Multiphase flow in porous media for digital rocks

jesantos · November 13, 2019, 9:22pm

Hi all,

I recently created a repository with my code for simulating multiphase flow through porous material (steady and unsteady approaches) Using the Shan Chen model. Feel free to visit it and let me know what you think.

video:

Comments and suggestions are appreciated

Best,
Javier.

liuqi · May 19, 2020, 3:28am

hi，
I have recently been studying multiphase flow in porous media，I read your code on guthub，You wrote it very well, but I have a question, has palabos incorporated the Gads parameters, the fluid and solid forces.I didn’t find the code for the Gads parameter call in palabos. Do I need to write the fluid and solid forces by myself?
Thank you for your reply
Best

jesantos · May 19, 2020, 5:03am

Hi @liuqi,

Thank you for your kind words. I have been working very hard to make a clean interface for people to build on top of it. For porous media applications, the G_ads parameter is very important, since it controls the wettability of the system, by setting a specific contact angle at the solid surface. Different contact angles can be set by printing different numbers at the solid surface. I currently support up to 5 different contact angles, by I can add more if necessary.

To do this, you will need to assign the numbers 1,3,5,6 through your geometry (as done here https://github.com/je-santos/MultiphasePorousMediaPalabos/blob/0ecb2d4977f69ae5b455c1de6a8d3c2bbdf03ecd/examples/steady_relperm_spherepack/create_geom_4_2phase.m#L42), and then add the contact angles that you want in those voxels here https://github.com/je-santos/MultiphasePorousMediaPalabos/blob/0ecb2d4977f69ae5b455c1de6a8d3c2bbdf03ecd/examples/steady_relperm_spherepack/input_spherepack_S25.xml#L39

An deeper explanation of the process is provided in this readme:

github.com

je-santos/MultiphasePorousMediaPalabos/blob/master/examples/readme.md

# English explanation of inputs for 2 phase flow XML file (input_sc.xml)
## followed by the explaination in Chinese

<img src="https://github.com/je-santos/MultiphasePorousMediaPalabos/blob/master/illustrations/LBM%20geometry%203D.png" align="middle" width="600" height="400" alt="geometry inputs explanation">

Figure 1- Geometry setup for 2-phase flow simulations


**The inputs are explained in the same sequence as the xml file**

**load_savedstated:** This input gives the user the option of loading a previous incomplete simulation (True), or starting a new simulation (False).

**geometry:** There are multiple inputs required under this heading.

**file_geom:** This input asks for the name of the geometry for simulation. PALABOS requires the input geometry be in .dat file format which can be created using the [pre-processing steps](https://github.com/je-santos/MultiphasePorousMediaPalabos/tree/master/pre-processing) The geometry file should be placed in the same folder as the [2-phase simulation code](https://github.com/je-santos/MultiphasePorousMediaPalabos/tree/master/src/2-phase_LBM) or if placed elsewhere, the path should be modified to point to the geometry.

**size:** This input requires the size (in voxels) in the X, Y, and Z directions of the geometry (Figure 1).

*Note:*

This file has been truncated. show original

Thanks for your interest

jesantos · May 25, 2020, 5:35pm

@liuqi if you want a simple example to start experimenting with different contact angles, change this value:

github.com

je-santos/MultiphasePorousMediaPalabos/blob/be6fafdc75277f7cb10ef9fb5b4143d637af0101/examples/validation_YoungLaplace/input_tubes_1e0.xml#L39


</init>





<fluids>

    <Gc> 0.9 </Gc>

    <omega_f1> 1 </omega_f1>

    <omega_f2> 1 </omega_f2>

    <force_f1> 0 </force_f1>

    <force_f2> 0 </force_f2>

    <!-- Wetting forces Huang et al., 2007. -->

    <G_ads_f1_s1> -0.4 </G_ads_f1_s1>

    <G_ads_f1_s2>  0.0 </G_ads_f1_s2>

    <G_ads_f1_s3>  0.0 </G_ads_f1_s3>

    <G_ads_f1_s4>  0.0 </G_ads_f1_s4>



	  <!-- Intial density of fluid 1 -->

    <rho_f1> 2 </rho_f1>

    <!-- Initial density of fluid 2 -->

    <rho_f2> 2 </rho_f2>

    <!-- Include a pressure gradient bc (for flow)-->

    <pressure_bc> True </pressure_bc>

The ranges of it should be from -0.4 to 0.4 (non-wetting to wetting, drainage->imbibition) according to Huang et al., 2007.

liuqi · May 26, 2020, 12:58am

hi，friend
I’ve figured out how to simulate forces by setting a virtual density on the boundary.Thank you very much, I learned a lot from your code, and I have a question about the two-phase boundary condition, which boundary is better for export in palabos when the import is constant speed?Or can you provide me with some code for boundary conditions that I can learn about?Thank you for your guidance and reply

Taihengliu · June 12, 2020, 8:31am

Hello, I’ve found your code base, and I think it’s good. It fits my needs.But I didn’t run it successfully. Can you help me?Thanks

jesantos · June 15, 2020, 4:13pm

Sorry @Taihengliu I did not get a notification for this.
Have you tried running the matlab code that creates the geometry?

Taihengliu · August 15, 2020, 6:58am

Hi, what am I going to do at the end of the unsteady_relperm_spherepack operation? I just looked at the generated data using ParaView.
Thank you for your reply.

lkadzungura · August 16, 2020, 4:14pm

Hallo guys, i am new to palabos and i am still trying to figure out some codes. i have encountered a hurdle as i was trying to run an old code on which runs well on the older version of palabos but failing to build using cmake on the recent palabos version. i am getting the following response on the terminal. please help.

mars · August 17, 2020, 1:58pm

Hello lkadzungura,

welcome tho the Palabos Forum!

Try to substitute the deprecated auto_ptr with the unique_ptr (or the shared_ptr), I think it could solve the issue.

I hope this helps,

Francesco

lkadzungura · August 18, 2020, 9:06am

Hie Mars

you have helped a lot, thanx. it worked but now i have to look for the executable file that it makes. that is not a problem though. thanx a lot.

Taihengliu · September 1, 2020, 8:13am

Hi,why do I use the new version of the package and compute slower than the old version?
Thanks

liuqi · October 15, 2020, 2:42am

Hi,Javier
I have run your code and found that when the density ratio and viscosity ratio are 1. omega1 and omega2 are both 1, the flow process conforms well to the requirements of Laplace’s Law, but when the viscosity ratio increases, omega1=1.6, omega2 =1, the pressure gradient is 0, and the flow also occurs. Why?Can you help me to explain the problem?Looking forward to your reply!

jesantos · October 29, 2020, 3:46pm

That’s a common set-back of the Shan-Chen method. Let’s wait until the color-gradient method it’s released.

Taihengliu · November 24, 2020, 3:00am

Hi,Javier
How is the surface tension of the two phase interface determined in the program? thanks

Taihengliu · November 26, 2020, 2:04pm

Hi Javier
How is the surface tension value 0.15 obtained?
Thank you for your reply.

jesantos · December 11, 2020, 3:52am

Howdy @Taihengliu,

Feel free to check Chapter 5 of my thesis:
https://repositories.lib.utexas.edu/handle/2152/69246

Best,
Javier

Taihengliu · January 25, 2021, 8:43am

Hi,Javier
I use your program, if I don’t change Gc, the interfacial tension doesn’t change either, it’s still 0.15.
Thank you for your reply.

Taihengliu · February 26, 2021, 1:09am

Hi,Javier
Can your program be modified to make fluid 2 drive fluid 1?If so, how should I modify it?
Thank you for your reply!

jesantos · April 26, 2021, 7:01pm

Howdy @Taihengliu,

Fluid 1 and 2 are interchangeable labels. I believe that you would like to change the contact angle and that can be achieved by changing the G_ads of the appropriate surface.

Thanks