Hi Robert, at first thank you for your help, I have solved the problem, and now I want to use "sfepy/examples/multi_physics/biot.py", I suspect to add Dirichlet (essential) boundary conditions in the biot. Here I want to fix the bottom and load a press on the left facet and right facet, the press can be same or different, actually I want more is to load a linear function(like,P = γgh) from Top to Bottom along the z axis. How can I achieve it.

r.

在 2015年5月23日星期六 UTC+8上午12:21:15，Robert Cimrman写道：

Hi Kid,

see [1] - you need to generate a structured grid and set some further options to have a fully hexahedral grid. (I have never used gmsh in this mode, so I have no more details.)

r.

[1] http://geuz.org/gmsh/doc/texinfo/gmsh.html#Mesh-module

On 05/21/2015 02:33 PM, Kid Guo wrote:

...

Hi

I'm using SfePy to solve a problem.Firstly, I creat a simple cube by

Gmsh

...

and save as my.mesh file. Now I am using sfepy/examples/multi_physics/biot.py to calculate.but it something wrong. Then I open the "/meshes/3d/cube_medium_hexa.mesh",in the "cube_medium_hexa.mesh" file,it consists Vertices and Hexahedra.But

in my

...

mesh file,it only consists Vertices,Edges and Triangles,no Hexahedra.

When

...

I run the biot.py, there aren't any results I want. How can I save .mesh file including Hexahedra? I think it is the main probelm.

Regards, Kid

at first thank you for your help, I have solved the problem.I have seen this example,sinbc()function.and I add a new function in biot.py. but it
something wrong. 1.I added a new facet named Right and select its region.Like "'Right' : ('vertices in (x > 0.999999)', 'facet')," 2.a new function added in biot.py. 3.I load a stress to the left and right facet. if the stress is a constant,it worked.but when I added a new function on the left facet and another new function on the right facet,it didn't work.

r. 在 2015年5月27日星期三 UTC+8下午3:41:59，Robert Cimrman写道：

On 05/26/2015 11:04 AM, Kid Guo wrote:

...

Hi Robert, at first thank you for your help, I have solved the problem, and now I

want

...

to use "sfepy/examples/multi_physics/biot.py", I suspect to add

Dirichlet

...

(essential) boundary conditions in the biot. Here I want to fix the

bottom

...

and load a press on the left facet and right facet, the press can be

same

...

or different, actually I want more is to load a linear function(like,P =

γgh) from Top

...

to Bottom along the z axis. How can I achieve it.

Check [1] - sinbc() function - it shows how to apply Dirichlet BC using a function.

r.

[1] http://sfepy.org/doc-devel/examples/diffusion/sinbc.html

On 06/03/2015 03:41 PM, Kid Guo wrote:

at first thank you for your help, I have solved the problem.I have seen this example,sinbc()function.and I add a new function in biot.py. but it something wrong. 1.I added a new facet named Right and select its region.Like "'Right' : ('vertices in (x > 0.999999)', 'facet')," 2.a new function added in biot.py. 3.I load a stress to the left and right facet. if the stress is a constant,it worked.but when I added a new function on the left facet and another new function on the right facet,it didn't work.

What exactly does not work? I am getting (with biot.py from your next message) the linear pressure distribution in z direction on the left and right facets, as the functions for Dirichlet conditions on pressure enforce.

BTW. what you apply is the fluid pressure in the pores. To apply a stress load on the boundary, you would need to use the 'dw_surface_ltr' term (as in [1]), not the Dirichlet BC on pressure.

r. [1] http://sfepy.org/doc-devel/examples/linear_elasticity/linear_elastic_tractio...

On 06/04/2015 04:27 AM, Kid Guo wrote:

Maybe I can not express clearly, I point out something that I don't understand, you will know what I mean.I put the pictures in attachment named figure 1-7. 1.From the figure 3 and figure 4.the stress is a constant,it worked.

ok

2.From figure 2, the left boundary loaded a porewater pressure function and the right loaded a pressure--a constant equal 1000.but the figure 2 show the right pressure is 0.

No, look at the scale (range of pressure in the colorbar in left-unequal-right_pore.png) - 1000 is about 100x smaller than the max. pressure (about 1e5), so it looks blueish. But its blue is lighter then the blue for the zero. Try to rerun that with a higher constant, e.g. 1e5 :)

3.From figure 6 and 7. if the value in two function is equal,both they are p=p2=9.8*10e3*z, it worked,we can see it from the figure 6.but if it is unequal, p2=9.8*10e2*z, the figure 7 show the right pressure is 0.

I guess it is the same here.

r.