SKM 2023 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
MM: Fachverband Metall- und Materialphysik
MM 33: Topical Session: Fundamentals of Fracture – Amorphous Metals
MM 33.7: Talk
Wednesday, March 29, 2023, 17:30–17:45, SCH A 215
Machine Learning of fracture in glasses — Francesc Font-Clos1, Marco Zanchi1, Stefan Hiemer2, •Silvia Bonfanti3, Roberto Guerra1, Michael Zaiser2, and Stefano Zapperi1,4 — 1Center for Complexity and Biosystems, Department of Physics, University of Milan, via Celoria 16, 20133 Milan, Italy. — 2Institute of Materials Simulation, Department of Materials Science Science and Engineering, Friedrich-Alexander-University Erlangen-Nuremberg, Dr.-Mack-Str. 77, 90762 Fürth, Germany — 3NOMATEN Centre of Excellence, National Center for Nuclear Research, ul. A. Sołtana 7, 05-400 Swierk/Otwock, Poland — 4CNR Consiglio Nazionale delle Ricerche, Istituto di Chimica della Materia Condensata e di Tecnologie per l'Energia Via R. Cozzi 53, 20125 Milan, Italy.
Being able to predict the failure of materials based on structural information is a fundamental issue with enormous practical and industrial relevance for the monitoring of devices and components. Thanks to recent advances in deep learning, accurate fracture predictions are becoming possible even for strongly disordered solids, but the sheer number of parameters used in the process renders a physical interpretation of the results impossible. Here we address this issue and use machine learning methods to predict the failure of simulated two-dimensional silica glasses from their initial undeformed structure. We show that our predictions can be transferred to samples with different shapes or sizes than those used in training, as well as to experimental images.