The activities described in this report focus on the application of data analysis models and real-time artificial intelligence techniques. The ultimate goal is to develop IT tools to support the real-time operation of power grids and monitoring of infrastructure malfunctions.

The first part of the report deals with the definition of a Digital Twin (DT) of the electricity distribution network. The two key aspects of developing a Digital Twin are identified and described: the choice and use of a defined data model and the identification of use cases to be solved.

Regarding the data model, the possible use of Knowledge Graph technologies for combining data from different systems is described. In particular, the semantic IEC CIM model is chosen as a reference as it is the standard for information exchanges in the electricity industry.

The functionalities of different Digital Twins developed by some industries active in the electricity sector are analyzed.

The final result is the identification of use cases to be considered for the development of a Digital Twin of an electricity distribution network.

Relating to components of energy networks, the specification is outlined for the implementation of a Digital Twin of a Storage System, installed in an industrial context and coupled with a photovoltaic system.

The second part of the report concerns the analysis and processing of images and video footage with artificial intelligence techniques to support the diagnostics of energy networks. In particular, the work concerns the characterization and monitoring of two phenomena that are detrimental to power system infrastructures: surface discharges that occur on insulator surfaces due to the passage of currents of leakage caused by the accumulation of pollutant deposits, and the formation of ice sleeves on high-voltage overhead lines due to wet-snow precipitation.