Re: NDSolve - Nice function but stiffness-problem
- To: mathgroup at smc.vnet.net
- Subject: [mg96472] Re: NDSolve - Nice function but stiffness-problem
- From: "slawek" <human at site.pl>
- Date: Sat, 14 Feb 2009 03:10:10 -0500 (EST)
- References: <gh8hih$qv5$1@smc.vnet.net> <ghb04q$ok5$1@smc.vnet.net>
U¿ytkownik "Alois Steindl" <Alois.Steindl at tuwien.ac.at> napisa³ w wiadomo¶ci news:ghb04q$ok5$1 at smc.vnet.net... > Hello, > I would guess, that Mathematica uses some kind of shooting algorithm to > solve > the BVP. Since the solutions of your ODE can become infinite at finite > time, because the right hand side increases rapidly with f(x), the > shooting algorithm isn't adequate for these problems. Collocation > methods (like bvp4c in another system) work fine with these examples. This ODE has finite solutions for finite time. Never derivative D[f[x],x] is inifite for finite f[x]. BTW, Substitute f[x] -> Log[g[x]]/A D[Log[g[x]]/A,x] == g[x] D[g[x],x]/g[x]/A == g[x] D[g[x],x] == A g[x]^2 slawek