electricitymaps
Create parser for Albania (source is available)
Carbon intensity of the electricity produced in Albania is currently 461 gCO2eq/kWh. I suspect it could be wrong, since 100% of Albania's electricity come from renewable sources (mainly hydro + a small share of solar).
The only thermal plant of the country (TEC Vlora) has never been functioning.
The installed capacity (listing only hydro though) is included in the daily data published by OST (https://www.ost.al/te-dhena).
I think Albania's value never got specified (I don't see it in zoneOverrides in https://github.com/electricitymap/electricitymap-contrib/blob/master/config/co2eq_parameters_all.json ) so the 461 g might be the default value for worldwide averages (it might be computed from the values in fallbackZoneMixes -> defaults -> powerOriginRatios in that file?).
If you or someone could link to recently monthly or yearly production data for Albania, we could add an override for this value.
In fact looks like it was previously added in https://github.com/electricitymap/electricitymap-contrib/pull/1264 but must have gotten lost somewhere along the way.
https://www.iea.org/countries/albania says 2020 production was 5281 GWh hydro and 32 GWh solar which is pretty clear cut.
However, the issue of exchanges was raised in #1264. IEA 2019 data has yearly total of 11435 TJ = 3176 GWh imports and 2774 TJ = 770 GWh exports, with 5184 GWh own hydro generation. So imports can make up a significant part of electricity supply in Albania, at least in some years (for example in 2018 imports were 850 GWh and hydro production was 8552 GWh - so it is quite variable). That's a bit of a problem if Electricity Map doesn't know Albania's current generation, because it's hard to estimate carbon intensity of what's being exported: when Albania is exporting some power, is it exporting mostly Albanian hydro, or half hydro and half Montenegrin coal?
Thanks @jarek. Yearly production data can be found on the website of ERE (energy regulator of Albania). For 2021: https://ere.gov.al/images/2022/04/Raporti%20Vjetor%202021%20perfundimtare.pdf.
- Total generation capacity for Albania as of 31/12/2021 is 2605 MW (p. 29).
- P. 39, paragraph 1.2.3 ("Status of Thermal Plant of Vlorë") confirms that the plant has not been functioning since 2012. For the record, capacity is 98 MW (cf. table p. 33).
- From p. 42 to p. 45, there are several lists of all the power plants of the country, according to the type of ownership. It includes the Thermal Plant of Vlorë (98 MW) and the solar capacities (23 MW). All the rest is hydro (2605 - 98 - 23 = 2484 MW).
I will create a PR with this data. Currently, in zones.json, biomass is also listed (1 MW) but I don't see what facility it can refers to in Albania.
The exchanges with neighboring countries are already on the map. So what seems to be missing is only live generation to get the share of hydro (and marginally, solar) in the power mix of Albania. Maybe it could be possible to use the forecasts included in the daily data published by OST as an estimate?
Sorry for taking so long to respond to this, I'm currently going through some of our older issues and this seems to be a simple change in the carbon intensity value. I'll take a look at it tomorrow and double check all the numbers and create a PR if it all checks out.
No worries, I've been quite busy myself. I've dug a bit deeper in the data published by OST (https://ost.al/te-dhena). This daily data is unfortunately published with some delay (last available set was published on March 16 2023) but it includes a table of the hourly production/exchanges/consumption of Day-2 (see line 81 of the XLS file).
I've compared the figures of the "Exchange" column of the table with the ones from ENTSO-E for the given day, and they match, which seems to give credit to the data published by OST.
Since an overwhelming share of the Albanian electricity is produced by hydro power plants, maybe it could help to include historical data for Albania in the app, and more precise estimates for the carbon intensity?
As stated before the capacities in Albania are overwhelmingly hydro, since the Thermal Power Plant of Vlorë is still not working, and other temporary thermal assets leased by the government (https://www.euractiv.com/section/politics/short_news/albanians-up-in-arms-over-oil-burning-power-plants-off-coast-of-tourist-hotspot) have not been working either so far (and it doesn't look like they will do in the near future from what I read locally). Solar generation is slowly becoming a thing in Albania but it's still very marginal, and unfortunately solar generation is not included in OST figures.
The delay isn't much of a issue anymore since we started showing historical data but the fact that most of the historical data it's published as a PDF will make automatic parsing extremely difficult if not impossible. The newer ones from February 2022 is published as excel documents though and should be able to be parsed.
It's unfortunate that solar is not included but it's not a dealbreaker as this would be a new data source. I'll see if I can create a parser for it this week (with a 1 month delay). After that we can look into what the unknown production is and try and find a good co2eq value for it.
EDIT: It might also be smart to reach out to ENTSO-E and OST directly and ask if this data can be published in the ENTSO-E transparency platform (the delay might be an issue).
Okay so I have looked into this some more.
- It should be fine that solar is excluded as it represent such a small share and is as far as it only represented 0.6% of the generation 2020. Which means it it's even less with imports.
- The source seems to have all the hydro generation.
- The source seems to have all exchange information needed.
- The source seems to have all the consumption data we need.
- The URL's seems to follow a pattern that can be automated.
- Excel files should facilitate parsing.
The delay is a slight issue as it won't show up on the live map, at least not right away but if we get enough historical data we might be able to estimate the live production. Though a month+ from the baseline might be inaccurate.
I am going to take a crack at making a parser for this one so we can finally get some data for Albania!
I found some issues with the data, the production for the individual units don't match the generation. So it looks like not all units are included, which makes sense cause the units that are included are the biggest ones.
But I think we should be able to create a estimation method for it instead based on the generation and consumption instead of parsing the production directly.
I'll still create necessary parser functions but use the production per unit instead.
Thanks a lot @VIKTORVAV99! I will also reach out to OST as suggested regarding the publication of its historical data on the ENTSOE platform.
@VIKTORVAV99, can you give an update here on other solution discussed in #5416? :)
@VIKTORVAV99, can you give an update here on other solution discussed in #5416? :)
That is a no go, it doesn't report all the production data but just a few of the biggest plants. We need to build an estimation model for it if we want data.