>Agreed, but what if I don't know before hand what the frequency is?
As you've probably gathered from Marc Roussel's post, you have
to know something about the local of the closed orbit to build a
Poincare map. If your numerical algorithm looks for crossings in
the way he described, then you don't need to know the frequency
a priori. You should be able to get a decent idea of the local
by running the model in simulation mode. I haven't heard back
from my friend at NCSU (I think he's giving finals) but I'm fairly
sure that he uses XPP--it's easy to run simulations in it and plot
the results.
I'm 90% certain of the following: If you're system is linear in
the dependent variables and it does have a periodic forcing
function (*any* periodic function) of period T, and you can show
that the eigenvalues of the Jacobian are always negative, then you
are guaranteed a stable, periodic solution of period T. Finding
it's local is trickier.
-Paul
"Saying Windows 95 is equal to Macintosh is like finding a
potato that looks like Jesus and believing you've witnessed
the second coming." -- Guy Kawasaki, Apple Fellow
