Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

correct initialization of learning parameters for electrolysis following Adrian's paper and improve initialization procedure by2025 #1051

Merged
merged 7 commits into from
Nov 28, 2022
Merged

correct initialization of learning parameters for electrolysis following Adrian's paper and improve initialization procedure by2025 #1051

merged 7 commits into from
Nov 28, 2022

Conversation

fschreyer
Copy link
Contributor

@fschreyer fschreyer commented Nov 4, 2022 •
edited
Loading

Purpose of this PR

This corrects learning parameters for electrolysis once more (follow-up to #997) and improves the initialization procedure to hit the desired capital cost values in 2025.

See results and description of open points in this Issue.

Type of change

(Make sure to delete from the Type-of-change list the items not relevant to your PR)

  • Minor change (default scenarios show only small differences)

Checklist:

  • My code follows the coding etiquette
  • I have performed a self-review of my own code
  • Changes are commented, particularly in hard-to-understand areas
  • I have updated the in-code documentation
  • The model compiles and runs successfully (Rscript start.R -q)
  • check learning initialization in test runs

Further information (optional):

  • Test runs are here:

/p/tmp/schreyer/Modeling/remind/Current/output/Base_Learn_2022-11-18_11.51.57

/p/tmp/schreyer/Modeling/remind/Current/output/NDC_Learn_2022-11-18_16.39.53

/p/tmp/schreyer/Modeling/remind/Current/output/PkBudg500_Learn_2022-11-19_01.27.07

  • Comparison of results (what changes by this PR?):

See plots in Issue.

@fschreyer fschreyer requested a review from aodenweller November 4, 2022 13:08
aodenweller
aodenweller reviewed Nov 4, 2022
View reviewed changes
Copy link
Contributor

@aodenweller aodenweller left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I attached some comments.

core/equations.gms Outdated Show resolved Hide resolved
core/input/generisdata_tech.prn Outdated Show resolved Hide resolved
core/input/generisdata_tech.prn Outdated Show resolved Hide resolved
@fschreyer fschreyer marked this pull request as ready for review November 23, 2022 11:14
aodenweller
aodenweller approved these changes Nov 27, 2022
View reviewed changes
Copy link
Contributor

@aodenweller aodenweller left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good.

@fschreyer fschreyer merged commit eb4a570 into remindmodel:develop Nov 28, 2022
@orichters
Copy link
Contributor

orichters commented Mar 31, 2023 •
edited
Loading

Dear @fschreyer, @aodenweller: Do you have a compareScenario document with runs before and after the changes you made? This PR is one of the candidates creating a drop in consumption and increased investment in 2010, 2015 that we are struggling to explain.

@aodenweller
Copy link
Contributor

I'm afraid I don't as Felix did all of these runs. Does this PR increase investment into electrolysis that much in 2010 and 2015?

@orichters
Copy link
Contributor

Somehow, we have dramatically reduced capital stocks. I'm not entirely sure your PR is the root cause, but at least it looks the most suspicious.

image

@aodenweller
Copy link
Contributor

I'm very sorry, but I don't think I'll be of much help here. Felix did both the code changes and the runs. I only checked that the parameters in generisdata_tech.prn looked plausible. Have you tried reverting this PR and re-running?

@robertpietzcker
Copy link
Contributor

Somehow, we have dramatically reduced capital stocks.

But isn't the capital stock (at least in 2005) an input to the calibration?
it changed in several regions (EUR, JPN, REF) in 2005, when there clearly was no H2 electrolysis capacity.

I think it might be more helpful to check the Macro inputs to the calibration? (I have to admit I have no ideas what determines the non-ESM macro capital stock in 2005 ...)

@orichters orichters mentioned this pull request Mar 31, 2023
Closed
@orichters
Copy link
Contributor

The bug was actually totally unrelated to this PR, but found in the data REMIND was calibrated to, so my suspicion was wrong. Hopefully be fixed with this PR in mrdrivers.

@fschreyer fschreyer deleted the TechLearn branch October 31, 2023 09:10
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@aodenweller aodenweller aodenweller approved these changes

@robertpietzcker robertpietzcker Awaiting requested review from robertpietzcker

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@fschreyer @orichters @aodenweller @robertpietzcker