journal article·2022-08-18

Quality monitoring in additive manufacturing using emission spectroscopy and unsupervised deep learning

Wenjing Ren, Guangrui Wen, Zhifen Zhang, Jyoti Mazumder

Publisher page

Abstract

Widespread adoption of additive manufacturing (AM) is hindered by the low quality and reproducibility of printed parts. In-situ quality monitoring technologies provide early defect detection and correction capability for AM processes and are crucial for part quality assurance. Research on in-situ monitoring and control of AM is exceptionally challenging due to the numerous variations and complex reactions in AM. This study has developed a novel unsupervised recognition model for the quality of parts printed by directed energy deposition (DED). This model consists of a long short-term memory-based autoencoder (LSTM-Autoencoder) and a K-means clustering. The LSTM-Autoencoder automatically extracts features from spectra collected during the DED process, and the K-means clustering model is employed for the deposition quality classification. Experimental verifications are conducted on the quality recognition of Al7075 alloy depositions printed under different conditions by varying the laser power, printing speed, and powder delivery rate. The results show that the proposed method can discern between the unqualified depositions with rough surfaces and high porosity from the qualified depositions correctly. This proposed LSTM-Autoencoder is data-driven and thus may be applied to printing conditions not tested within this work.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Additive Manufacturing Materials and Processes · Additive Manufacturing and 3D Printing Technologies · Industrial Vision Systems and Defect Detection

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Department of Mechanical Engineering, Xi’an Jiaotong University, Xi’an, China

Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI, USA

Funding

NIST

References (28)

Show source-provided references
1. Brandt M. · 2016
2. 10.1088/1361-6501/aa5c4f10.1088/1361-6501/aa5c4f ↗
3. 10.1016/j.addma.2015.07.00210.1016/j.addma.2015.07.002 ↗
4. 10.1016/j.procir.2015.01.00910.1016/j.procir.2015.01.009 ↗
5. 10.1016/j.phpro.2014.08.03810.1016/j.phpro.2014.08.038 ↗
6. 10.1115/1.402854010.1115/1.4028540 ↗
7. 10.1016/j.addma.2018.10.02010.1016/j.addma.2018.10.020 ↗
8. Kruth, J.P.; Mercelis, P. Procedure and apparatus for in-situ monitoring and feedback control of selective laser powder processing. US Patents: US20090206065A1, June 20, 2009.
9. 10.1016/j.addma.2018.11.01010.1016/j.addma.2018.11.010 ↗
10. 10.1016/j.addma.2018.08.01410.1016/j.addma.2018.08.014 ↗
11. 10.1038/s41467-018-03734-710.1038/s41467-018-03734-7 ↗
12. 10.1016/j.addma.2017.11.01210.1016/j.addma.2017.11.012 ↗
13. 10.1038/s41598-019-56847-410.1038/s41598-019-56847-4 ↗
14. 10.1088/0963-0252/15/4/S01.1410.1088/0963-0252/15/4/S01.14 ↗
15. Griem H. R. · 1961
16. 10.1109/tie.2016.260831810.1109/tie.2016.2608318 ↗
17. 10.1117/12.90826410.1117/12.908264 ↗
18. 10.1109/tie.2016.260831810.1109/tie.2016.2608318 ↗
19. Mazumder, J.; Song, L. Smart Additive Manufacturing System (Sams). U.S. Patent: 9981341, May 29, 2018.
20. 10.1016/j.jmapro.2020.02.01610.1016/j.jmapro.2020.02.016 ↗
21. 10.1016/j.addma.2018.03.01710.1016/j.addma.2018.03.017 ↗
22. 10.1080/24725854.2019.165952510.1080/24725854.2019.1659525 ↗
23. Arthur D. · 2006
24. Goodfellow I. · 2016
25. Graves, A.; Supervised sequence labelling with recurrent neural networks. Ph.D. Dissertation, Technical University of Munich, Germany, 2008.
26. Scikit-learn developers. Tuning the hyper-parameters of an estimator. https://scikitlearn.org/stable/modules/grid_search.html (accessed Sept 1, 2020).
27. Ralchenko Y. · 2008
28. Malika C. · J. Stat. Softw · 2014 · 61 · 1