Deep learning methods for acoustic emission evaluation

0549679 - ÚT 2022 RIV CZ eng C - Konferenční příspěvek (zahraniční konf.)
Kovanda, M. - Chlada, Milan
Deep learning methods for acoustic emission evaluation.
SPMS 2020/21 Stochastic and Physical Monitoring Systems. Praha: Czech Technical University in Prague, 2021, s. 111-118. ISBN 978-80-01-06922-6.
[SPMS 2020/21. Malá Skála (CZ), 24.06.2021-28.06.2021]
Institucionální podpora: RVO:61388998
Klíčová slova: acoustic emission * deep learning * machine learning * plastic deformation * time series classification
Obor OECD: Materials engineering

The goal of this paper is to show the possibilities of state-of-the-art deep learning methods for ultrasound signals evaluation. Several neural network architectures are applied to
acoustic emission signals measured during the tensile tests of metallic specimen to determine the beginning of plasticity in the material. Plastic deformation is accompanied by microscopic
events such as a slip of atomic plane dislocations which is hardly detectable by other methods. The potential of machine learning is demonstrated on two tensile tests where the material is
strained until it collapses. The examined networks proved well to reliably predict the risk of collapse together with changes in the ultrasound emission signals.
Trvalý link: http://hdl.handle.net/11104/0327457