How to use previous results of a non-linear function as the initial "guess" for the next step?

Hi all,

I am working on a model that requires 21 initial guesses, right now I have my initial guess set at what works best for the very first instance (i = 1). However, over the course of this model, the non-linear function solver is run 500+ times using the same initial guess.

My question is, is there anyway that I can use the previous results ( = i - 1) as the initial guess for subsequent solving of the function. I would like to do this because my values change very little from step to step.

Thanks!

Answer 1

[Mohammad Abouali]

Instead of hard coding the initial guess take that in as an input

P=[initial values]

P=myNLfuction(someinput, P) % this uses the initial values for your first

some more code here

P=myNLfuction(someinput, P) % this uses the last output for as initial value.