Distributed Artificial Intelligence The resilient and optimized operation of fully renewable and digitalized energy systems requires new paradigms in system operations. Intelligent multi-agent systems with self-x properties are therefore the focus of the work of the Distributed Artificial Intelligence (DAI) group.
The complete conversion of our energy supply to renewable energies is a central challenge of our time and an indispensable contribution to climate protection and energy autonomy. To achieve this, not only must the generation and consumption of energy be reconciled on the market, but energy systems must also be operated reliably and efficiently. With increasing digitalization, the historically evolved energy supply structures are becoming a complex and dynamic cyber-physical energy system in which thousands of components interact with each other. The digital systems must be able to adapt to different operating conditions and compensate for faults in operation as independently as possible, while meeting the requirements and needs of human users in a trustworthy manner.
Vision: self-organising, self-optimising and self-healing cyber-physical energy systems
The DAI group is therefore working on enhancing the distributed components of a cyber-physical energy system with intelligence and autonomy and networking them with each other based on concepts of the so-called Organic Computing. In research and development projects, we investigate the possibilities and limits of self-organising, self-optimising and self-healing cyber-physical energy systems.
Our main research topics
In the context of these characteristics of digitalized systems, collectively referred to as "self-x properties," we address the following key issues:
Resilient Self-X-Systems
Resilience is a system's ability to quickly restore stable system operation and become more robust to disruptions in the long term.
Learning in Self-X-Systems
The ability of agents to learn is a basic prerequisite for adapting system behaviour to new or changing conditions.
Transparency and Acceptance of Self-X-Systems
The acceptance of autonomously deciding, digitalized systems also depends on their ability to explain decisions in a transparent and comprehensible way. Find out more..
Modelling and Agent-based Management of Flexibility
Flexibility is the basis of almost all decision-making processes in digitized energy systems, especially when it comes to the use of decentralized energy plants in the distribution network. Find out more..
Open Science – freely available and open scientific results
As a research group, we are committed to transparent science - Open Science, which makes research results freely and openly available. The artefacts, data and publications we produce should be able to be used collectively and without hurdles in order to improve the quality of research and contribute to progress in society. An essential aspect of this is also the possibility of reproducing research results. Further information..
Persons 
F
H
Dr. rer. nat. Stefanie Holly
E-Mail: stefanie.holly(at)offis.de, Phone: +49 441 9722-732, Room: Flx-E
K
O
R
S
T
Dr. rer. nat. Martin Tröschel
E-Mail: martin.troeschel(at)offis.de, Phone: +49 441 9722-150, Room: Flx-E
Projects
D
DEER
Dezentraler Redispatch (DEER): Schnittstellen für die Flexibilitätsbereitstellung
Duration: 2022 - 2025E
I
M
N
NFDI4Energy
National Research Data Infrastructure for the Interdisciplinary Energy System Research
Duration: 2023 - 2028R
REMARK
Resilienz im digitalisierten Stromsystem: Toolbox zur Bewertung von Systemdienstleistungsmärkten
Duration: 2022 - 2024T
W
WWNW
WärmewendeNordwest – Digitalisierung zur Umsetzung von Wärmewende- und Mehrwertanwendungen für Gebäude, Campus, Quartiere und Kommunen im Nordwesten
Duration: 2021 - 2026 Publications
2026
Agent-Based Flexibility Aggregation for a Distributed Redispatch
Radtke, Malin and Stark, Sanja and Holly, Stefanie; Energy Informatics; 2026
2025
A Comprehensive Toolbox for Evaluating Reactive Power Market Designs
Holly, Stefanie and Pechan, Anna and Schulte, Eike and Palovic, Martin and Buchmann, Marius; 2025 IEEE Kiel PowerTech; June / 2025
A multi-agent approach with verifiable and data-sovereign information flows for decentralizing redispatch in distributed energy systems
Heess, Paula and Holly, Stefanie and Körner, Marc-Fabian and Nieße, Astrid and Radtke, Malin and Schick, Leo and Stark, Sanja and Strüker, Jens and Zwede, Till; Energy Informatics; February / 2025
Amplify: Multi-purpose flexibility model to pool battery energy storage systems
Paul Hendrik Tiemann, Marvin Nebel-Wenner, Stefanie Holly, Emilie Frost and Astrid Nieße; Applied Energy; 2025
Communication Modeling Approaches in Energy System Applications: A Systematic Overview
Radtke, Malin and Frost, Emilie and Nieße, Astrid and Lehnhoff, Sebastian; IEEE Access; 2025
Effect of the Quantity and Penetration of Energy Resources on the Incentivization of Grid-Beneficial Behavior in Prosumer Households
Sager, Jens and Wegkamp, Carsten and Niesse, Astrid and Engel, Bernd; ETG Kongress 2025; Voller Energie-heute und morgen.; 2025
Online Meta-Modeling of Communication Network Simulations for Agent-Based Energy Systems
Radtke, Malin and Lehnhoff, Sebastian; 2025 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm); 2025
Simulative Analysis of Multi-Agent Systems in Energy Systems: Impact of Communication Networks
Radtke, Malin and Frost, Emilie; Proceedings of the 17th International Conference on Agents and Artificial Intelligence (ICAART 2025) - Volume 1; 01 / 2025
Ten Recommendations for Engineering Research Software in Energy Research
Ferenz, Stephan and Frost, Emilie and Schrage, Rico and Wolgast, Thomas and Beyers, Inga and Karras, Oliver and Werth, Oliver and Nieße, Astrid; Proceedings of the 16th ACM International Conference on Future and Sustainable Energy Systems; June / 2025
Understanding the Diversity of Communication Scenarios in Cyber-Physical Energy Systems
Malin Radtke and Sebastian Lehnhoff; ACM SIGENERGY Energy Informatics Review; September / 2025
