3D Printing Monitoring System Based on Supervised Machine Learning

March 31, 2023  |  Vol.9, No.3  |  PP. 1-10  | PDF

AUTHORS:

Ho Jong Song, Department of Industrial Engineering, Kumoh National Institute of Technology, Republic of Korea

Tae Sung Kim, Department of Industrial Engineering, Kumoh National Institute of Technology, Republic of Korea

Goo Sang Jung, Conception Co., Ltd., Republic of Korea

Jin Ung Jeon, Conception Co., Ltd., Republic of Korea

KEYWORDS:

Tensile Strength, Overlap Rate, Scan Speed, Laser Power

Abstract

One of the additive manufacturing processes that is quickly evolving is metal 3D printing. Selective laser melting (SLM) is one of the popular methods for metal 3D printing. Although it has many advantages and capabilities, there are still unsolved problems like process monitoring and printing reliability. Printing high-quality robust products require appropriate parameter settings and real-time monitoring. In order to satisfy the requirement, we propose a new monitoring system based on multiple sensors that can measure the index of different quality affecting parameters of SLM 3D printing. The system serves to improve printing quality; it involves supervised machine learning to predict the expected tensile strength of the printed product. We trained the machine learning model on our new “tensile strength” dataset which includes multiple sensing data and indexes of tensile strength. While collecting data we printed products that have a tensile strength between 449 and 506 MPa. A number of SLM 3D printing tests are carried out to show the viability of the proposed approach. After testing the tensile strength of the printed product, test results were compared to the results of the tensile strength predicting model. According to experiments, the monitoring system showed satisfactory results predicting expected tensile strength. The highest accuracy has been achieved with Multiple Linear Regression, recording 97%. The monitoring system helps not only to predict the tensile strength of the printing product but also to find optimal parameter settings of the SLM printer.

Menu

APJCRI

    The purpose of this journal is to share and exchange the results of recent research by Asia-Pacific national researchers, and the purpose of the project is to publish the results of convergence research with the humanities and social sciences, including advanced research in specific fields.

References:

[1] G. M. Shashi, Md A. R. Laskar, H. Biswas, A. K. Saha, A Brief Review of Additive Manufacturing with Applications, Proceedings of 14th Global Engineering and Technology Conference, (2017).
DOI: https://doi.org/10.6084/m9.figshare.12520667
[2] J. Qin, F. Hu, Y. Liu, Research and application of machine learning for additive manufacturing, Additive manufacturing, (2022), Vol.52, 102691.
DOI: https://doi.org/10.1016/j.addma.2022.102691
[3] D. Ratna, Recent Advances and Applications of Thermoset Resins, Elsevier, (2022).
DOI: https://doi.org/10.1016/C2020-0-02814-8
[4] E. Yasa, Selective laser melting: principles and surface quality, Additive manufacturing, (2021), pp.77-120.
DOI: https://doi.org/10.1016/B978-0-12-818411-0.00017-3
[5] C. Y. Yap, C. K. Chua, Z. L. Dong, Z. H. Liu, Review of selective laser melting, Applied physics Review, (2015), Vol.2, No.4.
DOI: https://doi.org/10.1063/1.4935926
[6] M. Amini, Process Monitoring of 3D Metal Printing in Industrial Scale, 13th International manufacturing science and engineering conference, ASME, pp.1-8, (2018).
DOI: https://doi.org/10.1115/MSEC2018-6332
[7] https://www.slm-solutions.com/fileadmin/Content/Machines/SLM_R_500_Web.pdf, (2015).
[8] P. Yadav, O. Rigo, In Situ Monitoring Systems of The SLM Process: On the Need to Develop Machine Learning Models for Data Processing, Crystal, (2020), Vol.10, No.6, pp.1-26.
DOI: https://doi.org/10.3390/cryst10060524
[9] C. Tan, K. Zhou, W. Ma, Selective laser melting of high-performance pure tungsten: parameter design, densification behavior and mechanical properties, Science and Technology of Advanced Materials (2018), Vol.19, No.1, pp.370-380.
DOI: https://doi.org/10.1080/14686996.2018.1455154
[10] A. H. Maamoun, Y. F. Xue, M. A. Elbestawi, S. C. Veldhuis, The Effect of Selective Laser Melting Process Parameters on the Microstructure and Mechanical Properties of Al6061 and AlSi10Mg Alloys, Materials, (2019), Vol.12, No.1, pp.1-24.
DOI: https://doi.org/10.3390/ma12010012
[11] D. Chicco, J. M. Warren, The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation, PeerJ Computer Science, (2021).
DOI: https://doi.org/10.7717/peerj-cs.623
[12] S Nakagawa, P. Johnson, H. Schielzeth, The coefficient of determination R2 and intra-class correlation coefficient from generalized linear mixed-effects models revisited and expanded, Journal of the Royal Society Interface, (2017), Vol.14, No.134.
DOI: https://doi.org/10.1098/rsif.2017.0213

Citations:

APA:
Song, H. J., Kim, T. S., Jung, G. S., Jeon, J. U. (2023). 3D Printing Monitoring System Based on Supervised Machine Learning. Asia-pacific Journal of Convergent Research Interchange (APJCRI), ISSN: 2508-9080 (Print); 2671-5325 (Online), KCTRS, 9(3), 1-10. doi: 10.47116/apjcri.2023.03.01.

MLA:
Song, Ho Jong, et al. “3D Printing Monitoring System Based on Supervised Machine Learning.” Asia-pacific Journal of Convergent Research Interchange, ISSN: 2508-9080 (Print); 2671-5325 (Online), KCTRS, vol. 9, no. 3, 2023, pp. 1-10. APJCRI, http://apjcriweb.org/content/vol9no3/1.html.

IEEE:
[1] H. J. Song, T. S. Kim, G. S. Jung, J. U. Jeon, “3D Printing Monitoring System Based on Supervised Machine Learning.” Asia-pacific Journal of Convergent Research Interchange (APJCRI), ISSN: 2508-9080 (Print); 2671-5325 (Online), KCTRS, vol. 9, no. 3, pp. 1-10, March 2023.


Information

Guidelines

Article

Copyright © 2023 APJCRI. All Rights Reserved