Atom topic feed | site map | contact | login | Protection des données personnelles | Powered by FluxBB | réalisation artaban
You are not logged in.
Pages: 1
Hi to all.
I am trying to use macro MACR_RECAL to obtain material parameters for Chaboche material model. But I am having some difficulties obtaining results. Results are totally wrong, I am not getting even right trends of data evolution.
What I have are three different strain controlled experimental curves and I am computing model responses using SIMU_POINT_MAT. Simulations are fast, but wrong .
I am including all necessary files to run the calibration. I have tried also different optimization method - HYBRIDE, to better map space of parameters but without any results.
Thank you for any advices.
Michal
Offline
Here is image with results.
Offline
Hi,
According to me the issue comes from the way you are using MACR_RECAL. You have to plot your results stress Vs time and strain Vs time in order to have curves as defined in paragraph 5 of documentation U4.73.02
A second point, is the use of the Chaboche model. This model is very powerful but you have to be sure that you need this one.
I emphasis this point, because based on the curves you posted, you don't have enough information in your results to fulfill all the Chaboche model parameters. Sometimes, a non-linear isotropic hardening model is enough (COMPORTEMENT=VMIS_ISO_TRAC).
Regards
Stephane
Offline
Thank you for your help. Now it works fine.
I have another question. Is there any way how to extract resulting (fitted) parameters from MACR_RECAL? I mean if i save reuslts to lets say:
RESULT = MACR_RECAL( ... ) then save just parameters to file or to variables using python.
Regards,
Michal
Offline
Thank you for your help. Now it works fine.
I have another question. Is there any way how to extract resulting (fitted) parameters from MACR_RECAL? I mean if i save reuslts to lets say:
RESULT = MACR_RECAL( ... ) then save just parameters to file or to variables using python.Regards,
Michal
Hello,
Example:
ReCall = MACR_RECAL(
...............
);
Values=ReCall.Valeurs()
"Values" is the array of fitted parameters. You can then handle it like any other array.
I'm interested in fitting parameters so if you could share your case I will be glad .
regards,
FilJan
Last edited by filjan (2017-03-17 14:29:35)
Offline
Hi,
I have also found this in documentation, but it does not work. I am getting enclosed error.
I am enclosing files of 'calibration'. This parth should work fine if you increase number if allowable iterations and population size. This is just testing version to get faster some reuslts and be able to save them using
-interact
to variables and work with them.
Regards,
Michal
Offline
Hi,
thank you for the attachment.
It's surprising what you've wrote about error with output-handle, because I use this trick with success. Exactly like:
Re = MACR_RECAL(
...
);
ValArr=Re.Valeurs()
Val=ValArr[0]
print("First component from Macr_Recal otuput: "+str(Val))
regards
FilJan
Last edited by filjan (2017-03-17 14:43:13)
Offline
See attachment, It works!
Offline
Well ... You are right. I was using this command in Batch window after computation. Your example works fine. Thank you.
What is you aim of calibration? I am trying to calibrate Chaboche model with other possible extension available in C-A.
Regards,
Michal
Offline
I usually use Macr_Recal to get the proper response force of the system by displacement constraint. Problems like tightening a screw (ssnv228) with specified torque.
Offline
Pages: 1