Next question - has anyone managed to find a way of importing or exporting time variant ID's to nodes on seepage faces?

[conesster]

Thanks very much. I agree that regionaliseatoin was not the correct choice of wording, I have to assign DFE's every 50 m throughout a large area progresssively with time and this is very manual!

Next question - has anyone managed to find a way of importing or exporting time variant ID's to nodes on seepage faces?

[Peter Schätzl]

For the boundary conditions you can use Assign - database - Import Time-varying IDs - Head BC - Automatic regionalization.

[conesster]

Thanks, is it possible to apply these to seepage face constraint conditions?/

[Peter Schätzl]

Yes, this option can also be used to set seepage face conditions. Just use Options - Head - Seepage Face before you do the regionalization.

[conesster]

Thanks, however this appears to work for the boundary condition.

I would like to do this for the flux constraint as this is how im managing to "turn on and off" the seepage face condition as required. 

So I wish to assign a flux constraint (min) via a TVID. However I am unable to import the TVID for a flux constraint condition onto the seepage face as i get the error box:

"Use of database for pre-defined time-varying functions makes no sense! Set the time-constant toggle. Returning..."

I am able to click on every node and assign the TVID manually. But it is going to give me arthritis in my clicking finger very quickly! Am I approaching this wrong, or missing a trick with respect to importing seepage face TVID BC constraints?

[Peter Schätzl]

OK, this is another task. WIthout checking in detail, I think it should work if you choose 'Time-constant' in the constraints settings, then 'Import Power Function ID' in the regionalization dialog and then an automatic regionalization. You will get, however, new functions with an internal ID (i) for all the nodes because of the regionalization.

[conesster]

Ah thankyou. Now I think I see -

set up TVID's in feflow as required
set max/min tvids to time constant
use database to import TVID numbers and rationalise as normal

Then feflow assigns individual an internal power ID's to each node, but this ID takes information from TVID number imported. I note that although the durations have worked correctly the flow rate in the internal power ID is only half of that entered in the TVID I created in FEFLOW and which the import file referenced - this is supiciously like an average?

[Peter Schätzl]

I would recommend to use 'Inverse Distance Weighting' with only one neighboring point as the regionalization method. Then the value should be directly transferred. With other methods, you'll get an interpolation between different functions that may lead to changed values.

[conesster]

Thanks for your help I have managed to do this now.

I would like to ask something - i was using a very small snap distance as I was matching to the exact node coordinates. However the regionalisation then doesnt work and occasionally assigns the tvid reference as the constraint rather than the prescribed flow rate.

This was fixed by increasing the regionalistion snap distance. The data was imported using only 1 neighbour in order to try to match to the exact node without averaging.

However this only worked once and I could not repeat the success, even for the same data input files.

[Wolfram Rühaak]

After discussing this matter at FEFLOW Services we have found the reason of the problem:
Additional to the Power Function ID's the user has had defined Time Constant Data too (in the <Assign> Database > Data regionalization dialog).
Therefore, as Time Constant Data were defined, FEFLOW uses these values for the regionalization, not as expected the Power Funtion ID's.

[giovanni]

For the boundary conditions you can use Assign - database - Import Time-varying IDs - Head BC - Automatic regionalization.

Peter,
I am facing the problem to assign a time variant seepage boundary condition

I am trying to follow your procedure but it seems that it does not work:

when I use the command:
Assign - database - Import Time-varying IDs

and when I select the *.trp file, I get this error message:
WARNING: Unknown time-varying function ID 385.   Data point will be skipped.
WARNING: Unknown time-varying function ID 385.   Data point will be skipped.
WARNING: Unknown time-varying function ID 385.   Data point will be skipped.
WARNING: Unknown time-varying function ID 385.   Data point will be skipped.
WARNING: Unknown time-varying function ID 385.   Data point will be skipped.
WARNING: Unknown time-varying function ID 385.   Data point will be skipped.
WARNING: Unknown time-varying function ID 385.   Data point will be skipped.
WARNING: Unknown time-varying function ID 385.   Data point will be skipped.
WARNING: There are MORE unknown time-varying functions. Skipping all ...

the same trp file works fine when I import it a a time-constant data file.

Giovanni

PS. As Feflow allows to assign the pressure instead of the head as 1st boundary condition, why isn't possible to assign a power function for the pressure ? I tryied that but the command "pressure" is disabled when I try to use a time-variant function
It would be easier to deal with the pressure as it would have a constant value along the seepage face and I shopuld not have to define a different time-variant function for each node.

[giovanni]

For the boundary conditions you can use Assign - database - Import Time-varying IDs - Head BC - Automatic regionalization.

I understood what was wrong. I have to define a power function for each node I import and in trp file I have to put the ID of the power function for the selected node. By the way this procedure is quite tricky and very hard to set up. As I asked before , is it possible to use the pressure instead?

Giovanni