- Master of Science (M.Sc.) in electrical engineering at htw saar - University of Applied Sciences, Saarbrücken, Germany
- Research Associate at Fraunhofer Institute for Nondestructive Testing IZFP, Saarbrücken, Germany. Materials Characterization Department, Method Development Group.
Microstructure- and stress-independent materials characterization in reactor safety research
Despite of the German phase-out from nuclear energy, the highest safety requirements for the operation of nuclear power plants in Europe are still vital for all countries. The German ministry BMWi, as a funding body, therefore supports research to maintain German expertise in the field of reactor safety. Reactor safety research aims at the safe operation of nuclear power plants during their remaining design service life (and possibly beyond). In this respect, the non-destructive testing technique 3MA, developed at Fraunhofer IZFP, has already made a significant contribution to the understanding of different aging mechanisms of component materials and their characterization. The basis of 3MA is the fact that microstructure and mechanical stress determine both the mechanical and magnetic material behavior. The correlation between parameters of magnetic and mechanical material behavior enables the micromagnetic prediction of mechanical properties and stress, both of which can decisively influence the service life. A challenge, especially under practical conditions, lies in handling the mutually superimposed microstructural and stress-dependent influences. This superposition leads to ambiguities of the micromagnetic parameters, which in consequence, significantly impair the prediction quality. However, the practical application requires an economical solution to reduce these ambiguities. Within the scope of a BMWi-funded research project, the 3MA testing method has been extended, in particular for the relevant conditions in reactor safety, such as superimposed microstructure and stress influences. Investigations dealing with the extension of the feature extraction and machine learning methods have led to a more precise distinction between microstructural and stress-dependent influences.