[CIG-SHORT] Dynamic simulation with absorbing boundary

[Brad Aagaard]

Birendra,

Imposing multiple traction related boundary conditions on a boundary 
will result in their effects being added together. The AbsorbingDampers 
BC applies tractions proportional to velocity. Superimposing a Neumann 
BC will cause motion that will be partially damped by the absorbing 
boundary condition. I doubt this is the type of behavior you are trying 
to simulate.

If all you need is prescribed tractions on the fault, then using 
traction perturbation prescribed on the fault is your best approach.

If you need stresses in the bulk medium too, then you can impose initial 
stress field in the bulk medium. This is what we use in the SCEC dyanmic 
rupture benchmarks with plasticity to setup consistent fault tractions 
and stresses in the bulk medium.

Regards,
Brad



On 09/17/2013 12:39 PM, Birendra jha wrote:
> Hi
>
> I am working on a problem similar to
> bar_shearwave/quad4/spontaneousrup.cfg. The domain is a square, fault
> is slightly inclined, and instead of driving the fault with traction
> perturbations, I am using Neumann BC on top and right. Given that it
> creates a normal faulting condition, I would like to use Absorbing
> Boundary on the left and right edges. Is it possible to have both
> Neumann and Absorbing BC on same boundary (like Dirichlet and
> Absorbing on the same boundary)? It seems not, because absorbing BC
> sets the boundary traction equal to that of the outgoing wave, and it
> does not have to be equal to the imposed Neumann tractions.
>
> For such a setup (dynamic simulation, explicit timestepping, fault
> driven by boundary tractions), how should I impose the Absorbing BC?
>
> Thanks and regards Birendra
> _______________________________________________ CIG-SHORT mailing
> list CIG-SHORT at geodynamics.org
> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>