Network datasets by region
Europe
| Name
|
Version
|
Year
Published
|
Represented year
|
Region
|
Num. Substations or Buses
|
Num. Lines
|
Contains
|
Direct download?
|
Licence
|
Format
|
| SciGRID
|
0.2
|
2015
|
2015
|
Germany, but in principle whole world
|
495
|
825
|
Topology, Impedances
|
Yes
|
Apache Licence, Version 2.0 (code, documentation). ODBL (data)
|
CSV (csvdata)
|
| Bialek European Model
|
2
|
2013
|
2009
|
Continental Europe
|
1494 buses
|
2322
|
Topology, Impedances, Loads, Generators
|
Yes
|
Public Domain Dedication
|
PowerWorld, Excel
|
| National Grid ETYS 2014 Model
|
|
2014
|
2014
|
Great Britain
|
365
|
316
|
Topology, Impedances, Loads, Generators
|
Yes
|
Unclear
|
|
Austrian Power Network Grid
|
|
2015
|
2015
|
Austria
|
|
~100
|
Topology, Impedances
|
Yes
|
Unclear
|
PDF
|
| ENTSO-E STUM
|
1
|
2015 and before
|
2020?
|
Continental Europe?
|
1000s
|
1000s
|
Topology, Impedances
|
Requires registration
|
Restrictive
|
CIM
|
| ENTSO-E STUM
|
2
|
2015
|
2030
|
GB, Ireland, Baltics, Finland, Continental Europe
|
1000s
|
1000s
|
Topology, Impedances
|
Requires registration
|
Restrictive
|
Excel
|
| ENTSO-E STUM
|
3
|
2016
|
2030
|
GB, Ireland, Baltics, Finland, Continental Europe
|
1000s
|
1000s
|
Topology, Impedances
|
Requires registration
|
Restrictive
|
Excel
|
| ENTSO-E STUM
|
2
|
2015
|
2030
|
GB, Ireland, Baltics, Finland, Continental Europe
|
1000s
|
1000s
|
Topology, Impedances
|
Requires registration
|
Restrictive
|
Excel
|
| PyPSA-Eur
|
0.1.0
|
2020
|
2020
|
GB, Ireland, Baltics, Scandinavia, Continental Europe
|
5000
|
6000
|
Topology, Impedances, Loads, Generators, Renewable Availability Time Series
|
Yes
|
CC BY 4.0 (Dataset), GNU GPL v3.0 (Code)
|
NetCDF, CSV (alternative)
|
SciGRID
SciGRID is a project which started in 2014 and will be running for three years. The aim of SciGRID is to develop an open and free model of the European transmission network based on data from the OpenStreetMap. It is carried out by NEXT ENERGY - EWE Research Centre for Energy Technology, an independent non-profit institute at the University of Oldenburg, Germany, and funded by the German Ministry of Education and Research, and the initiative Zukunftsfähige Stromnetze.
An unofficial, post-processed version of SciGRID version 0.2 for Germany with attached load, generation and transformers is available as a PyPSA example, see also screenshots.
GridKit European Dataset
GridKit uses spatial and topological analysis to transform map objects from OpenStreetMap into a network model of the electric power system. It has been developed in the context of the SciGRID project at the NEXT ENERGY - EWE Research Centre for Energy Technology, to investigate the possibility of 'heuristic' analysis to augment the route-based analysis used in SciGRID. This has been implemented as a series of scripts for the PostgreSQL database using the PostGIS spatial extensions.
Data extracts are provided for Europe and North America in a similar CSV format to SciGRID.
osmTGmod Model
osmTGmod is a load-flow model of the German transmission-gird, based on the free geo-database OpenStreetMap (OSM). The model, respectively the heuristic abstraction process employs a PostgreSQL-database extended by PostGIS. The key part of the abstraction process is implemented in SQL and ProstgreSQL's procedural language pl/pgSQL. The abstraction and all additional modules are controlled by a Python-environment.
Bialek European Model
The 2nd version of the Bialek European Model is downloadable as an Excel file and in the format of the proprietary modelling software PowerWorld. The model covers voltages from 110 kV (a single line in the Balkans) up to 380 kV. It is released under a Public Domain Dedication.
The 1st version was released in 2002-2004 and is no longer available (see Archive mirror). The 1st version did not contain the Balkans region.
The methodology and validation for the 1st version of the model can be found in the paper Approximate model of European interconnected system as a benchmark system to study effects of cross-border trades by Zhou and Bialek, 2005.
The model contains the impedances and number of circuits of each line, but not the length (which can in principle be determined from the impedance and number of circuits, given standard line parameters). Only cross-border lines are assigned thermal capacities.
There is currently no coordinate dataset for the buses. The PowerWorld file contains spatial data, but in an unknown projection. The georef-bialek github project is an attempt to fix this; there is also a geo-referenced version from Tue Vissing Jensen.
DIW ELMOD-DE open model of Germany
ELMOD-DE is an open model of the German electricity system developed at DIW and TU Berlin, which includes both a model of the high voltage transmission network, power plants, hourly load and weather data for the year 2012 and GAMS code to run linear optimisation simulations. It contains 438 geo-referenced network nodes and 697 transmission lines at 380 kV and 220 kV. Transformers are not modelled but per unit line series impedances are adjusted to the voltage level.
The model includes 47 pages of documentation.
The transmission data was, according to the documentation, derived from the VDE and TSO maps and from OpenStreetMap. The data is provided as-is without the code that generated it.
National Grid Model
National Grid Electricity Ten Year Statement 2014 Model
Shapefiles and maps of tower, lines, cables and substations here.
Austrian Power Network Grid Model
Austrian Power Network Grid
Danish Power Network Grid Model
Danish Transmission network data
The data are not directly available, but rather a registration form is required before obtaining access.
It has features not present in the ENTSO-E STUM (see below):
- It's a full non-linear model with all the reactive power demand, P and Q capabilities of gens and shunt reactive power compensation.
- It lists the power capabilities of the generators and their fuel type (wind/solar/gas etc), not just the dispatch.
- They seem to have separated RE feed-in from the load, which wasn't the case for STUM where wind and solar are lumped with the load as residual load.
What's missing are geocoordinates for the substations (which can be read off roughly from the JPG map) and time-dependence of the loads and/or variable generators. For Denmark, which has many CHP units, it would also be useful to know the heat demand and how the CHP units are operated.
RTE Network Dataset for France
RTE network dataset
Elia Network Dataset for Belgium
Elia network dataset
TenneT NL Network Dataset for the Netherlands
TenneT NL
TenneT DE Network Dataset for Central Germany
Tennet DE
Amprion Network Dataset for Western Germany
Amprion, interactive map of the grid extension projects
TransnetBW Network Dataset for Southwest Germany
TransnetBW
50 Hertz Network Dataset for Eastern Germany
50 Hertz statistisches Netz
Ceps Network Dataset for Czech Republic
CEPS
ENTSO-E Interactive Grid Map
ENTSO-E announced its Interactive ENTSO-E Transmission Network Map in March 2016.
The map uses OpenStreetMap as a background and Mapbox for displaying the map data.
The map is based on the ENTSO-E static grid map, which is based on the TSOs' own maps. It is known to be an approximate artistic representation rather than an accurate geographical map. Some power plants may be incorrectly labelled (e.g. fuel type may not be accurate).
The map includes information on the number of circuits and the voltage levels of transmission lines.
Information, including all geographical coordinates, can be extracted from the web API, but requires further topological processing to be turned into an electrical network model. Lines need to be connected, etc. The GridKit project provides code for this purpose and has released an unofficial dataset, which forms an electrical network model complete with buses, links, generators and transformers, full geographic coordinates, as well as all electrical metadata contained in the ENTSO-E map.
ENTSO-E Static Grid Map
ENTSO-E releases maps of the European transmission grid, both electronically and in paper form.
The maps for the whole ENTSO-E system are in the projection EPSG 3034, which is a Lambert Conformal Conic projection. The lower left corner is approximately at (lon,lat) = (-9.5,28) and the upper left corner is at (75.5,58.5). This was checked in the georef-bialek github project.
ENTSO-E STUM
ENTSO-E makes available a model of the European transmission system. Registration is required to download it on the ENTSO-E STUM page. It is not totally clear what one may and may not do with it (e.g. whether it is possible to publish results derived from it or an aggregation of the nodes, etc.).
The first version of the model was released in the CIM XML-based format for the old UCTE area. The model was a winter snapshot for 2020, including TYNDP projects. The node names were obscured so that the model was unusable. Line capacities were missing.
The second version, published in June 2015 as Excel spreadsheets, is more useful. It contains the whole ENTSO-E area with the exception of Norway, Sweden, Cyrus and Iceland. The node names are the same as those used by the TSOs. Quoting from the documentation: "It represents the power system of the ENTSO-E members for 2030 in Vision I of the TYNDP 2014", i.e. it includes planned TYNDP projects. It includes all nodes, lines, transformers and aggregated loads and generators at each node for one snapshot. Line data includes series reactance and resistance, but not line length or capacity or number of circuits or wires per circuit bundle. Geolocation data for the nodes is missing. Node names are recognisable from the substation names on the ENTSO-E map. The model is intended for a linear load flow only. It is not clear which wind/solar/load snapshot the model represents (it is an "exemplary scenario"). Generators are not distinguished by generation source.
The third version, published in February 2016 as Excel spreadsheets has in addition thermal ratings for most transformers and most transmission lines, along with reactive power feed-in, consumption and compensation, so that a full non-linear power flow can be run on the grid.
ENTSO-E Initial Dynamic Model of Continental Europe
ENTSO-E Initial Dynamic Model of Continental Europe
Requires registration. Can model "the main frequency response of the system as well as the main inter-area oscillation modes".
Flow-based market coupling data by Joint Allocation Office
The joint allocation office hosts various official data (including PTDFs) around the Flow-based market coupling algorithm in use in Europe.
http://utilitytool.jao.eu/
http://utilitytool.jao.eu/CascUtilityWebService.asmx
PyPSA-Eur: An Open Optimisation Model of the European Transmission System
PyPSA-Eur is a model/dataset of the European power system at the transmission network level.
The transmission network data is based on a cleaned up extraction of the Interactive ENTSO-E Transmission Network Map, extracted using GridKit.
The model covers the ENTSO-E area and contains all alternating current lines at and above 220 kV voltage level and all high voltage direct current lines, substations, an open database of conventional power plants, time series for electrical demand and variable renewable generator availability, and geographic potentials for the expansion of wind and solar power.
The model only includes freely available and open data. It provides a fully automated free software pipeline to assemble the load-flow-ready model from the original datasets. The model is suitable both for operational studies and generation and transmission expansion planning studies.
Current versions of the code and dataset can be found on zenodo:
Documentation is available at https://pypsa-eur.readthedocs.io
Development takes place on Github at https://github.com/pypsa/pypsa-eur
The netcdf files (.nc) can be imported with PyPSA. Documentation for this is available at https://pypsa.readthedocs.io/en/latest/import_export.html#import-from-netcdf.
Some basic validation is provided in a paper describing the dataset:
Jonas Hörsch, Fabian Hofmann, David Schlachtberger, and Tom Brown. PyPSA-Eur: An open optimisation model of the European transmission system. Energy Strategy Reviews, 22:207-215, 2018. https://arxiv.org/abs/1806.01613, https://doi.org/10.1016/j.esr.2018.08.012.
Australia
Lines and substations
United States
There is raster graphic of the US transmission grid at https://www.e-education.psu.edu/geog469/book/export/html/111.
Western Electricity Coordinating Council
Apparently there is a a WECC Transmission Expansion Planning Policy Committee (TEPPC) 2024 Common Case GridView dataset, but the exact link seems elusive.
The WECC Transmission Expansion Planning has links to Excel files.
Western US Power Grid
The Western US Power Grid dataset has 4941 nodes and 6594 lines, but apparently these are not well enough labelled to distinguish where and what the nodes/lines are.
GridKit North American Dataset
GridKit uses spatial and topological analysis to transform map objects from OpenStreetMap into a network model of the electric power system. It has been developed in the context of the SciGRID project at the NEXT ENERGY - EWE Research Centre for Energy Technology, to investigate the possibility of 'heuristic' analysis to augment the route-based analysis used in SciGRID. This has been implemented as a series of scripts for the PostgreSQL database using the PostGIS spatial extensions.
Data extracts are provided for Europe and North America in a similar CSV format to SciGRID.
US grid dataset by Breakthrough Energy
US grid dataset by Bill Gates' Breakthrough Energy
Global
OpenStreetMap
The global OpenStreetMap (OSM) power grid data is visible at ITO World Electricity Distribution and Enipedia has nightly extracts of the power grid from OSM.
GridKit Datasets
GridKit uses spatial and topological analysis to transform map objects from OpenStreetMap into a network model of the electric power system. It has been developed in the context of the SciGRID project at the NEXT ENERGY - EWE Research Centre for Energy Technology, to investigate the possibility of 'heuristic' analysis to augment the route-based analysis used in SciGRID. This has been implemented as a series of scripts for the PostgreSQL database using the PostGIS spatial extensions.
Data extracts are provided for Europe and North America in a similar CSV format to SciGRID.
See IRENA News Announcement
Non-Region Specific
University of Washington Power Systems Test Case Archive
Power Systems Test Case Archive
IEEE PES Power Grid Library
Overview
Optimal Power Flow Cases
RWTH Aachen Transmission Expansion Problem Benchmark Case
RWTH Aachen has published A Benchmark Case for Network Expansion, which is "derived from the IEEE 118 bus network and modified in accordance with European standards such as a nominal frequency of 50Hz, the use of conventional voltage levels, and conductor dimensions."
Registration is required to download the model.
The paper describing the model is A benchmark case for network expansion methods, 2015.
Other lists of network datasets
Free software for power system analysis
PyPower in Python
PyPSA: Python for Power System Analysis
PowerGAMA in Python
MATPOWER in Matlab or Octave
OpenDSS in Delphi
PSAT in Matlab or Octave
PowerModels.jl in Julia
AutoGridComp in Python for the comparison of power grid models
Other lists of power system analysis software
https://wiki.openelectrical.org/index.php?title=Power_Systems_Analysis_Software
https://nkloc.wordpress.com/2011/11/11/power-system-simulation-software-list/
http://www2.econ.iastate.edu/tesfatsi/ElectricOSS.htm
Typical electrical parameters for transmission infrastructure
Calculating cable impedances
See http://www.openelectrical.org/wiki/index.php?title=Cable_Impedance_Calculations and electrical engineering textbooks.
Generalities on overhead alternating current transmission lines
Three-phase power
In almost all of the world electrical power is transmitted using alternating current with three phases separated by 120 degrees, see Wikipedia: Three-phase electric power.
For this reason the cables on power lines are bundled in groups of three.
(Exceptions include: direct current power lines and some transmission systems for supplying trains, which are e.g. in Germany two-phase and at 16.7 Hz.)
Current I and current limits are almost always quoted per phase.
Voltage in the transmission system is almost always quoted as the phase-to-phase potential difference, often called line-to-line voltage V_{LL}, since this is the easiest value to measure. It is related to the line-to-ground or line-to-neutral potential difference V_{LN} by V_{LL} = \sqrt{3} V_{LN}.
The apparent power transported in each phase is give by I*V_{LN}, so that for a complete transmission circuit the power is three times this value:
S = 3*I*V_{LN} = \sqrt(3)*I*V_{LL}
Often it is assumed that the voltage and current magnitudes are the same in each phase, i.e. that the system is balanced and symmetric. This should be the case in the normal operation of the transmission system. The impedances and limits below are quoted assuming that the system is balanced, so that only positive sequence impedances are given. See Wikipedia: Symmetrical components.
In an unbalanced system, the three phases can be described using the positive-, negative- and zero-sequence components, where the impedances are different for each sequence.
Bundled conductors
See Wikipedia: Overhead power line: Bundle conductors.
Often the conducting wires for each phase are separated into bundles of several parallel wires, connected at intervals by spacers. This has several advantages: the higher surface area increases the current-carrying capacity, which is limited by the skin effect, it reduces inductance and it helps to cool the wires.
Circuits
Each group of three phases is called a circuit. Power-carrying capability can be increased by having several circuits on a single pylon, so that wire bundles always appear in multiples of 3 in power lines.
European 50 Hz transmission lines
The main European alternating current (AC) electricity system is operated at 50 Hz. (Other networks, such as those for electrified trains, operate at other frequencies and some transmission lines use direct current.)
On the continent AC transmission voltages are typically 220 kV or 380 kV (sometimes quoted as 400 kV, since network operators often run their grid above nominal voltage to reduce network losses).
220 kV overhead lines are typically configured with a bundle of 2 wires per phase with wires of cross-section Al/St 240/40.
380 kV overhead lines are typically configured with a bundle of 4 wires per phase with wires of cross-section Al/St 240/40.
We now list the impedances of the transmission lines, which can be used for example in the lumped pi model.
Electrical properties for single circuits
| Voltage level (kV)
|
Type
|
Conductors
|
Series resistance (Ohm/km)
|
Series inductive reactance (Ohm/km)
|
Shunt capacitance (nF/km)
|
Current thermal limit (A)
|
Apparent power thermal limit (MVA)
|
| 220
|
Overhead line
|
2-wire-bundle Al/St 240/40
|
0.06
|
0.301
|
12.5
|
1290
|
492
|
| 380
|
Overhead line
|
4-wire-bundle Al/St 240/40
|
0.03
|
0.246
|
13.8
|
2580
|
1698
|
In the table the thermal limit for the current is calculated as 645 A per wire at an outside temperature of 20 degrees Celsius.
The thermal limit for the apparent power S is derived from the per-phase current limit I and the line-to-line voltage V by S = \sqrt{3}VI.
Sources for the electrical parameters:
Oeding and Oswald Elektrische Kraftwerke und Netze, 2011, Chapter 9
See also comparable parameters in:
- DENA Distribution Network Study, 2012, Table 5.6
- DIW Data Documentation 72, 2014, Table 15, taken from Kießling, F., Nefzger, P., Kaintzyk, U., "Freileitungen: Planung, Berechnung, Ausführung", 2001, Springer
- KIT Electrical Parameters Reading Sample, 2013
European 50 Hz high voltage transformers
Typical 380/220 kV transformers have a nominal power of around 400-500 MVA and a per unit series reactance of around 0.08-0.1.
- TODO: references
Combining electrical parameters for multiple circuits
In the table above, the impedances are quoted for a single circuit. The resistance and inductive reactance decrease proportional to the number of parallel circuits (with small modifications to the inductance due to the different geometry of the parallel circuits). Similarly the capacitance increases proportional to the number of parallel circuits (again, roughly because of changing geometry).
Standard Test Test Networks
http://sites.ieee.org/pes-testfeeders/resources/
https://github.com/e2nIEE/pandapower/tree/develop/pandapower/networks
