Solution Non-optimal

Hi all,

I am writing a code in gams for Battery scheduling, in which have modeled the battery charging, discharging and idle condition of the battery as integer variable. I have chosen MINLP solver for solving the optimization problem. The battery maximum output power possible and charging and discharging limits are also specified. The formula says: battery_soc_t+1 = battery_soc_t+(battery_power/battery_capacity)ubattdel_t. where ubatt is the integer variable, describing the state of the battery.

The problem I am solving it using SBB/Conopt and Baron for comparison. But unfortunately the both the solver giving different result. Under close observation I found that solution using SBB the integer variable remain at zero and its marginal vale remains at 0. But using Baron solves the problem. Why such an huge error? Also in SBB I have manually set one variable to some integer value other than 0 and I am getting objective function closer to the optimal. Why is this problem coming?

regards,
Subir


To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at http://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

I have Attached the .gms file. Can anyone please look into that?

Regards,
Subir


Untitled_12.gms (2.1 KB)