Fuel cost

[jabarivelisdeh]

Dear all,

We are using VEDA-TIMES to address the low-carbon production of ammonia. Accordingly, we defined different technologies for ammonia production (e.g. steam methane reforming with carbon capture, water electrolysis, ...), such that the demand for the ammonia production is supplied with minimum cost through different emission reduction scenarios. For each of the production technologies, corresponding CAPEX and OPEX are specified (as $2020/ton produced ammonia). However, we are not sure about the way that we defined the model inputs to calculate the total production cost associated with each technology. 
We know that besides CAPEX and OPEX, the fuel costs also play a key role in each technology's production cost and thus its contribution in addressing ammonia demand. For each technology different fuel energy inputs (mainly natural gas, coal and electricity) are required coming from the supply side. My question is that when we input the cost data (CAPEX, OPEX, ...) should we input the price of each required fuel energy commodity (e.g. by FLO_COST for each input commodity?!) to have a proper definition of the each technology's costs? If the electricity is itself the output of the power sector (grid or any dedicated renewable installations), are we duplicating the electricity costs in the calculations? It is not clear for me whether the price of each input fuel should be considered individually as CAPEX and OPEX when we define each ammonia production technology or it is calculated based on the upstream supply side (from the marginal cost of producing each commodity). What is the best way to define the input fuel/electricity cost? From that, we also would like to obtain the levelized ammonia production cost from each technology.
I would be grateful to have any advises about that...

Best
Banafsheh

[Antti-L]

> My question is that when we input the cost data (CAPEX, OPEX, ...) should we input the price of each required fuel energy commodity (e.g. by FLO_COST for each input commodity?!) to have a proper definition of the each technology's costs?

It somewhat depends on your model and your "system boundary".  If you don't have each of those fuel energy commodities included in the model (and commodity balances for them), you would need to include their price in the process operating costs. However, one should basically never use FLO_COST for defining the prices of input commodities even in that case, because then you would need to define the prices at multiple places and make sure the price data is consistent over those multiple processes, and that is not good practice. FLO_COST should be used only for additional operating costs related to individual input/output commodities.

If you have commodity balances for the input commodities, then you have some processes supplying the commodities, and the equilibrium price is derived by the model. At simplest, the only supplying processes might be exogenous import processes, and then you would define the prices at those import processes. But more generally, energy system models include various technologies for producing the energy commodities that are subsequently consumed by other processes. In that case the equilibrium prices are fully endogenous.

> If the electricity is itself the output of the power sector (grid or any dedicated renewable installations), are we duplicating the electricity costs in the calculations?

If you mean that your model includes the power sector, then your model would produce equilibrium prices for at least some electricity commodity/commodities at each timeslice.  If you link the consuming processes to these commodities, then the consumed electricity will have the endogenous price.  But it seems that you may want to use exogenous price assumptions for electricity?  If that is indeed the case, then just do NOT link the consuming processes to the endogenously modelled electricity commodities, but instead, introduce some other electricity commodity, supplied by an exogenous import process, and then define the electricity price at that import process.  But then the question remains, why have you modelled also the power sector, if you don't wish to use the endogenous prices derived by the model?

> What is the best way to define the input fuel/electricity cost?

As explained above, it somewhat depends on your model and your "system boundary". However, most energy system models include more or less the full energy chains from primary production to end-use, and then the prices of energy commodities are endogenous and need not be specified via exogenous assumptions at all, except for some exogenous trade processes where necessary. The idea is then to define all the investment and operating costs, delivery costs, and possible taxes/subsidies for the various energy conversion and distribution processes, together with the various technical parameters, and let the model derive the prices. The commodity price (plus any additional overhead cost defined on the input flow) will be a cost to the consumer, while it will be a revenue to the producer. No need to define that revenue for the producer nor the corresponding input cost for the consumer.

[jabarivelisdeh]

Thanks a lot for the comprehensive reply. Appreciated.

Best
Banafsheh