Federica Bragone

Ciao! I am a PhD student at the CST Division, and my research is in Physics-Informed Machine Learning applied to the circular reuse of power components. The objectives of the project are to estimate the state of component ageing when it is close to the end of its projected life and to reuse parts of older components that have only several preserved parts for the repair of other units. The focus is given mainly to the insulation system of power transformers, which is defined by insulating paper and insulation oil. The degree of polymerization (DP) is used to evaluate the cellulose degradation and other chemical factors accumulated in the insulation oil to monitor transformer insulation paper condition. My interest is in applying physics-based machine learning models, like Physics-Informed Neural Networks (PINNs), to overcome the lack of data in this area and to exploit the knowledge provided by the physical equations governing the degradation and the thermal behavior of the power components.

I obtained my BSc Degree in Mathematics at the University of Aberdeen and spent my second year in exchange at the Hong Kong University of Science and Technology (HKUST). Finally, I pursued my MSc Degree at KTH in Applied and Computational Mathematics.