基于分子动力学及格子玻尔兹曼理论的多相渗流模拟新方法

Absolute permeability tensor and relative permeability curve, which are the two most important properties in prediction of subsurface flow dynamics, are both obtained from lab experiments traditionally. Although the measuring approaches have been widely used and results are largely accepted for many years in most cases, the lab experiments are usually expensive and cannot be repeated for different fluids or under different flow scenarios..Multi-phase modeling based on geometry information of cores obtained by Micro X-ray Computed Tomography becomes an emerging technology that tries to yield rock properties directly. Among all the methods of pore scale modeling, Lattice Boltzmann Method (LBM) shows an obvious advantage in terms of computational efficiency, readiness for parallel computing, and capability of modeling flow with complex boundary conditions. In spite that it has shown promising applications in single-phase flow modeling, the conventional LBM method suffers from tackling actual multi-phase problems due to theoretical lack in characterizing multi-phase flowing systems..In our research, we propose to connect Molecular Dynamic (MD) simulation with Lattice Boltzmann method to solve this problem. The basic idea is to first construct the molecular model based on the actual components of the rock-fluid system. Then MD simulation is performed to compute the equilibrium state of the system, upon which to obtain cohesive and repulsive forces between the rock and the fluid and fluids of different densities. MD simulation results indicate that the composition of the forces is a surface force as a nonlinear function of fluid density. It is in the form of cohesion when the fluid density is low, and repulsion when the fluid density is high. This correlation of force and density is then used in LBM modeling to obtain the absolute permeability of reservoir rocks and relative permeability of rock-fluid systems. In this way, a bridge linking flow simulation in micro, meso and macro scales is built up. This is of great value for theoretical studies as well as practical applications..We have validated this approach by simulating a two-phase separation process. The success in MD-LBM results in align with published EOS solution results demonstrated a breakthrough in pore-scale, multi-phase flow modeling. Based on an actual x-ray CT image of a reservoir core, we applied our workflow to calculate absolute permeability of the core, vapor-liquid H2O relative permeability and capillary pressure curves. With a more complex model considering realistic factors, the ultimate goal is to develop an accurate method for prediction of permeability tensor, relative permeability and capillary curves based on 3D CT image of the rock, actual fluid and rock components.

多孔介质中多组分多相流体的渗流问题，是油气田开发、化工生产、CO2及核废料等地下埋存的基础理论问题。与常规流体力学问题相比，多相渗流问题由于理论不够完备，符合孔隙精细微观结构分布的模拟计算无法开展。本研究将从数字岩心成像描述出发，以格子波尔兹曼方法为模拟手段，通过扩展的状态方程及分子动力学分别计算多孔介质中流体及流固分子间的作用力，并以此为基础建立多组分多相格子波尔兹曼模型，最终提出一套将分子动力学、微格子玻尔兹曼理论，以及达西定律相结合的理论方法，用精细的力学计算获得岩心的物性参数以及岩石-流体的耦合作用特征曲线，克服传统多相流实验时间长、样品稀缺、精度低、不可重复的弱点，为油藏数值模拟提供高精度、更加符合实际的渗透率、相对渗透率及毛管压力数据输入；同时为深入研究油气藏提高采收率机理如三元复合驱、二氧化碳驱、气体滑脱等，以及煤层气、页岩气开发中的流固耦合过程提供系统、科学的理论工具。

格子Boltzmann方法可以处理复杂边界内的流体流动，且程序结构简单、并行性好，非常适合进行孔隙尺度的流动模拟。目前的格子Boltzmann模型能够有效处理单组分气液两相流问题，并获得与热力学理论一致的结果。但现有的多组分格子Boltzmann模型还无法使模拟物质与实际物质（例如原油和天然气中的烷烃、芳烃等）进行对应，同时目前的格子Boltzmann方法无法正确处理流固边界间的相互作用力。这极大的限制了格子Boltzmann方法在实际中的应用。.鉴于此，本文以多组分烃类状态方程为基础，开发出一个新的多组分气液两相流格子Boltzmann模型，该模型能够准确计算烃类混合物气液平衡时气相和液相的相关性质。同时通过引入烃类混合物粘度模型，正确处理了混合物粘度对流动的影响。同时，将分子动力学模拟与格子Boltzmann模型结合，用于精确计算矿物表面与流体之间的相互作用力。.为验证该模型的有效性，本文用其进行了烃类混合物气液相分离模拟，并将模拟获得的气液相密度、表面张力系数、接触角等与实验值或理论解进行对比，证明其能够正确模拟烃类混合物气液相平衡行为；在此基础上，利用其进行了重力驱动的油气两相层流模拟，并将模拟获得的速度剖面与理论解进行对比，证明其能够正确处理混合物粘度对流动的影响。.在模型建立完成后，本文以实际油田的井流物组成及岩心孔隙结构为基础，利用新模型对压差驱动下的多组分单相流、重力驱动下的多组分气液两相流、干气驱替原油以及原油脱气等不同机理的流动现象进行了模拟和分析。模拟结果表明，该模型能够很好地捕捉到孔隙尺度上油气流动的特殊现象，反映了油藏开发中的相关规律。因此，该模型能够满足石油工业中孔隙尺度上多组分烃类气液两相流模拟的需要。.利用该模型，除可以真实重现原油和天然气在岩心多孔介质内的流动过程，从而对气体滑脱、非牛顿流、非达西流、滞后效应、启动压力等现象的微观机理进行研究外，还能够在一定程度上替代传统的岩心渗流实验，计算岩心的相渗曲线、毛管力曲线等。因此本文工作对实际有一定的指导意义，且为后续的研究奠定了基础。