UC problem: specify comm share in a FLO

[JM Cayla]

Good morning,

I would like to model Plug-in hybrid electric vehicles technologies that might produce different types of distances travelled thanks to fuel and electricity.

For each type of commodity-out, the share of electricity is supposed to vary: from 100% for short-distance travels to almost 0% in long-distance travels.

Then I would like to build a set of UC that specify the share of electricity consummed for each flow in the process- not the global share for the process- But I don't know how to declare two sets of commodities in a UC file or if it is even possible. Because I need to declare both the Cset relative to COMM-IN and COMM-OUT: the share of the COMM-IN (electricity) in a particular FLO (specific COMM-OUT travels) of the PHEV process.

I hope my question is quite clear and I thank you for your answer

Best Regards

Jean-Michel Cayla

CMA-MinesParisTech

[Antti-L]

In an earlier topic in this Forum, there was already an example given for the modeling of a plug-in hybrid vehicle with two inputs (electricity and gasoline) and two outputs (short and long distance travel). In that example, the average annual share of electricity in the inputs was assumed to be fixed to 55%, and the charging was assumed to occur 75% during the night time and 25% during the daytime.

Using this example, one could reasonably easily replace the fixed 55% average share of electricity by a user constraint defining the share to be dependent on the output flows. Assuming that the share of electricity could be 95% in short-distance travel and 10% in long-distance travel, you could define the combined share constraint as follows:

Code:

FLOW(ELC)*EFFC(ELC) <= 0.95*FLOW(CAR_SD)/EFFC(CAR_SD) + 0.1*FLOW(CAR_LD)/EFFC(CAR_LD)

This constraint would thus limit the share to at most 95% of the inputs (after fuel efficiency) for short-distance travel and to at most 10% of the inputs (after fuel efficiency) for long-distance travel.  The main drawback in this approach is that you need to embed the efficiencies into the UC coefficients. An additional smaller drawback is that the shares are not exactly input energy shares, but shares after the fuel efficiencies.

Such a process can be modeled also without the need for defining any user constraints for the shares, but by using FLO_SHAR parameters constraining the exact input energy shares. However, that would require introducing some dummy commodities into the process topology. I'll leave that approach as an exercise to you.

[JM Cayla]

Thank you for your answer.

It is not a a big issue to embed the efficiencies in fact.
But in the case you are suggesting, we are defining the global elec share for the process and it seems not possible to specify a certain input share relative to a certain output.

Nevertheless this method should provide results that are close to a real PHEV utilization.

[Antti-L]

Sure it is possible. I am using such in many processes.
Directly, it is possible by using FLO_SHAR (simple shares). More complex shares like the one in my example can be defined via UC constraints.  You only need to define the UC_FLO coefficients on both sides, and set the RHS to zero. It should be actually quite easy.

But I guess you are having some problem with it, as you said that it seems not possible. What is the problem you are facing?