[CIG-SHORT] Initial stress interaction with Dirichlet bc

Tue Jan 24 14:49:33 PST 2017

Dear Brad,

I have a Drucker-Prager material.

"You can either (a) use a static simulation or initial stresses to create
the stress distribution at t=0 (assuming you have a linear elastic
material) or (b) use Neumann BC in combination with Dirichlet BC."

I've tried both (a) and (b).

The problem with (a) (initial stresses db in my case) is that Dirichlet bc on +y and -y start working BEFORE the block got compressed from the initial stresses and boundaries themselves stay fixed.

The problem with (b) is that it seems there's no displacement compatible with normal compression. If I compress the -y boundary to any extent the displacement due to normal traction on +x and -x then changes from the original one and I end up with non universal compression.

Thanks!

Brad Aagaard baagaard at usgs.gov <mailto:cig-short%40geodynamics.org?Subject=Re%3A%20%5BCIG-SHORT%5D%20Initial%20stress%20interaction%20with%20Dirichlet%20bc&In-Reply-To=%3C9e5899d1-0cfc-bd85-3697-16d578e7c7a8%40usgs.gov%3E>
Fri Jan 20 14:09:16 PST 2017

  *   Previous message: [CIG-SHORT] Initial stress interaction with Dirichlet bc<http://lists.geodynamics.org/pipermail/cig-short/2017-January/002693.html>
  *   Next message: [CIG-SHORT] Initial stress interaction with Dirichlet bc<http://lists.geodynamics.org/pipermail/cig-short/2017-January/002697.html>
  *   Messages sorted by: [ date ]<http://lists.geodynamics.org/pipermail/cig-short/2017-January/date.html#2694> [ thread ]<http://lists.geodynamics.org/pipermail/cig-short/2017-January/thread.html#2694> [ subject ]<http://lists.geodynamics.org/pipermail/cig-short/2017-January/subject.html#2694> [ author ]<http://lists.geodynamics.org/pipermail/cig-short/2017-January/author.html#2694>

________________________________

Ekaterina,

It sounds like you are trying to do the following:

t=0: Apply Dirichlet BC that constrain the normal displacements on the
+x, -x, +y, and -y boundaries.

t>0: Apply Dirichlet BS on -y boundary constraining the x and y
displacements; apply Dirichlet BC on +y that constrains the y
displacement (constant rate as a function of time).

PyLith does not currently support changing which DOF are constrained as
a function of time.

You can either (a) use a static simulation or initial stresses to create
the stress distribution at t=0 (assuming you have a linear elastic
material) or (b) use Neumann BC in combination with Dirichlet BC.

For (b) I think you would do something like:

i. Neumann BC with normal tractions on -x, +x, and +y.

ii. Dirichlet BC on -y as in t>0 with x displacement compatible with the
compression in the x direction associated with the Neumann BC. Dirichlet
BC on x component of +y as in t>0 compatible with the  compression in
the x direction associated with the compression.

If you have a homogeneous linear isotropic material, then I would just
impose the initial stresses associated with the compression. Otherwise,
I think you will need to use a static simulation and take the output and
apply them as initial stresses.

Regards,
Brad

On 01/20/2017 11:43 AM, Ekaterina Yevgenyevna Bolotskaya wrote:
> Dear Developers,
>
> I'm currently working on a 2D simulation where I need to initially
> compress a 2D block in both directions and then fix both displacements
> of lower horizontal boundary and apply a horizontal displacement rate on
> the upper horizontal boundary.
> The problem is that my initial compression interacts with my Dirichlet
> bc: I have the whole block compressed but the horizontal boundaries
> themselves (I'm talking about time 0, when there's no rate yet). I get
> smth like this:
>
>
> While I want to compress the whole block uniformly and only after have
> the compressed block fixed.
> Could you tell me if I can do that and how?
>
> Thanks!
>
>
>
> Best regards,
> Ekaterina Bolotskaya
>
> PhD in Geophysics,
> Earth, Atmospheric and Planetary Science Department,
> Massachusetts Institute of Technology
> E-mail. bolee at mit.edu<http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short>
> Mob. +1 (857) 284-2805 <tel:%2B1%20%28857%29%20284-2805>
>          +7 (963) 995-36-33 <tel:%2B7%20%28963%29%20995-36-33>
>
>
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org<http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short>
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.geodynamics.org/pipermail/cig-short/attachments/20170124/e7f60740/attachment.html>