Hello,
I would like to solve an optimization problem with the casadi ipopt solver and have written my code with opti stack.
My objective function is a sum over T with a variable number of steps (according to the optimization horizon) and thus my opti.variables also have T different iterations.
This T should be able to go up to 300 iterations, while I'm having up to 11 different sets of variables depending on the complexity level, that I want my objective function to be.
Now the solver is really slow (for T=96) it took 136 min to solve the problem, even though I had the very basic objective function with only 7 different variable sets. On plus, I had a constraint so that 3 of the variable sets would equal another 3 variable sets.
Now I tried to speed the code up using
opts = {"ipopt.hessian_approximation":"limited-memory"}
but then I got the error message:
Problem with interger stack size 1 1 6.
Thus, I removed it again and ended up with the same message somehow.
Is it possible, that I have a mistake in the code, or is it normal that the solver need so much time?
I would appreciate it a lot, if you would have some hints for me. Also I could share more code, if that is helpful.
Thank you!