Abstract
Laser directed energy deposition (L-DED) is an additive manufacturing (AM) technology that offers unique advantages for creating and repairing metallic parts. The quality of the final print is highly dependent on the consistency of the printing parameters, and as a result, the deposited bead geometry along the printing pathway. Inconsistencies, particularly in the bead width, can produce defects between adjacent beads, increasing the material susceptibility to failure. Therefore, it is crucial to monitor and predict the printing parameters in real time during the process. The objective of this study is to develop a machine learning model that accurately predicts the width of the deposited track in real time. To achieve this, a dataset with approximately 12,379 melt pool images was used to train the algorithm and predict the width of the deposited track. Each one of the deposited track was measured by a profilometer at 10 points along the pathway, and intermediate labels were predicted using a spline curve, making the model semi-supervised. To implement the machine learning model, pre-trained frozen networks based on convolutional neural networks (CNN) architectures VGG, ResNet, and DenseNet were employed, using transfer learning principles. These networks were integrated into a dense, fully connected network containing trainable parameters. The results demonstrate a good model fit, with a mean absolute error of 4.5% and a mean absolute error of 0.0358 mm. Moreover, the processing frequency of the best model, at 55 Hz, enables real-time control of the L-DED manufacturing process. It is important to highlight, however, that the VGG-based model shows a frequency of 250 Hz and similar results. Thus, accurately predicting the width of the deposited bead has the potential to significantly improve the quality of the final part by reducing overall defects, such as lack of fusion and porosity, whenever this data is used as input to a feedback control system. This is particularly valuable in industries where mechanical properties and fatigue life are critical, such as automotive, aerospace, and biomedical sectors. The models developed in this study offer a promising approach to effectively provide data to control the L-DED manufacturing process in real time, enabling the reliable production of high-quality components associated with reduced manufacturing costs.
Similar content being viewed by others
Availability of data and materials
The data collected has been made available to support future research and training of algorithms and is available at github.com/henriquenunez/nir-ded-dataset.
References
DebRoy T, Wei HL, Zuback JS, Mukherjee T, Elmer JW, Milewski JO, Beese AM, Wilson-Heid A, De A, Zhang W (2018) Additive manufacturing of metallic components - process, structure and properties. Progress in Materials Science 92:112–224. https://doi.org/10.1016/jpmatsci201710001
Armstrong M, Mehrabi H, Naveed N (2022) An overview of modern metal additive manufacturing technology. Journal of Manufacturing Processes 84:1001–1029. https://doi.org/10.1016/jjmapro202210060
Majeed A, Ahmed A, Lv J, Peng T, Muzamil M (2020) A state-of-the-art review on energy consumption and quality characteristics in metal additive manufacturing processes. Journal of the Brazilian Society of Mechanical Sciences and Engineering 42(5):249. https://doi.org/10.1007/s40430-020-02323-4
52910:2018(E) I (2018) Additive manufacturing – design – requirements, guidelines and recommendations 1st edn,ISO/ASTM International 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959 United States pp 1–15
Blakey-Milner B, Gradl P, Snedden G, Brooks M, Pitot J, Lopez E, Leary M, Berto F, du Plessis A (2021) Metal additive manufacturing in aerospace: a review. Materials & Design 209: 110008. https://doi.org/10.1016/jmatdes2021110008
Liu R, Wang Z, Sparks T, Liou F, Newkirk J (2017) Aerospace applications of laser additive manufacturing. In: Brandt M (ed) Laser Additive Manufacturing Woodhead Publishing Series in Electronic and Optical Materials, Woodhead Publishing, Cambridge, MA, USA, pp 351–371. https://doi.org/10.1016/B978-0-08-100433-300013-0
Agarwal R, Gupta V, Singh J (2022) Additive manufacturing-based design approaches and challenges for orthopaedic bone screws: a state-of-the-art review. J Braz Soc Mech Sci Eng 44(1): 37. https://doi.org/10.1007/s40430-021-03331-8
Modi YK, Sanadhya S (2018) Design and additive manufacturing of patient-specific cranial and pelvic bone implants from computed tomography data. J Braz Soc Mech Sci Eng 40(10): 503. https://doi.org/10.1007/s40430-018-1425-9
Zhang J, Wang C, Shareef N (2023) Microstructure and properties of Ti-Zr-Mo alloys fabricated by laser directed energy deposition. Materials (Basel) 16(3):1054
Tan ZE, Pang JHL, Kaminski J, Pepin H () Characterisation of porosity, density, and microstructure of directed energy deposited stainless steel aisi 316l. Additive Manuf 25: 286–296. https://doi.org/10.1016/jaddma201811014
Wang S, Ning J, Zhu L, Yang Z, Yan W, Dun Y, Xue P, Xu P, Bose S, Bandyopadhyay A (2022) Role of porosity defects in metal 3d printing: formation mechanisms, impacts on properties and mitigation strategies. Materials Today 59: 133–160. https://doi.org/10.1016/jmattod202208014
Ribeiro KSB, Núñez HHL, Venter GS, Doude HR, Coelho RT (2023) A hybrid machine learning model for in-process estimation of printing distance in laser directed energy deposition. Int J Adv Manuf Technol. https://doi.org/10.1007/s00170-023-11582-z
Wang H, Pfefferkorn FE, Wolff SJ (2022) Investigation of pore formation mechanisms induced by spherical-powder delivery in directed energy deposition using in situ high-speed x-ray imaging. Addit Manuf Lett 3: 100050. https://doi.org/10.1016/jaddlet2022100050
Ribeiro KSB, Núñez HHL, Jones JB, Coates P, Coelho RT (2021) A novel melt pool map** technique towards the online monitoring of directed energy deposition operations. Procedia Manuf 53: 576–584. https://doi.org/10.1016/jpromfg202106058. 49th SME North American Manufacturing Research Conference (NAMRC 49, 2021)
Bohlen A, Freisse H, Hunkel M, Vollertsen F (2018) Additive manufacturing of tool steel by laser metal deposition. Procedia CIRP 74: 192–195. https://doi.org/10.1016/jprocir201808092. 10th CIRP Conference on Photonic Technologies [LANE 2018]
Jamnikar ND, Liu S, Brice C, Zhang X (2022) In-process comprehensive prediction of bead geometry for laser wire-feed ded system using molten pool sensing data and multi-modality cnn. Int J Adv Manuf Technol 121(1): 903–917. https://doi.org/10.1007/s00170-022-09248-3
Sampson R, Lancaster R, Sutcliffe M, Carswell D, Hauser C, Barras J (2020) An improved methodology of melt pool monitoring of direct energy deposition processes. Opt Laser Technol 127: 106194. https://doi.org/10.1016/joptlastec2020106194
Jeon I, Yang L, Ryu K, Sohn H (2021) Online melt pool depth estimation during directed energy deposition using coaxial infrared camera, laser line scanner, and artificial neural network. Addit Manuf 47: 102295. https://doi.org/10.1016/jaddma2021102295
Cunha FG, Santos TG, Xavier J (2021) In situ monitoring of additive manufacturing using digital image correlation: a review. Materials 14(6): 1511. https://doi.org/10.3390/ma14061511
Bevans B, Ramalho A, Smoqi Z, Gaikwad A, Santos TG, Rao P, Oliveira JP (2023) Monitoring and flaw detection during wire-based directed energy deposition using in-situ acoustic sensing and wavelet graph signal analysis. Materials & Design 225: 111480. https://doi.org/10.1016/jmatdes2022111480
Chen L, Yao X, Tan C, He W, Su J, Weng F, Chew Y, Ng NPH, Moon SK (2023) In-situ crack and keyhole pore detection in laser directed energy deposition through acoustic signal and deep learning. Addit Manuf 69: 103547. https://doi.org/10.1016/jaddma2023103547
Gajbhiye RV, Rojas JGM, Waghmare PR, Qureshi AJ (2022) In situ image processing for process parameter-build quality dependency of plasma transferred arc additive manufacturing. Int J Adv Manuf Technol 119: 7557–7577. https://doi.org/10.1007/s00170-021-08643-6
Sestito GS, Venter GS, Ribeiro KSB, Rodrigues AR, da Silva MM (2022) In-process chatter detection in micro-milling using acoustic emission via machine learning classifiers. Int J Adv Manuf Technol 120(11): 7293–7303. https://doi.org/10.1007/s00170-022-09209-w
Ouidadi H, Guo S, Zamiela C, Bian L (2023) Real-time defect detection using online learning for laser metal deposition. J Manuf Process 99:898–910. https://doi.org/10.1016/jjmapro202305030
Ciccone F, Bacciaglia A, Ceruti A (2023) Optimization with artificial intelligence in additive manufacturing: a systematic review. J Braz Soc Mech Sci Eng 45(6): 303. https://doi.org/10.1007/s40430-023-04200-2
**ang S, Nie F, Zhang C (2010) Semi-supervised classification via local spline regression. IEEE Trans Pattern Anal Mach Intell 32: 2039–53. https://doi.org/10.1109/TPAMI201035
Ribeiro KSB, Mariani FE, Coelho RT (2020) A study of different deposition strategies in direct energy deposition (ded) processes. Procedia Manuf 48: 663–670. https://doi.org/10.1016/jpromfg202005158. 48th SME North American Manufacturing Research Conference, NAMRC 48
Haley JC, Zheng B, Bertoli US, Dupuy AD, Schoenung JM, Lavernia EJ (2019) Working distance passive stability in laser directed energy deposition additive manufacturing. Mater Des 161: 86–94. https://doi.org/10.1016/jmatdes201811021
Sun Z, Guo W, Li L (2020) In-process measurement of melt pool cross-sectional geometry and grain orientation in a laser directed energy deposition additive manufacturing process. Opt Laser Technol 129: 106280. https://doi.org/10.1016/joptlastec2020106280
Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. In: International conference on learning representations (ICLR). https://doi.org/10.48550/ar**v14091556
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), pp 770–778. https://doi.org/10.1109/CVPR201690
Huang G, Liu Z, Weinberger KQ (2017) Densely connected convolutional networks. Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), pp 4700–4708. https://doi.org/10.48550/ARXIV160806993
Deng J, Dong W, Socher R, Li L-J, Li K, Fei-Fei L, ImageNet: a large-scale hierarchical image database. https://doi.org/10.1109/CVPR20095206848
Acknowledgements
The authors are thankful to Tatiana Dias Santana, from Mahr do Brasil, for aiding this study with accurate measurements of the workpiece. This work was fully supported by grant #2016/11309 – 0, #2019/00343 – 1 and #2020/03119 – 2, São Paulo Research Foundation (FAPESP).
Funding
Grant #2016/11309–0 and #2019/00343–1, São Paulo Research Foundation (FAPESP).
Author information
Authors and Affiliations
Contributions
Victor H. Mochi contributed to the study methodology, software, investigation and analysis. Kandice S. B. Ribeiro contributed to the study conceptualization, methodology and experiment investigation. Henrique H. L. and Giuliana S. Venter aided the conceptualization and methodology for both the experiments and software. Data curation and writing were performed by Victor H. Mochi, Giuliana S. Venter and Kandice S. B. Ribeiro. Reginaldo T. Coelho contribution was related to the research resources. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Mochi, V.H., Núñez, H.H.L., Ribeiro, K.S.B. et al. Real-time prediction of deposited bead width in L-DED using semi-supervised transfer learning. Int J Adv Manuf Technol 129, 5643–5654 (2023). https://doi.org/10.1007/s00170-023-12658-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00170-023-12658-6