natural boundary conditions over an i... Log Out | Topics | Search
Moderators | Register | Edit Profile

FlexPDE User's Forum » User Postings » natural boundary conditions over an internal boundary « Previous Next »

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

Mark Flegg (z_wasa)
New member
Username: z_wasa

Post Number: 1
Registered: 10-2005
Posted on Monday, October 17, 2005 - 09:16 pm:   

I have a question regarding the equation that I use and how the natural boundary condition is generated from it.

I have two regions, the first has the value of say v as 0 and the other has the value of v as 1. The PDE looks something like (ommiting constants and things like that)...

dzz(H)+v*H+(1-v)*dz(H)=0

What happens here over the boundary if I say make natural(H)=0. is the only thing that effects this boundary condition the dzz term? does this mean that it will match the value and derivative over the boundary?

Thanks heaps!
Top of pagePrevious messageNext messageBottom of page Link to this message

Mark Flegg (z_wasa)
New member
Username: z_wasa

Post Number: 2
Registered: 10-2005
Posted on Monday, October 17, 2005 - 09:19 pm:   

Oh and another thing with this!

if it is not the case that it will match the derivative and the value at this point, is there anyway for me to make it do so.
Top of pagePrevious messageNext messageBottom of page Link to this message

Robert G. Nelson (rgnelson)
Moderator
Username: rgnelson

Post Number: 471
Registered: 06-2003
Posted on Tuesday, October 18, 2005 - 02:18 am:   

Natural boundary conditions are discussed in several places in the documentation. See "Natural" in the Help Index.

Quickly summarized:

FlexPDE integrates second-order derivative terms by parts. This gives rise to boundary terms. These boundary terms comprise the Natural boundary condition at outer boundaries, and define the continuity conditions at interior boundaries.

In your case, the surface term arising from dzz(H) is dz(H). This term is assumed continuous across interior boundaries. That is, no surface term is computed at interior boundaries, so continuity of the derivative is built in to the discretization.

You omitted constants and things, but be aware that in the term dz(K*dz(H)), it is K*dz(H) that is continuous, not dz(H).

Values are continuous by default, since the mesh nodes are shared between adjoining cells.
Top of pagePrevious messageNext messageBottom of page Link to this message

Robert G. Nelson (rgnelson)
Moderator
Username: rgnelson

Post Number: 472
Registered: 06-2003
Posted on Tuesday, October 18, 2005 - 02:21 am:   

PS
Natural(H)=0 is the default, interior or exterior, so you don't need to specify Natural(H)=0 on any boundary, but particularly on interior boundaries.

If you specify Natural(H) different from zero on an interior boundary, it represents a line source.

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