Re: NDSolve backward in time
- To: mathgroup at smc.vnet.net
- Subject: [mg109599] Re: NDSolve backward in time
- From: Bruno Silva <bfasilva at gmail.com>
- Date: Fri, 7 May 2010 06:30:35 -0400 (EDT)
- References: <hrtvse$s6q$1@smc.vnet.net> <4BE2ACA5.3080201@umbc.edu>
Hi Kevin,
Thanks for the tip (in fact I was misplacing the initial conditions at LH).
It's working.
I'm a bit far from QM scattering, working on Tokamak plasma transport model
(thank you for that too!).
Best wishes,
Bruno
On Thu, May 6, 2010 at 12:48 PM, Kevin J. McCann <Kevin.McCann at umbc.edu>wrote:
> Hi Bruno,
>
> I use NDSolve in the "backwards" direction for quantum mechanical
> scattering. Just do NDSolve as usual, i.e. {t,-10,10}, for example, but
> specify the initial conditions at the right-hand end.
>
> I have a notebook on QM scattering I can send you if you are interested.
>
> Kevin
>
> Bruno Silva wrote:
>
>> Hi,
>>
>> It is a common strategy to integrate an ode with time flowing in the
>> opposite direction, to get an idea of possible unstable equilibria.
>> Does anyone know how to ask NDSolve to integrate "backward" in time?
>>
>> Thanks,
>> Bruno
>>
>