[Coin-ipopt] Termination code 11.....

[Rodrigo López Negrete de la Fuente]

Hello,

I posted before because I had a strange termination error (Termination code
11). At first glance I thought it was the linear solver since I had just
changed it, but now I've acutally isolated this to something else. The
problem resides in the option: constr_mult_init_max, I changed the value
from its default of 1000 to 10000. Which I changed at the same time I
changed the linear solver (my bad, too many changes at once).

I've tried this with the PARDISO solver, and also with the MA27 solver. In
both cases when using this altered option I get the termination code 11
error, but if I use the default constr_mult_init_max value the error does
not happen (again with both solvers).

I've also reproduced this with two very different models one for a reactive
distillation column the other one for a semibatch polymerization reactor. I
should point out that when I use the default value for the option I can get
optimal solutions for both models.

The reason for posting this in the mailing list is that I'm not sure if i
should open a new ticket, or if I should leave it as is (ticket #14).
Anyway, I hopefully this will help in the improvement of the code.

Please email me if more information is needed.

Thanks,
   Rodrigo Lopez-Negrete

-- 
Rodrigo López Negrete
http://muon.blogdns.org/~rush/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.coin-or.org/pipermail/ipopt/attachments/20060623/26e560a3/attachment.html