VOF(Volume Of Fluid)方法能够通过在欧拉网格上使用离散的体积分数域表示光滑界面,在不可混合流体的数值模拟中得到广泛应用。针对多相流仿真中的液滴曲率计算问题,开发了一种计算界面曲率的算法。首先提出了一种新的数据生成方法,在...VOF(Volume Of Fluid)方法能够通过在欧拉网格上使用离散的体积分数域表示光滑界面,在不可混合流体的数值模拟中得到广泛应用。针对多相流仿真中的液滴曲率计算问题,开发了一种计算界面曲率的算法。首先提出了一种新的数据生成方法,在液滴界面上进行随机采样,增强网格内体积分数的信息量,并调整取值范围以覆盖正负曲率。然后改进了传统的深度神经网络(DNN)模型,使其在计算曲率时保持对称性。基于VOF方法与该模型,利用目标单元及邻近单元体积分数计算曲率。最后选取最优模型并应用于Basilisk软件中,以提高计算曲率的准确性和稳定性。测试结果表明,其计算曲率时准确稳定。在计算小半径液滴曲率时,误差减小了25%至50%,并能用于液滴融合仿真,证明了其应用价值。展开更多
Gas-liquid two-phase flow in fractal porous media is pivotal for engineering applications,yet it remains challenging to be accurately characterized due to complex microstructure-flow interactions.This study establishe...Gas-liquid two-phase flow in fractal porous media is pivotal for engineering applications,yet it remains challenging to be accurately characterized due to complex microstructure-flow interactions.This study establishes a pore-scale numerical framework integratingMonte Carlo-generated fractal porousmedia with Volume of Fluid(VOF)simulations to unravel the coupling among pore distribution characterized by fractal dimension(Df),flow dynamics,and displacement efficiency.A pore-scale model based on the computed tomography(CT)microstructure of Berea sandstone is established,and the simulation results are compared with experimental data.Good agreement is found in phase distribution,breakthrough behavior,and flow path morphology,confirming the reliability of the numerical simulation method.Ten fractal porous media models with Df ranging from 1.25~1.7 were constructed using a Monte-Carlo approach.The gas-liquid two-phase flow dynamics was characterized using the VOF solver across gas injection rates of 0.05-5m/s,inwhich the time-resolved two-phase distribution patternswere systematically recorded.The results reveal that smaller fractal dimensions(Df=1.25~1.45)accelerate fingering breakthrough(peak velocity is 1.73 m/s at Df=1.45)due to a bimodal pore size distribution dominated by narrow channels.Increasing Df amplifies vorticity generation by about 3 times(eddy viscosity is 0.033 Pa⋅s at Df=1.7)through reduced interfacial curvature,while tortuosity-driven pressure differentials transition from sharp increases(0.4~6.3 Pa at Df=1.25~1.3)to inertial plateaus(4.8 Pa at Df=1.7).A nonlinear increase in equilibrium gas volume fraction(fav=0.692 at Df=1.7)emerges from residual gas saturation and turbulence-enhanced dispersion.This behavior is further modulated by flow velocity,with fav peaking at 0.72 under capillary-dominated conditions(0.05 m/s),but decreasing to 0.65 in the inertial regime(0.5 m/s).The work quantitatively links fractal topology to multiphase flow regimes,demonstrating the critical role of Df in governing preferential pathways,energy dissipation,and phase distribution.展开更多
文摘VOF(Volume Of Fluid)方法能够通过在欧拉网格上使用离散的体积分数域表示光滑界面,在不可混合流体的数值模拟中得到广泛应用。针对多相流仿真中的液滴曲率计算问题,开发了一种计算界面曲率的算法。首先提出了一种新的数据生成方法,在液滴界面上进行随机采样,增强网格内体积分数的信息量,并调整取值范围以覆盖正负曲率。然后改进了传统的深度神经网络(DNN)模型,使其在计算曲率时保持对称性。基于VOF方法与该模型,利用目标单元及邻近单元体积分数计算曲率。最后选取最优模型并应用于Basilisk软件中,以提高计算曲率的准确性和稳定性。测试结果表明,其计算曲率时准确稳定。在计算小半径液滴曲率时,误差减小了25%至50%,并能用于液滴融合仿真,证明了其应用价值。
基金funded by the National Key R&D Program of China,China(Grant No.2023YFB4005500)National Natural Science Foundation of China,China(Grant Nos.52379113 and 52379114).
文摘Gas-liquid two-phase flow in fractal porous media is pivotal for engineering applications,yet it remains challenging to be accurately characterized due to complex microstructure-flow interactions.This study establishes a pore-scale numerical framework integratingMonte Carlo-generated fractal porousmedia with Volume of Fluid(VOF)simulations to unravel the coupling among pore distribution characterized by fractal dimension(Df),flow dynamics,and displacement efficiency.A pore-scale model based on the computed tomography(CT)microstructure of Berea sandstone is established,and the simulation results are compared with experimental data.Good agreement is found in phase distribution,breakthrough behavior,and flow path morphology,confirming the reliability of the numerical simulation method.Ten fractal porous media models with Df ranging from 1.25~1.7 were constructed using a Monte-Carlo approach.The gas-liquid two-phase flow dynamics was characterized using the VOF solver across gas injection rates of 0.05-5m/s,inwhich the time-resolved two-phase distribution patternswere systematically recorded.The results reveal that smaller fractal dimensions(Df=1.25~1.45)accelerate fingering breakthrough(peak velocity is 1.73 m/s at Df=1.45)due to a bimodal pore size distribution dominated by narrow channels.Increasing Df amplifies vorticity generation by about 3 times(eddy viscosity is 0.033 Pa⋅s at Df=1.7)through reduced interfacial curvature,while tortuosity-driven pressure differentials transition from sharp increases(0.4~6.3 Pa at Df=1.25~1.3)to inertial plateaus(4.8 Pa at Df=1.7).A nonlinear increase in equilibrium gas volume fraction(fav=0.692 at Df=1.7)emerges from residual gas saturation and turbulence-enhanced dispersion.This behavior is further modulated by flow velocity,with fav peaking at 0.72 under capillary-dominated conditions(0.05 m/s),but decreasing to 0.65 in the inertial regime(0.5 m/s).The work quantitatively links fractal topology to multiphase flow regimes,demonstrating the critical role of Df in governing preferential pathways,energy dissipation,and phase distribution.
