Finding optimal parameters in ODE
- From: mathieu <mathieu.malaterre@xxxxxxxxx>
- Date: Tue, 06 Nov 2007 13:44:14 -0000
Hi there,
I am thinking that I am reinventing the wheel here. I have two
discretized functions (n known values over time): X(t) and Y(t). The
behavior is defined by:
Y'(t) = A*X(t) + B*Y(t), Y(0) = 0
If I were to solve this ODE I would use a RK4...well my problem here
is not solving the ODE but finding the A and B parameter. So instead I
have (n - 1) set of linear equations with 2 unknows.
y(n+1) = y(n) + h/6 (k1+2*k2+2*k3+k4)
Right now I am brute forcing the expansian of the 4 RK coefficients
(k1,k2,k3,k4) to expand them in A and B, but I am thinking that
someone else must have done that. Right ?
Thanks for suggestion,
-Mathieu
.
- Follow-Ups:
- Re: Finding optimal parameters in ODE
- From: JCH
- Re: Finding optimal parameters in ODE
- From: pnachtwey
- Re: Finding optimal parameters in ODE
- From: BEC Charlie
- Re: Finding optimal parameters in ODE
- From: Peter Spellucci
- Re: Finding optimal parameters in ODE
- From: jdturner
- Re: Finding optimal parameters in ODE
- From: Helmut Jarausch
- Re: Finding optimal parameters in ODE
- Prev by Date: Re: Question regarding QR Factorization
- Next by Date: Re: Finding optimal parameters in ODE
- Previous by thread: DIGITAL LOGIC SOLUTIONS
- Next by thread: Re: Finding optimal parameters in ODE
- Index(es):