SwarmGrid

 

Safe operation of electric grids with swarm-enabled system services

In the context of the energy transition in Germany, the project SwarmGrid is concerned with the important question of how the stability of future energy grids can be assured. In the future, the share of decentralized energy conversion units will strongly increase whereas the amount of large power stations will decrease. Today, the energy grid is mainly stabilized by large power stations. That is why new strategies for the future have to be developed incorporating decentralized generators, for example wind turbines or photovoltaic units, into the stabilization of the energy grid. 

 

In the course of the project, ACS develops a simulation tool that enables the simulation of distribution grids with a large amount of distributed energy resources (DER). Every DER within the distribution grid is simulated as a “swarm particle”. Swarm particles organize themselves in swarms through communication and their inherent decision making behavior. Thereby, system services such as voltage or frequency control are provided to support the stability of the energy grid.

The simulation of DER swarms requires a mapping of their characteristics related to power engineering as well as communication and decision making. Classical simulation tools reach their limits when it comes to the simulation of such large and complex cyber-physical systems. In order to control the complexity of swarm simulation and to investigate the scalability of swarm-based solution approaches, techniques and methods from high performance computing are applied. A test simulation system incorporating several computation nodes is installed in ACS laboratory to enable large-scale swarm simulations.

Grateful acknowledgement is made to BMBF (German Federal Ministry of Education and Research) for providing financial support (promotional reference 03EK3568A). SwarmGrid is a joint research project among the RWTH Aachen institutes ACS, Institute for High Voltage Technology (IFHT), and Institute for Theoretical Information Technology (TI) and the industry partners QSC AG, Rheinische NETZGesellschaft mbH and the associated partner Amprion GmbH.

Final project report (only available in German)