Fracture Detection and Numerical Modeling for Fractured Reservoirs

The subsurface fractures could impact the fluid mechanisms dramatically, which makes the modeling of the hydraulic and natural fractures an essential step for fractured reservoirs simulations. However, because of the complexities of fracture patterns and distributions, it is difficult to detect and quantify the fracture networks. In this study, line detection techniques are designed and applied to quantify the fracture segments from fracture figures. Using this fracture detection algorithm, the fracture segments could be located by detecting the endpoints and the intersections of fractures, thus that the fracture patterns could be accurately captured and characterized. The proposed method is applied to two previous well-known field cases and the pressure distribution results are consistent with the micro-seismic data profiles. These two field cases are simulated and computed by using a semianalytical model and Embedded Discrete Fracture Model (EDFM) respectively. The third case is constructed by the fracture outcrop figure and simulated by a numerical simulator with EDFM implemented. The simulation results are accurate and clearly illustrate the important role fractures play in unconventional reservoirs. The technology proposed in this study could be used to quantify the fracture input data for reservoir simulations and be easily expanded for fracture detection and characterization problems in other fields.