-
Notifications
You must be signed in to change notification settings - Fork 965
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.
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
French regions #4853
French regions #4853
Conversation
This reverts commit d088702.
I would suggest we keep the 3rd party maps in the PR (and repository) for now until we are happy with both the zone and aggregated geometries. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good for the most part, just some small notes and one question.
In the current return objects it returns nan, should this not be Null/None if it's missing?
"France": "FR", | ||
**{ | ||
subzone.replace("FR-", ""): subzone | ||
for subzone in ZONES_CONFIG["FR"].get("subZoneNames", []) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This creates some type errors as the Literal "FR" can't be assigned to the ZoneKey type but should not affect the functionality.
} | ||
|
||
|
||
def query(url_type_arg, session=None, target_datetime=None): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could be smart to add types here (and the other functions) for better linting and DX but it's technically not required.
Co-authored-by: Viktor Andersson <[email protected]>
Co-authored-by: Viktor Andersson <[email protected]>
Co-authored-by: Viktor Andersson <[email protected]>
From what I know, nans will be turned to NULL before being inserted in the database (I tested locally without our backend). I can change that if needed though. |
Then it should be fine to keep as is. 🙂 |
Just putting some thoughts here as we are trying to achieve something similar in #4894. In a perfect world our aggregated data should match the data from the national feed but high chances of having some differences. |
@@ -290,3 +290,16 @@ sources: | |||
: link: https://unece.org/sites/default/files/2022-04/LCA_3_FINAL%20March%202022.pdf#page=37, | |||
https://proceedings.windeurope.org/biplatform/rails/active_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--c25280a7345257bd91bfaf6d64ddb75c55eef799/Windeurope-Wind-energy-in-Europe-2021-statistics.pdf?content_type=application%2Fpdf&disposition=inline%3B+filename%3D%22Windeurope-Wind-energy-in-Europe-2021-statistics.pdf%22%3B+filename%2A%3DUTF-8%27%27Windeurope-Wind-energy-in-Europe-2021-statistics.pdf | |||
timezone: Europe/Paris | |||
subZoneNames: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just wondering, how will the flowtracing pipeline react?
It will see data from France from both directly measured and also compute the aggregation data. So I guess we will have two datapoints for the same timestamp. One will probably replace the other as we are indexing by datetime. Since the aggregation is computed last, it will probably be the aggregation that will be in the database.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could you check how it works for Sweden? It should both have working parsers and Aggregates the subzones.
Note: for Sweden the source is exactly the same as we sum the bidding zones in the ENTSO-E parser.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ok thanks! I didn't know we were producing the SE data like this.
So the way I see it, the last step of the flowtracing pipeline computes spatial aggregations, which I think would be inserted after the flowtracing of parsed data. Therefore if a flowtraced data from parsed data exists with the same timestamp as the flowtraced data coming from the spatial aggregation step, it will be overwritten.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
So as it stands I think having a main zone defined like this would most likely overwrite the data obtained through the parsers and have it replaced with data obtained through the aggregation step. Is it the intended behavior ?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Our aggregations should give the same as the national level (especially in the case of Sweden since we are doing it at parser level). But there might be differences we would need to explain.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
So as it stands I think having a main zone defined like this would most likely overwrite the data obtained through the parsers and have it replaced with data obtained through the aggregation step.
We could remove the parser level aggregation for Sweden and see what happens? ENTSOE-E has country level data as well but it was missing solar and gas but it would be nice to do this anyway as it would save us a few API calls.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@unitrium yes it's correct that the country-wide data will be overwritten by the aggregation if the subzones pass the quality criterion (for now we just check if all subzones have been flowtraced). We could change the quality check to make sure the country load is equal to the aggregated load.
✅ Deploy Preview for phenomenal-syrniki-c5ceaa ready!
To edit notification comments on pull requests, go to your Netlify site settings. |
@corradio I just merged master into this branch and will try and get the geometries updated today but is there anything else blocking this from getting merged? |
Mmh I think the only thing missing is proper data validation as our FR data
is under high scrutiny.
…On Wednesday, February 8, 2023, Viktor Andersson ***@***.***> wrote:
@corradio <https://github.com/corradio> I just merged master into this
branch and will try and get the geometries updated today but is there
anything else blocking this from getting merged?
—
Reply to this email directly, view it on GitHub
<#4853 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAMUIKB4U254QGCNPCG5TJDWWNRRLANCNFSM6AAAAAAS4MM2CI>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
--
Olivier Corradi
Founder, CEO
Njalsgade 76
2300 Copenhagen S
electricitymaps.comlinkedin.com/electricitymaps
<https://www.linkedin.com/company/electricitymaps>[image: Electricity Maps]
|
Is this still blocked? If so I think we should close the PR until it becomes unblocked. |
Closing this for now as there is more and more merge conflicts, it would most likely be easier to start from scratch than resolving all the merge conflicts. |
Issue
#2777
Description
French regional data pulling directly from ECO2MIX. Uses the same data as displayed in the RTE app.
⚠️ Lumps all thermal into
unknown
as thermal breakdown is unknown. This affects the country as well, as it is computed as an aggregation of subzones.Outstanding issues
Data validation
TBD.
Preview