Example of application seismic dynamic rupture simulation using moose

[Xiao Ma]

Hi,

I am wondering is there some example regrading solving dynamic rupture simulation using moose ? 

Or as a start solving a dynamic problem in elasticity ? 

Thank you 

[Xiao Ma]

Maybe an example solving

rho*u_dotdot=DIV(Sigma)+bodyforce

[Derek Gaston]

Could someone in solid mechanics respond with how to do a dynamics problem with MOOSE?  Are there any examples of doing that with Tensor Mechanics yet?

Derek

--
You received this message because you are subscribed to the Google Groups "moose-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to moose-users...@googlegroups.com.
Visit this group at http://groups.google.com/group/moose-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/5984ff95-ddc3-4558-9e9c-897488b33002%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Michael Tonks]

I have never done a dynamics problem with Tensor Mechanics. I don't know if anyone else has.

Mike

To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/CAFfxPjpcSBfY%3DsQZh5qitHPma_GP29iur7kN5SYHu6ZbbOtfvQ%40mail.gmail.com.

[Daniel Schwen]

We do have an intern implementing a dynamic model in SolidMechanics (merged) _and_ TensorMechanics (pull request is being reworked currently).

https://github.com/idaholab/moose/issues/5537

https://github.com/idaholab/moose/issues/5559

Ben Spencer can probably comment more on that.

To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/CAJGmYifuUtXb%3Dhxt0LXfmrpzW4bpGK_Wii0sxae0-dtMXnchTA%40mail.gmail.com.

[Xiao Ma]

Thank you Daniel. 

I am wondering is there an existing example problem for solving the wave equations ? 

[Xikai Jiang]

Hi,

I've solved a simple 1D wave equation couple of months ago, and tested both central difference method and newmark method for discretizing the second time derivative. Newmark method performs better which allows larger time step and meanwhile converse the energy, while in central differencing I need to use smaller time step to avoid numerical energy dissipation.

Attached you could find my moose files and simple test for 1D wave equation using Newmark method and Central difference method. I think it could be a good idea to have a wave-equation example in moose package, since it shows a simple example to deal with second-time derivatives.

Regards,

Xikai

[Cody Permann]

Thanks Xikai, we'll take a look.

To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/560b6fd8-3ff5-49f7-b5fd-12b0635f5722%40googlegroups.com.

[Benjamin Spencer]

Xikai,

  Sorry for the late response -- I'm behind on emails.

  It looks like you already found the beginnings of a dynamics capability and the tests that we have for that in MOOSE because some of the code you sent for the Newmark method is copied from the existing capability. I agree that we need more tests to demonstrate its usage.  We have a post-doc at INL who is primarily working on the dynamics capability.  She has already implemented HHT for solid mechanics and tensor mechanics, and we are in the process of merging that in.  She will also be adding more wave propagation tests in the near future.

  Could you elaborate on what you have in mind regarding dynamic rupture? We have a number of fracture modeling capabilities, including phase field, smeared cracking, domain integrals, and XFEM either implemented or in development.  Any one of those could in principle be used on dynamics problems, at least problems that make sense for implicit time integration, although we haven't yet done that.

-Ben

Benjamin Spencer
Fuels Modeling and Simulation
Idaho National Laboratory
PO Box 1625
Idaho Falls, ID 83415-3840

To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/560b6fd8-3ff5-49f7-b5fd-12b0635f5722%40googlegroups.com.

[Xikai Jiang]

Dear Ben,

Thanks for letting us know the current dynamics capability in MOOSE.

Yes, the Newmark method in codes attached in my last email was copied from those in HHT_time_integrator tests in SolidMechanics module and then reduced from 3D to 1D. I did these tests mainly for exercise purposes, and I sent them since Xiao is asking for wave-equation examples. But maybe my codes are too introductory.. Regarding the dynamic rupture, Xiao may have better answers.

Regards,
Xikai

[Xiao Ma]

Dear Xikai and Ben,

Thank you for the instructive examples and the knowledge of the capability of moose regarding dynamic rupture. 

For my dynamic rupture problem , is that we want to develop an application for seismic rupture simulation, which has a displacement continuity on the fault surface , which is governed by a friction model. 

XFEM and phase filed approach are both suited for our problem .

Thank you ,

Xiao

On Saturday, August 22, 2015 at 2:35:02 PM UTC-5, Xiao Ma wrote:

[Benjamin Spencer]

We have a phantom node based XFEM capability for fracture in progress that we are cleaning up in preparation to be merged with MOOSE.  We're still a ways off from being able to enforce contact constraints across the cut surfaces, though.

-Ben

--

You received this message because you are subscribed to the Google Groups "moose-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to moose-users...@googlegroups.com.
Visit this group at http://groups.google.com/group/moose-users.

To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/137dc6b4-df59-46b8-b65c-305af0d7612f%40googlegroups.com.