PhD Defense | Hussein Zahran | Prediction of Fatigue Behaviour in Fusion and Generation IV Reactor Materials: Numerical, Experimental, and Machine Learning Study

Prediction of Fatigue Behaviour in Fusion and Generation IV Reactor Materials: Numerical, Experimental, and Machine Learning Study

This dissertation investigates the fatigue behaviour of Reduced Activation Ferritic/Martensitic (RAFM) steels and CuCrZr alloys under conditions relevant to nuclear fusion reactors, with emphasis on DEMO structural components. The research combines physics-based viscoplasticity modelling, finite element implementation, and machine learning (ML) techniques to predict fatigue life and cyclic softening for both irradiated and non-irradiated materials. Experimental low-cycle fatigue tests informed the development of an enhanced constitutive model, implemented in a finite element framework for accurate simulation of cyclic deformation. A comprehensive fatigue database supported the evaluation of empirical life prediction methods and the development of data-driven models. ML approaches were employed to predict fatigue life using minimal input features, while deep learning strategies, including physics-informed neural networks, were applied to model cyclic softening behaviour under limited data availability. The integration of these methods provides scalable and efficient tools for fatigue assessment, reducing reliance on costly irradiation experiments. The outcomes contribute to improving the structural integrity and lifetime assessment of fusion reactor components and support the broader goal of achieving commercial fusion energy.

Promoters:

• Abdel Wahab Magd (UGENT)

SCK CEN mentors:

• Aleksandr Zinovev
• Dmitry Terentyev