An Integrated Approach for Simulation of Two-phase flows in Naturally Fractured Reservoirs using Embedded Discrete Fracture Model and Aerial Image Processing

Naturally fractured reservoirs are ubiquitous worldwide and form majority of the oil and gas fields in Iran. Fracture networks in some cases can be modeled using simplistic approaches such as the so-called dual-medium models. However, these models perform poorly in cases where fracture dimensions are comparable to those of the reservoir itself. In such cases, the state-of the-art computational approaches use the Discrete Fracture model. This paper introduces an integrated approach for simulating two-phase flows in naturally fractured reservoirs. The proposed computational process comprises two key steps: 1) generating a representation of the fracture network from aerial images image processing techniques, 2) conducting two-phase flow simulations using the Embedded Discrete Fracture Method (EDFM). The reservoir contains only oil and water as immiscible incompressible fluids, with the assumption that the fracture system matches surface outcrops observations. Python libraries are utilized for Image processing to convert aerial images into a fracture system defined by line segments. The effect of various tuning parameters, such as image resolution and edge detection, on the final fracture model is investigated. Once a robust fracture system representation is achieved, the reservoir geometry is defined and boundary conditions are set. The problem is solved using the Matlab Reservoir Simulation Toolbox (MRST), which includes an EDFM module for accurate analysis of naturally fractured reservoirs by incorporating detailed information, yielding precise fluid flow predictions. All types of fracture-matrix and fracture-fracture intersections and fluid exchanges between matrix and fractures are incorporated in the EDFM. Simulation results are verified against a benchmark problem, and the influence of rock and fracture physical properties on the flow field is examined. The study finds that rock permeability significantly affects fluid flow compared to fracture aperture or permeability. For instance, a 3 percent decrease in oil production is observed when increasing fracture aperture from 0.15 (mm) to 0.21 (mm), while increasing the matrix permeability from 1 to 20 millidarcy leads to a 22 percent increase.