Your opinion on a class of ODE problems

[Namir]

(01-27-2016 02:45 AM)gestrickland Wrote:  Namir,

In your example equations, if you add the three together, you get d(A+B+C)/dt=0, which means A+B+C=M, M being a constant throughout the solution. I suppose this is your conservation of mass, actually embedded in the equations. If I am not missing something, you could use this equation to eliminate one of the unknowns, say A, from the remaining two equations, which you could then solve for B and C with the numerical method of your choice, at each step calculating A from the above conservation relation. Would not this work?

Gordon

Good point! I can replace [A] with = M - [B] - [C]. I think it will work and make the solution a bit simpler. I can still do my optimization for a nonlinear curve fit by minimizing the square root of:

Sum of ([A(i)]obs - [A(i)]calc)^2 + ([B(i)]obs - [B(i)]calc)^2 + ([C(i)]obs - [C(i)]calc)^2 for i=1 to Number of Data points

Will give it a try!

Thanks,

Namir