Message Boards Message Boards

GROUPS:

Avoid error with dependent variable while solving a PDE with NDSolve?

Posted 17 days ago
219 Views
|
1 Reply
|
1 Total Likes
|

Hello,

I'm getting the following error when trying to numerically solve a PDE:

NDSolve::fembdnl: The dependent variable in (P^(1,0))[0,t]==0 in the boundary condition DirichletCondition[(P^(1,0))[0,t]==0,x==0.] needs to be linear.

Does anyone know what this means or how to avoid the error?

The code is provided below as an image and attachment.

Thank you,

Abed

enter image description here

Attachments:

You shouldn't have "True" in NDSolve anywhere. I would say cleaning up your equation and re-post?

(it means that P, x, or t are defined so that they Evaluated to True - and as a multiple against p^{1,0} this is not a function, that is, True * f[x] is not acceptable anywhere in the PDE)

I can guarantee (you you can just see in examples in mm's help), that derivatives of P are allowed in IC or BC (allowed, not necessarily giving solutions in all cases, but would not be an error)

(in other words: Clear[P,s,x,t] or reset Quit[] kernel and remember if you give P x s t values they are no longer variables. also: x>1 cannot be multiplied times a function, see Help as to how "Solve" conditions can be used properly with PDE)

Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract