Mapping Energy Infrastructure
Mobilizing artificial intelligence tools to understand the energy access playing field
Big data offers big opportunities to understand enduring problems in new ways, and Duke University’s interdisciplinary Energy Data Analytics Lab is at the forefront, developing methods and tools to evaluate electricity access in developing countries using machine learning techniques applied to aerial imagery data.
Our understanding of the drivers and impacts of electrification on health, land use, the environment and the local economy relies heavily on household surveys or national-level data sources. Collecting and maintaining data on access to energy and the location of energy infrastructure is overly time-consuming.
Beginning with the development of tools to measure electricity access in developing countries through machine learning techniques applied to satellite image data – far beyond the relatively common “Lights at Night” data, and adding the automatic detection of features such as transmission lines, substations, and off-grid generation assets such as solar photovoltaic panels – the project has provided a much-needed base for understanding the path to electrification in underserved areas, as well as its impacts. Moving forward, faculty and students are building on this work by employing state-of-the-art machine learning tools to detect critical energy infrastructure in satellite imagery, including power distribution lines and off-grid distributed energy generation sources such as solar photovoltaics. This work will help to deepen understanding of both the drivers and impacts of electricity access and reliability in underserved areas, such as the productive use of power and improved development outcomes.
Students: Brian Wong, Wendell Cathcart, Fangge Deng, Shamikh Hossain, Shijia Hu, Prithvir Jhaveri, Ashley Meuser, Harshvardhan Sanghi, Joseph Squillace, Anuj Thakkar, Xiolan You, Ben Alexander, Yutao Gong, Xinchun Hu, Varun Nair, Lin Zuo
Publication: Automating Electricity Access Prediction Using Satellite Imagery, presented at Bass Connections Showcase, April 18, 2018, and Visible Thinking, April 19, 2018
- Duke Expert Touts Transformative Potential of Energy Data Analytics in New Book on Digital Decarbonization
- Three Graduate Students Honored for Outstanding Bass Connection Project Team Mentorship
- Crossing Boundaries to Meet Our Energy Needs
- Gauging Renewable Energy Generation Using Satellite Imagery
- Energy Data Analytics Lab Team Takes Top Prize at 2018 Duke Research Computing Symposium with Electricity Access Project
- Mapping Electricity Access for a Sixth of the World’s People
- Student Team’s Success in Energy Case Competition Is Powered by the Unique Duke Experience
- Duke Undergraduates Use Machine Learning Techniques to Evaluate Electricity Access in Developing Countries