"Iteration limit reached. Current point is infeasible" - are both messages true?

Skip to first unread message

Lothar Löwer

Nov 11, 2021, 9:11:44 AM11/11/21
to Artelys Knitro forum
Using AMPL with Knitro 12.2.2 trying to solve an optimization problem (optimal power flow / electrical grid calculation), I get the message "Iteration limit reached. Current point is infeasible" (solve_result = limit; solve_result_num = 410). Nevertheless, using the calculated data in another "basic" calculation (normal power flow), all defined limits in the prior calculated optimization problme are kept.

Of course, I am not doubting that the calculated data are not the optimal solution due to too less iterations, but the calculated result seems to be a feasible solution.

Thus, would it be ok to use the calculated data?

Best regards,

Richard Waltz

Nov 11, 2021, 9:20:10 AM11/11/21
to kni...@googlegroups.com

Dear Lothar,

Certainly if the solution satisfies your requirements as being "feasible enough" you are free to use it.  However, it might be good to understand why Knitro is claiming the point is infeasible.  What is the final "feasibility error"?  If it is very small, then it may have just missed the tolerance for being feasible and may be close enough to feasible for you.  However, if it is large, then there is some significant constraint violation somewhere.  You can try to find out which constraints are infeasible by using the AMPL "display" command to display the constraint body value at the solution and comparing to the constraint bounds.

If you run with outlev=4 and share the output, it may be more clear what is going on. Also, if you want to share your model, we would be happy to look into it further.

Best regards,

Richard Waltz

Richard WALTZ
Senior Scientist
Artelys USA

From: 'Lothar Löwer' via Artelys Knitro forum <kni...@googlegroups.com>
Sent: Thursday, November 11, 2021 4:53:37 AM
To: Artelys Knitro forum
Subject: [Knitro] "Iteration limit reached. Current point is infeasible" - are both messages true?
You received this message because you are subscribed to the Artelys "Knitro Nonlinear Optimization Solver" google group.
To post to this group, send email to kni...@googlegroups.com
To unsubscribe from this group, send email to knitro-un...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/knitro?hl=en
Thank You,
You received this message because you are subscribed to the Google Groups "Artelys Knitro forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to knitro+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/knitro/bbd00072-a07f-4fdf-afac-4d841cf69f60n%40googlegroups.com.
Reply all
Reply to author
0 new messages