Re: NDSolve. SolveDelayed->True

• To: mathgroup at smc.vnet.net
• Subject: [mg57028] Re: NDSolve. SolveDelayed->True
• From: "Jens-Peer Kuska" <kuska at informatik.uni-leipzig.de>
• Date: Thu, 12 May 2005 03:53:08 -0400 (EDT)
• Organization: Uni Leipzig
• References: <d5uv4p\$983\$1@smc.vnet.net>
• Sender: owner-wri-mathgroup at wolfram.com

```Hi,

insert the scaline explicit in your equations and
solve the system
for the mass of the chemicals and not for the
concentrations.
Regards
Jens

"Chris Fernando" <ctf20 at sussex.ac.uk> schrieb im
Newsbeitrag news:d5uv4p\$983\$1 at smc.vnet.net...
>I am simulating a simple cell containing
>chemicals. I have a set of kinetic equations for
>each chemical in the cell. However, the cell
>grows at a certain rate, and this changes the
>concentration of the chemicals in it. This means
>I have to scale the concentrations of chemicals
>at each timestep of the numerical simulation,
>according to the factor [(S(t)/S(t+1)]^(3/2).
>When I was using Eular integration in a C program
>this was easy. I just did it every timestep. But
>with NDSolve, how can I scale all the equations
>for substrate concentration appropriately.
>
> In general terms, how do I deal with having to
> scale the r.h.s. of a set of differential
> equations on the basis of the function above,
> where S'[t] is also a differential equation that
> is being solved in that system!!??
>
> Yours
> Chrisantha
>

```

• Prev by Date: Re: More problems with SetPrecision[] and/or \$MinPrecision,...
• Next by Date: Re: function definition difficulty
• Previous by thread: NDSolve. SolveDelayed->True
• Next by thread: GramSchmidt problem