Duke undergraduates use machine learning techniques to evaluate electricity access in developing countries
For the past few summers, the Energy Initiative has mentored undergraduate student teams tackling energy data challenges. This year, a team of students led by researchers in the Energy Initiative's Energy Data Analytics Lab and the Sustainable Energy Transitions Initiative developed means to evaluate electricity access in developing countries through machine learning techniques applied to aerial imagery data.
The team's work will provide a much-needed basis for researchers and practitioners (at Duke and beyond) who are interested in understanding the path to electrification in underserved areas.
The students were participants in Data+, a Bass Connections program administered by the Information Initiative at Duke (iiD). Data+ is a 10-week summer research experience for Duke undergraduates interested in exploring new data-driven approaches to interdisciplinary challenges. Students join small project teams, working alongside other teams in a communal environment. They learn how to marshal, analyze, and visualize data, while gaining broad exposure to the modern world of data science.
Students on this year's Energy Data Analytics Lab team identified features of satellite imagery that can be used to demonstrate whether a community has access to electricity, created a reference dataset of key features, and applied machine learning methods to a large dataset.
Kyle Bradbury (managing director of the Energy Data Analytics Lab and project manager for the Data+ team) noted, "This year's team not only created two incredibly valuable datasets for research into energy access in developing countries, but also created a tool for crowdsourcing this data collection, tremendously increasing the efficiency of our research. The team has been remarkable in both their devotion to the research and their agility to learn and grow to overcome hurdles that naturally arise throughout the research process."
This team's work built on the accomplishments of several previous Data+ teams and Bass Connections in Energy projects (interdisciplinary teams that collaborate during the school year). Bradbury comments, "We've been pushing the bounds of what you can learn about energy systems using satellite imagery, starting in the U.S. and expanding globally. Two years ago, the team explored behind-the-meter electricity generation through identifying solar photovoltaic arrays. Last year, the team estimated energy consumption in buildings from the size of buildings. And now this year, we're bridging the gap to the developing world and looking for relevant aspects of satellite imagery that may indicate whether or not a village has access to electricity."
Faculty Leads for the summer 2017 Data+ team were Kyle Bradbury (Energy Initiative), Leslie Collins (Pratt School of Engineering), Timothy Johnson (Nicholas School of the Environment), Marc Jeuland (Sanford School of Public Policy), and Guillermo Sapiro (Pratt School of Engineering). The student team included Ben Brigman, Gouttham Chandrasekar, Shamikh Hossain, Boning Li, and Trishul Nagenalli.
The Energy Initiative's Energy Data Analytics Lab is seeking partners for future projects. We'd welcome a chance to talk with professionals about how a Duke University team could contribute to upcoming energy data projects at your company or agency. Participation in our student teams is highly competitive among Duke undergraduates with quantitative skills.
Contact Energy Data Analytics Lab managing director Dr. Kyle Bradbury for more information.