Data-Driven Prediction and Uncertainty Quantification of Process Parameters for Directed Energy Deposition
Materials · 24.11.2023 · 16 (23) · 7308
DOI: 10.3390/ma16237308 ↗Kurz gefasst
Fachartikel in der Zeitschrift Materials, online erschienen am 24.11.2023. Verfasst von Florian Hermann und 5 weiteren Autor:innen. Beteiligte Einrichtung: Graduate School of Excellence Advanced Manufacturing Engineering (GSaME), University of Stuttgart, Nobelstraße 12, 70569 Stuttgart, Germany. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Federal Ministry of Education and Research, Landesministerium für Wissenschaft, Forschung und Kunst Baden-Württemberg, Deutsche Forschungsgemeinschaft.
Abstract (Originalsprache)
Laser-based directed energy deposition using metal powder (DED-LB/M) offers great potential for a flexible production mainly defined by software. To exploit this potential, knowledge of the process parameters required to achieve a specific track geometry is essential. Existing analytical, numerical, and machine-learning approaches, however, are not yet able to predict the process parameters in a satisfactory way. A trial-&-error approach is therefore usually applied to find the best process parameters. This paper presents a novel user-centric decision-making workflow, in which several combinations of process parameters that are most likely to yield the desired track geometry are proposed to the user. For this purpose, a Gaussian Process Regression (GPR) model, which has the advantage of including uncertainty quantification (UQ), was trained with experimental data to predict the geometry of single DED tracks based on the process parameters. The inherent UQ of the GPR together with the expert knowledge of the user can subsequently be leveraged for the inverse question of finding the best sets of process parameters by minimizing the expected squared deviation between target and actual track geometry. The GPR was trained and validated with a total of 379 cross sections of single tracks and the benefit of the workflow is demonstrated by two exemplary use cases.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Graduate School of Excellence Advanced Manufacturing Engineering (GSaME), University of Stuttgart, Nobelstraße 12, 70569 Stuttgart, Germany.
- TRUMPF Laser- und Systemtechnik GmbH, Johann-Maus-Straße 2, 71254 Ditzingen, Germany.
- Institut für Strahlwerkzeuge (IFSW), University of Stuttgart, Pfaffenwaldring 43, 70569 Stuttgart, Germany.
- Institut für Wasser- und Umweltsystemmodellierung (IWS), University of Stuttgart, Pfaffenwaldring 5a, 70569 Stuttgart, Germany.
ORCID-Profile: Florian Hermann, Peter Reimann, Thomas Graf
Förderung
- Federal Ministry of Education and Research
- Landesministerium für Wissenschaft, Forschung und Kunst Baden-Württemberg
- Deutsche Forschungsgemeinschaft
- German Federal Ministry of Education and Research (BMBF)
Zitieren
APA
Hermann, F., Michalowski, A., Brünnette, T., Reimann, P., Vogt, S., & Graf, T. (2023). Data-Driven Prediction and Uncertainty Quantification of Process Parameters for Directed Energy Deposition. Materials, 16(23), 7308. https://doi.org/10.3390/ma16237308
Vancouver
Hermann F, Michalowski A, Brünnette T, Reimann P, Vogt S, Graf T. Data-Driven Prediction and Uncertainty Quantification of Process Parameters for Directed Energy Deposition. Materials. 2023;16(23):7308. doi:10.3390/ma16237308
BibTeX
@article{Hermann2023,
author = {Hermann, Florian and Michalowski, Andreas and Brünnette, Tim and Reimann, Peter and Vogt, Sabrina and Graf, Thomas},
title = {Data-Driven Prediction and Uncertainty Quantification of Process Parameters for Directed Energy Deposition},
journal = {Materials},
year = {2023},
volume = {16},
number = {23},
pages = {7308},
doi = {10.3390/ma16237308}
}
EUVIMED