Model not solving

[Antti-L]

I am not surprised that the model solves almost equally bad as before.  I checked with the latest files you uploaded (model (2).zip), and I can see that the dummy import processes are again enabled in the model!

I cannot see any START=2200 defined for them, although you specifically claimed to have disabled the dummy imports in that way.  I can see from the listing file that they are, in fact, enabled once again, and have high costs. Disabling them would be the first thing to do if you want a well-solvable model (but not the only thing, as mentioned before).

[EDIT:] I tested quickly the latest version you uploaded (model (2).zip) with my assorted small fixes, and the model solved in less than six minutes.  If interested to see, the fixes are attached (I am not at all sure that they are all relevant, though). For example, you seem to have ignored my earlier remark that OILRES, OILCOM, GASRES and GASCOM apparently had a bad timeslice level (DAYNITE), because the fuel techs were ANNUAL.
  ALFIXES.zip (Size: 966 bytes / Downloads: 3)

[MohammedAbiAfthab]

I am not surprised that the model solves almost equally bad as before.  I checked with the latest files you uploaded (model (2).zip), and I can see that the dummy import processes are again enabled in the model!

I cannot see any START=2200 defined for them, although you specifically claimed to have disabled the dummy imports in that way.  I can see from the listing file that they are, in fact, enabled once again, and have high costs. Disabling them would be the first thing to do if you want a well-solvable model (but not the only thing, as mentioned before).

[EDIT:] I tested quickly the latest version you uploaded (model (2).zip) with my assorted small fixes, and the model solved in less than six minutes.  If interested to see, the fixes are attached (I am not at all sure that they are all relevant, though). For example, you seem to have ignored my earlier remark that OILRES, OILCOM, GASRES and GASCOM apparently had a bad timeslice level (DAYNITE), because the fuel techs were ANNUAL.

Dear Antti,

Thanks a lot for your support. Day and night, you have been helping me.

I incorporated your suggestions (attached are the changes I made in the model and the .DD and .Run files). The model solving time has improved drastically. Although it's not finishing in 6 minutes like in your case, it got solved in 24 minutes for me. May be this has to do with RAM of the PC?

Could you have a final look  at my files, if you have some time? If the solving time is even lesser, I could do my sensitivity analysis a lot easier.

Regards
Abi Afthab

[MohammedAbiAfthab]

Dear Antti,

I simplified my model a bit. Along with your suggestions, I have reduced the number of storages in my model. Now it's solving in 11 minutes. I guess, this simplification is better than spending a lot of time to represent all the complexities of the process.

Thanks a lot for your support and your time. I will be back with many more questions on other aspects and hopefully not again on this solving time.

Regards
Abi Afthab

[MohammedAbiAfthab]

Antti,

The model again started behaving like earlier one. If at all, I make some changes it runs for 2 hours, 3 hours and yesterday night it's been running for more than 7 hours. Still it's running and not solved. I will not make anymore changes. Could you have a look at the files? I have made all the changes earlier suggested by you. I can't afford this much time for solving because I have to run with different scenarios and I have to do sensitivity analysis. I have only very few days with me.

In the latest run window it's showing some singularity issue with EPLT_NucGenIIIPlus_new in the subRes files "ELCconventionalTechs". But this was there like this from the beginning. I have not done any changes with it. 

Regards
Abi Afthab

[MohammedAbiAfthab]

Dear Antti,

I am puzzled about this model. A model which was solving in 33 minutes just stopped solving (attached is the snapshot, it is been running for 10442 seconds and never ended) when I just switched between two scenario files which had similar values. 

I have simplified the model maximum which I could and the model was being solved in 33 minutes. But now if I change a small value here and there, I don't know why this is happening. I am literally down now. All the hard work I have taken for modelling is being in vain just because I can't take out the results properly. You have been helping me a lot and I am very thankful to you. I am now in the last lap of my thesis. Next week I have my presentation. I request you to extend your help at least for one more week in this situation. This is something out of the hands of the modeler.

Regards
Abi Afthab