osm-fr
Wrong voltage suggestion for réséda network
https://github.com/osm-fr/osmose-backend/blob/master/analysers/analyser_merge_power_substation_minor_FR.py#L30 This analysis suggest 20 000 volts for minor_distribution this voltage is used by Enedis But other operator use different voltage, in my case réséda use 17 500 volts
I think the best way is to enumerate operator with voltage different of Enedis, ORE Agence can be these informations?
20000 volts is the default value for medium voltage in France, that's why we use it in Osmose.
Voltage actually depends and varies at each medium voltage bay in distribution substations, as show on capareseau website for each substation. There is no voltage per operator, such a list doesn't exist. It can be changed in the middle of a line in the country.
As we don't have open data for voltages of each line, it' not possible to automatize osmose on this tag. However if you are knowledgeable about a given area, it is possible to set a particular voltage, see what was done near of Falaise https://openinframap.org/#11.26/48.8544/-0.3231
I'm just thinking, if the voltage is unknown from the open data set, it may be better to not give an explicit value in a fix proposal (but e.g. describe it in the sidebar text). Otherwise it's likely people will assume the 20000 is known to be correct from the open data and add it to OSM?
It would be good to move voltage=20000 from static1 to static2 in this analyzer, just like substation=minor_distribution we are unable to define from opendata currently.
Data will come, maybe.
I never worked on the merge analysers, so I don't know what exactly this changes, but if I look at the output it seems they're just merged, e.g. for the user nothing would change, correct?
@flacombe why do you want to set voltage if unknown ? We can just remove it.
for the user nothing would change, correct?
Putting voltage=20000 in static2 won't produce additional warnings when user will define another voltage (like 17500) when applicable.
@flacombe why do you want to set voltage if unknown ?
Because it is really important to allow distinction with other substations with higher voltage.
The most common term would be voltage=medium or voltage=hta but it's a mess to process with math expressions.
As the default value is define by law, we just use it waiting for more extensive data.
Because it is really important to allow distinction with other substations with higher voltage.
Hmm, I would still be in favor of not proposing a value if it's unknown. I would argue a missing value is less bad than a wrong value.
But as an intermediate solution: how about giving multiple fix suggestions, like voltage=17500, voltage=20000, voltage=25000. In that way it'll be more clear for the user that it's not a fact that's proposed, just a possibility. (Not sure if Merge analysers support this)
And definitely static2 then :)
how about giving multiple fix suggestions, like voltage=17500, voltage=20000, voltage=25000.
It's not suitable here since the mapper isn't able to find the information at all (it's not readable on ground).
Data has been asked to operators a few days ago to reduce the amount of undetermined voltages. https://www.agenceore.fr/forum-expert/tension-dexploitation-des-reseaux-hta
PR for static2 is on its way.
PR for static2 is on its way.
I merged it, as minimum requirement change.
Rather than just a wrong voltage, maybe add a note tag to let know that the voltage is uncertain, but high.
I think it can be closed for now, we can't doing more things for now