부산대학교 도서관

The present study reports a numerical investigation of oil extraction from a pore-scale perspective using water as the injection fluid. The pore network is constructed from the statistical realization of the pore space of the reservoir rock. Conceptually, the pore network model used in the study miniaturizes the porous reservoir containing oil/gas onto a microfluidic platform, capturing the actual pore-level length scale and it complex features. The validity of the numerical model is established through an experimental investigation of single-phase flow. The experimentally calculated absolute permeability based on the Darcy law shows an excellent agreement with the numerically attained value. The two-phase numerical model uses the phase field technique to track the development of the interface between the two immiscible phases, i.e., oil and water. The numerical model shows a piston-like displacement and captures interesting pore-level phenomena like snap-off and trapping. Implementing the complete network for simulation reveals the unstable nature of the flooding, which is persistent with invasion percolation. The complete network simulation reveals the discontinuous flood front with a segregated flow configuration. Two-phase experiments conducted on a polydimethylsiloxane test chip with an equivalent pore level network also showed similar flow features, thereby establishing the credibility of our two-phase simulations. The recovery factor obtained from the simulation was found to be 0.78, which is in close agreement with experimental data reported in the literature. The modified Darcy law applied to the numerical model generates relative permeability plots similar to the experimental core flooding plots reported in the literature. The numerical model presented here provides valuable insight into the oil recovery process and its implications at field scale. To the best of our knowledge, this is the first instance involving numerical analysis of the full-scale system of Reservoir on a Chip system detailing the pore-level flow dynamics. [ABSTRACT FROM AUTHOR]