EUROPEAN CONGRESS AND EXHIBITION ON ADVANCED MATERIALS AND PROCESSES - EUROMAT 2021

Phase transformations and the partitioning of carbon play a key role for the evolution of the microstructures of high speed steels. In particular the microstructural contributions to the hot properties of high-speed steels (matrix strengthening and strengthening due to secondary hardening carbides) have to be revealed. The quenching process after austenitization of high speed steels is a key processing step investigated in this work. To this end in-situ X-ray diffraction experiments of the fast heating and cooling process are used to complement dilatometer tests. In-situ data, however, may exhibit a low signal-to-noise ratio in particular due to the limited time available for recording the diffractograms. Detection of fine microstructural changes is often complicated by noise in dilatometric data. Regularization terms - as they have been developed in the field of machine learning and as they are used in image processing - are applied to suppress the noise coming along with the experimental data. Real phenomena are distinguished from artifact e.g. false peak detection due to tube tails for X-ray diffraction analysis. In this case imperfect in-situ X-ray diffractograms can be advantageously analyzed by the Bayesian approach with a Markov Chain Monte Carlo (MCMC) algorithm. It is expected that these measuring and evaluation tools have future potential in steel characterization and design.