search

coin-or / Bonmin

Basic Open-source Nonlinear Mixed INteger programming
https://coin-or.github.io/Bonmin
Eclipse Public License 1.0
130 stars 22 forks source link

Same problem working with IPOPT but failing with BONMIN #1

Open sourav003 opened 7 years ago

sourav003 commented 7 years ago

Hi I am trying to solve a mixed-integer non-linear programming problem by using BONMIN package through AMPL platform. The objective function is linear and have several other linear constraints. Only one constraint is non-linear but it is twice or more differentiable. When I tried to solve it using IPOPT solver by issuing the following commands, it works but doesn't produce an integer optimal solution, as expected.

model hyb_cld.mod;
data hyb_cld.dat;
option solver ipopt;
option ipopt_options "linear_solver Pardiso";
solve;

However, when I am using the BONMIN solver next, the solver is halting with the following output.

model hyb_cld.mod;
data hyb_cld.dat;
option solver bonmin;
solve;

Bonmin 1.8.4 using Cbc 2.9.7 and Ipopt 3.12.4

This program contains Ipopt, a library for large-scale nonlinear optimization. Ipopt is released as open source code under the Eclipse Public License (EPL). For more information visit http://projects.coin-or.org/Ipopt

NLP0012I Num Status Obj It time Location NLP0014I 1 FAILED 583.30161 1 22.484 NLP0014I 1 FAILED 583.30161 1 22.216 resolve robustness NLP0014I 3 FAILED 583.30161 1 21.167 totot Cbc0006I The LP relaxation is infeasible or too expensive

"Finished"

bonmin: Infeasible problem

Can anyone please help me in finding a way out from this situation? Do I need to change any particular option for the solver? I have attached my model and data files as well. AMPL.zip

Thanks in advance. Regards, Sourav

svigerske commented 1 year ago

If this is still relevant, then try increasing the log level of the NLP solver to see why Ipopt failed. (https://coin-or.github.io/Bonmin/options_list.html) Maybe you need to set linear_solver Pardiso as Bonmin option, too?

I cannot reproduce with the attached model and data files, as I don't have an AMPL license (and the instance goes beyond their demo limits).