e-Energy Workshop 2023
International Workshop on Energy Data and Analytics
Orlando, Florida, United States
June 20, 2023
- Paper Registration and Submission: April 05, 2023
- Notification of Acceptance: May 12, 2023
- Final Manuscript Due: May 26, 2023
“Using Data and Machine Learning to Understand the Power Grid”
The transition towards renewable energy generation raises several questions about control, stability
and operation and therefore requires a solid understanding of existing and future power systems.
The power grid frequency is the central observable in power system control, as it measures the
balance of electrical supply and demand. Here, we use a data-driven approach, analysing the power
grid frequency to work towards a quantitative understanding of the power grid. We present an open
database with time series from various synchronous areas such as Continental Europe, Great Britain,
the Western and Texas Interconnection, as well as several European islands. We analyse the data
and highlight significant deviations from Gaussianity in several regions, scaling laws and spatio-temporal dynamics.
Furthermore, we utilize state-of-the-art machine learning approaches to forecast trajectories and identify risks and drivers for frequency stability.
Overall, we offer a model-free and data-centered perspective on understanding power systems.
Benjamin Schäfer is leading a Young Investigator Group at Karlsruhe Institute of Technology, Germany.
Before joining KIT, he worked as an associate professor at NMBU in Ås, Norway, as a Marie-Curie Fellow
at Queen Mary University of London and a postdoc at TU Dresden. He obtained his PhD in 2017 at University
of Göttingen, after conducting research at the Max Planck Institute for Dynamics and Self-Organization in Göttingen,
Queen Mary University of London and the University of Tokyo.
He is a theoretical scientist analyzing various complex systems with an emphasis on the energy transition,
sustainability and climate change. Method-wise, he combines data analysis, stochastic modelling and machine learning
to understand complex systems. In his research he stresses the need for transparency and openness,
e.g. via open data bases and the usage of interpretable (white-box) machine learning models.
- 13:00 - 13:40: "Using Data and Machine Learning to Understand the Power Grid", Benjamin Schäfer, Karlsruhe Institute of Technology
Session: Data Analytics and Forecasting
- 13:45 - 14:00: "Improving Building Energy Efficiency through Data Analysis", DiAndra Philipp (City College of New York), Jin Chen (Nearable Inc.), Fani Maksakuli (Nearable Inc.), Zhigang Zhu (City College of New York), Arber Ruci (Nearabl Inc.), E'edresha Sturdivant (Nearabl Inc.)
- 14:00 - 14:20: "A new Data-Driven Approach for Comparative Assessment of Baseline Load Profiles Supporting the Planning of Future Charging Infrastructure", Johannes Galenzowski, Simon Waczowicz, Veit Hagenmeyer (Karlsruhe Institute of Technology)
- 14:20 - 14:40: "Meta-regression analysis of errors in short-term electricity load forecasting", Konstantin Hopf, Hannah Hartstang, Thorsten Staake (University of Bamberg)
- 14:40 - 15:00: "The Impact of Forecast Characteristics on the Forecast Value for the Dispatchable Feeder", Dorina Werling, Maximilian Beichter, Benedikt Heidrich, Kaleb Phipps, Ralf Mikut, Veit Hagenmeyer (Karlsruhe Institute of Technology)
Session: Data Sets and Case Studies
- 15:05 - 15:20: "Realtime temperature-adjusted Natural Gas Savings of European private Households: A Study on the German Gas Market in 2022", Fabian Kächele, Oliver Grothe (Karlsruhe Institute of Technology)
- 15:20 - 15:40: "Towards closing the data gap: A project-driven distributed energy resource dataset for the U.S. Grid", Rabab Haider (MIT), Yixing Xu (Breakthrough Energy), Weiwei Yang (Microsoft Research)
- 15:40 - 16:00: "Datasheets for Energy Datasets: An Ethically-Minded Approach to Documentation", Ilana Heintz (Synoptic Engineering)
Scope and Topics
The design of future energy systems that are efficient, ecologically friendly, robust and scalable is a core
concern of our societies. Another very relevant development in recent years is the one towards a data-driven
perspective on system design. In the context of energy systems, a broad variety of data, often huge in
volume, is available. For instance, each smart meter is generating data streams, which often are recorded
and archived. On the other side, this is not the case for all aspects of energy systems, even though the
availability of data is crucial for the development of new methods. The questions how data describing energy
systems can be captured and processed, how its availability can be increased, and what can be learned from
it are fundamentally important. This last aspect includes predictions of various kinds of supply and demand,
predictive maintenance of energy infrastructures, the processing of energy-consumption data in a way that
respects the privacy of the individuals involved as well as business secrets etc.
This workshop is interdisciplinary in nature, i.e., brings together individuals interested in both data
management/data analytics and energy systems. Its objectives are the following ones:
- The workshop wants to draw attention to the fact that data-driven approaches often are possible and tend
to be promising when designing and operating energy systems.
- The workshop wants to give researchers in databases/KDD communities the opportunity to subject their
ideas, concepts and solutions to a critical perspective by experts for energy systems.
- The workshop wants to help bringing researchers on energy systems close to the state-of-the-art on what
data-oriented approaches can do for the design and operation of such systems. It wants to provide
support to individuals who want to broaden up methodologically.
- The workshop wants to serve as a networking platform, with an eye on funding opportunities in
- The workshop aims to expose researchers to a diverse audience eager to learn about novel data sets,
which relate to emerging research topics in particular.
The workshop solicits submissions on the following topics – all of them specific to energy data/energy
systems and their characteristics:
- New approaches and techniques to analyze energy data
- data reduction
- data science for energy data
- infrastructures for/techniques/best principles for the administration, management and archiving of
- data and measurements from real-world energy systems
- data from simulations of energy systems
- synthetic data generation
- data integration and data quality
- data privacy and anonymization
- modeling and representing energy-specific knowledge
On a methodological level, the workshop is open to any kind of submission:
- research papers
- vision papers
- comparative studies
- descriptors of energy data sets
- case studies and experience reports.
Two types of contributions are solicited:
- Full papers, up to 8 pages in 9-point ACM double-column format (i.e., excluding references) and
unlimited number of pages for appendices and references, single-blind.
- Short papers, up to 4 pages in 9-point ACM double-column format (i.e., excluding references) and
unlimited number of pages for appendices and references, single-blind.
The submission must be in PDF format and be formatted according to the official ACM Proceedings format.
Papers that do not meet the size and formatting requirements may not be reviewed. Word and LaTeX templates
are available at http://www.acm.org/publications/article-templates/proceedings-template.html.
The proceedings of the workshop will be published by ACM Digital Library along with the e-Energy conference
Submissions are made by HotCRP: https://eda23.hotcrp.com/
- Klemens Böhm, KIT, Germany
- Nicole Ludwig, University of Tübingen, Germany
- Andreas Reinhardt, TU Clausthal, Germany
- Aniket Chakrabarti, Microsoft, USA
- Bijay Neupane, Siemens Gamesa, Denmark
- Charlotte Debus, KIT, Germany
- Erik Buchmann, Universität Leipzig, Germany
- Jorge Ángel González-Ordiano, Universidad Iberoamericana Ciudad de México, Mexico
- Jorge Ortiz, Rutgers University, USA
- Mario Bergés, CMU, USA
- Marnie Shaw, ANU, Australia
- Martin Arlitt, Micro Focus, Canada
- Oliver Grothe, KIT, Karlsruhe
- Pandarasamy Arjunan, Berkeley Education Alliance for Research in Singapore, Singapore
- Philipp Staudt, University of Oldenburg, Germany
- Priya Donti, MIT, USA
- Stephen Haben, Energy Systems Catapult, UK
Please turn to Klemens Böhm (klemens dot boehm at kit dot edu)
for any questions or comments.