Investigation of laser surface remelting supported by acoustic emission analysis and machine learning

Dunja Ravnikar (Author), Bor Mojškerc (Author), Roman Šturm (Author)

Abstract

In the present work, laser surface remelting (LSR) was carried out on C45 carbon steel using an Nd:YAG pulse laser. The effect of process parameters, such as different laser pulse durations and the absence or presence of a graphite absorber, on the microstructure, remelting depth, and microhardness was examined. In most cases, the graphite coating enhanced the laser energy absorption into the surface, resulting in greater depths of the remelted zone (RZ). RZ depth increased ranging from 8 to 350 pct, depending on the laser pulse duration. An increase in the surface microhardness by a factor of 2.6 was achieved in comparison with the substrate material microhardness, namely 559 HV 0.05 versus 211 HV 0.05. Concurrently, the LSR treatment parameters were also investigated using the in process generated acoustic emission (AE) signals. AE characteristics, such as AE peak amplitude, signal duration, count, and energy, were evaluated. A correlation of the AE characteristics was established for the various LSR treatment parameters. The LSR treatment classification results confirm the feasibility of using AE in combination with machine learning (ML) for monitoring LSR and the resulting surface properties of the hardened material.

Keywords

laser surface remelting;microstructure;acoustic emission;machine learning;

Data

Language:	English
Year of publishing:	2022
Typology:	1.01 - Original Scientific Article
Organization:	UL FS - Faculty of Mechanical Engineering
UDC:	620.1/.2:621.785:534
COBISS:	96993283
ISSN:	1073-5623
Views:	200
Downloads:	33
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	Slovenian
Secondary keywords:	lasersko površinsko pretaljevanje;mikrostruktura;akustična emisija;strojno učenje;
Type (COBISS):	Article
Embargo end date (OpenAIRE):	2023-02-04
Pages:	str. 837-849
Issue:	ǂVol. ǂ53
Chronology:	2022
DOI:	10.1007/s11661-021-06552-7
ID:	14519277