SlopeCube Example problem 1a - different results

A discussion forum for users of the new HYDRUS 2D/3D. HYDRUS is a software package for simulating water, heat and solute movement in two- and three-dimensional variably saturated media. Happy Posting!
Post Reply
DandD
Posts: 14
Joined: Wed Jan 15, 2020 12:11 pm
Location: Germany

SlopeCube Example problem 1a - different results

Post by DandD » Thu Jan 16, 2020 12:56 pm

Hi everyone,

I am new to the forum and would like to discuss some difficulties I have while learning to work with the Slope Cube module (so far only 2D).
I already have a problem with the first example problem "Example 1a" - I basically rebuilt and then simulated the model but I am getting different results compared to the model from the PC-Progres website. I already went through all settings and can not find any diffences in the setup (I changed everything according to the original example 1a model). The only difference occurs in the mesh (3807 instead of 3769 Nodes) - I used the same mesh parameters but I get a slightly different mesh (which is normal I believe). I think this should not have a big impact on the results. One more thing is different - In my model there are 203 Boundary Mesh Nodes and Elements displayed, which is not the case in the example model.
I get similar results for pressure head and water content, different streamlines and velocities and the slope cube results differ quite a lot (e.g. see example attached).
Any Idea, why the differences occur?

Unfortunately, my model file was to big to upload it here as an attachment. It can be accessed vie the following link (if necessary - I hope the external link is OK for you to use):
https://cloudstore.zih.tu-dresden.de/in ... 97FwjBcz5A

Thank you for your help and I am looking forward to hearing from you!

Frieder
Attachments
Problem_Test1a.docx
(323.32 KiB) Downloaded 46 times

Mirek
Posts: 428
Joined: Sat Mar 16, 2002 4:00 pm
Location: Czech Republic
Contact:

Re: SlopeCube Example problem 1a - different results

Post by Mirek » Thu Jan 16, 2020 1:34 pm

Hi Frieder,

I see that you are using „Custom Scale“ for the graphical display of LFSF. Can you try to use the „Standard Scale“, which corresponds to real min/max values? See the attached figure. Right-click on the scale to open the menu. Let us know if you then still have substantially different results.
M.
CustomScale.jpg
CustomScale.jpg (140.08 KiB) Viewed 1531 times

DandD
Posts: 14
Joined: Wed Jan 15, 2020 12:11 pm
Location: Germany

Re: SlopeCube Example problem 1a - different results

Post by DandD » Thu Jan 16, 2020 1:46 pm

Hello Mirek,

thanks for the fast reply. I am using the custom scale according to the one displayed in the manual (same numbers for the original model as well as my model). The only difference is the maximum LFSF (original 140.3 compared to 51.1 in my model -> discrepancy). The problem with the standard scale is that you can not really see any results. Anyways - the differences still occur.

Frieder
Attachments
StandardScale_Frieder.PNG
StandardScale_Frieder.PNG (187.17 KiB) Viewed 1529 times
StandardScale_original.PNG
StandardScale_original.PNG (128.83 KiB) Viewed 1529 times

DandD
Posts: 14
Joined: Wed Jan 15, 2020 12:11 pm
Location: Germany

Re: SlopeCube Example problem 1a - different results

Post by DandD » Thu Jan 16, 2020 2:23 pm

Hi again,

a colleague gave me a useful hint - I now have rerun the original model with the newest Hydrus version 3.02 and it now shows different results a) to the original results (downlaoded from the PCProgres website) as well as b) my model.
It is closer to my model now (showing LFSF <1 at the top of the slope as well) though. But the maximum LFSF after 15 h still differs with a value of 77.5 (compared to 51.1).
The interesting parts (slope boundary areas) show similar results now.

I will go more into detail with the stability modules in the next weeks...

Best regards
Frieder
Attachments
RerunOriginalModel.PNG
RerunOriginalModel.PNG (292.52 KiB) Viewed 1521 times

Mirek
Posts: 428
Joined: Sat Mar 16, 2002 4:00 pm
Location: Czech Republic
Contact:

Re: SlopeCube Example problem 1a - different results

Post by Mirek » Thu Jan 16, 2020 2:30 pm

Yes, example Test1a (see https://www.pc-progress.com/en/Default.aspx?h3d2-Slope3) was created in HYDRUS 2.05 and results should be recalculated when working with this project in HYDRUS 3.02. The reason is that SlopeCube module has changed a little bit – for more details, please contact Dr. Alexandra Wayllace (alexandra.wayllace@gmail.com or Prof. Ning Lu (https://www.pc-progress.com/en/Default.aspx?ning-lu). This example was replaced by Slope_Stability\Slope_Cube_Test3.h3d3 in HYDRUS 3.x – please have a look it it.

DandD
Posts: 14
Joined: Wed Jan 15, 2020 12:11 pm
Location: Germany

Re: SlopeCube Example problem 1a - different results

Post by DandD » Thu Jan 16, 2020 2:34 pm

Thanks Mirek, I will dig into the topic!

Post Reply