pv-camecon
commented
2 weeks ago This is related to the limited number of 3D variables we plug into FORTRAN. To bypass this limitation, I've added the final output of the life cycle inventory calculation in the cost matrix and it's a mix of things (see below). The values are determines in the raw material routine and used in the LCOS routine.
The additional columns represent:
- Final output of crude steel per unit (always 1)
- Demand for intermediate (iron product) inputs (produced iron, imported iron - in case of scrap shortage, and scrap) per unit of crude steel production
- Demand for precursor materials (pellets, cokes, sintered iron, oxygen)
- Final energy demand
So, in the end, this provides the information on the amount of precursor materials needed based on the scrap vs produced iron inputs to estimate the final energy demand. In essence, it's just a slightly more complicated scaling exercise.
In Python we don't have this limitation, so the values can be stored in a separate variable(s).
In BSTC, some of the data seems to be missing. From column 33 (starting from "Hard Coal"), all the entries are zeros. When I run the code in FORTRAN, there does seem to be 1s in some of these values, so that the calculation for SJEF makes sense. Do you know where that information comes from?