open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

It cannot work like this because when you move to primitive equations the linear system becomes non symmetrical and thus the conjugate gradient cannot work, you need to have also :

SOLVER : 3 or 7

which is normal equation or GMRES (if you have many iterations of solver 3 try 7).

With best regards,

Jean-Michel Hervouet

Dear Hervouet, Thank you very much for suggesting, I would try it again.

Best regards,

Viola

Dear Hervouet,

I had done my estimation of my case, but without the proper estimation results(the initialised bottom friction value did not change after computation).
All of the content of file rfo_estimation.txt is as below:

' '
'PARAMETER ESTIMATION'
'IDENTIFICATION OF FRICTION'
'ITERATION'
'COST'
'ERROR ON H (M)'
'ERROR ON U (M/S)'
'ERROR ON V (M/S)'
1 0.0000000000000000 113.57110595703125 1.8386944532394409 1.7108145952224731

I had read the code, it seems the error is too large? I'm not sure where is the problem lies? Would you give me some suggestion?

Thank you in advance.

Best regards,

Viola

Hello,

Normally the program should iterate to decrease the error. Did you try the test case called "estimation", in this case you see that the program does several times the same run with different values of friction. It seems here that you have only one iteration.

With best regards,

JMH

Dear Hervouet,

I have tryed that case, thank you for suggestion, I'll try more.

Best regards,

Viola

Hello,
it's been a while, I'm trying to make a parameter estimation of bottom friction by having water depth measurements for some points in time. Just like your case. I read the posts in the forum but I can't get the model to work. Can you help me? I'm trying the "estimation" test case. I will attach my files.

It is the declarations you do at the beginning of your fortran files.


When I ran it I got this message:
Error: Symbol 'lng' at (1) conflicts with symbol from module 'declarations_special', use-associated at (2)

If you comment out the lines:
INTEGER LNG,LU
COMMON/INFO/LNG,LU
in def_zones.f and mesures.f files it runs (you would need to check if it does what you want).

Hope that helps,
Patrick

Hello JMH,
I follow all your comments here and still couldn't find the solution, I have same problem.

I hope to get why the problem show up at

Error: Symbol 'lng' at (1) conflicts with symbol from module 'declarations_special', use-associated at (2)
runCAS: failed to compile
|processExecutable: could not compile your FORTRAN (runcode=1).
|

Hello again kero_feuz,

As told recently in another post, people in grey will not answer (their accounts are not active anymore or were deleted).
JMH is now retired after several decades working for the TELEMAC system.

Anyway, as told in rule #2 of this forum: "Do not cross-post the same question to multiple places."

Chi-Tuan