Electricitymap-contrib: Data for Mexico MX

[alixunderplatz]

Promising news!

There is total generation (no breakdown yet), load data and power flow data for Mexico!
Seems like this site has been online since 15th March, so that's an accidental find :)

Power flows and total generation and consumption per region (hit button on the right to switch):
http://www.cenace.gob.mx/Paginas/Publicas/Info/DemandaRegional.aspx

And generation and consumption as a time series (seems like it is updated every 15 min, see timestamp on the left):
http://www.cenace.gob.mx/GraficaDemanda.aspx

All live exchanges with the U.S. in the North are there, also with Belize (new) and Guatemala in the south.

Seems like the two regions on the left part of another system/synchronous area: "National Interconnected System, does not include the regions of Baja California and Baja California Sur."
So California is only linked with these and not with the Mexican "mainland", but Texas is.

---

Seems like this is the open data platform of the Mexican government:
https://datos.gob.mx/libreusomx

Maybe there will be or is more info on breakdown, so please take a look!

Terms of use: https://datos.gob.mx/libreusomx

[systemcatch]

Great find as usual.

Going on http://www.cenace.gob.mx/Paginas/Publicas/Info/DemandaRegional.aspx and clicking 'intercambio entre regiones' (exchange between regions) seems to show live flows between Mexico and the USA. We could use this for #959!

[alixunderplatz]

@systemcatch right, works for CAISO imports!
The MX region "Baja California" in the upper left corner is not connected with the rest of the Mexikan mainland, so the influence on CAISO will be really low (24 MW at the moment, I've found up to ~100 MW on EIA's site in March)

[brunolajoie]

What about adding it to the map as 100% "unknown", taking 2017 average mix to compute CO2 intensity of "unknown", and iterate from there when we'll have the proper breakdown per fuel?

[alixunderplatz]

@brunolajoie do you mean entire Mexico or just the upper left corner?

[brunolajoie]

I meant all regions 😃 (and the exchanges btw them).
I know it hasn't been the usual practice, but Mexico is a major country, and I'm wondering if adding it to the map with an unknown power mix might not help convince them to disclose the breakdown as a next step! thinking out loud here!

[alixunderplatz]

According to this source, there were around 4 GW of wind and 0.46 GW of PV by middle of 2017 in MX.
Installed PV could have doubled in the meantime and wind can contribute up to 1/7 of the total demand, which will "break" the intensity factor. Their hydro generation will most likely follow the load scheme as well.

Of course the "unknown" assumption will make some areas with "total generation only"-data look horribly bad, which could raise the desire to publish the actual generation mix. But it will also negatively influence the intensity of zones in importing from MX.

---

In general, I support including generation data with an unknown intensity, because it is a way to make people become aware of when consumption in their area is the highest (which often is the equivalent to "the dirtiest" or "most stress for the system"="higher risk of outage"). But this is rather an option to further develop the functions of electricitymap, aside from the main goal of showing carbon intensity.
Coloring them in stripes on the map to highlight the "unknown" live generation mix could be a way to do it. _(Just leaving some other candiadtes here: Japan, South Korea, Iceland (demand = generation in these 3), The Netherlands)_

[systemcatch]

I'm looking at adding the exchange between California and Baja California Norte, but it seems like the time is stuck at 12:00:00 while the exchange value is changing. Does anyone else see this?

[alixunderplatz]

FYI: I accidentally found a website with some more general and regional info on the profile of mexican electricity generation and the location of power plants - images are from around 2016:
http://www.conacytprensa.mx/index.php/tecnologia/energia/6597-cenace-energia-potencia-electrica

[systemcatch]

MX exchanges are failing due to SSL verification of of the sites certificate.

[alixunderplatz]

Total generation for all Mexican zones could still be added to the map as well (like for Japan), although there is no value added regarding carbon intensity. But it's always interesting just to have it and visualize it.

https://www.cenace.gob.mx/Paginas/Publicas/Info/DemandaRegional.aspx

[cuauhtemoc11]

There hasn't been any progress for this country in almost a year. Maybe joining the regions would make it easier for the time being?

[systemcatch]

@cuauhtemoc11 I don't see how that makes it easier, we lack any production data for Mexico, however we do have some exchange data which is useful.

[scriptator]

I found a page which looks to me like it has production data for the whole country: https://www.cenace.gob.mx/SIM/VISTA/REPORTES/EnergiaGenLiqAgregada.aspx

What do you think - is this valuable?

[jarek]

That looks quite good, however do I see correctly that it's currently (mid-October) only data for September available?

[scriptator]

Yes, sadly you are right. I did not notice that before because I had only looked at the publication date. Would it be still interesting to have a parser for historical data?

[alixunderplatz]

@scriptator nice! imo, it's never wrong to have national historical data available.
Maybe @corradio could include this data in the visual review for the entire year 2019 ;)
(Like for 2018: https://www.youtube.com/watch?v=GxB91InvxiY)

[scriptator]

Cool, then I am going to give this a shot. Nice Video btw - I didn't know that so far.

[scriptator]

In the fetch_* functions of a parser, how much data before or after the provided target_datetime should or can be returned? As fetching the CSV file takes quite a bit of time and returns a whole month it would make sense here to return as much as possible IMO.

[corradio]

For historical data it doesn't really matter (the backend will adapt)
For real-time data the parser should be fast enough so 24hrs in practice is reasonable.