MPPICInterFoam中的bug？

李东岳

一楼中提到的bug，fvc::ddt(alpha)的值取决于时间格式，将其换为backward就可以了

更新下：问题在于alphac.oldTime()。

在Euler中，计算的是

(alphac - alphac.oldTime())/delta T

Euler不起作用的原因在于alphac.oldTime()=0。

backward是可以解决问题。但是在backward中调用的是三层时间步。即vf、vf.oldTime()和vf.oldTime().oldTime()。起作用的是vf.oldTime().oldTime()。alphac.oldTime()还依然是0。

但是，问题在于alphac，其他的场oldTime()正确。

李东岳

在我自己的求解器中，我验证了双向耦合中界面交换相、相分数的耦合作用大小。在我测试的算例中，相分数的耦合作用较小。耦合主要体现在彼此之间的力交互。

但这绝对是一个bug

linhan.ge

@东岳当颗粒位于界面处时，MPPICInterFoam是不是没考虑颗粒对界面的影响？因为MPPICInterFOAM中的相方程还是跟interFOAM中一样的吧？MULES那块不怎么看得懂。

linhan.ge

@东岳这里有个问题，mppicInterFOAM中，alpha1的表达式是什么？是占整个流体的体积分数alpha1=V1/( V1+V2)。还是占三相的？ alpha1 = V1/(V1+V2+Vparticle).

linhan.ge

@东岳这个bug能否通过手动赋值解决，alphac.oldTime() = alphac ?

李东岳

现在手头没有DPM算例，要不你测试一下

李东岳

简单测试了一下，alphac.oldTime() = alphac不能解决，留给其他感兴趣的朋友们，谁有兴趣修复一下。

但是DPMFoam中没有这个问题。

溯光

@linhan-ge 请问这是哪篇文献中的控制方程

linhan.ge

@溯光 Jing, L., Kwok, C.Y., Leung, Y.F., Sobral, Y.D., 2016. Extended CFD-DEM for free-surface flow with multi-size granules. International Journal for Numerical and Analytical Methods in Geomechanics 40, 62-79.

溯光

@linhan-ge 在 MPPICInterFoam中的bug？中说：

Extended CFD-DEM for free-surface flow with multi-size granules.

谢谢

白礼耕

东岳老师，我的印度同事也经常逛CFD中文网，平时都是用谷歌翻译阅读。他也发现了这个问题（气泡不会让交界面的位置升高），想问您下面的问题：
When particles are injected, the liquid level is not changing. What could be the reason? Is it the ddt(alphac)=zero the reason?
多谢您了！

李东岳

@白礼耕 Check the alpha.Eqn and p.Eqn:

fvScalarMatrix p_rghEqn
(
    fvm::laplacian(Dp, p_rgh)
  ==
    fvc::ddt(alphac) + fvc::div(alphacf*phiHbyA)
);

even ddt(alphac)=0, it considers only the continuous phase (water) because alphac = 1 - alpha.air. It makes sense if you do the following to address the effects of particles: 1) include the sink terms in alpha.Eqn, 2) alphac = 1 - alpha.air - alpha.particle 3) manage the momentum interface exchange terms in U.Eqn. In this way a two-way coupling can be implemented, but not quite easy. Currently the implementation is quite simple, but it only works for dilute particle flow (one-way coupling).

白礼耕

@东岳 Thanks for your help!

In the paper given above, the definition of alphac (=alpha.air + alpha.water) is different. I want to confirm exactly what it is. Is it not following the paper?

comment on 3rd point: I have already done that.

李东岳

@白礼耕在 MPPICInterFoam中的bug？中说：

In the paper given above, the definition of alphac (=alpha.air + alpha.water) is different. I want to confirm exactly what it is. Is it not following the paper?

What is different?

白礼耕

@东岳 The expression in the paper (Fig. 2 in paper "Extended CFD–DEM for free-surface flow with multi-size granules", which is given as alphaf) is alphac = 1-alpha.particle which is different from what you provided (alphac = 1 - alpha.air - alpha.particle). I just wanted to clarify which one is in the code?

Thank you again!

李东岳

code: alpha.water = 1-alpha.air. Particles play no role

白礼耕

@东岳 Thanks again for your help! 谢谢

zhe

@东岳 Currently, I am working on MPPICInterFoam. I thought alphac is the volume fraction of fluid ( not just water or air). If there is no particles in the system, alphac = alpha1 + alpha 2 = alpha.water + alpha.air=1. If with particles, alphac = 1-theta = 1- alpha.particles. The particles are accounted up. We can not just say alphac is alpha.water (which alphac = 1 - alpha.air - alpha.particle). Please fix it if I said wrong. Thank you.

AppleKiller

@linhan-ge 我觉得也应该是这样，但我还没理解李老师说的，您理解了吗？