Need to replace MINPACK ODE solver in an old Fortran code of Nuclesythesis in Supernovae II. What could be the best choice for sparse matrices?
Thanks for your suggestion.
Hi Roberto,
It strongly depends on your system. If you have a stiff system then you should find some special algorithm. In a normal case (a non-stiff) I prefer the Predictor-Corrector methods. They have much better accuracy than single step methods (Runge-Kutta...etc) and they need only a few function call (like a single step method). The implementation of Adams-Bashoford (predictor) -- Adams Moulton (corrector) is quite easy. I have recently implemented it in Fortran for arbitrary order. See the following paper, where the generation of the necessary coefficient is described for arbitrary order:
B. Seidu, Int. J, Comp. Appl. Math., 2001, Vol. 6., pp. 215-220
For the details of stiff ODEs check this webpage:
http://lh3lh3.users.sourceforge.net/solveode.shtml
I was looking for something less generic, to avoid lot of work i customization and to have good performances. It was a long time ago since I used Numerical Recipes, I'll refresh my memory and take a look at it anyhow.
Thank you both!
- Badges
- Science topic