Digital twin-based hybrid framework for steam generator clogging prognostics

We present a hybrid framework to support prognostics of the clogging degradation phenomenon in tube support plates for digital twins of steam generators in pressurized water reactors. The proposed approach combines a physics-based simulation code, heterogeneous and sparse observational data, and several uncertainty quantification techniques to obtain a robust estimate of the steam generator remaining useful life associated with the clogging rate. The proposed framework is compatible with a digital twin platform to assist maintenance planning of EDF steam generators.