Machine learning framework advances detection and identification of coordinated cyber-attacks on wind turbines: New study published in Sustainable Energy, Grids and Networks

A newly published study presents a robust machine learning-based framework that does more than detect cyber-attacks on wind turbines—it identifies their precise nature. As wind energy systems become increasingly digitalized, they face growing threats from stealthy, coordinated attacks capable of disrupting operations and triggering cascading failures. In such scenarios, early detection alone is no longer enough—accurate identification is essential for swift and targeted mitigation.
This research introduces a two-part framework: the first stage learns and monitors normal operational behavior to detect anomalies, while the second accurately classifies specific coordinated attack patterns.
Tested on a high-fidelity offshore wind turbine benchmark under realistic conditions, the method achieved over 96% detection accuracy within 10 seconds and maintained >99% identification accuracy for known attacks—while generalizing effectively to previously unseen scenarios. These results mark a significant step toward building wind energy infrastructure that is resilient by design.