Energy Data Analytics Lab: Electricity Access in Developing Countries from Aerial Imagery (2017-2018)

Background

Access to reliable electricity is strongly correlated with economic prosperity and health. Over 15% of humanity has no access to it at all, and far more have access only to intermittent supplies that preclude most productive energy uses. According to the World Bank, in some nations like South Sudan, Chad and Burundi, fewer than 7% of people have any access at all.

Research on sustainable energy transitions being conducted at Duke and elsewhere is trying to better understand the drivers and impacts of electrification on health, land use, the environment and the local economy. However, current methods for assessing access rely almost completely on household surveys or highly aggregated (e.g., national-level) data sources.

Project Description

This Bass Connections project will develop means to evaluate electricity access in developing countries through machine learning techniques applied to aerial imagery data. This work will provide the needed basis for research groups at Duke and elsewhere interested in understanding the path to electrification in underserved areas.

The project team will fill critical gaps by piloting and then applying automated algorithms for generating spatially-disaggregated data on electricity access in developing countries using aerial imagery. The team will develop new machine learning techniques to extract information on energy infrastructure and consumption using high-resolution aerial imagery.

In high-resolution aerial imagery, power transmission and distribution lines are often visible to the naked eye, and the poles and pylons supporting them and substations linking them are clearly visible. Utilizing the information from these images, the team will develop estimate network topologies of electric power lines. Additionally, imagery of lights at night can provide indicators of energy consumption.

The initial phase of this work will involve a related Data+ summer research project, which will launch the larger Bass Connections project by compiling, curating and publishing an initial ground-truthed dataset. The Bass Connections team will then use this as the basis of its work.

Anticipated Outcomes

Dataset of fully-identified electricity infrastructure in aerial imagery; published version of the dataset made available on the data repository Figshare; collection of code implementing the algorithm for identifying the power system infrastructure; conference or journal paper based on the methods developed and/or the dataset

Timing

Fall 2017 – Spring 2018

  • Fall 2017: Intensive short course on data analytics, energy systems, energy access, research techniques and project management; Kaggle machine learning competition; clearly defined subset of goals
  • Spring 2018: Project execution; reflections and presentations to wider community

This Team in the News

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

See earlier related team, Energy Data Analytics Lab (2016-2017), and a Data+ summer project, Electricity Access in Developing Countries from Aerial Imagery.

Faculty/Staff Team Members

Kyle Bradbury, Duke University Energy Initiative*
Leslie Collins, Pratt School - Electrical & Computer Engineering*
Marc Jeuland, Sanford School of Public Policy and Duke Global Health Institute
Timothy Johnson, Nicholas School - Earth & Ocean Sciences*
Guillermo Sapiro, Pratt School - Electrical & Computer Engineering

Graduate Team Members

Brian Wong, Master of Environmental Management, Global Environmental Change

Undergraduate Team Members

Fangge Deng, Computer Science (AB), Environmental Sci/Policy (AB2)
Shamikh Hossain, Computer Science (BS), Economics (BS2)
Prithvir Jhaveri, Computer Science (AB)
Ashley Meuser, Electrical & Computer Egr(BSE), Computer Science (AB2)
Harshvardhan Sanghi, Mechanical Engineering (BSE)
Joseph Squillace, Computer Science (AB)
Anuj Thakkar, Mechanical Engineering (BSE)
Xiaolan You, Computer Science (AB), Statistical Science (AB2)

* denotes team leader

Status

Active