SKM 2023 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
MM: Fachverband Metall- und Materialphysik
MM 12: Poster I
MM 12.37: Poster
Montag, 27. März 2023, 18:15–20:00, P2/OG1+2
An in situ crack detection approach in additive manufacturing based on acoustic emission and machine learning — •Viktoriia Nikonova1, Denys Kononenko1, Dmitry Chernyavsky1, Mikhail Seleznev2, and Jeroen van den Brink1,3 — 1Institute for Theoretical Solid State Physics, IFW Dresden, 01069 Dresden, Germany — 2Institute of Materials Engineering, Technische Universitaet Bergakademie Freiberg, Gustav-Zeuner-Straße 5, D-09599, Freiberg, Germany — 3Institute for Theoretical Physics, TU Dresden, 01069 Dresden, Germany
Laser Powder Bed Fusion (LPBF) is a state-of-the-art solution for producing metal elements with complex shapes in various industries, from automotive to aerospace. One of the crucial practical drawbacks of LPBF is the presence of structural defects in the printed part, especially cracks. Here we propose an in situ monitoring system of cracks formation during the LPBF process utilizing acoustic emission and machine learning. We demonstrate that the representation of acoustic emission events in the space of principal components (PC) of the spectra yields a robust differentiation of crack events from the noise bursts. The ML classification algorithms achieve high accuracy of ~99 % for the PC-based descriptors. The presented approach advances the method of in situ quality control systems and brings it closer to practical implementations.