Hello Yvan,
I have just installed version 5, it has two new fonctions, I would like to know whether the automatic scheme is better than the first order scheme, and if I use iterative process error estimators (with / without volume contribution), can i get a better convergence result?
When I start to calculate it, it can not run, it shows:
'code_saturne' n'est pas reconnu en tant que commande interne
That is:
'Code_saturne' is not recognized as an internal command
How can i solve it?
By the way, it matches which version of med? Is this a problem with med version does not match? I used MED 3.0 files.
Thank you very much
Best Regards
XING Jian
How to plot residuals in Windows?
Forum rules
Please read the forum usage recommendations before posting.
Please read the forum usage recommendations before posting.
-
Yvan Fournier
- Posts: 4077
- Joined: Mon Feb 20, 2012 3:25 pm
Re: How to plot residuals in Windows?
Hello,
This looks like an installation problem.
Could you move the post to the installation section (otherwise I can do it later) and provide the information recommended in the installation section ?
Do you have other versions of the code installed ?
Are you running on Windows ? If so, you are probably one if the first people to test this (but the build has not changed significantly from 4.3, which had more feedback, so issues would be surprising).
Regards,
Yvan
This looks like an installation problem.
Could you move the post to the installation section (otherwise I can do it later) and provide the information recommended in the installation section ?
Do you have other versions of the code installed ?
Are you running on Windows ? If so, you are probably one if the first people to test this (but the build has not changed significantly from 4.3, which had more feedback, so issues would be surprising).
Regards,
Yvan
Re: How to plot residuals in Windows?
Hello Yvan,
I have moved it to the installation page, and could you please answer the question I mentioned earlier?
Best Regards
XING Jian
I have moved it to the installation page, and could you please answer the question I mentioned earlier?
Best Regards
XING Jian
-
Yvan Fournier
- Posts: 4077
- Joined: Mon Feb 20, 2012 3:25 pm
Re: How to plot residuals in Windows?
Hello,
I am not quite sure about your first question. Are you talking of the error estimators that may now be post-processed ? Or something else. Same for the scheme.
You did not tell me whether you have another install or not. Also, did you run from the GUI or try running directly SCRIPTS/runcase.bat ? If you did not try it, can you try ?
The MED version is 3.2, so it should be able to read MED 3.0 files.
Regards,
Yvan
I am not quite sure about your first question. Are you talking of the error estimators that may now be post-processed ? Or something else. Same for the scheme.
You did not tell me whether you have another install or not. Also, did you run from the GUI or try running directly SCRIPTS/runcase.bat ? If you did not try it, can you try ?
The MED version is 3.2, so it should be able to read MED 3.0 files.
Regards,
Yvan
Re: How to plot residuals in Windows?
Hello Yvan,
I'm sorry I did not say it clearly. I'm talking about the question I asked on the second page. Like I have a fluctuating flow, I can only treat it as it is convergence, and you said it may be useful to define time averages for postprocessing, could you remind me how to define time averages for postprocessing in Salom?
And I have already uploaded the XML files, the problem about the heat exchange.
I also would like to know the two new functions in version5.0: automatic scheme and iterative process error estimators (with / without volume contribution), will they help the results converge?
Thank you very much
Best Regards
XING Jian
I'm sorry I did not say it clearly. I'm talking about the question I asked on the second page. Like I have a fluctuating flow, I can only treat it as it is convergence, and you said it may be useful to define time averages for postprocessing, could you remind me how to define time averages for postprocessing in Salom?
And I have already uploaded the XML files, the problem about the heat exchange.
I also would like to know the two new functions in version5.0: automatic scheme and iterative process error estimators (with / without volume contribution), will they help the results converge?
Thank you very much
Best Regards
XING Jian
-
Yvan Fournier
- Posts: 4077
- Joined: Mon Feb 20, 2012 3:25 pm
Re: How to plot residuals in Windows?
Hello,
You can define time averages from the Code_Saturne GUI, in the calculation control section (this was already available in V3.0 at least, though it has been improved a bit).
As I already answered, the error estimators are simply a postprocessing output, which has been available for a long time, but only recently exposed through the GUI. They have no influence on convergence.
I'm not sure about the automatic variable schemes (I believe something to do with the new gradient limiters, which may help, but I'm not a specialist).
Regards,
Yvan
You can define time averages from the Code_Saturne GUI, in the calculation control section (this was already available in V3.0 at least, though it has been improved a bit).
As I already answered, the error estimators are simply a postprocessing output, which has been available for a long time, but only recently exposed through the GUI. They have no influence on convergence.
I'm not sure about the automatic variable schemes (I believe something to do with the new gradient limiters, which may help, but I'm not a specialist).
Regards,
Yvan
Re: How to plot residuals in Windows?
Hello Yvan,
Thank you so much, this time averages is very helpful to my situation, I never thought about it before, thank you.
Just like my results of calculations, after calculating a thousand steps, all the variables of the monitoring points fluctuate in their respective regions, and the start of my time averages is set at 1000, is that right?
By the way, for heat exchange:
These two are both using the heat exchange coefficient of 50, k-w SST, and reference time step is 0.0001, and calculated 16000 times.
Best Regards
XING Jian
Thank you so much, this time averages is very helpful to my situation, I never thought about it before, thank you.
Just like my results of calculations, after calculating a thousand steps, all the variables of the monitoring points fluctuate in their respective regions, and the start of my time averages is set at 1000, is that right?
By the way, for heat exchange:
Best Regards
XING Jian
Re: How to plot residuals in Windows?
Hello Yvan,
Here are the XML files, most of their settings are the same, only a small part is different. And the setting of the Periodic Boundary of my test model is wrong because it is a test. Another point, the biggest difference is that the mesh of the test model is very rough, and I do not have a boundary layer on the test model, but I have done boundary layers on all boundaries for the formal model.
Thank you very much
Best Regards
XING Jian
Here are the XML files, most of their settings are the same, only a small part is different. And the setting of the Periodic Boundary of my test model is wrong because it is a test. Another point, the biggest difference is that the mesh of the test model is very rough, and I do not have a boundary layer on the test model, but I have done boundary layers on all boundaries for the formal model.
Thank you very much
Best Regards
XING Jian
- Attachments
-
- Test(have exchange).xml
- (11.59 KiB) Downloaded 263 times
-
- Steady_SIMPLEC_0.0001(non_exchange).xml
- (12.86 KiB) Downloaded 201 times
-
Yvan Fournier
- Posts: 4077
- Joined: Mon Feb 20, 2012 3:25 pm
Re: How to plot residuals in Windows?
Hello,
There are several differences in your setups, but the one with probably the most impact is the fact that one of the 2 has a sliding wall, while the other does not, so this may affect heat exchange quite a bit.
Regards,
Yvan
There are several differences in your setups, but the one with probably the most impact is the fact that one of the 2 has a sliding wall, while the other does not, so this may affect heat exchange quite a bit.
Regards,
Yvan
Re: How to plot residuals in Windows?
Hello Yvan,
I can confirm that has no effect on heat exchange, I have calculated one with sliding wall, the result is still no heat exchange。
Is it possible to explain this because the mesh is refined and the boundary layer is added? They lead to no heat exchange.
You carefully observe the first picture, the color of the middle of the cardboard changed a little bit, is it because I calculated the time is not enough?
Thank you very much
Best Regards
XING Jian
I can confirm that has no effect on heat exchange, I have calculated one with sliding wall, the result is still no heat exchange。
Is it possible to explain this because the mesh is refined and the boundary layer is added? They lead to no heat exchange.
You carefully observe the first picture, the color of the middle of the cardboard changed a little bit, is it because I calculated the time is not enough?
Thank you very much
Best Regards
XING Jian