. This paper adds to these investigations.…”

“…Sensitivity studies have investigated whether the total fracture volume, aperture, and porosity are sensitive to the fracture length [25] and the effect that cohesion, the in situ-stress ratio, the internal friction angle [26] and injection rate [26,27] have on the natural fractures. It was found that the total fracture volume, aperture, and porosity are not sensitive to the fracture length.…”

Hydraulic Fracturing

“…The mechanical interference can be demonstrated by experiments, eg, in Reference 42. Both analytical and numerical models of various degree of sophistication have been developed to model interference (43)(44)(45) A discrete element formulation (46) is capable of handling nonplanar fracture growth, toughness, and viscosity-controlled propagation regimes, and even the situation where fluid pressure is not sufficient to drive the fluid into the newly developed fracture tip (fluid lag.) (Fracture network) Inspired by the real-time experience provided by diagnostics, considerable efforts have been focused on describing the evolution of fracture networks in heterogeneous environment.…”

A review of stimulated reservoir volume characterization for multiple fractured horizontal well in unconventional reservoirs

“…if S/N is big enough, the inversion of source mechanism is applied to the seismic data to characterize the failure-plane orientation and slip for almost all the detected seismic traces. Williams-Stroud et al, 2013) represents the result of final processed events of a well. Events showed in the image are colored with respect to fracture stage and sized with respect to different relative energy.…”