This project develops a data-driven decision support system for the pre-dictive maintenance of the electrical assets by integrating the information at component level and evaluating the impact on risk and resilience at network level. This project goes beyond the state-of-the-art because of (a) predictive maintenance algorithms for multiple grid assets, and (b) the holistic picture of the electric grid and its components. A close collabora-tion with Swissgrid facilitates access to relevant data and physical and statistical models in use. In addition, testing of the developed algorithms in Swissgrid’s environment ensures their applicability and relevance. The proposed tools has expected impacts on (a) science (predictive mainte-nance algorithms), (b) TSO (decreasing the maintenance expenses and the operational costs), (c) equipment manufacturer (business model) and (d) society (increase reliability of power supply).
