[CIG-SHORT] 回复: Re: An error when running Pylith .
faqidiao
faqidiao at asch.whigg.ac.cn
Tue Nov 1 20:08:23 PDT 2011
Dear All:
Thank you all for your kindly help in pylith using. As a Pylith user, here I want to offer
some suggestions from my experience of Pylith learning and using.
Firstly, Pylith is a very useful software for simulation of co- and post-seismic deformation
which can consider different earth model. Secondly, All of the Pylith developers are very
kind in helping the users.
However, I still think that the Pylith is not very convenient for geo-scientists to use.
One important reason, in my opinion, perhaps is the complexity and difficult of the mesh
construction. As to real problem with intricate fault geometry and earth structure, the mesh
construction will take much time and general the user can not generate perfect mesh.
This limits the wide use of Pylith to a great extent.
For users we want to pay more attention on the scientific problems itself ,not the tools
though it is also very important. So, I don't know if you can add another function: mesh
construction. In fact, I think it is not very difficult for you because the procedure is similar
for different model (earthquake).
For example, recently I use Pylith to model the co-seismic deformation of 2008 Mw7.9 Wenchuan
earthquake and the 2011 Mw9.0 Japan earthquake, by considering the real earth structure, and
invert the slip distribution with 3D Green function. If one of you are interested in this problem,
we can make a cooperation. I do the simulation ,inversion and analysis, and you construct
the mesh. In addition, you can make cooperation with scientist over the world who use Pylith.
Here is just my hope and suggestions, and I reckon on your support !
Best wishes,
Faqi
发件人： Brad Aagaard
发送时间： 2011-11-02 09:55
收件人： faqidiao
抄送： Charles Williams; Matthew Knepley; faqidiao
主题： Re: [CIG-SHORT] An error when running Pylith .
Faqi,
Following up on Charles's comment. You may have nearly coincident points
in your mesh that LaGriT is connecting together, resulting in slivers
elements. You may want to try removing coincident points and see if that
removes all of the bad cells. If you still have bad cells, then use some
of the other mesh quality tools to improve the mesh. The minimum aspect
ratio in LaGriT should be about 0.2 or larger (ParaView uses the inverse
so with it you want an aspect ratio less than about 5).
Brad
On 11/01/2011 06:03 PM, Charles Williams wrote:
> Hi Faqi,
>
> After looking at your mesh, the problem appears to be extremely poor mesh
> quality at the fault. In the attached figure, I calculated the mesh quality
> using ParaView, and then thresholded to show the really bad elements. There are
> a bunch of elements with essentially zero thickness right on the fault. Note
> that I created this by outputting the mesh in AVS format and looking at it in
> ParaView, but you can also check the mesh quality in LaGriT. Try improving the
> mesh quality and see what happens.
>
> Cheers,
> Charles
>
> [see attached file: wc_mesh_quality.jpg]
>
> On 1/11/2011, at 8:37 PM, faqidiao wrote:
>
>> Dear all:
>> I try to solve my problem following your suggestions. However, the error still
>> exist.
>> This time I construct a simple fault mesh with no curvature and just a single
>> fault.
>> I have checked my model again and again and did not find where the problem is.
>> Without any other way, I have to ask for your help again. Attached is my model,
>> I don't know if you have time checking my model. In addition, please have a check
>> of my 'cfg' file at the same time.
>> Thank you all again.
>> Best wishes,
>> Faqi
>> --------------------------------------------------------------------------------
>> *From:* Charles Williams<mailto:willic3 at gmail.com>
>> *Date:* 2011-10-27 03:49
>> *To:* Matthew Knepley<mailto:knepley at mcs.anl.gov>
>> *CC:* faqidiao<mailto:faqidiao at asch.whigg.ac.cn>;faqidiao
>> <mailto:faqidiao at gmail.com>;cig-short<mailto:cig-short at geodynamics.org>
>> *Subject:* Re: [CIG-SHORT] An error when running Pylith .
>> It can also happen if there is too much fault curvature. For this to happen,
>> there must be 3 faces on the fault, which means pretty extreme curvature or
>> your cell size is not fine enough for the fault geometry you are representing.
>>
>> Charles
>>
>>
>> On 27/10/2011, at 4:29 AM, Matthew Knepley wrote:
>>
>>> On Wed, Oct 26, 2011 at 10:13 AM, faqidiao<faqidiao at asch.whigg.ac.cn
>>> <mailto:faqidiao at asch.whigg.ac.cn>>wrote:
>>>
>>> __
>>> Brad,
>>> Thanks for your rapid response. I try my problem with new version
>>> (pylith-1.6.1). However, error still occurred.
>>> Please find the attached error information. I agree with you, this
>>> perhaps induced by the fault topology. BUT,
>>> I have checked my Lagrit running and it's result , all seems well. So, I
>>> still puzzled about this.
>>> I also attached my problem, and I don't know whether you have time to run
>>> my model and check where the
>>> problem is?
>>>
>>>
>>> Here is the problem. Somewhere in your mesh, there is a terahedron with all 4
>>> vertices on a fault. This can happen
>>> because
>>>
>>> a) there is a problem in your fault description
>>>
>>> b) You have two faults which get too close to each other
>>>
>>> Thanks,
>>>
>>> Matt
>>>
>>> Thank you again!
>>> Faqi Diao
>>> --------------------------------------------------------------------------------
>>> *From:* Brad Aagaard<mailto:baagaard at usgs.gov>
>>> *Date:* 2011-10-26 12:45
>>> *To:* faqidiao<mailto:faqidiao at asch.whigg.ac.cn>
>>> *CC:* cig-short<mailto:cig-short at geodynamics.org>
>>> *Subject:* Re: [CIG-SHORT] An error when running Pylith .
>>> Faqi,
>>> The error message suggests that the nodeset/pset you are using to create
>>> the fault has a problem. This can happen if the fault has holes in it or
>>> there is a problem with the topology of the mesh. Another possibility is
>>> that there is a bug. We have made a lot of improvements since v1.4.2, so
>>> I recommend that you use the current release (v.1.6.1) and see if you
>>> still get the same error message.
>>> Regards,
>>> Brad
>>> -----faqidiao<faqidiao at asch.whigg.ac.cn
>>> <mailto:faqidiao at asch.whigg.ac.cn>> wrote: -----
>>> =======================
>>> To: cig-short<cig-short at geodynamics.org<mailto:cig-short at geodynamics.org>>
>>> From: faqidiao<faqidiao at asch.whigg.ac.cn<mailto:faqidiao at asch.whigg.ac.cn>>
>>> Date: 10/25/2011 07:07PM
>>> Subject: [CIG-SHORT] An error when running Pylith .
>>> =======================
>>> Hi all,
>>> Recently, I met a problem when running Pylith1.4.2, and I don't know how
>>> to fix
>>> it, though I have paid much time.
>>> Attached is the error message, can you have a look and give some suggestions
>>> on how to solve the problem?
>>> Thank you in advance!
>>> Best regards,
>>> Faqi Diao_______________________________________________
>>> CIG-SHORT mailing list
>>> CIG-SHORT at geodynamics.org<mailto:CIG-SHORT at geodynamics.org>
>>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>> [attachment(s) "error message.png" removed by Brad Aagaard/GD/USGS/DOI]=
>>>
>>> _______________________________________________
>>> CIG-SHORT mailing list
>>> CIG-SHORT at geodynamics.org<mailto:CIG-SHORT at geodynamics.org>
>>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>
>>>
>>>
>>>
>>> --
>>> What most experimenters take for granted before they begin their experiments
>>> is infinitely more interesting than any results to which their experiments lead.
>>> -- Norbert Wiener
>>> _______________________________________________
>>> CIG-SHORT mailing list
>>> CIG-SHORT at geodynamics.org<mailto:CIG-SHORT at geodynamics.org>
>>> http://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>
>>
>> <wenc-test.zip><readme.doc><error message.png>
>
> 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>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://geodynamics.org/pipermail/cig-short/attachments/20111102/0eb962e9/attachment-0001.htm
More information about the CIG-SHORT
mailing list