Stochastic Pore-Scale Network Modeling And Multiphase Flow Simulation

To study microscopic fluid flow mechanism in porous media is a critical measure to reveal the effect of rock microstructure behavior on fluid multiphase flow. Network model and multiphase flow analog model of pore level are the platform for researching microscopic flow theory and is currently the hot international research topic which is systematacially and deeply investigated in this issue.The already existed generation algorithm of stochastic pore-scale network modeling is analysed in this issue and the modified generation algorithm of stochastic pore-scale network modeling of equivalent topology is proposed. The modified generation algorithm can be combined with analog pore-scale network multiphase flow modeling which can characterize complex phenomena in multiphase flow, such as the hysteresis in both capillary pressure and relative permeability and wettability reversal phenomena, to form a stochastic network simulation software of core flow and also build the platform for the micro-scale study of fluids flow mechanism in porous media.Pore-scale network simulation of multiphase flow reveals the mechanism of micro-scale displacement. For non-wetting phase displacing wet phase, the dominant piston like displacement is formed, while the displacement method for wetting phase displacing non-wetting phase becomes very cumbersome because of the complex structure of the pore space, then piston type, the snap-off type and pore filling are possible. Different ways of displacement is to form different pattern of remaining oil.To carry out the study properties of multiphase flow in the real core by applying the pore network model is the development direction of the pore-scale network modeling. The method of getting topological and wettability pore-scale network modeling of the real core is proposed in this paper by matching the experimental capillary pressure, wettability index. The capillary pressures of two different pore configuration cores are matched and the relative permeability is predicted respectively. The comparative analysis of predicted results and physical experiment on relative permeability suggests that pore-scale network modeling obtained by matching method can accurately characterize the pore structure of the real core. The comparative analysis results demonstrate the feasibility and practicality of this method and find a scientific method for the quantity test of microscopic flow by the pore-scale network modeling.The effect of pore those factors including pore throat size, topology of pore space, wettability, degree of wetting and primary water saturation on the relative permeability capillary pressure and degree of reserve recovery by developing the stochastic pore-scale network modeling of equivalent topology based on the modified generation algorithm have been analyzed. In comparison with macroscopic numerical simulation results (E100), the network generation algorithm mentioned in the issue has validity and reliability and can build a bridge to convert the microscopic analog results into the macroscopic available parameters.Finally, clay swelling modeling with wettability is developed and combined with the pore network modeling, from the microscopic scale, damage of clay swelling due to wettability on the formation have been investigated.Through research the formation of micro-flow simulation software provide a basic research platform for carrying out micro-scale flow simulation in porous media. Further deepening the understanding of the micro-scale displacement mechanism and providing a bridge for macro-micro simulation and a theoretical basis for enhanced oil recovery study.