| Author |
Message |
   
shi liang (caleb_shi)
Member Username: caleb_shi
Post Number: 6 Registered: 12-2004
| | Posted on Wednesday, January 12, 2005 - 07:24 am: | |
Hi, Sir. last time you said :"FlexPDE uses Newton-Raphson iteration automatically whenever the problem is nonlinear. You don't have to do anything." if we have 2 equations to be solved in flexpde. 1) the schrodinger equation, with V, phi, E 2) poisson equation, with V, phi and E V with boundary specified. Can this be solved automatically by Flexpde? |
   
Robert G. Nelson (rgnelson)
Moderator Username: rgnelson
Post Number: 304 Registered: 06-2003
| | Posted on Wednesday, January 12, 2005 - 05:19 pm: | |
Your script requests an eigenvalue calculation, but the equations are nonlinear. Your term V*Phi requires the product of eigenfunctions. FlexPDE cannot solve nonlinear eigenvalue problems. My remarks about automatic Newton iteration apply only to direct solution of PDE's. The best I can suggest is to solve the PDE system directly, with lamda as a parameter. Use STAGEs and sweep through values of lambda as a parameter study. Solution of a linear eigenvalue problem of similar characteristics might give you a range of lambdas to search.
|
|