Effects of Numerical Treatment of Viscous and Surface Tension Forces on Predicted Motion of Interface
Effects of Numerical Treatment of Viscous and Surface Tension Forces on Predicted Motion of Interface
复制标题
DOI：
10.1260/1757-482x.6.2.111
复制
发表时间：
2014-06
期刊：
影响因子：
--
通讯作者：
K. Hayashi;A. Tomiyama
中科院分区：
文献类型：
--
作者：
K. Hayashi;A. Tomiyama
研究了粘性力和表面张力的数值处理对界面预测运动的影响。粘性力和切向表面张力（马兰戈尼力）有两种不同的处理方法：一种是涂抹界面法，另一种是鬼流体法。在抹去界面法中，采用算术平均值和调和平均值来评估粘滞应力。用这些方法模拟了线性剪切流、单振荡液滴和表面张力波。这些基准试验表明，鬼流体法对粘性力和马兰戈尼力给出了最准确的评价。谐波平均值能较好地评价清洁界面的切向粘性应力，而污染界面的切向粘性应力误差较大。虽然相对于鬼流体法，该算法对洁净界面和污染界面会产生一定的误差，但可以通过增加空间密度来减小误差。
Effects of numerical treatment of viscous and surface tension forces on predicted motion of an interface are investigated. The viscous force and tangential surface tension force (Marangoni force) are treated in two different ways: one is a smeared-out interface method and the other is a ghost fluid method. In the smeared-out interface method, the arithmetic mean and harmonic mean are tested for evaluation of the viscous stress. Linear shear flows, single oscillating drops and surface tension waves are simulated using these methods. These benchmark tests show that the ghost fluid method gives the most accurate evaluations of the viscous and the Marangoni forces. The harmonic mean can give good evaluations of the tangential viscous stress at clean interface, whereas large errors in the viscous stress are caused for contaminated interface. Although the arithmetic mean causes some errors for clean and contaminated interfaces compared to the ghost fluid method, the error can be reduced by increasing spatial re...