[Ipopt] slow convergence

[Mostafa.B]

Hi Juan,

I have had similar problems with other solvers in AIMMS when the number of
variables and constraints in my model increased, I am not a professional
some tips that might help you are:
1- reduce the number variables and constraints, combine them as much as
possible if you are using a modelling environment such as AIMMS, AMPL,
GAMS, Pyomo, etc.
2- provide good initial points for your variables, something that you
expect them to be close to at the end of the computations. if you are
running independent iterations, good initial points would be results from
the previous iteration.
3- investigate the possibility of parallel computation in your modelling
environment, if you are using any.

On Fri, Jun 29, 2012 at 7:57 AM, Juan Carlos Lopez Alfonso <
jc.atlantis at gmail.com> wrote:

> Hi all,
>
> I have changed the stop conditions and I am sure that the
> implementation of my algorithm is correct. I have tested the results
> with a small problems and the results are correct, but when I execute
> my big problem the convergence is very slow and I dont know the
> reason. Is very difficult to test my algorithm, and for example I have
> one of them running for more than 2 days.
>
> someone can give any idea to solve this problem? Is normal this behavior?
> Best regards and thank you in advance
> Juan Carlos
> _______________________________________________
> Ipopt mailing list
> Ipopt at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/ipopt
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/ipopt/attachments/20120629/b72bf763/attachment.html>