Artificial Intelligence & Machine Learning , Endpoint Security , Governance & Risk Management
New AI System Shields Virtual Power Plants From Cyberattacks
German Government-Funded Detection System Stops Attacks at Individual Facilities
A new artificial intelligence-based protection system developed by the German government-funded SecDER project is revolutionizing the security of virtual power plants by detecting cyberattacks and predicting failures.
See Also: The SIEM Selection Roadmap: Five Features That Define Next-Gen Cybersecurity
The system uses AI and machine learning to monitor communication data between decentralized energy facilities, eliminating the need for detailed knowledge of specific energy installations, according to an announcement by Fraunhofer SIT Institute Darmstadt.
The SecDER project is a research initiative focused on developing an AI-based protection system for virtual power plants. It has received funding from the Federal Ministry for Economic Affairs and Climate Action and includes collaboration among several institutions, including Fraunhofer Institute, Hanover University of Applied Sciences, DECOIT GmbH, ENERTRAG AG and ANE GmbH and Co. KG.
Virtual power plants, which aggregate and manage energy flows from decentralized sources such as wind turbines, photovoltaic systems and hydroelectric power plants, function similarly to large power plants to meet the required pool size for electricity market demand.
Researchers in the SecDER project initially assessed the security of virtual power plants and simulated cyberattacks on a model plant. Their findings indicated that even successful attacks on individual facilities often go unnoticed by operators.
Conventional monitoring systems typically do not react to failures in single units.
To address this, the project consortium developed an intrusion detection system that uses machine learning to identify cyberattacks and technical disruptions. This system dynamically adjusts the entire setup into a secure state, preventing the execution of any unsafe control actions.
The system responds precisely to different threat scenarios, including fires and denial-of-service attacks, to ensure reliable electricity generation even during ongoing disruptions.
The SecDER system relies on general communication data shared by facilities with their virtual power plant rather than specific network and system data from individual facilities. This approach makes the solution manufacturer-independent and effective across various proprietary technologies, network architectures and protocols.
The project, which began in April 2021 and lasted 36 months, received 2.7 million euros in funding from the Federal Ministry for Economic Affairs and Climate Action and support from Project Management Jülich.
Project leader Tobias Schellien emphasized that while cyberattacks on energy systems are inevitable, the SecDER project has equipped systems to react in ways that prevent total failure.
