The Use of Acoustic Emission and Neural Network in the Study of Phase Transformation below MS

Małgorzata Łazarska; Tadeusz Z Wozniak; Zbigniew Ranachowski; Andrzej Trafarski; Szymon Marciniak

doi:10.3390/ma14030551

The Use of Acoustic Emission and Neural Network in the Study of Phase Transformation below M_S

Materials (Basel). 2021 Jan 24;14(3):551. doi: 10.3390/ma14030551.

Authors

Małgorzata Łazarska^{1

2}, Tadeusz Z Wozniak², Zbigniew Ranachowski¹, Andrzej Trafarski², Szymon Marciniak³

Affiliations

¹ Institute of Fundamental Technological Research, Polish Academy of Sciences, Pawińskiego 5B, 02-106 Warszawa, Poland.
² Institute of Materials Engineering, Kazimierz Wielki University, Chodkiewicza 30, 85-064 Bydgoszcz, Poland.
³ Faculty of Materials Science and Engineering, Warsaw University of Technology, Wołoska 141, 02-507 Warszawa, Poland.

Abstract

Acoustic emission and dilatometry were applied to investigate the characteristics of phase transformations in bearing steel 100CrMnSi6-4 during austempering below the martensite start temperature (M_S 175 °C) at 150 °C. The aim of this study is to characterize the product of transformation occurring below the M_S temperature using various research methods. Analysis of the dilatometric curves shows that, after the formation of athermal martensite below the M_S temperature, the austenite continues to undergo isothermal transformation, indicating the formation of bainite. Additionally, tests were carried out with the use of acoustic emission during isothermal hardening of the adopted steel. The obtained acoustic emission signals were analyzed using an artificial neural network. The results, in the form of a graph of the frequency of acoustic emission (AE) event occurrence as a function of time, make it possible to infer about the bainite isothermal transformation. The results of this research may be used in the future to design optimal heat treatment methods and, consequently, may enable desired microstructure shaping.

Keywords: acoustic emission; austempering; bainite; dilatometry; neural networks.