[CIG-SHORT] Pylith problem

[Brad Aagaard]

Qingbiao,

If you look at the aspect ratio information from LaGriT, you will see
that you have some cells with aspect ratios that are < 0.01. In fact,
some are even < 1.0e-5. These flat cells are likely causing problems
when PyLith attempts to create cohesive cells. These will also affect
the solution. You should work on your mesh until the minimum aspect
ratio from LaGriT is greater than about 0.2-0.4.

Regards,
Brad


On 09/25/2013 08:13 PM, 范清彪 wrote:
>     
>     I checked again, but still can not find the problem. Maybe something was neglected.  The .gmv (ASCII) and .pset files are attached.
>     Really appreciate your help.
>     Best regards.
> 
> Qingbiao
> 
> ----- Original Message -----
> From: Brad Aagaard
> To: 范清彪
> Date: 2013-09-25, 23:04:16
> Subject: Re: [CIG-SHORT] Pylith problem
> 
> 
> Before you post the files, check to make sure you call the following
> commands after connecting the points.
> 
> settets / geometry
> resetpts/cell_color
> 
> resetpts
> rmpoint / compress
> 
> If you do call these commands, then post the .gmv (ASCII) and .pset files.
> 
> Brad
> 
> 
> On 09/25/2013 07:57 AM, 范清彪 wrote:
>> I am using the Lagrit. What files should I post to you ?
>>
>> Qingbiao Fan
>>
>>
>> ----- Original Message -----
>> From: Brad Aagaard
>> To: 范清彪
>> Date: 2013-09-25, 22:47:13
>> Subject: Re: [CIG-SHORT] Pylith problem
>>
>>
>> On 09/25/2013 06:41 AM, 范清彪 wrote:
>>> I can sure that the vetices selected for fault f001 correspond to an
>>> interior surface with no holes or gaps. Each cell has only one face
>>> on the fault. Is there any other problem cause this error ? What
>>> does "partially interpolated" mean ? Thanks. Best wishes. Qingbiao
>>> Fan
>>
>> "Partially interpolated" refers to how many levels (vertices, edges,
>> faces, cells) the mesh topology has. The lower dimension mesh PyLith
>> assembled from your list of vertices has more than just vertices and faces.
>>
>> What mesh generator are you using? If you are using CUBIT/Trelis, then
>> please post your Exodus file where I can download it, so I can take a
>> look to see what the problem might be.
>>
>> Regards,
>> Brad
>>
>>
>>>
>>>
>>> ----- Original Message ----- From: Brad Aagaard To:
>>> cig-short,qbfanvan Date: 2013-09-24, 23:06:57 Subject: Re:
>>> [CIG-SHORT] Pylith problem
>>>
>>>
>>> This error shows up when there are inconsistencies in setting up the
>>> fault from the group of vertices. I suspect that the vertices you
>>> selected for fault 'f001' do not correspond to an interior surface.
>>> Make sure the group of vertices corresponds to a simply connected
>>> surface (no holes or gaps) and that each cell has only one face (3-D)
>>> or edge (2-D) on the fault.
>>>
>>> Regards, Brad Aagaard
>>>
>>> On 09/24/2013 01:16 AM, 范清彪 wrote:
>>>> Hi I am using Pylith to simulate earthquake postseismic
>>>> deformation. A problem occurs:
>>>>
>>>> RuntimeError: Error occurred while adjusting topology to create
>>>> cohesive cells for fault 'f001'. Cannot handle partially
>>>> interpolated meshes
>>>>
>>>> I don't known what the problem is with the meshes. Please help.
>>>> Thanks and regards. Qingbiao Fan
>>>>
>>>>
>>>>
>>>> _______________________________________________ CIG-SHORT mailing
>>>> list CIG-SHORT at geodynamics.org
>>>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>>