Switches for convergence of regional prices for non-learning technologies

[lecfab]

Purpose of this PR

We look at regional prices of non-learning technologies (all energy technologies except solar, wind, related storage and electrolysis). Until now, the price in each region would linearly converge, from 2020 to 2070, to a unique value fm_dataglob("inco0",te).

Two updates allow for more flexibility in that regard:

1. a different convergence year can be set with the new switch c_teNoLearngConvEndYr (default still 2070).
2. convergence can be removed by setting the existing switch (or flag?) cm_techcosts to to the new value REG2040: the prices are still regionally differentiated, but each region follows prices read from /core/input/p_inco0.cs4r until 2040 then stagnate at the 2040 value.

To-dos:

1. under switch on, make an exception for nuclear to fix the CHA nuclear renaissance issue
2. implement newer regional cost data WEO2021

With their default values, the new switches do not have any impact.

Type of change

• [x] New feature
• [x] Minor change (default scenarios show only small differences)

Checklist

• [x] My code follows the coding etiquette
• [x] I performed a self-review of my own code
• [x] I explained my changes within the PR, particularly in hard-to-understand areas
• [x] I checked that the in-code documentation is up-to-date
• [x] All automated model tests pass (FAIL 0 in the output of make test)
• [ ] The changelog CHANGELOG.md has been updated correctly

Further information

• Test runs are here: /p/tmp/fabricel/switchRemind (the ones before March 20th are useless)
• Comparison of scenarios NPi (with default convergence in 2070, convergence in 2030, and no convergence): /p/tmp/fabricel/switchRemindcompScen-techConvNoconv-2024-03-27_19.25.18-H12.pdf
• Cascade for no convergence (REG2040): /p/tmp/fabricel/switchRemind/compScen-cascadeNoTechConv-2024-04-03_11.36.34-H12-short.pdf

[cchrisgong]

Thanks, @lecfab! As we already discussed, under non-convergence scenario, the scenario mainly differs from default in that there is a strong nuclear growth in China that push growth of wind off track, we perhaps need to revisit current nuclear cost development in China, because the old cost trajectories are from IEA WEO 2016, and might need an update

[lecfab]

The most visible effect of the 2 switches is for nuclear energy capital costs (see image below).

[lecfab]

The non-convergence switch (REG2040, scenario techNoConv) follows projections of WEO2016 pasted below (/p/projects/rd3mod/inputdata/sources/IEA_WEO/WEO_2016_PG_Assumptions_NPSand450_Scenario.xlsx). However, I wonder if these numbers make sense:

• they decrease in Europe from 2015 while I understood that prices were only going up
• in other regions, they are constant or roughly
• in China, the cost is cheap which leads to much higher nuclear shares (and lower wind) in the techNoConv scenario

Nuclear capital costs ($/kW)	2015	2020	2030	2040
Europe	6600	6000	5100	4500
United States	5000	5000	4800	4500
China	2000	2800	2750	2500
India	2800	2800	2800	2800
...

Update: the WEO2023 report (page 301) has the same values, except higher for current EU, and the projection for 2040 is pushed to 2050.

The capital costs for nuclear power represent the “nth-of-a-kind” costs for new reactor designs, with substantial cost reductions from the first-of-a-kind projects.

Do you think we should update the values (maybe WEO2023?) or consider a different source? maybe @robertpietzcker?

@cchrisgong thinks we could keep convergence of capital costs (in 2070 or later) only for nuclear.

[cchrisgong]

@robertpietzcker for now the new switch is off - so the trunk stays the same, we may put the nuclear cost convergence as default in a subsequent PR under the newly implemented switch (when it is activated) (we can discuss further in this PR when you are back from vacation ;))

[lecfab]

The cascade for non converging cost is done, and its cs2 can be found here: /p/tmp/fabricel/switchRemind/compScen-cascadeNoTechConv-2024-04-03_11.36.34-H12-short.pdf. @cchrisgong I think we can now merge and address the data update / nuclear convergence choice later?

[cchrisgong]

@lecfab thanks a lot! I made minor comments, let me know what you think :)