Data-driven models for predictions of geometric characteristics of bead fabricated by selective laser melting

Thai Le-Hong, Pai Chen Lin, Jian-Zhong Chen, Thinh Duc Quy Pham and Xuan Tran ()
Additional contact information
Thai Le-Hong: Thu Dau Mot University
Pai Chen Lin: National Chung-Cheng University
Jian-Zhong Chen: National Chung-Cheng University
Thinh Duc Quy Pham: Thu Dau Mot University
Xuan Tran: Thu Dau Mot University

Journal of Intelligent Manufacturing, 2023, vol. 34, issue 3, No 16, 1257 pages

Abstract: Abstract In this paper, the effects of two key process parameters of the selective laser melting process, namely laser power and scanning speed, on the single-track morphologies and the bead characteristics, especially the depth-to-width D/W and height-to-width H/W ratios, were investigated using both experimental and Machine Learning (ML) approaches. A total of 840 single tracks were fabricated with several combinations of laser power and scanning speed levels. Surface morphologies of the single tracks and bead profiles were thoroughly investigated, providing a track-type map and the evolutions of the bead characteristics as a function of laser power and scanning speed. The results indicate neither severe balling nor keyholing effect for all combinations of laser power and scanning speed. Besides, simple relationships cannot accurately describe the evolutions of the D/W and H/W ratios as a function of laser power and scanning speed. Two Machine Learning-based regression models, Random Forest and Artificial Neural Network, were chosen to estimate the D/W and H/W ratios using laser power and scanning speed. The Bayesian optimization algorithm was employed to optimize the model hyperparameter selection. Both Machine Learning-based models appear to be able to predict reasonably well the two aspect ratios, D/W and H/W, with an overall R2 value reaching about 90%, evaluated on the cross-validation dataset after a few seconds of training time, respectively.

Keywords: Selective melting laser; Bead geometry; Single-track morphology; Machine learning; Artificial neural network; Bayesian optimization (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10845-021-01845-5 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:joinma:v:34:y:2023:i:3:d:10.1007_s10845-021-01845-5

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10845

DOI: 10.1007/s10845-021-01845-5

Access Statistics for this article

Journal of Intelligent Manufacturing is currently edited by Andrew Kusiak

More articles in Journal of Intelligent Manufacturing from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().