雷射粉床熔融等積層製造技術已經徹底改變了生物醫學到航太或汽車等廣泛的產業。然而，製造過程中最大的挑戰之一為控制與優化工件的機械性能與孔隙形成；由於系統的不穩定性與雷射粉床熔融的隨機性，影響零件孔隙率的主要因素為加工參數（如雷射參數、粉層厚度、前一層的表面粗糙度）的變化。控制製造零件品質最有效方法之一是通過監測系統和機器學習技術互相結合的缺陷檢測系統建構。
本研究提出了一種基於人工智慧的方法，利用熔池形狀特徵和噴濺物角度特徵來預測雷射粉床熔融過程中孔隙的形成。監測模組的核心組件是高速及高解析度的CMOS 相機和一套用於熔池形狀和噴濺物角度為特徵檢測的統計分析之機器學習計算法。對於孔隙預測模組，採用自我組織映射 (SOM) 方法來探索熔池形狀、噴濺物角度特徵和孔隙形成可能性之間的相關性；所提出系統的最終預測模型可為工件內部孔隙的位置和大小做出準確預測。使用CT孔隙結果可用來評估預測模型結果，以確定所提出方法的準確性；結果顯示，孔隙位置預測準確率可達79.17 % 及孔隙體積預測準確率可達97.42 %。據我們所知，這是該領域首次嘗試在整合的孔隙度預測模型框架中，同時使用熔池形狀和噴濺物角度為特徵的機器學習研究。

Additive manufacturing technology such as Laser Powder Bed Fusion (L-PBF) process has been revolutionizing a wide range of industries from biomedical to aerospace or automotive ones. However, one of the biggest challenges in the process is to control and optimize the mechanical properties and porosity formation of the built parts during the fabrication. The primary factors that influence the porosity of the parts are the variations of processing parameters (the laser parameters, the powder layer thickness, the surface roughness of the previous layer) due to system instability and the stochastic nature of L-PBF process. One of the most effective methods to control the quality of the fabricated part is the defect detection via a combination of an in-situ monitoring system and machine learning techniques.
This study proposed an Artificial Intelligent-based approach utilizing the melt-pool shape and spatter angle features to predict the formation of pores during L-PBF process. The core component of the monitoring module is the high-speed and high-resolution CMOS camera and a set of in-house algorithms for melt-pool shape and spatter angle detections. For the pore-prediction module, self-organizing map (SOM) method was deployed to explore the correlation between melt-pool shape and spatter angle features and the possibility of pore formation. The final predicted outputs of the proposed system are the location and size of the porosity inside the built part. The prediction results are assessed using micro-CT results to determine the accuracy of the proposed method. As a result, the accuracy in pore position predication is 79.17 % and that in pore volume predication is 97.42%. As our best knowledge, this is the first attempt in the field to use both the melt-pool shape and spatter angle features in one unified framework for porosity prediction.

[1] J. C. Heigel and B. M. Lane, "Measurement of the melt pool length during single scan tracks in a commercial laser powder bed fusion process," Journal of Manufacturing Science and Engineering, vol. 140, no. 5, 2018.
[2] B. Lane et al., "Transient Laser Energy Absorption, Co-axial Melt Pool Monitoring, and Relationship to Melt Pool Morphology," Additive Manufacturing, vol. 36, p. 101504, 2020.
[3] Y. Zhang, C. Zhang, L. Tan, and S. Li, "Coaxial monitoring of the fibre laser lap welding of Zn-coated steel sheets using an auxiliary illuminant," Optics & Laser Technology, vol. 50, pp. 167-175, 2013.
[4] M. Luo and Y. C. Shin, "Vision-based weld pool boundary extraction and width measurement during keyhole fiber laser welding," Optics and Lasers in Engineering, vol. 64, pp. 59-70, 2015.
[5] L. Caprio, A. G. Demir, and B. Previtali, "Nonintrusive estimation of subsurface geometrical attributes of the melt pool through the sensing of surface oscillations in laser powder bed fusion," Journal of Laser Applications, vol. 33, no. 1, p. 012035, 2021.
[6] T. Craeghs, S. Clijsters, E. Yasa, F. Bechmann, S. Berumen, and J.-P. Kruth, "Determination of geometrical factors in Layerwise Laser Melting using optical process monitoring," Optics and Lasers in Engineering, vol. 49, no. 12, pp. 1440-1446, 2011.
[7] S. Li et al., "Melt-pool motion, temperature variation and dendritic morphology of Inconel 718 during pulsed-and continuous-wave laser additive manufacturing: A comparative study," Materials & design, vol. 119, pp. 351-360, 2017.
[8] Z. A. Young et al., "Types of spatter and their features and formation mechanisms in laser powder bed fusion additive manufacturing process," Additive Manufacturing, vol. 36, p. 101438, 2020, doi: 10.1016/j.addma.2020.101438.
[9] D. Wang, W. Dou, Y. Ou, Y. Yang, C. Tan, and Y. Zhang, "Characteristics of droplet spatter behavior and process-correlated mapping model in laser powder bed fusion," Journal of Materials Research and Technology, vol. 12, pp. 1051-1064, 2021, doi: 10.1016/j.jmrt.2021.02.043.
[10] J. Yin et al., "Correlation between forming quality and spatter dynamics in laser powder bed fusion," Additive Manufacturing, vol. 31, 2020, doi: 10.1016/j.addma.2019.100958.
[11] A. Stefaniak et al., "Insights into emissions and exposures from use of industrial-scale additive manufacturing machines," Safety and health at work, vol. 10, no. 2, pp. 229-236, 2019.
[12] L. Scime and J. Beuth, "Anomaly detection and classification in a laser powder bed additive manufacturing process using a trained computer vision algorithm," Additive Manufacturing, vol. 19, pp. 114-126, 2018.
[13] L. Scime and J. Beuth, "A multi-scale convolutional neural network for autonomous anomaly detection and classification in a laser powder bed fusion additive manufacturing process," Additive Manufacturing, vol. 24, pp. 273-286, 2018.
[14] J. Winn, A. Criminisi, and T. Minka, "Object categorization by learned universal visual dictionary," in Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1, 2005, vol. 2: IEEE, pp. 1800-1807.
[15] A. Krizhevsky, I. Sutskever, and G. E. Hinton, "Imagenet classification with deep convolutional neural networks," Advances in neural information processing systems, vol. 25, pp. 1097-1105, 2012.
[16] W. Shen, M. Zhou, F. Yang, C. Yang, and J. Tian, "Multi-scale convolutional neural networks for lung nodule classification," in International conference on information processing in medical imaging, 2015, vol. 9123: Springer, pp. 588-599.
[17] S. Chowdhury, K. Mhapsekar, and S. Anand, "Part build orientation optimization and neural network-based geometry compensation for additive manufacturing process," Journal of Manufacturing Science and Engineering, vol. 140, no. 3, 2018.
[18] A. Noriega, D. Blanco, B. Alvarez, and A. Garcia, "Dimensional accuracy improvement of FDM square cross-section parts using artificial neural networks and an optimization algorithm," The International Journal of Advanced Manufacturing Technology, vol. 69, no. 9-12, pp. 2301-2313, 2013.
[19] M. Khanzadeh, P. Rao, R. Jafari-Marandi, B. K. Smith, M. A. Tschopp, and L. Bian, "Quantifying geometric accuracy with unsupervised machine learning: using self-organizing map on fused filament fabrication additive manufacturing parts," Journal of Manufacturing Science and Engineering, vol. 140, no. 3, 2018.
[20] M. Samie Tootooni, A. Dsouza, R. Donovan, P. K. Rao, Z. J. Kong, and P. Borgesen, "Classifying the dimensional variation in additive manufactured parts from laser-scanned three-dimensional point cloud data using machine learning approaches," Journal of Manufacturing Science and Engineering, vol. 139, no. 9, 2017.
[21] J. Xiong, G. Zhang, J. Hu, and L. Wu, "Bead geometry prediction for robotic GMAW-based rapid manufacturing through a neural network and a second-order regression analysis," Journal of Intelligent Manufacturing, vol. 25, no. 1, pp. 157-163, 2014.
[22] J. zur Jacobsmühlen, S. Kleszczynski, G. Witt, and D. Merhof, "Detection of elevated regions in surface images from laser beam melting processes," in IECON 2015-41st Annual Conference of the IEEE Industrial Electronics Society, 2015: IEEE, pp. 001270-001275.
[23] H. Wu, Z. Yu, and Y. Wang, "Real-time FDM machine condition monitoring and diagnosis based on acoustic emission and hidden semi-Markov model," The International Journal of Advanced Manufacturing Technology, vol. 90, no. 5-8, pp. 2027-2036, 2017.
[24] H. Wu, Z. Yu, and Y. Wang, "A new approach for online monitoring of additive manufacturing based on acoustic emission," in International Manufacturing Science and Engineering Conference, 2016, vol. 49910: American Society of Mechanical Engineers, p. V003T08A013.
[25] D. Ye, G. S. Hong, Y. Zhang, K. Zhu, and J. Y. H. Fuh, "Defect detection in selective laser melting technology by acoustic signals with deep belief networks," The International Journal of Advanced Manufacturing Technology, vol. 96, no. 5-8, pp. 2791-2801, 2018.
[26] S. A. Shevchik, C. Kenel, C. Leinenbach, and K. Wasmer, "Acoustic emission for in situ quality monitoring in additive manufacturing using spectral convolutional neural networks," Additive Manufacturing, vol. 21, pp. 598-604, 2018.
[27] I. A. Okaroa, S. Jayasingheb, C. Sutcliffeb, K. Blackb, P. Paolettia, and P. L. Greena, "Automatic Fault Detection for Selective Laser Melting using Semi-Supervised Machine Learning," Preprints, vol. 1, 2018, doi: 10.20944/preprints201809.0346.v1.
[28] N. Alqahtani, R. T. Armstrong, and P. Mostaghimi, "Deep learning convolutional neural networks to predict porous media properties," in SPE Asia Pacific oil and gas conference and exhibition, 2018: Society of Petroleum Engineers.
[29] R. Snell et al., "Methods for rapid pore classification in metal additive manufacturing," The Journal of The Minerals, Metals & Materials Society (TMS), vol. 72, no. 1, pp. 101-109, 2020.
[30] M. Khanzadeh, L. Bian, N. Shamsaei, and S. M. Thompson, "Porosity detection of laser based additive manufacturing using melt pool morphology clustering," in Annual International Solid Freeform Fabrication Symposium (SFF), 2016, pp. 8-10.
[31] C. Wu, J. Gao, X. Liu, and Y. Zhao, "Vision-based measurement of weld pool geometry in constant-current gas tungsten arc welding," Proceedings of the institution of mechanical engineers, Part B: Journal of Engineering Manufacture, vol. 217, no. 6, pp. 879-882, 2003.
[32] M. Zhang, G. Chen, Y. Zhou, and S. Li, "Direct observation of keyhole characteristics in deep penetration laser welding with a 10 kW fiber laser," Opt. Express, vol. 21, no. 17, pp. 19997-20004, 2013/08/26 2013, doi: 10.1364/OE.21.019997.
[33] A. J. Harvey, "Correlating In-Situ Monitoring Data with Internal Defects in Laser Powder Bed Fusion Additive Manufacturing," Wright State University, 2020.
[34] J. zur Jacobsmühlen, S. Kleszczynski, D. Schneider, and G. Witt, "High resolution imaging for inspection of laser beam melting systems," in 2013 IEEE international instrumentation and measurement technology conference (I2MTC), 2013: IEEE, pp. 707-712.
[35] T.-N. Le, M.-H. Lee, Z.-H. Lin, H.-C. Tran, and Y.-L. Lo, "Vision-based in-situ monitoring system for melt-pool detection in laser powder bed fusion process," Journal of Manufacturing Processes, vol. 68, pp. 1735-1745, 2021.
[36] C.-Y. Chen, T.-N. Le, Z.-H. Lin, and Y.-L. Lo, "Numerical and Experimental Investigation into Gas Flow Field and Spattering Phenomena in Laser Powder Bed Fusion Processing of Inconel 718," under review by Materials & Design, 2021.
[37] S. Lu, G. Tsechpenakis, D. N. Metaxas, M. L. Jensen, and J. Kruse, "Blob analysis of the head and hands: A method for deception detection," in Proceedings of the 38th Annual Hawaii International Conference on System Sciences, 2005: IEEE, pp. 20c-20c.
[38] P. Bholowalia and A. Kumar, "EBK-means: A clustering technique based on elbow method and k-means in WSN," International Journal of Computer Applications, vol. 105, no. 9, 2014.
[39] M. Syakur, B. Khotimah, E. Rochman, and B. D. Satoto, "Integration k-means clustering method and elbow method for identification of the best customer profile cluster," in IOP Conference Series: Materials Science and Engineering, 2018, vol. 336, no. 1: IOP Publishing, p. 012017.
[40] Z. Xiang, R. Yan, X. Wu, L. Du, and Q. Yin, "Surface morphology evolution with laser surface re-melting in selective laser melting," Optik, p. 164316, 2020.
[41] A. Du Plessis, I. Yadroitsev, I. Yadroitsava, and S. G. Le Roux, "X-Ray Microcomputed Tomography in Additive Manufacturing: A Review of the Current Technology and Applications," 3D Printing and Additive Manufacturing, vol. 5, pp. 227-247, 2018, doi: 10.1089/3dp.2018.0060.
[42] T.-N. Le, Y.-L. Lo, and Z.-H. Lin, "Numerical simulation and experimental validation of melting and solidification process in selective laser melting of IN718 alloy," Additive Manufacturing, vol. 36, p. 101519, 2020.
[43] U. S. Bertoli, A. J. Wolfer, M. J. Matthews, J.-P. R. Delplanque, and J. M. Schoenung, "On the limitations of volumetric energy density as a design parameter for selective laser melting," Materials & Design, vol. 113, pp. 331-340, 2017.
[44] J. Ning, D. E. Sievers, H. Garmestani, and S. Y. Liang, "Analytical modeling of part porosity in metal additive manufacturing," International Journal of Mechanical Sciences, vol. 172, p. 105428, 2020.
[45] J. Gockel, J. Beuth, and K. Taminger, "Integrated control of solidification microstructure and melt pool dimensions in electron beam wire feed additive manufacturing of Ti-6Al-4V," Additive Manufacturing, vol. 1, pp. 119-126, 2014.
[46] B. Cheng and K. Chou, "Melt pool geometry simulations for powder-based electron beam additive manufacturing," in 24th Annual International Solid Freeform Fabrication Symposium-An Additive Manufacturing Conference, Austin, TX, USA, 2013, pp. 644-654.
[47] V.-P. Matilainen, H. Piili, A. Salminen, and O. Nyrhilä, "Preliminary investigation of keyhole phenomena during single layer fabrication in laser additive manufacturing of stainless steel," Physics Procedia, vol. 78, pp. 377-387, 2015.
[48] T.-Y.-F. Chen, Y.-L. Lo, Z.-H. Lin, and R.-Y. Lin, "Simultaneous Extraction of Profile and Surface Roughness of 3D SLM Components Using Fringe Projection Method," minor revision for Rapid Process Journal, 2021.
[49] E. Mirkoohi, D. E. Sievers, H. Garmestani, K. Chiang, and S. Y. Liang, "Three-dimensional semi-elliptical modeling of melt pool geometry considering hatch spacing and time spacing in metal additive manufacturing," Journal of Manufacturing Processes, vol. 45, pp. 532-543, 2019.
[50] V. Gunenthiram et al., "Experimental analysis of spatter generation and melt-pool behavior during the powder bed laser beam melting process," Journal of Materials Processing Technology, vol. 251, pp. 376-386, 2018.
[51] E. Alabort, D. Barba, and R. C. Reed, "Design of metallic bone by additive manufacturing," Scripta Materialia, vol. 164, pp. 110-114, 2019.
[52] Y. Zhang, X. Cao, P. Wanjara, and M. Medraj, "Fiber laser deposition of Inconel 718 using powders," Materials Science and Technology (MS&T), 2013.
[53] S. Visa, B. Ramsay, A. L. Ralescu, and E. Van Der Knaap, "Confusion matrix-based feature selection," MAICS, vol. 710, pp. 120-127, 2011.
[54] S. Pattanayak, Pattanayak, and S. John, Pro deep learning with tensorflow. Springer, 2017