Import/Catalogue/Nordic Charging Station Import

From OpenStreetMap Wiki
Jump to navigation Jump to search


Introduction

Electric and hybrid cars have quickly become popular in the Nordic countries, particularly in Norway where electric cars now account for 26% of all new cars sold plus plug-in hybrids accounting for an additional 20% (update 2023: 80% combined). An electric car is only useful as long as it can be charged, of course, so a wide network of charging stations have rapidly been rolled out by private and public operators. There are now more than 4000 charging stations in Norway, Sweden and Finland.

The need for a database of charging stations to facilitate and promote the use of electric vehicles was quickly evident. Cooperation between the governmental entity Enova and the Norwegian Electric Vehicle Association resulted in the development of an open, publicly owned database that allows everyone to build services using a standardized API free of charge. The Nordic Council of Ministers later included also the other Nordic countries into this program.

As of 23 September 2018 the database contains 4089 charging stations (2346 in Norway, 1449 in Sweden, 289 in Finland and 4 in Denmark). There are currently 1153 charging stations in OSM in Norway, 276 in Sweden and 109 in Finland. The current charging stations are displayed on this map.

Update: Iceland has later been included. As of 23 March 2023 the database contains 8527 charging stations.

Goals

The goals of this import are to:

  • Add and update charging stations in Norway and Sweden.
    • Add 1193 new stations in Norway to the current 1153 stations (as of 23 September 2018)
    • Add 1179 new stations in Sweden to the current 276 stations
  • Add charger details such as connector type, number of connectors and charging capacity to all 4089 charging stations.
    • Currently only 321 (20%) charging stations in OSM in the three countries have any information about sockets or output

Schedule

The plan is to import Norway first, then discuss the import with Sweden and Finland afterwards. The two latter imports will only be done if the local communities approves it.

  • 22 March 2018 - Explicit permission and data obtained from Nobil.no
  • 12 October 2018 - Import plan discussed on the Norwegian OSM forum - link
  • 18 October 2018 - Import plan submitted to the Import maling list for review - link
  • 26 October 2018 - Start of import for Norway into OSM - progress here
  • 01 November 2018 - Import plan discussed on the Swedish mailing list - link. Later discussion in forum 2022/23: link.

Import data

Background

The charging station database NOBIL is developed and maintained by the Norwegian Electric Vehicle Association. Information is collected from EV-users, charging stations owners and operators and other contributors. NOBIL is financed and owned by the governmental entity Enova, an agency under the Norwegian Ministry of Climate and Environment. Swedish charging stations in NOBIL are being maintained by Energimyndigheten.

Data is available through an API which is described in this document.

Key license information:

  • Data source site: Dataset provided from https://nobil.no/api/server/
  • Data license: CC-BY 3.0
  • Type of license (if applicable): CC-BY 3.0
  • Link to permission (if required): Explicit permission obtained in 2012 (link) and confirmed 12 March 2018 (link)
  • OSM attribution (if required): Nobil: http://info.nobil.no/index.php/english
  • ODbL Compliance verified: Yes, explicit permission obtained

Explicit permission was obtained in 2012 and has been confirmed in 2018, including for Sweden and the other Nordic countries. An API key has been provided for OSM.

OSM data files

All data files in this project are available in this folder.

  • nobil.json - Complete set of original data i json format.
  • nobil.osm - Converted to tagged osm file.

Data qualiy

NOBIL is the de facto database for charging stations in Norway and Sweden. It is being used by a number of apps and web services, including by the operators and by several in-car services in electric cars. Unlike many other public databases, the content is being actively maintained and updated by a small staff. The quality is therefore quite good, but there are a couple of limitations to be aware of:

  • All of the coordinates which have been inspected will lead navigators to the right locations. In some cases, however, the exact location of the chargers on that location could have been even more precise. For example the coordinates could pinpoint a large parking lot but not the correct corner of that parking lot. Or the coordinates could be in the middle of a hotel building, when in fact the chargers are on the North side.
  • A few cases of duplicate charging stations in the dataset has been observed.
  • The given number of sockets is higher than the capacity for simultaneous charging because two or more sockets may share the same power line. Typically CHAdeMO and CSS (and to some extent Type2) sockets share the same line.
  • A few inconsistencies have been discovered, for example a Schucko outlet with too high output power. These have been reported back to NOBIL and will hopefully be fixed.
  • Charging station names often need editing.
  • Payment and authentication information is not complete.

The NOBIL database is not being actively maintained in Finland, Denmark and Iceland, so data quality may be questionable for those countries.

Import type

The local OSM community in Norway has discussed the import plan and local people will carry out the import.

Data preparation

Data reduction and simplification

The original data contains one record for each physical sockets, so if one site/charging station consists of 100 sockets, there will be 100 records for that charging station. A python program, nobil2osm.py, has been created to reduce that into OSM records which only need information for each type of sockets, e.g. socket:type2=5 to denote 5 chargers with a Type 2 socket. The python program is available on Github: nobil2osm.

NOBIL also contains real time information about how many chargers are in use and how many are vacant, however, this information will not be used for OSM.

Tagging plans

Station information

All charging station will have tagging according to the following two tables. Only attributes in the "OSM tagging" column will be tagged. Some of the other attributes will be included (with keys in CAPITAL letters) to aid during import, but should not be imported into OSM.

Update: Building on learnings from the first import, some of the tags below are not being imported anymore, for example description, contact information and authentication/payment (see prepared OSM file).

Original data Values OSM tagging Comment
Id ref:nobil=* Unique Nobil reference for the charging station, e.g. "361". To be used for later updates.
name name=* Name of the charging station, e.g. "Tesla Vestby Supercharger". For commercial, national charging stations, the brand name is included in the name tag.
Street Address street name
House_number Address house number
Zipcode Address post code
City Address city, capital letters.
Municipality_ID Municipality reference. Same as ref=* in boundary relation for the municipality.
Municipality Name of the municipality. Capital letters.
County_ID County reference. Same as ref=* in boundary relation for the county.
County Name of the county, capital letters
Description_of_location Text describing how to get to the location, in local language
Owned_by brand=* Owner/operator of charging station.

The commercial, national operators will be tagged in brand=*

Number_charging_points capacity=* Rather than using this attributes directly, capacity=* will be estimated to take into account the fact that CCS and CHAdeMO sockets often cannot be used simultaneously (also Type2 for certain charging points).
Position Latitude, longitude (Latitude, longitude)
Image "Id".jpg Filename for picture of location. Full URL is e.g.https://nobil.no/img/ladestasjonbilder/1682.jpg for filename 1682.jpg (station with Id/ref 1682).
Available_charging_points Real time information, not used.
User_comment Text with practical information from operator or users
Contact_info Text with phone, e-mail and/or website + comments, not formatted
Created Date created
Updated Date updated
Station_status 1 Always 1, not used
Land_code
  • NOR
  • SWE
  • FIN
  • DAN
Country code
International_id Country code and ID concatenated, e.g. "NOR_00361".
2: Availability
  • 1: Public
  • 2: Visitors
  • 3: Employees
  • 4: By appointment
  • 5: Residents
The three categories Public, Visitors and By appointment are regarded as public charging stations.

Tesla V3 chargers are tagged as

3: Location
  • 1: Street
  • 2: Car park
  • 3: Airport
  • 4: Shopping center
  • 5: Transport hub
  • 6: Hotel & restaurants
  • 7: Gas station
Type of location.
6: Time limit
  • Yes
  • No
If "yes", then the time limit is also specified.
7: Parking fee
  • Yes
  • No
parking:fee=yes/no Additional cost for parking.
21: Real-time information
  • Yes
  • No
Does the charing station deliver real time information about availability.
22: Public funding Either "None" or name of funding organization.
24: Open 24h
  • Yes
  • No

Socket information

There is one record for each outlet in the dataset (if 100 outlets are available for a site, there will be 100 records). Each socket type will have socket:"type"=* tagging according to the following translation table:

Original ref Original data OSM tagging Comment
1: Accessibility
  • 1: Open
  • 2: Standard key
  • 4: RFID
  • 5: Payment
  • 6: Cellular phone
  • 3: Other
Method of authentication before charging.
4: Connector
  • 14: Schuko/Schuko CEE 7/4
  • 29: Tesla Connector Roadster
  • 34: Blue industrial 3-pin
  • 35: Blue industrial 4-pin
  • 36: Red industrial 5-pin
  • 31: Type 1
  • 32: Type 2
  • 60: Type1/Type2
  • 50: Type 2 + Chucko CEE 7/4
  • 39: CCS/Combo
  • 41: Combo + CHAdeMO
  • 42: CHAdeMO + Type 2
  • 30: CHAdeMO
  • 40: Tesla Connector Model
  • 43: CHAdeMO + Combo + AC-Type 2
  • 70: Hydrogen
  • 80: Biogas LBG
  • 81: Biogas CBG
  • 0: Unspecified
Number of outlets for each available socket type.

The tesla_supercharger tag is used for earlier V1 and V2 versions. V3 is type2_combo.

5: Charging capacity
“<watt> – <volt> max <ampere>”
  • 7: 3.6 kW - 230V 1-phase max 16A
  • 16: 6.3 kW - 230V 3-phase max 16A
  • 8: 7.4 kW - 230V 1-phase max 32A
  • 10: 11 kW - 400V 3-phase max 16A
  • 17: 13 kW - 230V 3-phase max 32A
  • 19: 20 kW - 500VDC max 50A
  • 11: 22 kW - 400 3-phase max 32A
  • 18: 25 kW - 230V 3-phase max 63A
  • 12: 43 kW - 400 3-phase max 63A
  • 13: 50 kW - 500VDC max 100A
  • 20: 50 kW - Less than 100 kW + 43 kW - 500VDC max 200A + 400V 3-phase max 63A
  • 21: 50 kW - Less than 100 kW + 22 kW - 500VDC max 50A + 400V 3-phase max 32A
  • 28: 50 kW - 400VDC max 125A
  • 29: 75 kW DC
  • 23: 100 kW - 500VDC max 200A
  • 22: 135 kW - 480 VDC max 270A
  • 24: 150 kW DC
  • 32: 200 kW DC
  • 30: 225 kW DC
  • 31: 250 kW DC
  • 33: 300 kW DC
  • 25: 350 kW DC
  • 26: 350 bar
  • 0: Unspecified
Output wattage for each available socket type.
17: Vehicle type
  • 1: All vehicles (excluding plug-in hybrids)
  • 2: Short vehicles
  • 3: Two-wheel mopeds
  • 5: Plug-in hybrids
  • 6: Van
18: Reservable
  • Yes
  • No
19: Payment method
  • 1: Cellular phone
  • 2: VISA
  • 21: VISA, Mastercard and Charging card
  • 20: Cellular phone and Charging card
  • 6: Other cards
  • 7: Subscription
  • 8: Coin machine
  • 9: Miscellaneous cards
  • 10: Miscellaneous
All charging stations accept payment with RFID and/or app.
20: Charge mode
  • 1: Mode 1
  • 2: Mode 2
  • 3: Mode 3
  • 4: Mode 4
Not used
23: Manufacturer Not used
25: Fixed cable
  • Yes
  • No
8, 9, 10, 16 Not used (real time connector status information)

Changeset Tags

The changesets will be uploaded using import accounts and tagged with:

description=NOBIL charging station import
source=Nobil.no
source:date=20xx-xx-xx

Data transformation

Transformation for the various attributes in the database are shown in the two earlier tables. A list of attributes have been provided here.

The python program nobil2osm transforms the source data in a number of areas:

  • Converts the charging data, e.g. socket:type2=5 to denote 5 chargers with a Type 2 socket and socket:type2:output=22 kW for output capacity.
  • Estimates car capacity, i.e. number of simultaneous charging sockets (pair of CHAdeMO and CSS assumed to belong to the same power line)
  • Extracts information about the branded networks and adds the brand to the name of the charging stations (similar to petrol stations).
  • A number of orthographic fixes to the stations names (lowercasing, removes superfluous words, spaces etc).
  • Extracts contact email, phone and website

Data merge workflow

Team approach

Import will be carried out by members of the local communities in Norway and eventually Sweden.

References

Aerials will be used to check locations and to correct the position of the stations if possible/needed.

Workflow

This is the proposed workflow per country. It will make sure that the coordinates of the existing stations are used by default and that any conflicting tagging will be highlighted for manual resolution.

  1. Download all existing charging stations in OSM into JOSM, for example by a amenity=charging_station in Norway query using Overpass.
  2. Copy a suitable part of the nobil.osm file to the same layer, for example one county.
    • Only include public stations, i.e. with access=yes/permissive/customers
  3. Use the Conflation plugin in JOSM to identify and merge new and existing stations:
    • In the Configure menu select all new charging stations as Subject and all existing stations as Reference (this way existing positions will be kept)
    • Experiment with various Distance numbers to get correct matches.
    • Then walk though the list of matching stations:
      • Click Remove for false matches
      • Click Conflate to merge the new tags with the existing station. If there are conflicting tags, you will be prompted to make a selection.
      • If needed, manually move the station to an even more precise location using aerials for guiding
      • If needed, remove duplicate stations
      • Pay the most attention to the top entries in the list (i.e. stations far apart) and to the conflicting/red stations
  4. Review the name of the charging station.
  5. Delete all tags with capital letters.
  6. Upload to OSM using changeset tagging as shown earlier. In particular make sure that source:date=* is set.
  7. Store the nobil.osm file so that it can be compared to later versions for future updates (described below)
  8. Please report back to the local community if you think a certain area needs more inspection and refinement

Conflation

Conflation is described in the workflow above. Approx. 800 charging stations were imported in Norway in 2012, but with little information beyond the name of the stations. These stations will now get a full specification of charging capabilities.

Updates

There is a rapid growth in new charging stations and stations are being upgraded with more capacity and with new socket options. For the time being there is no automated script for updating the stations. Updates may be discovered by selecting stations with a Updated date after the source:date=* of the previous import. These stations may then be conflated with the existing stations in OSM using the same workflow as described above.

An alternative approach is to first compare the previous OSM file vs the new file and use the Conflation plugin to discover updates between them. That way it can be determined what has been updated at NOBIL versus what has been updated in OSM since the previous import.

See also

nobil.no - The database + the Norwegian site for maintenance of the database

NOBIL API documentation

nobil2osm on GitHub

ladestasjoner.no - Map for charging stations in the database

laddinfra.se - The Swedish site for maintenance of the database

Charging stations in OSM - Map showing existing charging stations in OSM