Publications

Numerous factors influence the prices of electricity and control power. We utilized the increasing amount of available data from energy systems to identify how certain generation types, load or prices influence power grid stability. Specifically, we focused on how to capture both determinsitic and probabilistic aspects of power systems.

Probabilistic and Explainable Machine Learning for Tabular Power Grid Data, Alexandra Nikoltchovska, Sebastian Pütz, Xiao Li, Veit Hagenmeyer, Benjamin Schäfer, E-Energy '25: Proceedings of the 16th ACM International Conference on Future and Sustainable Energy Systems

Paper

Constructing new lines in power grids may reduce the system's performance.  We proposed an approach for the prediction of edges lowering system performance and define potential constraints for grid extensions.

Understanding Braess’ Paradox in power grids, Benjamin Schäfer, Thiemo Pesch, Debsankha Manik, Julian Gollenstede, Guosong Lin, Hans-Peter Beck, Dirk Witthaut & Marc Timme, Nature Communications volume 13, Article number: 5396 (2022)
Paper Press release (in German) Spektrum (in German)

In modern power grids, knowing the required electric power demand and its variations is necessary to balance demand and supply. We developed a data-driven approach to extract the trend and characterise demand fluctuations. 

Data-driven load profiles and the dynamics of residential electricity consumption, Mehrnaz Anvari, Elisavet Proedrou, Benjamin Schäfer, Christian Beck, Holger Kantz & Marc Timme, Nature Communications volume 13, Article number: 4593 (2022) 

Paper News (in German)