search

ufz / ogs5

OpenGeoSys THM/C simulator version 5
https://www.opengeosys.org/ogs-5
Other
39 stars 97 forks source link

Incorrect nodal stress/strain in partitioned coupling with flow #73

Open norihiro-w opened 8 years ago

norihiro-w commented 8 years ago

Nodal stress and strain values (usually used only for output) can be wrong if mechanics and flow are coupled in a partitioned way. This is because nodal strain values are always summed up for each solving DEFORMATION process. However, in each coupling iteration, the values should be reset to previous time step values. Nodal displacement and gauss point stress/strain values are still correct.

UPDATE: it seems nodal stresses are correct.

ddettrick commented 8 years ago

Hi Norihiro,

What is an example of mechanics and flow being coupled in a partitioned way?

Thanks for your help,

Cheers David

Please visit our new AMD website: www.acidmetalliferousdrainage.com

David Dettrick Earth Systems | Principal Environmental Engineer/Surface and Groundwater Hydrologist

14 Church Street, Hawthorn, Victoria, 3122, Australia

T +61 (0)3 9810 7500 | T +61 (0)3 9810 7505 (Direct) | M +61 (0)4 1756 1074

E david.dettrick@earthsystems.com.au | F +61 (0)3 9853 5030 | www.earthsystems.com.au http://www.earthsystems.com.au/

This email and its attachments are confidential. If you received this email in error, please notify the sender and delete.

P Please consider the environment before printing this e-mail

On 30 August 2016 at 00:43, Norihiro Watanabe notifications@github.com wrote:

Nodal stress and strain values (usually used only for output) can be wrong if mechanics and flow are coupled in a partitioned way. This is because nodal strain values are always summed up for each solving DEFORMATION process. However, in each coupling iteration, the values should be reset to previous time step values. Nodal displacement and gauss point stress/strain values are still correct.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/ufz/ogs5/issues/73, or mute the thread https://github.com/notifications/unsubscribe-auth/ANdsqLkWoNMVC3QUn4kAwQzgSmRyyH3Pks5qkvAcgaJpZM4Jvi-S .

norihiro-w commented 8 years ago

@ddettrick you can get an example from rect.zip. Change $OVERALL_COUPLING from 1 1 to 2 2 and you will see a difference in nodal strain.

I've just noticed that nodal stress is still correct. The description has been updated.

ddettrick commented 8 years ago

Thanks very much Norihiro.

Cheers David

Please visit our new AMD website: www.acidmetalliferousdrainage.com

David Dettrick Earth Systems | Principal Environmental Engineer/Surface and Groundwater Hydrologist

14 Church Street, Hawthorn, Victoria, 3122, Australia

T +61 (0)3 9810 7500 | T +61 (0)3 9810 7505 (Direct) | M +61 (0)4 1756 1074

E david.dettrick@earthsystems.com.au | F +61 (0)3 9853 5030 | www.earthsystems.com.au http://www.earthsystems.com.au/

This email and its attachments are confidential. If you received this email in error, please notify the sender and delete.

P Please consider the environment before printing this e-mail

On 30 August 2016 at 14:29, Norihiro Watanabe notifications@github.com wrote:

@ddettrick https://github.com/ddettrick you can get an example from rect.zip https://github.com/ufz/ogs5/files/444008/rect.zip. Change $OVERALL_COUPLING from 1 1 to 2 2 and you will see a difference in nodal strain.

I've just noticed that nodal stress is still correct. The description has been updated.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ufz/ogs5/issues/73#issuecomment-243330013, or mute the thread https://github.com/notifications/unsubscribe-auth/ANdsqPeSywaKEa4bWRnM96wK7Zo34W5aks5qk7GagaJpZM4Jvi-S .