Numerical solution not converging

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
Saheli
Posts: 2
Joined: Wed May 29, 2019 8:04 pm
Location: USA

Numerical solution not converging

Post by Saheli » Wed Jun 19, 2019 7:52 pm

Hi,

I am trying to simulate the pressure heads in an infiltration basin for a variable flux but I keep getting the same error that says "Numerical solution has stopped because it has not converged in 10 consecutive time steps". I am unable to figure out where I have gone wrong. Could you help fix the problem? Some information specific to the project has been attached here.
Attachments
time output info.JPG
time output info.JPG (35.97 KiB) Viewed 234 times
initial condition.JPG
initial condition.JPG (71.19 KiB) Viewed 234 times
error.JPG
error.JPG (103.01 KiB) Viewed 234 times
domain property.JPG
domain property.JPG (41.88 KiB) Viewed 234 times
boundary condition.JPG
boundary condition.JPG (57.86 KiB) Viewed 234 times

Jirka
Posts: 4760
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Numerical solution not converging

Post by Jirka » Wed Jun 19, 2019 9:44 pm

Please read “Rules about spatial and temporal discretization (FAQ 8: http://www.pc-progress.com/Documents/No ... zation.pdf) and do not do things, which we clearly say that should not be done, such as changing Iteration Criteria. Otherwise I cannot comment on this project since I do not know what your hydraulic properties are, what you boundary fluxes are, etc. J.

Saheli
Posts: 2
Joined: Wed May 29, 2019 8:04 pm
Location: USA

Re: Numerical solution not converging

Post by Saheli » Thu Jun 27, 2019 10:54 pm

Hi again,

I have read the instructions for temporal and spatial discretization and have tried to follow them. But the solution still doesn't converge. Please find the file for my model attached here. I would really appreciate it if you could help me out!

Thanks,
Saheli
Attachments
steady state model.zip
(249.87 KiB) Downloaded 17 times

Jirka
Posts: 4760
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Numerical solution not converging

Post by Jirka » Thu Jun 27, 2019 11:31 pm

First of all, I do not think that you read those instructions on temporal and spatial discretization since your project clearly does not follow the recommendations.

Anyway, even if it did the program would not be able to run since the problem is not physically realistic. How can you expect that you can enforce a boundary flux that is more than ten times larger than the saturated hydraulic conductivity of the surface layer and hundred times larger than the saturated hydraulic conductivity of the underlying layer? Similarly, since there is no outflow allowed at the bottom and is allowed only on the sides, you should be able to calculate (using simple back of the envelope calculations) what kind of gradient you would need to divert this flux to sides for given hydraulic conductivity. Again, clearly not realistic.

You cannot define an unrealistic problem and expect that HYDRUS will be able to calculate it.

J.

Post Reply