ERROR: Infeasible constraint deduced from presolve.

9 views
Skip to first unread message

Adedeji Isaac Adeloye

unread,
Jun 10, 2024, 4:52:35 PMJun 10
to Artelys Knitro forum
Hello,

I am trying to solve a Non-linear optimization problem. I have been using Knitro to solve non-linear problems, but this is my first time encountering this issue.

My lb range from 0 to 10,000 or -10,000 to 0 or -10,000 to 10,000. For an initial point, i use:

X=(lb+(ub-lb)).*rand(Nvar,1))'.


Here is how I call knitro:

options = knitro_options('maxit', 1000, 'outlev', 4, 'xtol', 1e-15, ...

                         'feastol', 1e-8, 'opttol', 1e-8, ...

                         'bar_maxcrossit', 5, 'honorbnds', 2, 'numthreads', 8);


[Xsqp, FUN, FLAG, ~,~,~,~] = knitro_nlp(@(X)SQP(X,Dat),X,[],[],Aeq,beq, ...

    lb,ub,@(X)SQPnolcon(X,Dat),[],options,[]);


Here is the error I get:

WARNING: 1 constraint is constant or undefined and will be ignored.

=======================================
             Trial License
       (NOT FOR COMMERCIAL USE)
         Artelys Knitro 14.0.0
=======================================

ERROR: Infeasible constraint deduced from presolve.
       Deduced constraint value: Aeq(3026,:) =    -1.34730000000000e-06
       violates the constraint lower bound =     0.00000000000000e+00
Knitro presolver has deduced that constraint Aeq(3026,:) cannot be satisfied.

EXIT: Problem determined to be infeasible with respect to constraint bounds.

Final Statistics
----------------
Final objective value               =   7.29751108473195e+03
Final feasibility error (abs / rel) =   8.46e+07 / 1.00e+00
Final optimality error  (abs / rel) =   1.80e+308 / 1.80e+308
# of iterations                     =          0
# of CG iterations                  =          0
# of function evaluations           =      10044
# of gradient evaluations           =          0
Total program time (secs)           =       0.39499 (     1.827 CPU time)
Time spent in evaluations (secs)    =       0.37890

Any help will be appreciated.






Richard Waltz

unread,
Jun 10, 2024, 5:09:35 PMJun 10
to kni...@googlegroups.com
Hi,

The Knitro presolver is claiming that the model as formulated is infeasible.  However, the infeasibility it deduces is quite small (1.3e-6) so it is possible that the infeasibility deduction is due to roundoff errors.  You can try setting "presolve_tol" to a larger value (e.g. 1e-5 or 1e-4) so that it only claims infeasibility for larger values:


You can also just try turning the Knitro presolver off by setting "presolve" to 0 to see what happens:


Regards,
-Richard


From: kni...@googlegroups.com <kni...@googlegroups.com> on behalf of Adedeji Isaac Adeloye <dej...@gmail.com>
Sent: Monday, June 10, 2024 12:29 PM
To: Artelys Knitro forum <kni...@googlegroups.com>
Subject: [Knitro] ERROR: Infeasible constraint deduced from presolve.
 
--
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,
Artelys
http://www.artelys.com/en/optimization-tools/knitro
---
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/b3c03213-084c-420a-b1af-b32405949004n%40googlegroups.com.

Adedeji Isaac Adeloye

unread,
Jun 10, 2024, 6:17:16 PMJun 10
to Artelys Knitro forum
This worked by setting preserve_tol to 1e-5. Thank you for your suggestion.

Regards
Reply all
Reply to author
Forward
0 new messages