[Couenne] Clp0006I

[Pietro Belotti]

Hi, this looks like a CLP problem, but it's hard to say. I tried running
the .py script but it's asking for a Data.csv. I think it would be much
easier if you could provide the .nl file.

Thanks,
Pietro

On Tue, Oct 16, 2018 at 10:41 AM Jasper Boekel <boekeljasper at gmail.com>
wrote:

> Dear sir, sirs,
>
> I have a problem while trying to solve a MINLP. I use couenne together
> with Pyomo, however when applying couenne directly to the .nl file I
> encounter the exact same problem.
>
> I added a reproducible example for generating and solving the problem as
> well as the couenne options file. So I put a time_limit of 30 seconds in
> the options file. Now what happens is that couenne starts optimizing until
> the 30 seconds boundary is reached. However, subsequently it just keeps
> infinitely printing rows like the ones below.
>
> Clp0006I 26150  Obj 9574.1254 Primal inf 604.9011 (27) Dual inf 261306.19
> (23)
> Clp0006I 26152  Obj 9574.1254 Primal inf 257.8018 (27) Dual inf 406.59172
> (22)
> Clp0006I 26155  Obj 9574.1254 Primal inf 620.87745 (29) Dual inf
> 5.6336085e+13 (24)
> Clp0006I 26156  Obj 9574.1254 Primal inf 609.66265 (29) Dual inf 2998.5172
> (23)
> Clp0006I 26158  Obj 9574.1254 Primal inf 255.92422 (27) Dual inf 353.53079
> (22)
> Clp0006I 26161  Obj 9574.1254 Primal inf 627.48252 (28) Dual inf
> 1.4600708e+14 (25)
>
> I have found that when I tweak some options in the couenne.opt file such
> as feasibility_bt, optimality_bt and aggressive_fbbt it solves the problem.
> However I have to solve this problem for every day for a 3 year period and
> every problem needs a different combinations of options.
>
> Any help would be greatly appreciated.
>
> Kind regards,
>
>
> Jasper Boekel
>
>
>
> _______________________________________________
> Couenne mailing list
> Couenne at list.coin-or.org
> https://list.coin-or.org/mailman/listinfo/couenne
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/couenne/attachments/20181110/393adc36/attachment.html>