Industrial networks are the technological backbone of many critical infrastructures, as they include all the functional components that enable proper monitoring and control of facilities and assets, such as power/energy systems, particularly those that use renewable energy sources (RES), which are becoming increasingly widespread thanks to their sustainability and energy independence benefits. Such widespread adoption makes RES a potentially lucrative target for cyberattacks, due to the potential catastrophic consequences such as: loss of energy production (and related revenues), permanent damage to assets and infrastructure, leakage of commercial information and damage to reputation, regulatory non-compliance and fines, and finally (for interconnected/dependent critical infrastructure) risks to health, safety, and the environment.  
Solutions that improve the resilience of the electricity system and the RES included therein represent a strategic advantage for economic and social security.

The proposed prototype, which uses AI/ML-based tools/functionalities, offers innovative anomaly detection strategies based on the observation of the physical behavior of the monitored industrial process. AI/ML-based algorithms extract certain measurements of the system’s physical parameters and process them using a neural network architecture to build a classifier that makes automatic decisions about the system’s behavior and detects faults and potential cyberattacks (e.g., man-in-the-middle, spoofing, etc.).
Areas of application and users include Essential Services Operators (ESO) sector, with a focus on electricity generation and renewable energy networks.