Tool for Mapping the Electrical Grid (Medium-Voltage Distribution)
Introduction
This model generates electricity grid paths of one kilometer for every known grid location. It includes distribution lines that connect high-voltage distribution lines to low-voltage lines serving the end consumer. The model is based mainly on publicly available data such as night time satellite imagery, roads, political boundaries and population centers. The model currently has an accuracy of 70% and can be further improved by integrating locally available data into the model.
[1] This model is a joint collaboration among Facebook, the Energy Sector Management Assistance Program (ESMAP) at the World Bank, KTH Royal Institute of Technology, World Resources Institute (WRI), and the University of Massachusetts Amherst.
Download
The model is openly available and can be replicated for other countries. The data is currently available for six different countries: Malawi, Nigeria, Uganda, Democratic Republic of Congo, Cote D’Ivoire, and Zambia.
- link to the tutorial: https://github.com/facebookresearch/many-to-many-dijkstra
- link to the data for different countries: https://energydata.info/dataset/medium-voltage-distribution-predictive