[CIG-SHORT] Stress concentration near the fault tip
Josimar Alves da Silva
jsilva.mit at gmail.com
Tue Nov 8 07:24:44 PST 2016
Dear Charles,
Thank you for the suggestions. After fixing the boundary conditions I now
observe slip (if let run for many time steps), however the linear solver
take many iterations to converge. I believe this was mostly due to my poor
mesh quality, as you mention. I will fix this following previous post
recommendations.
Thank you
Josimar
On Mon, Nov 7, 2016 at 9:40 PM, Charles Williams <willic3 at gmail.com> wrote:
> Dear Josimar,
>
> After removing the z-constraint, are there still many iterations for the
> first time step? If so, things that might make for slow convergence:
>
> 1. Poor mesh quality. Make sure you check this.
> 2. Incorrect solver settings. Unfortunately, this can be a bit of an art
> sometimes.
> 3. Poor problem setup/initial conditions.
>
> Check all of these, and then look at the solution and convergence over
> time. Also, are you now able to run the problem until slip occurs?
>
> Cheers,
> Charles
>
>
> On 8/11/2016, at 1:55 PM, Josimar Alves da Silva <jsilva.mit at gmail.com>
> wrote:
>
> Dear Charles and Brad,
>
> I removed the z-constraint on the upper right boundary of my domain the
> problem seems to be fixed now.
>
> You mention that the "first time step takes a huge number of iterations,
> while subsequent steps only seem to need one iteration". Could you give
> some hints on how to improve this aspect of my simulation ?
>
> Thank you
> Josimar
>
>
>
> On Mon, Nov 7, 2016 at 6:05 PM, Charles Williams <willic3 at gmail.com>
> wrote:
>
>> Dear Josimar,
>>
>> See if removing the z-constraint helps your solution. Looking at your
>> log, the first time step takes a huge number of iterations, while
>> subsequent steps only seem to need one iteration.
>>
>> I did not realize that you had gravity turned on. That would provide
>> sufficient normal stress to clamp the fault. Also, you would not be able
>> to apply tractions in the same place you are applying displacement BC.
>>
>> If the only purpose of having gravity in your problem is to provide
>> normal stress on the fault, it is probably simpler to compute the
>> overburden stress and apply it to the fault using db_initial_stress, and
>> turn off gravity. That would also mean you wouldn’t need the traction BC
>> on the left side of your domain.
>>
>> Cheers,
>> Charles
>>
>>
>> On 8/11/2016, at 4:54 AM, Brad Aagaard <baagaard at usgs.gov> wrote:
>>
>> Josimar,
>>
>> Why are you constraining the tangential (z) displacement on the right
>> hand side boundary? I think this may be causing the large tractions around
>> the fault on the right hand side.
>>
>> You should not mark the buried edges of the fault on edges of the fault
>> that reach the edges of the domain.
>>
>> Brad
>>
>>
>> On 11/06/2016 04:31 PM, Josimar Alves da Silva wrote:
>>
>> Dear Charles,
>>
>> Thank you so much for the reply. I am attaching a figure with the
>> correct boundary conditions.
>>
>> My questions is:
>>
>> 1) What causes the high normal and shear stresses near the fault tip ?
>>
>> Would it be possible that there is something wrong on the way I am
>> parameterizing the fault edge ? I have tried turning on the "edges"
>> command on the set up but I still observe the high stresses.
>>
>> My simulation only runs for 15 years. At the end of the simulation there
>> is no fault displacement. However, only the high stress values indicated
>> above. I wonder if this behavior is normal because the failure criteria
>> was not met ?
>>
>> Best,
>> Josimar
>>
>>
>>
>>
>>
>>
>>
>> On Sun, Nov 6, 2016 at 1:25 PM, Charles Williams <willic3 at gmail.com
>> <mailto:willic3 at gmail.com <willic3 at gmail.com>>> wrote:
>>
>> Dear Josimar,
>>
>> I am somewhat confused about your boundary conditions. Are any
>> conditions being placed on the top and bottom boundaries? Also, it
>> appears that you are applying Neumann BC on the left boundary,
>> rather than the fixed displacement BC shown in your figure. Is this
>> correct, or is something mislabelled?
>>
>> Cheers,
>> Charles
>>
>>
>> On 5/11/2016, at 11:24 AM, Josimar Alves da Silva
>> <jsilva.mit at gmail.com <mailto:jsilva.mit at gmail.com
>> <jsilva.mit at gmail.com>>> wrote:
>>
>> Dear All,
>>
>> I am running PyLith with a 2D fault friction model, including
>> gravitational effects.
>>
>> I am finding stress concentration near the tip of the fault and I
>> would like to understand why this occurs. See attached figure.
>>
>> On the link below Brad seems to address this issue, however it is
>> still not clear to me why this happens.
>>
>> http://lists.geodynamics.org/pipermail/cig-short/2014-Nov
>> ember/001924.html
>> <http://lists.geodynamics.org/pipermail/cig-short/2014-No
>> vember/001924.html>
>>
>> Details about my simulation are attached (log files, boundary
>> conditions...)
>>
>> Thank you in advance for your time.
>> Best,
>> Josimar
>> <pylith_problem.pdf><log.out>____________________________
>> ___________________
>> CIG-SHORT mailing list
>> CIG-SHORT at geodynamics.org <mailto:CIG-SHORT at geodynamics.org
>> <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>
>>
>>
>> Charles A. Williams
>> Scientist
>> GNS Science
>> 1 Fairway Drive, Avalon
>> PO Box 30368
>> Lower Hutt 5040
>> New Zealand
>> ph (office): 0064-4570-4566
>> fax (office): 0064-4570-4600
>> C.Williams at gns.cri.nz <mailto:C.Williams at gns.cri.nz
>> <C.Williams at gns.cri.nz>>
>>
>>
>>
>>
>> _______________________________________________
>> CIG-SHORT mailing list
>> CIG-SHORT at geodynamics.org
>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>
>>
>> _______________________________________________
>> CIG-SHORT mailing list
>> CIG-SHORT at geodynamics.org
>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>
>>
>> Charles A. Williams
>> Scientist
>> GNS Science
>> 1 Fairway Drive, Avalon
>> PO Box 30368
>> Lower Hutt 5040
>> New Zealand
>> ph (office): 0064-4570-4566
>> fax (office): 0064-4570-4600
>> C.Williams at gns.cri.nz
>>
>>
>> _______________________________________________
>> CIG-SHORT mailing list
>> CIG-SHORT at geodynamics.org
>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>
>
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>
>
> Charles A. Williams
> Scientist
> GNS Science
> 1 Fairway Drive, Avalon
> PO Box 30368
> Lower Hutt 5040
> New Zealand
> ph (office): 0064-4570-4566
> fax (office): 0064-4570-4600
> C.Williams at gns.cri.nz
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.geodynamics.org/pipermail/cig-short/attachments/20161108/9621f69a/attachment-0001.html>
More information about the CIG-SHORT
mailing list