| Autor: |
Zouhri, Wahb, Dantan, Jean-Yves, Häfner, Benjamin, Eschner, Niclas, Homri, Lazhar, Lanza, Gisela, Theile, Oliver, Schäfer, Martin |
| Zdroj: |
Procedia CIRP; 2020, Vol. 99, p319-324, 6p |
| Abstrakt: |
Additive manufacturing (AM), and particularly metal laser powder bed fusion (L-PBF) processes are rapidly being industrialized. Still, the current L-PBF process lacks both process quality and reproducibility. For that reason, robust process monitoring needs to be developed to reduce the process variation and ensure quality. Accordingly, to deal with this issue, this work proposes a new approach to predict the quality of L-PBF products. The approach consists of selecting relevant statistical features from optical data and validating these features by assessing their ability to predict the different products' density classes. The approach was applied on cubical specimens produced with different process parameters. Support vector machines (SVMs) were used as classification tools, and the first results are promising with a prediction accuracy higher than 90%." [ABSTRACT FROM AUTHOR] |
| Databáze: |
Supplemental Index |
| Externí odkaz: |
|
