an iteration for boundary conditon Log Out | Topics | Search
Moderators | Register | Edit Profile

FlexPDE User's Forum » User Postings » an iteration for boundary conditon « Previous Next »

Author Message
Top of pagePrevious messageNext messageBottom of page Link to this message

lee (leemu215)
New member
Username: leemu215

Post Number: 1
Registered: 06-2008
Posted on Tuesday, June 10, 2008 - 06:01 am:   

hello,i just beginn to learn this software,and i got a problem here.
is it possible to do an iteration on the boundary?
for example,for 1 D problem,the boundary is from (0) line to (x),and x changes with the last solution.
Top of pagePrevious messageNext messageBottom of page Link to this message

Robert G. Nelson (rgnelson)
Moderator
Username: rgnelson

Post Number: 1128
Registered: 06-2003
Posted on Tuesday, June 10, 2008 - 02:19 pm:   

FlexPDE automatically iterates on nonlinear boundary conditions, so you don't need to do anything special about that.

However, it appears that what you want is not a boundary condition modification, but a redefinition of the problem domain.

You can do this in two ways:

1) Use a STAGED definition to sweep through a set of predefined sizes. You can plot a History of some evaluation parameter, and use this plot to manually select the preferred domain size. See "Staged Geometry" in the Help Index, and the example "Samples | Misc | Stage_Geom.pde".

2) Construct a moving mesh problem. It would probably be best to use a pseudo-time-dependent model and define a boundary velocity that depends on internal conditions. This velocity should, of course, fall to zero for some achievable condition. See "Moving Meshes" in the Help Index and the examples in the "Samples | Moving_Mesh" folder.

Top of pagePrevious messageNext messageBottom of page Link to this message

lee (leemu215)
New member
Username: leemu215

Post Number: 2
Registered: 06-2008
Posted on Tuesday, June 17, 2008 - 09:24 am:   

thx ,you are right,i want a redefinition of the problem domain.
Moving mesh seems not able to change the boundary value,or could you please show me what is wrong with this code?
BTW,how to get the coodinate value of the intersection point?
application/octet-stream
1d_problem01.pde (1.1 k)
Top of pagePrevious messageNext messageBottom of page Link to this message

Marek Nelson (mgnelson)
Moderator
Username: mgnelson

Post Number: 46
Registered: 07-2007
Posted on Tuesday, June 17, 2008 - 07:33 pm:   

This script is not posed as a moving mesh problem.

You have given FlexPDE a region path that is not constant. This is not how you pose a moving mesh problem. You construct the path with stationary points, then define the velocity or trajectory of the points on the path. You also need to define a surrogate variable for each coordinate that you want to move. See the Help section "User Guide : Moving Meshes" and the example scripts mentioned above. These will show you how to construct a moving mesh problem.

I am not sure what you are asking about "the coordinate value of the intersection point".

PS - Your region 1 is completely overlayed by regions 2 & 3 and therefore ignored.
Top of pagePrevious messageNext messageBottom of page Link to this message

lee (leemu215)
Junior Member
Username: leemu215

Post Number: 3
Registered: 06-2008
Posted on Monday, June 23, 2008 - 03:58 pm:   

it is true,this script is not posed as a moving mesh problem,and I don't want to do a moving mesh problem.I want to make a calculation whose boundary lies a nonconstant path.
for the" the coordinate value of the intersection point" I mean when two curves meet each other,how can I know the where is the intersection point,coz the graph does not show the coordinates of the meeting point.Thx a lot !
Top of pagePrevious messageNext messageBottom of page Link to this message

Robert G. Nelson (rgnelson)
Moderator
Username: rgnelson

Post Number: 1143
Registered: 06-2003
Posted on Wednesday, June 25, 2008 - 02:36 pm:   

You said you had a 1D problem with a domain from 0 to x, and you want to move x.
Since FlexPDE builds a computation mesh spanning your domain, moving the end point moves the mesh. Ergo, moving mesh.

Also, in 1D you don't have any curves, so what is intersecting?
Top of pagePrevious messageNext messageBottom of page Link to this message

lee (leemu215)
Member
Username: leemu215

Post Number: 5
Registered: 06-2008
Posted on Sunday, June 29, 2008 - 06:12 am:   

sorry for the misunderstanding.
It is a 1D problem.I draw some curves .one axis is dimension,the other is the stress.two curves at a certain x,have the same stress.I want to know the stress value and coodinates.
Top of pagePrevious messageNext messageBottom of page Link to this message

Robert G. Nelson (rgnelson)
Moderator
Username: rgnelson

Post Number: 1146
Registered: 06-2003
Posted on Sunday, June 29, 2008 - 05:47 pm:   

There is no mechanism for finding intersections of plot curves.
You'll have to do it visually.

Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action:

Topics | Last Day | Last Week | Tree View | Search | Help/Instructions | Program Credits Administration