To obtain sustainable economical oil production and recovery of investment, commingling production has been widely used in multi-layer oil reservoirs. However, the characteristics of oil-water flow in porous media have long been neglected, making variations in multi-layer co-production (MLCP) difficult to anticipate. This paper concentrates on complex seepage and pore throat characterizations, as well as the construction of a prediction model capable of monitoring the dynamic behavior of MLCP in microscopically variable porous media. More specifically, high-pressure mercury injection (HPMI) and nuclear magnetic resonance (NMR) were used to characterize pore throat sizes and distributions, and a capillary bundle model was used to assess water displacement seepage resistance. In the process of continuous water parallel displacement, the changes in seepage resistance induced by throat altering and coupling boundary layer effects were especially explored. As a consequence, using the time-node analysis approach, a thorough mathematical model was built and confirmed by comparing experimental results. With errors of 3.94 % and 1.62 %, the projected oil recovery and water cut from the created model are in excellent agreement with actual findings.
Keywords Multi-layer co-production, porous media, Seepage resistance, Boundary layer, Dynamic prediction mathematical model.