Integer Solution Does not Exist

[Robert Owusu-Mireku]

Hello,

What are some of the reason why CPLEX will return that integer
solution does not exist. I tried a problem I have already solved in
CPLEX IDE in AIMMS and i get the warning that an "Integer solution
does not exist". However it is the same code the only difference is
the one in AIMMS is over a longer horizon.
Kind Regards
Bobby

[Leonardo Moraes]

Robert,

try to increase the "time limit" of your problem. Maybe CPLEX can't find a feasible integer solution in this time.

Remember the bigger a problem is, the harder it is to be solved, usually.

Best regards,

Leonardo.

2011/4/11 Robert Owusu-Mireku <rom...@gmail.com>

--
You received this message because you are subscribed to the Google Groups "AIMMS - The Modeling System" group.
To post to this group, send email to ai...@googlegroups.com.
To unsubscribe from this group, send email to aimms+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/aimms?hl=en.

--
Enviado do meu gmail.

[Robert Owusu-Mireku]

Thanks Lenardo, but it did not help matters in anyway. The interesting
thing here is it comes to this conclusion after zero iterations.
So I don't really know what is wrong here. Also apart from the
distritution problem none of my codes works on AIMMs.
Kind Regards
Bobby

On Apr 11, 4:58 pm, Leonardo Moraes <lmorae...@gmail.com> wrote:
> Robert,
>
> try to increase the "time limit" of your problem. Maybe CPLEX can't find a
> feasible integer solution in this time.
>
> Remember the bigger a problem is, the harder it is to be solved, usually.
>
> Best regards,
>
> Leonardo.
>

> 2011/4/11 Robert Owusu-Mireku <rom0...@gmail.com>

>
>
>
>
>
> > Hello,
>
> > What are some of the reason why CPLEX will return that integer
> > solution does not exist. I tried a problem I have already solved in
> > CPLEX IDE in AIMMS and i get the warning that an "Integer solution
> > does not exist". However it is the same code the only difference is
> > the one in AIMMS is over a longer horizon.
> > Kind Regards
> > Bobby
>
> > --
> > You received this message because you are subscribed to the Google Groups
> > "AIMMS - The Modeling System" group.
> > To post to this group, send email to ai...@googlegroups.com.
> > To unsubscribe from this group, send email to
> > aimms+un...@googlegroups.com.
> > For more options, visit this group at
> >http://groups.google.com/group/aimms?hl=en.
>
> --

> Enviado do meu gmail.- Hide quoted text -
>
> - Show quoted text -

[Leonardo Moraes]

Robert,

it's difficult to me to say what is wrong. Maybe you should try to run in AIMMS the very same problem you already solved by using CPLEX IDE.

If it still doesn't work, you can send us a more detailed description of your problem.

Believe me, AIMMS do works! :)

Best regards,

Leonardo.

Enviado via iPhone

[mingl. fang]

Hi Robert,

How about pressing down ctrl+m to see whether Aimms generates a more detailed message? I am guessing it might be linked to out of memory issue.

Best,

Ming
Sent via blackberry

[Robert Owusu-Mireku]

 

Hello,

I still didn't work so I don't know what to do.I have attached the the code and error message to this posting.

Basically it is a scheduling problem over a period of 24 hours. I am trying to schecule electrical power and heat generation between, two combine heat and power plants, two boilers, two steam generators that use the heat produced, a battery and electrical power from the main grid.

 

I know the demand for heat, electricty, and power from other fixed sources like hydro. I know the price of electricity from the grid at each hour, the price of natural gas for the CHP and Boilers. I am just trying to determine which units to turn on or off at each hour and hour much heat and electricity should be generated by each unit.

 

I hope I am not disturbing your sleep noticed you are on BB. We can attend to it tomorrow morning.

 

Kind Regards

 

Robert

[Marcel Hunting]

Hi,

Based on your attachments it seems that CPLEX concludes that the model
is (integer) infeasible. You could try to switch off the CPLEX option
‘Presolve’ and run your model again; maybe that will give you more
information.

What does the progress window show (opens by pressing Ctrl+p)?

Your attachments seem to miss the data which is needed to reproduce
your observations. If you do not want to post the data here, could you
send it to our support e-mail address?

Best regards,

Marcel Hunting
AIMMS Software Developer

On Apr 12, 6:29 am, Robert Owusu-Mireku <rom0...@gmail.com> wrote:
> Hello,
> I still didn't work so I don't know what to do.I have attached the the code
> and error message to this posting.
> Basically it is a scheduling problem over a period of 24 hours. I am trying
> to schecule electrical power and heat generation between, two combine heat
> and power plants, two boilers, two steam generators that use the heat
> produced, a battery and electrical power from the main grid.
>

>  code.docx
> 27KViewDownload
>
>  error.docx
> 28KViewDownload

> >http://groups.google.com/group/aimms?hl=en.- Hide quoted text -

[Marcel Roelofs]

Did you actually consider that there may be a problem with your input
data for the extended horizon, which was not there for the smaller
horizon? You might consider using the VIOLATION PENALTIES attribute of
your math program as to relax your constraints, and get an idea which
constraints prevent CPLEX from reaching integer feasibility.

I also noticed that your model is full with explicit coefficients in
constraint definitions, which is not really considered good practice.
Also, you may want to replace some of the scalar variables in your
model by indexed ones (even if sets of heater and power units only
consist of two units :). The less bulky your symbolic model is, and
the more you strive for separation of model and data, the better one
is able verify the correctness of both your model and its input and
output data.

Cheers,
Marcel

On Apr 12, 6:29 am, Robert Owusu-Mireku <rom0...@gmail.com> wrote:

>  code.docx
> 27KViewDownload
>
>  error.docx
> 28KViewDownload

[Robert Owusu-Mireku]

Hello Marcel,

 

I tried out your suggestion of switching CPLEX off by chosing baron as my solver and activiating the nonlinear presolve function and i got below message from baron "******* Least Cost is infeasible"

I have attached the data i am using and a shot of the progress window. I am new at AIMMs so i suspect it might be a feature that I have to change or something. All the data is for the planning horizon. The rest are zero for now. Also the only variable that needs a fixed data before the horizon is BattSOC (Period-00) which is suppose to be "0.3".

 

 

Kind Regards

Bobby

[Marcel Hunting]

Hi Bobby,

 

Thanks for the data. I am not sure why you used BARON as your solver because the model is linear. Anyway, your model is infeasible which you can check by switching on the general solvers option ‘Infeasibility finder’ and run the model with CPLEX as the default MIP solver (CPLEX option ‘Presolve’ should be switched off to run the infeasibility finder). The constraint listing file (.lis) in the log folder will then show an inconsistency.

 

Best regards,

 

Marcel Hunting
AIMMS Software Developer

 

 

 

CONFIDENTIALITY: This e-mail and any attachments are confidential and may be privileged. If you are not a named recipient, please notify the sender immediately and do not disclose the contents to another person, use it for any purpose or store or copy the information in any medium.