[CIG-LONG] 回复： Re: What are the differences between staticRight, wrapRight,etc.?

[Qingwen Zhang]

Hey, Walter,
Thank you for your reply and it's a important hint for me to deal with moving boundary. I have run Gale again and get it testified.

Regards,
Qingwen

--- 原始邮件 ---

发件人： "Walter Landry" <wlandry at caltech.edu>
已发： 2015年3月3日 上午12:08
收件人： zhangqingwen at live.cn
抄送： cig-long at geodynamics.org
主题： Re: What are the differences between staticRight, wrapRight,etc.?

Hi Qingwen,

Cc'ing cig-long in case this is helpful to anyone else.

ZhangQingwen <zhangqingwen at live.cn> wrote:
> Hi, Walter,
> I have  been working on Gale for two months, but I am puzzled by the parameters
> like static, wrap, WallVC and left_equation/right_equation.
>
> For example, in the input file compression.json in cookbook examples  (
> Gale-Linux64-2_0_1\input\geologic\compression.json),
>
> ...
>     "EulerDeform":
>     {
>         "systems": [
>             {
>                 "mesh": "v-mesh",
>                 "p-mesh": "p-mesh",
>                 "displacementField": "DisplacementField",
>                 "remesher": "velocityRemesher",
>                 "velocityField": "VelocityField",
>                 "wrapTop": "True",
>                 "staticBottom": "True",
>                 "staticLeft": "True",
>                 "staticRight": "True",
>                 "left_equation": "0.01/(365*86400)*1e21 * t",
>                 "right_equation": "3200000 - 0.01/(365*86400)*1e21 * t"
>             }
>         ]
>     },
> ...
>
> As put in the User Manual (version 2.0.1, p.41), the term "staticLeft": "True"
> and "staticRight": "True" keep the left and right boundary fixed, so does it
> means that the left and right side of FEM grid  or materials are fixed?
> If so, however, the term "left_equation": "0.01/(365*86400)*1e21 *
> t" and "right_equation": "3200000 - 0.01/(365*86400)*1e21 * t" move
> these boundary sides again.

If 'staticLeft' is True, then the motion of the FEM grid is determined
by 'left_equation'.  The default for 'left_equation' is zero.  The
'static' variable names are a bit misleading, unfortunately.

> So, in my view, in case of shortening or compression experiments,
> 1. "wrapxxx": "True" make the materials move with the boundary or not,
> 2. "staticxxx": "True" fixes the grid on the boundary with respect to the
> materials,
> 3. "left_equation"and "right_equation"sets the absolute velocity of boundary
> of grid, and
> 4. "WallVC" in the "velocityBCs" sets the absolute velocity of materials on
> the boundary

This looks correct.  Is there a particular problem you are having?

> I  am not very sure about those parameters and how it controls the deformation
> of the model. The information in the User Manual seems a little concise for
> me.
>
> So do you have any instructions or suggests for this problem? Thanks for your
> time!
>
> Qingwen ZHANG
> Key Laboratory of Isotope Geochronology and Geochemistry
> Guangzhou Institute of Geochemistry of the Chinese Academy of Sciences (GIGCAS)
> +86 13822107405
> zhangqingwen at live.cn

Cheers,
Walter Landry