Core and field scale modeling of miscible injection processes in fractured porous media using Random Walk and Particle Tracking methods

Abstract

In this thesis, we introduced and applied non-classical techniques to simulate miscible flow in fractured porous media. First, the Random Walk technique was modified to simulate miscible displacement in 2D fractured porous media at the lab-scale. The method was validated using a series of laboratory solvent injection experiments obtained from literature. Then, this model was modified to apply it for field-scale simulations and a sensitivity analysis was performed to identify the most critical parameters of the process. To validate the model, a tracer test done in the naturally fractured Midale field was used. Subsequently, the same fracture network system, which was calibrated against the tracer test results, was used to simulate the pilot CO2 injection applied in the same field. In this exercise, additional modifications to the algorithm were made including diffusive transfer between matrix and fracture. In the last part of the thesis, an approach was presented to scale up the production profiles obtained for a fractured reservoir. The exponents in the scaling equation were correlated to the fracture network properties such as fracture density, box-counting fractal dimension, mass fractal dimension, and fracture volume ratio.