How to use variable polynomial order in compressible solver
******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Hi everyone, I would like to test the efficiency of p-adaptive method in CompressibleFlowSolver. But I could not find any sample which shows me how to modify the session file. I have tried the same set up found in incompressible solver Tests, as follows, <SOLVERINFO> <I PROPERTY="Driver" VALUE="Adaptive" /> </SOLVERINFO> <PARAMETERS> <P> AdaptiveMaxModes = 7 </P> <P> AdaptiveMinModes = 4 </P> </PARAMETERS> The CompressibleFlowSolver ran successfully (simulated forward step problem). But I didn't find any difference between the results of "Adaptive Driver" and "Standard Driver". Any suggestions are appreciated. Regrads, Chengeng
Hi Chengeng, Did you set up the number of adaptive iterations that you want to perform? For example, if you want to run 2000 time steps in total and adapt the mesh every 500 time steps, so three times, you might set it up as: <PARAMETERS> <P> NumSteps = 500 </P> <P> NumRuns = 4 </P> </PARAMETERS> Cheers, Margarida On Tue, 4 Apr 2023 at 11:35, Chengeng Qian <qiancg0313@gmail.com> wrote:
This email from qiancg0313@gmail.com originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list <https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email stamping for this address.
Hi everyone, I would like to test the efficiency of p-adaptive method in CompressibleFlowSolver. But I could not find any sample which shows me how to modify the session file. I have tried the same set up found in incompressible solver Tests, as follows,
<SOLVERINFO> <I PROPERTY="Driver" VALUE="Adaptive" /> </SOLVERINFO>
<PARAMETERS> <P> AdaptiveMaxModes = 7 </P> <P> AdaptiveMinModes = 4 </P> </PARAMETERS>
The CompressibleFlowSolver ran successfully (simulated forward step problem). But I didn't find any difference between the results of "Adaptive Driver" and "Standard Driver".
Any suggestions are appreciated.
Regrads, Chengeng _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/nektar-users
-- Margarida Moragues Ginard MSCA-IF Researcher *BCAM - *Basque Center for Applied Mathematics Alameda de Mazarredo, 14 E-48009 Bilbao, Basque Country - Spain Tel. +34 946 567 842 mmoragues@bcamath.org | www.bcamath.org/mmoragues *(**matematika mugaz bestalde)*
Hi Margarida, Thanks for your response. It worked when I used your set up. It seems that the solver reads the data every NumSteps timesteps and projects the solutions to new expansions. I guess that two parameters " AdaptiveLowerTolerance" and "AdaptiveUpperTolerance" can control the region where the new expansions are used. But we can't know the results from figures, right? There is no sensor which shows us where the higher order expansions and lower expansions are used, respectively. Chengeng Margarida Moragues Ginard <mmoragues@bcamath.org> 于2023年4月4日周二 19:30写道：
Hi Chengeng,
Did you set up the number of adaptive iterations that you want to perform?
For example, if you want to run 2000 time steps in total and adapt the mesh every 500 time steps, so three times, you might set it up as:
<PARAMETERS>
<P> NumSteps = 500 </P>
<P> NumRuns = 4 </P>
</PARAMETERS>
Cheers,
Margarida
On Tue, 4 Apr 2023 at 11:35, Chengeng Qian <qiancg0313@gmail.com> wrote:
This email from qiancg0313@gmail.com originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list <https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email stamping for this address.
Hi everyone, I would like to test the efficiency of p-adaptive method in CompressibleFlowSolver. But I could not find any sample which shows me how to modify the session file. I have tried the same set up found in incompressible solver Tests, as follows,
<SOLVERINFO> <I PROPERTY="Driver" VALUE="Adaptive" /> </SOLVERINFO>
<PARAMETERS> <P> AdaptiveMaxModes = 7 </P> <P> AdaptiveMinModes = 4 </P> </PARAMETERS>
The CompressibleFlowSolver ran successfully (simulated forward step problem). But I didn't find any difference between the results of "Adaptive Driver" and "Standard Driver".
Any suggestions are appreciated.
Regrads, Chengeng _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/nektar-users
-- Margarida Moragues Ginard MSCA-IF Researcher *BCAM - *Basque Center for Applied Mathematics Alameda de Mazarredo, 14 E-48009 Bilbao, Basque Country - Spain Tel. +34 946 567 842 mmoragues@bcamath.org | www.bcamath.org/mmoragues
*(**matematika mugaz bestalde)*
Dear Chengeng, I think that the value of the Sensor variable and, as you mention, parameters "AdaptiveLowerinModes" and "AdaptiveUpperTolerance" are used to decide the expansion that will be used on each location. The Sensor variable is part of the output solution. Cheers, Margarida On Wed, 5 Apr 2023 at 12:46, Chengeng Qian <qiancg0313@gmail.com> wrote:
Hi Margarida, Thanks for your response. It worked when I used your set up.
It seems that the solver reads the data every NumSteps timesteps and projects the solutions to new expansions. I guess that two parameters " AdaptiveLowerTolerance" and "AdaptiveUpperTolerance" can control the region where the new expansions are used. But we can't know the results from figures, right? There is no sensor which shows us where the higher order expansions and lower expansions are used, respectively.
Chengeng
Margarida Moragues Ginard <mmoragues@bcamath.org> 于2023年4月4日周二 19:30写道：
Hi Chengeng,
Did you set up the number of adaptive iterations that you want to perform?
For example, if you want to run 2000 time steps in total and adapt the mesh every 500 time steps, so three times, you might set it up as:
<PARAMETERS>
<P> NumSteps = 500 </P>
<P> NumRuns = 4 </P>
</PARAMETERS>
Cheers,
Margarida
On Tue, 4 Apr 2023 at 11:35, Chengeng Qian <qiancg0313@gmail.com> wrote:
This email from qiancg0313@gmail.com originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list <https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email stamping for this address.
Hi everyone, I would like to test the efficiency of p-adaptive method in CompressibleFlowSolver. But I could not find any sample which shows me how to modify the session file. I have tried the same set up found in incompressible solver Tests, as follows,
<SOLVERINFO> <I PROPERTY="Driver" VALUE="Adaptive" /> </SOLVERINFO>
<PARAMETERS> <P> AdaptiveMaxModes = 7 </P> <P> AdaptiveMinModes = 4 </P> </PARAMETERS>
The CompressibleFlowSolver ran successfully (simulated forward step problem). But I didn't find any difference between the results of "Adaptive Driver" and "Standard Driver".
Any suggestions are appreciated.
Regrads, Chengeng _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/nektar-users
-- Margarida Moragues Ginard MSCA-IF Researcher *BCAM - *Basque Center for Applied Mathematics Alameda de Mazarredo, 14 E-48009 Bilbao, Basque Country - Spain Tel. +34 946 567 842 mmoragues@bcamath.org | www.bcamath.org/mmoragues
*(**matematika mugaz bestalde)*
-- Margarida Moragues Ginard MSCA-IF Researcher *BCAM - *Basque Center for Applied Mathematics Alameda de Mazarredo, 14 E-48009 Bilbao, Basque Country - Spain Tel. +34 946 567 842 mmoragues@bcamath.org | www.bcamath.org/mmoragues *(**matematika mugaz bestalde)*
Hi Margarida, OK, I see. I thought that the Sensor variable is only for artificial viscosity. Thanks. Chengeng Margarida Moragues Ginard <mmoragues@bcamath.org> 于2023年4月7日周五 00:01写道：
Dear Chengeng,
I think that the value of the Sensor variable and, as you mention, parameters "AdaptiveLowerinModes" and "AdaptiveUpperTolerance" are used to decide the expansion that will be used on each location. The Sensor variable is part of the output solution.
Cheers,
Margarida
On Wed, 5 Apr 2023 at 12:46, Chengeng Qian <qiancg0313@gmail.com> wrote:
Hi Margarida, Thanks for your response. It worked when I used your set up.
It seems that the solver reads the data every NumSteps timesteps and projects the solutions to new expansions. I guess that two parameters " AdaptiveLowerTolerance" and "AdaptiveUpperTolerance" can control the region where the new expansions are used. But we can't know the results from figures, right? There is no sensor which shows us where the higher order expansions and lower expansions are used, respectively.
Chengeng
Margarida Moragues Ginard <mmoragues@bcamath.org> 于2023年4月4日周二 19:30写道：
Hi Chengeng,
Did you set up the number of adaptive iterations that you want to perform?
For example, if you want to run 2000 time steps in total and adapt the mesh every 500 time steps, so three times, you might set it up as:
<PARAMETERS>
<P> NumSteps = 500 </P>
<P> NumRuns = 4 </P>
</PARAMETERS>
Cheers,
Margarida
On Tue, 4 Apr 2023 at 11:35, Chengeng Qian <qiancg0313@gmail.com> wrote:
This email from qiancg0313@gmail.com originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list <https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email stamping for this address.
Hi everyone, I would like to test the efficiency of p-adaptive method in CompressibleFlowSolver. But I could not find any sample which shows me how to modify the session file. I have tried the same set up found in incompressible solver Tests, as follows,
<SOLVERINFO> <I PROPERTY="Driver" VALUE="Adaptive" /> </SOLVERINFO>
<PARAMETERS> <P> AdaptiveMaxModes = 7 </P> <P> AdaptiveMinModes = 4 </P> </PARAMETERS>
The CompressibleFlowSolver ran successfully (simulated forward step problem). But I didn't find any difference between the results of "Adaptive Driver" and "Standard Driver".
Any suggestions are appreciated.
Regrads, Chengeng _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/nektar-users
-- Margarida Moragues Ginard MSCA-IF Researcher *BCAM - *Basque Center for Applied Mathematics Alameda de Mazarredo, 14 E-48009 Bilbao, Basque Country - Spain Tel. +34 946 567 842 mmoragues@bcamath.org | www.bcamath.org/mmoragues
*(**matematika mugaz bestalde)*
-- Margarida Moragues Ginard MSCA-IF Researcher *BCAM - *Basque Center for Applied Mathematics Alameda de Mazarredo, 14 E-48009 Bilbao, Basque Country - Spain Tel. +34 946 567 842 mmoragues@bcamath.org | www.bcamath.org/mmoragues
*(**matematika mugaz bestalde)*
Hi Chengeng, If I am right, the Sensor variable is used both for the Arrificial Viscosity and for p-adaptivity. Cheers, Margarida On Fri, Apr 7, 2023, 4:05 PM Chengeng Qian <qiancg0313@gmail.com> wrote:
Hi Margarida, OK, I see. I thought that the Sensor variable is only for artificial viscosity.
Thanks.
Chengeng
Margarida Moragues Ginard <mmoragues@bcamath.org> 于2023年4月7日周五 00:01写道：
Dear Chengeng,
I think that the value of the Sensor variable and, as you mention, parameters "AdaptiveLowerinModes" and "AdaptiveUpperTolerance" are used to decide the expansion that will be used on each location. The Sensor variable is part of the output solution.
Cheers,
Margarida
On Wed, 5 Apr 2023 at 12:46, Chengeng Qian <qiancg0313@gmail.com> wrote:
Hi Margarida, Thanks for your response. It worked when I used your set up.
It seems that the solver reads the data every NumSteps timesteps and projects the solutions to new expansions. I guess that two parameters " AdaptiveLowerTolerance" and "AdaptiveUpperTolerance" can control the region where the new expansions are used. But we can't know the results from figures, right? There is no sensor which shows us where the higher order expansions and lower expansions are used, respectively.
Chengeng
Margarida Moragues Ginard <mmoragues@bcamath.org> 于2023年4月4日周二 19:30写道：
Hi Chengeng,
Did you set up the number of adaptive iterations that you want to perform?
For example, if you want to run 2000 time steps in total and adapt the mesh every 500 time steps, so three times, you might set it up as:
<PARAMETERS>
<P> NumSteps = 500 </P>
<P> NumRuns = 4 </P>
</PARAMETERS>
Cheers,
Margarida
On Tue, 4 Apr 2023 at 11:35, Chengeng Qian <qiancg0313@gmail.com> wrote:
This email from qiancg0313@gmail.com originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list <https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email stamping for this address.
Hi everyone, I would like to test the efficiency of p-adaptive method in CompressibleFlowSolver. But I could not find any sample which shows me how to modify the session file. I have tried the same set up found in incompressible solver Tests, as follows,
<SOLVERINFO> <I PROPERTY="Driver" VALUE="Adaptive" /> </SOLVERINFO>
<PARAMETERS> <P> AdaptiveMaxModes = 7 </P> <P> AdaptiveMinModes = 4 </P> </PARAMETERS>
The CompressibleFlowSolver ran successfully (simulated forward step problem). But I didn't find any difference between the results of "Adaptive Driver" and "Standard Driver".
Any suggestions are appreciated.
Regrads, Chengeng _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/nektar-users
-- Margarida Moragues Ginard MSCA-IF Researcher *BCAM - *Basque Center for Applied Mathematics Alameda de Mazarredo, 14 E-48009 Bilbao, Basque Country - Spain Tel. +34 946 567 842 mmoragues@bcamath.org | www.bcamath.org/mmoragues
*(**matematika mugaz bestalde)*
-- Margarida Moragues Ginard MSCA-IF Researcher *BCAM - *Basque Center for Applied Mathematics Alameda de Mazarredo, 14 E-48009 Bilbao, Basque Country - Spain Tel. +34 946 567 842 mmoragues@bcamath.org | www.bcamath.org/mmoragues
*(**matematika mugaz bestalde)*
_______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/nektar-users
participants (3)
-
Chengeng Qian
-
Margarida Moragues Ginard
-
Margarida Moragues Ginard