Percolation Network Study On Hydrodynamic Dispersion

Hydrodynamic dispersion,as one of the basic modes of mass transport,originates from the study of groundwater,and has subsequently been widely used in chemical engineering,chromatography,hydrogeology,and petroleum engineering.Domestic and foreign scholars have conducted extensive researches on the mechanisms of hydrodynamic dispersion and the non-Fickian behavior of solute transport,but there is a lack of research on the influencing factors of hydrodynamic dispersion.For this reason,the upscaling simulation of hydrodynamic dispersion is realized under the periodic boundary conditions.At the same time,based on the three-dimensional percolation network model,a large number of numerical simulations are implemented to study the basic laws of solute transport in single phase flow and to clarify the influence of the microscopic pore structures or fluid flow characteristics on the miscible displacement.The specific research works and results are as follows:(1)The microscopic pore structures of reservoir rock(i.e.,pore radius,the distribution of pore radii,pore connectivity,pore length)were qualitatively or quantitatively analyzed using the methods such as cast thin section,electron microscope scanning,mercury porosimetry and nuclear magnetic resonance,which is the prerequisite for the construction of the percolation network model.(2)The influences of the flow mechanisms(i.e.,average flow or Taylor-Aris dispersion)in the single pipe and the mixing mechanisms(i.e.,complete mixing or streamline rounting)at the node on the solute transport are clarified in the 2D pore network backbone model.Based on this,the upscaling network simulation method of hydrodynamic dispersion was determined.(3)The migration process of solute particles under different pore structures was simulated in the three-dimensional percolation network backbone model.The method of moments was used to determine the longitudinal and transverse dispersion coefficietns.The universal relationships between the longitudinal and transverse dispersion coefficietns and microstructure characteristics of porous media were studied by percolation theory.Then the quantitative models of the longitudinal and transverse dispersion coefficietns were established.According to the method of moments,it is found that the third moment of the solute cloud position distribution is negative,indicating that the asymptotic transport regime was intrinsically non-Fickian behavior.At the same time,the effects of pore heterogeneity and pore connectivity on the non-Fickian behavior are discussed.(4)The Euler method was used to count the fluid velocity,fluid flow rate and Taylor-Aris dispersion coefficient at the pore scale.The distribution types of these three properties were analyzed under the different disorder of porous media.It is found that the distribution of flow rate determines the main path of fluid and solute transport.The Lagrangian method was used to count the velocity,flow rate,and time distribution of solute transport.Besides,the effects of high flow velocity and low flow velocity on non-Fickian were discussed.(5)Through the coupling solution of the pressure field and concentration field of the network model,the dynamic network simulation method of miscible displacement considering hydrodynamic dispersion is proposed.By comparing the results of miscible displacement with or without hydrodynamic dispersion,the effects of hydrodynamic dispersion on miscible displacement were determined.Besides,the influences of the pore heterogeneity,pore connectivity,viscosity ratio,and displacement speed on miscible displacement with hydrodynamic dispersion were discussed.Depending on all the studies above,the upscaling network simulation method of hydrodynamic dispersion was proposed,and the quantitative relationships between the microscopic structures of porous media and the longitudinal and transverse dispersion coefficietns was derivated,and the effects of hydrodynamic dispersion on miscible displacement were clarified.Theses studies will further enrich and improve the existing hydrodynamic dispersion theory.