Optimisation error for Discrete Investment

[Lesh]

I'm simulating CHP units using discrete investments, in order to account for economies of scale. The CHP technologies are shown below:






The CHPs are all similar except INVCOST and NCAP_DISC. By running the simulation without discrete investments, the results for new CHP capacity is 0.05GW, which is the correct number. However, when I run in discrete investments, the model gives CHP08 (15GW), which is a lot more in total investments, compared to CHP03 or CHP04, which would satisfy the capacity requirements at lower costs.



Can someone suggest why the model is not providing the least total cost to satisfy the capacity requirements ?

Thanks
Lesh

[Antti-L]

Well, to me it looks like you have such high subsidies on the LOWDAICHPELEC output flow produced by LOWDAIADCHP08, that it will be profitable to invest as much as possible in the technology. The annual subsidies appear to be even many times higher than the combined annual investment and fixed O&M cost payments. Or maybe I am missing something? Anyway, to get a true least cost solution, you should remove all subsidies from the model. Subsidies will divert the model away from the least cost solution.

[Lesh]

Hi Antti,

Forgot to upload the subsidies sheet, but the subsidies apply equally to all CHP technology.

Lesh

[Antti-L]

Ok, I noticed that you uploaded the model for me.

Looking at it, I found out that you have just forgotten to set the optimality tolerance OPTCR (see the control panel). In your model, you need to set a zero or a very tiny value for OPTCR to get optimal MIP results.  I tested with zero, and got quite reasonable results. I hope that helps.

[Lesh]

Yes, changing the OPTCR to 0 seems to have solved the issue.

Thanks Antti,
Lesh