Error in calculation of stresses

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
Naveenk
Posts: 4
Joined: Fri Mar 26, 2021 5:19 am
Location: India

Error in calculation of stresses

Post by Naveenk » Tue Mar 30, 2021 5:55 am

We are trying to model an earthen dam under filling up condition. The soil strata below the base of the dam is saturated and modeled by giving pressure head value equilibrium with the lowest node. The filling up of reservoir starts from day 1 with 2.8m rise for 5 days. The filling up is modelled using variable head varH1. The model is analysed in Hydrus 2d and slope cube module. From the output of slope cube module, values of Total stress -maximum, suction stress and effective stress - maximum at the base of the model on upstream side are:
Day 0 : 175kPa, 94.5kPa, -8.15kPa
Day 3 : 175kPa, 168kPa, -83kPa (Total rise in water level on day 3 is 3x2.8=8.4m)

As can be seen from the output that the total stress has not changed with rise of water level and suction stress has increased very much. Also the effective stress - maximum is negative on both days. Out of these values only total stress and suction stress on Day 0 are correct. All other values are not correct.
Kindly guide in this matter to resolve issue in the analysis and output from slope cube. Do let me know if any more detail is required.

Thanks.
Attachments
Analysis.jpg
Analysis.jpg (262.36 KiB) Viewed 779 times

Mirek
Posts: 447
Joined: Sat Mar 16, 2002 4:00 pm
Location: Czech Republic
Contact:

Re: Error in calculation of stresses

Post by Mirek » Tue Mar 30, 2021 6:34 am

To answer your question, we need to have your project (* .h3d3 file). Send this project to support@pc-progress.com as a ZIP archive (if its size is smaller than 15 MB) or send a link for downloading this file from your FTP server or another source. You can also upload the ZIP archive here to this discussion forum.

Naveenk
Posts: 4
Joined: Fri Mar 26, 2021 5:19 am
Location: India

Re: Error in calculation of stresses

Post by Naveenk » Tue Mar 30, 2021 7:48 am

Thanks for the reply. I have mailed the project file.

Naveenk
Posts: 4
Joined: Fri Mar 26, 2021 5:19 am
Location: India

Re: Error in calculation of stresses

Post by Naveenk » Mon Apr 12, 2021 5:51 am

Awaiting reply to the query.

Two issues that we are facing are:
1. The calculation of effective stress is incorrect. It is not coming equal to the difference of total stress and suction stress (pore pressure) [the calculation as described in the manual].
2. Although the pore pressure is increasing, total stress in not increasing proportionally to the increase in water head (modelled as variable head with time).

Jirka
Posts: 5268
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Error in calculation of stresses

Post by Jirka » Mon Apr 12, 2021 2:49 pm

We have contacted the developers of the “stress” part of the module, forwarded them your question and project, and are awaiting their response. J.

Jirka
Posts: 5268
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Error in calculation of stresses

Post by Jirka » Wed Apr 14, 2021 11:22 pm

The response from the developers of the "Slope Cube" module:
"Currently Slope Cube does not have the capability to implement water pressure from a hydrological boundary to the stress boundary condition. We will work on implementing it for the next version. One general way to go around this, is for the user to implement the equivalent boundary as a force boundary. However, please be mindful that on the other side of the dam we recommend "no-displacement" boundary, and this may cause significant noise on the total stress distribution."

Naveenk
Posts: 4
Joined: Fri Mar 26, 2021 5:19 am
Location: India

Re: Error in calculation of stresses

Post by Naveenk » Sat Apr 24, 2021 2:23 pm

This answers the second query of no increment in the total stress, but not the first query. The effective stress values should have been the difference of Total stress and Suction stress. Can this be resolved?

We thought of giving additional equivalent force on the upstream side. However we could not find option of variable force – varying force with time. Modelling for each day with varying force as separate file will be cumbersome. Hope that the noise due to boundary condition (no displacement) can be reduced by increasing the model width. Can you kindly enlighten how varying force with time can be modelled?

Mirek
Posts: 447
Joined: Sat Mar 16, 2002 4:00 pm
Location: Czech Republic
Contact:

Re: Error in calculation of stresses

Post by Mirek » Thu Apr 29, 2021 7:31 am

The response from the developers of the "Slope Cube" module:

"1. The values of effective stress are very close to the difference between total stress and suction stress. The small difference comes from obtaining values at the nodes versus the centroid of each element. For example, at the top of the dam, right in the middle the values obtained are:

Total stress-maximum = 22.5 kPa
Suction stress = -15.8 kPa
Effective stress - maximum= 38kPa

The difference between total and suction stress is 38.3kPa, the model gives 38 kPa

2. I don't think assigning free displacement to the dam are the best boundary conditions. That will affect the stress values at the boundaries

3.Unfortunately, right now the software does not have the capability of variable mechanical boundary conditions, that will be implemented in the next version."

Post Reply