WATCH: Unsupervised Learning - Foundations for Energy Data Analytics

Dive into this foundational video on unsupervised learning by Dr. Jordan Malof. This talk introduces common unsupervised learning techniques and how they can be applied to energy challenges. You’ll hear about mixed clustering, dimensionality reduction, and more!

Part of the Foundations for Energy Data Analytics Series. This talk was originally presented during a workshop of the Energy Data Analytics Ph.D. Student Fellows Program, organized by the Energy Data Analytics Lab at Duke University. The Fellows Program is funded by a grant from the Alfred P. Sloan Foundation, Grant-G2020-13922.
(Note: Conclusions reached or positions taken by researchers or other grantees represent the views of the grantees themselves and not those of the Alfred P. Sloan Foundation or its trustees, officers, or staff).

Add to Slider: 
WATCH: Computer Vision - Foundations for Energy Data Analytics
Image of an atypical, blooming, purple flower

Dive into this foundational talk on how computer vision turns images into information. The talk by Dr. Kyle Bradbury introduces common problems in the space: image classification, segmentation, and object detection. Bradbury explains how computer vision techniques can be applied to remotely sensed data to derive insights about energy systems. He also discusses how to get started in this space, noting tools and resources that can help you explore further.

Part of the Foundations for Energy Data Analytics Series. This talk was originally presented during a workshop of the Energy Data Analytics Ph.D. Student Fellows Program, organized by the Energy Data Analytics Lab at Duke University. The Fellows Program is funded by a grant from the Alfred P. Sloan Foundation, Grant-G2020-13922.
(Note: Conclusions reached or positions taken by researchers or other grantees represent the views of the grantees themselves and not those of the Alfred P. Sloan Foundation or its trustees, officers, or staff).

Add to Slider: 
Add To Slider
WATCH: Energy Infrastructure Detection with Satellites: Synthetic Imagery for Finding Wind Turbines
wind turbines in a field

Duke students’ Bass Connections research on energy access and data analytics comes together in a final energy presentation on synthetic imagery used to improve automated wind turbine detection in satellite imagery, especially when applied to diverse locations.

Efforts to ensure energy access across the globe are often hampered by a lack of critical information to guide decision-making and electricity system planning. Information on village-level electricity access and reliability, as well as the location and characteristics of power system infrastructure, is especially scarce. Decision-makers require this information to determine the optimal strategies for deploying energy resources, like where to prioritize development and whether electrification should be accomplished through grid expansion, micro-grids, or distributed generation.

During the 2020-2021 school year, a Bass Connections research team at Duke University aimed to develop deep learning techniques that can automatically and rapidly scan massive volumes of remotely sensed data, such as satellite imagery, to develop detailed maps of energy infrastructure. These deep learning approaches may provide powerful tools for researchers, policy-makers, and governments to collect energy systems information. This video captures the Bass Connections team’s end-of-year presentation in April 2021.

The team used machine learning to create a model that detected wind turbines solely from satellite imagery by training it first with real images of turbines. Since these images are scarce and in practice the machine learning techniques need to be applied to different locations than from where the training data are available, this approach was compared to data resulting from a model which also was trained on synthetic images of wind turbines. Synthetic images, while they might look real to the machine, are generated images and are not genuine photos. Feeding the model synthetic images of wind turbines increased the accuracy or “average precision” of the predicted turbine location.

Bass Connections is a unique Duke University program that brings together faculty, postdocs, graduate students, undergraduates, and external partners to tackle complex societal challenges in interdisciplinary research teams.

Student Team Members: Ada Ye (T'23), Jessie Ou (T'22), Wendy Zhang (T'21), Eddy Lin (T'22), Tyler Feldman (T'23), and Jose Moscoso (MIDS '21)

Faculty Team Leaders: Kyle Bradbury (Pratt School of Engineering and Managing Director of the Energy Data Analytics Lab at the Duke University Energy Initiative) and Jordan Malof (Pratt School of Engineering)

Learn more about the project:

Add to Slider: 
Add To Slider
Duke Students, Apply for the 2021 Clean Energy Prize
Triangles with Duke colors highlighting wind turbines, solar installation, Duke Chapel, and a woman wearing a VR headset

The Duke University Energy Initiative (EI) and Innovation & Entrepreneurship Initiative (I&E) announce the return of the Clean Energy Prize to support the development of energy technologies, products, and services at Duke. The Prize makes a $10,000 award to support novel ideas, potential products, and services that advance an accessible, reliable, affordable, and clean energy future.

The Prize invites all Duke students, including May 2021 graduates, to propose innovative projects that could lead to new products or services that will advance a clean energy future. This could mean demonstrating the feasibility of an idea or innovation for a commercial or social venture; developing a working software, service, or device prototype; or developing new applications or markets for a technology in development.

“Securing a clean energy future is an urgent imperative—and one that demands transformative approaches,” said Energy Initiative director Dr. Brian Murray, a faculty member at the Nicholas School of the Environment and Sanford School of Public Policy. “The Clean Energy Prize aims to spur Duke University students’ creativity and fuel new solutions for one of our world’s greatest challenges.”   

Past recipients of the Clean Energy Prize include Arsheen Allam MBA ’17 and Towqir Aziz MA ’18, who received the prize in 2018 for GOLeafe, which developed a new production process for graphene, a promising nanomaterial with potential applications in solar energy production and energy storage. Allam, a Forbes “30 Under 30 in Energy” honoree, was recently named to the fifth cohort of Chain Reaction Innovations, the U.S. Department of Energy’s elite entrepreneurship program at Argonne National Laboratory.

The deadline for submissions is 11:59 pm EDT on Friday, May 28, 2021. For complete details and application instructions, download the Request For Applications.

Contact Suellen Aldina, Energy Initiative Director of Engagement and Administration, with any questions.

Add to Slider: 
Energy Data Analytics Ph.D. Student Fellows Program announces 2021 cohort with students from Duke University, NC State University, and UNC Charlotte
Lightning bolt on a blue and green background

The Duke University Energy Initiative has unveiled its newest cohort of Energy Data Analytics Ph.D. Student Fellows, which, for the first time, includes doctoral students beyond Duke as part of the program’s expansion. This year’s seven Ph.D. Fellows are from Duke University, North Carolina State University, and the University of North Carolina at Charlotte.

This one-of-a-kind fellows program is designed to produce scholars with expertise in both data science and energy application domains and enables collaboration across universities in the region.

“The Energy Data Analytics Ph.D. Student Fellows Program continues to break the mold of a typical doctoral training venture,” says Brian Murray, director of the Duke University Energy Initiative. “It is interdisciplinary in nature and it opens connections to other universities, so we’re able to innovate in key areas that broaden our ability to build a more accessible, affordable, reliable, and clean energy system.”

Each fellow will conduct a related research project applying data science techniques to energy applications this summer, working with faculty advisors from multiple disciplines. Fellows will participate in regular mentorship workshops as they develop their research. Learn more about the fellows’ backgrounds and their 2021-2022 research projects. The 2021 cohort includes:

  • Yang Deng
    Yang is a Ph.D. student in electrical and computer engineering at Duke University. He is interested in accelerating inverse design of revolutionary energy materials based on all-dielectric metasurfaces through deep learning. 
  • Qian Luo
    Qian is a Ph.D. candidate in environmental engineering at North Carolina State University. Her research focuses on developing strategies to mitigate negative human health impacts associated with power sector emissions by studying the interactions between power sector emissions, air quality, and human health. 
  • Suhas Raju
    Suhas is a Ph.D. student in electrical engineering at the University of North Carolina at Charlotte. His research focuses on energy efficiency through the use of artificial intelligence techniques, seeking to identify solutions to reduce greenhouse gas emissions in the commercial building sector.
  • Josh Randall
    Josh is a Ph.D. candidate in parks, recreation, and tourism management at North Carolina State University. He is a geographer interested in using spatial analysis to realize the equitable access of resources in society, particularly through informing policy and community action.
  • Simiao (Ben) Ren
    Ben is a Ph.D. student in electrical and computer engineering at Duke University. Working in Duke’s Applied Machine Learning Lab, he is interested in applying advanced deep learning techniques to develop novel algorithms that can extract energy systems information from unmanned aerial vehicle (UAV) imagery
  • Celine Robinson
    Celine is a Ph.D. candidate in civil and environmental engineering at Duke University. She is interested in applying deep learning techniques and Bayesian statistics to model and assess natural-hazard-triggered technological (natech) risk in complex and interconnected systems.
  • Zhenxuan Wang
    Zhenxuan is a Ph.D. student in the University Program of Environmental Policy at Duke University with an economics concentration. Sitting at the intersection of environmental economics, climate change, and industrial organization, his research employs innovative data and methods to explore human and firm behavioral responses to environmental changes, and evaluate the efficiency and distributional effect of environmental policies.

In addition to a stipend and partial tuition remission during the summer, fellows will receive up to $1,500 in funding for research and professional development.

Students will receive research mentorship, training on a wide array of energy and data science topics, and research communication advice to broaden the impact of their work. Research outputs can include publications, datasets or code repositories, and video presentations. The first two cohorts of fellows have collectively produced 19 journal or conference papers/presentations, 11 video presentations, and 6 code repositories and datasets to-date.

The program is organized by Duke’s Energy Data Analytics Lab, a collaboration among three of the university’s signature interdisciplinary units: Duke University Energy Initiative (which houses it), Rhodes Information Initiative, and Social Science Research Institute (SSRI).

Duke’s Energy Data Analytics Ph.D. Student Fellows Program is funded by a grant from the Alfred P. Sloan Foundation. (Note: Conclusions reached or positions taken by researchers or other grantees represent the views of the grantees themselves and not those of the Alfred P. Sloan Foundation or its trustees, officers, or staff.)

Add to Slider: 
Add To Slider
Thumbnail Image: 
Energy Data Analytics Ph.D. Fellows Program announces 2021 cohort with students from Duke University, NC State University, and UNC Charlotte
Doctoral students from Carnegie Mellon, UMass Amherst, and Duke land top prizes for energy data analytics research talks
Illustration of trophy on top of a motherboard

Today the Energy Data Analytics Lab at Duke University announced the winners of the Lightning Talks Competition at the  Energy Data Analytics Symposium (Dec. 8-9, 2020).

The competition highlights research by emerging scholars in energy data analytics, attracted 21 entries from 12 universities and organizations. Judges assessed participants’ five-minute “lightning talks” on 1) compelling communication of the core ideas and outcomes of the project to an interdisciplinary audience; and 2) innovation and potential for impact of the energy application and data science methodology.

Congratulations to the 2020 winners: 
 

FIRST PRIZE ($500): Priya Donti, Ph.D. student in computer science and public policy at Carnegie Mellon University, for a talk titled, “Inverse Optimal Power Flow: Assessing the Vulnerability of Grid Data”
 

 

SECOND PRIZE ($250): Akansha Singh Bansal, Ph.D. student in electrical and computer engineering at the University of Massachusetts Amherst, for a talk titled, “See the Light: Modeling Solar Performance using Multispectral Satellite Data”


THIRD PRIZE ($100): Tongshu Zheng, Ph.D. student in environmental engineering (and Energy Data Analytics Ph.D. Student Fellow) at Duke University, for a talk titled, “Estimating Solar PV Soiling Using a Satellite-Based Remote Sensing Approach”

 

HONORABLE MENTIONS: 

McKenna Peplinski, Ph.D. student in environmental engineering at the University of Southern California, for a talk titled, “Predicting Changes in Southern California's Residential Electricity Consumption using Machine Learning Models.”


Noman Bashir, Ph.D. student in electrical and computer engineering at the University of Massachusetts Amherst, for a talk titled, “Solar-TK: A Data-driven Toolkit for Solar PV Performance Modeling and Forecasting"

View a YouTube playlist that includes all competition entries.

“While these students are still early in their careers, their research has high potential for impact,” remarked Duke University Energy Initiative director Dr. Brian Murray, who served as a judge. “They are using advanced methods in data science to develop fresh approaches to our world’s great energy challenges—and they are adept at communicating their research clearly and succinctly.”  

The two-day Energy Data Analytics Symposium (Dec. 8-9, 2020) organized by the Duke University Energy Data Analytics Lab focused on how machine learning and other data science innovations can help transform energy systems to become more accessible, affordable, reliable, and clean. The event, which featured insights from established experts and emerging scholars, was supported by a grant from the Alfred P. Sloan Foundation.

The Energy Data Analytics Lab, which organized the symposium and competition, is a collaboration among the Duke University Energy Initiative (which houses it), the Rhodes Information Initiative at Duke, and the Social Science Research Institute.

(Note: Conclusions reached or positions taken by researchers or other grantees represent the views of the grantees themselves and not those of the Alfred P. Sloan Foundation or its trustees, officers, or staff).

For more information about the competition, contact Trey Gowdy.

Add to Slider: 
Add To Slider

Pages

Subscribe to RSS - research