Numerical model sensitivity to matrix grid refinement in dual porosity fractured reservoirs application to: improved oil recovery in waterflooding

Abstract

In this thesis, I present numerical model sensitivity to matrix refinement in dual-porosity formulation of fractured reservoirs. A dual-porosity fractured reservoir consists of a dominant, interconnected network of fractures embedded in a large number of isolated permeable sediments known as matrix blocks. Two practical situations are considered. The first is the effect of matrix refinement on the unsteady-state pressure solution, and the second situation is modeling water-oil, Buckley-Leverett (BL) displacement in waterflooding a fracture-dominated flow domain. The BL displacement was formulated in an ideal, 1D waterflood situation involving either water imbibition into water-wet matrix or low-salinity osmotic water imbibition into matrix. It will be shown that the amount of water entering a matrix block is highly sensitive to grid refinement of the matrix blocks. Based on physical reasoning, matrix refinement is consistent with dominance of mass transfer in a narrow region inside of the fracture-matrix interface. In absence of grid refinement, the mass transfer will be diffused into the entire matrix—not consistent with physics of flow. The results of the 1D formulation of grid refinement in this thesis clearly indicates its importance in any IOR or EOR process—water imbibition, chemical osmosis, or otherwise. The sensitivity of matrix refinement is illustrated in a one-ring matrix block refinement in a heterogenous reservoir system using three different matrix block sizes in each computational grid block. Finally, our results indicate that matrix refinement is significant in tight matrix in fractured reservoirs compared to the high-permeability matrix because refinement captures the true nature of mass transport at the fracture-matrix interface consistent with physical reasoning. In summary, any credible dual-porosity, matrix-refined model is a great tool to study mass transport across fracture-matrix interface. Examples include capillary imbibition, chemical osmosis, and molecular diffusion mass transport. However, the models developed in this thesis include only the code for capillary imbibition (and/or chemical osmosis imbibition). Chemical osmosis is a special case of molecular diffusion mass transport across a semi-permeable membrane which can be formulated in a similar manner to the capillary imbibition mass transport between fracture and matrix (Fakcharoenphol et al. 2014; Torcuk et al. 2019). In summary, a practical contribution of my research is to assess the contribution of matrix grid refinement in evaluating conventional waterflooding and improved oil recovery by low-salinity waterflooding.