The Application of Acoustic Emission and Artificial Neural Networks in an Analysis of Kinetics in the Phase Transformation of Tool Steel During Austempering

Abstract

AbstractDuring the course of the study it involved tool steel C105U was used. The steel was austempered at temperatures of 130°C, 160°C and 180°C respectively. Methods of acoustic emission (AE) were used to investigate the resulting effects associated with transformations and a large number of AE events were registered. Neural networks were applied to analyse these phenomena. In the tested signal, three groups of events were identified of: high, medium and low energy. The average spectral characteristics enabled the power of the signal spectrum to be determined. After completing the process, the results were compiled in the form of diagrams of the relationship of the AE incidence frequency as a function of time. Based on the results, it was found that in the austempering of tool steel, in the first stage of transformation midrib morphology is formed. Midrib is a twinned thin plate martensite. In the 2nd stage of transformation, the intensity of the generation of medium energy events indicates the occurrence of bainite initialised by martensite. The obtained graphic of AE characteristics of tool steel austempering allow conclusions to be drawn about the kinetics and the mechanism of this transformation.