金屬積層製造包含多種技術，其中選擇性雷射熔融（SLM）被廣泛用於生產複雜、精密且獨特形狀的金屬部件。SLM的好處包括顯著的設計靈活性、製造效率以及與多種金屬材料的兼容性。然而，準確地確定產品特性往往涉及複雜的實驗，並且可能會損害產品。因此迫切需要開發一種自動預測產品特性的方法。在本研究中，記錄了與金屬積層製造產品相關的細節，包括製程參數和影像紋理特徵。這些特徵是從產品的縱剖面和逐層影像中使用灰階共生矩陣（GLCM）提取的。然後利用機器學習（ML）模型，如支援向量迴歸（SVR）、XGBoost和LightGBM，來預測產品特性並比較它們的表現。實驗結果顯示，製程參數與縱剖面影像的紋理特徵之間的相關性比逐層影像中的較高。研究結果顯示這些模型展示了高預測準確性，尤其是XGBoost和LightGBM，其R²分數接近0.9。這些結果顯示了本研究方法的優勢和實用性。此外也展示了其準確預測各種產品屬性的能力，滿足了多種應用場景的需求。

Metal additive manufacturing incorporates several techniques, with Selective Laser Melting (SLM) being widely used to produce complex, precise, and distinctively shaped metal parts. The benefits of SLM encompass considerable design flexibility, manufacturing efficiency, the elimination of molds, and compatibility with a range of metal materials. Nevertheless, accurately determining product characteristics often involves intricate experimentation that could potentially harm the products. Therefore, there is an imperative need to develop an automated approach for predicting product characteristics. In this research, details pertaining to metal additive manufacturing products were recorded, including process parameters and textural features. These features were derived from the products' longitudinal sectional and layer-by-layer images using the gray-level co-occurrence matrix (GLCM). Machine learning (ML) models such as Support Vector Regression (SVR), XGBoost, and LightGBM were then utilized to forecast product properties and assess their performance. The experimental outcomes revealed more significant correlations between process parameters and textural features in longitudinal section images than layer-by-layer images. Additionally, the models exhibited high predictive accuracy, especially XGBoost and LightGBM, with R² score nearing 0.9 for all properties. These results underscore the advantages and practicality of the suggested method. Moreover, this technique demonstrates its capability to accurately predict various product properties, meeting the demands of diverse application scenarios.

1. Pham, D.T., & Dimov, S.S. (2001). Technical Characteristics and Technological Capabilities of Rapid Prototyping Systems. In S. Dimov (Ed.), Rapid Manufacturing (pp. 43-69). Springer.
2. Sing, S. L., & Yeong, W. Y. (2020). Laser powder bed fusion for metal additive manufacturing: perspectives on recent developments. Virtual and Physical Prototyping , 15 (3), (pp. 359–370).
3. Shokrollahi, H., & Janghorban, K. (2007). Soft magnetic composite materials (SMCs). Journal of Materials Processing Technology , 189 (Issues 1–3), (pp. 1-12) .
4. Wu, S.-T., Huang, P.-W., Chang, T.-W., Jiang, I.-H., & Tsai, M.-C. (2021). Application of Magnetic Metal 3-D Printing on the Integration of Axial-Flow Impeller Fan Motor Design. IEEE Transactions on Magnetics , 57 (2), (pp. 1-5).
5. Huang, P.-W., Tsai, M.-C., & Jiang, I.-H. (2018). 3-D Structure Line-Start Synchronous Reluctance Motor Design Based on Selective Laser Melting of 3-D Printing. IEEE Transactions on Magnetics , 54 (11), (pp. 1-4).
6. Jhong, K. J., Huang, W.-C., & Lee, W. H. (2016). Microstructure and magnetic properties of magnetic material fabricated by selective laser melting. Physics Procedia , 83 , (pp. 818-824).
7. Jhong, K.-J., Chang, T.-W., Lee, W.-H., Tsai, M.-C., & Jiang, I.-H. (2019). Characteristic of high frequency Fe-Si-Cr material for motor application by selective laser melting. AIP Advances , 9 (3), (pp. 1-4).
8. Kamath, C. (2016). Data mining and statistical inference in selective laser melting. International Journal of Advanced Manufacturing Technology , 86 , (pp. 1659–1677).
9. Lee, A.-C., Huang, R. Y., Nguyen, T. D., Cheng, C.-W., & Tsai, M. C. (2020). Laser powder bed fusion of multilayer thin-walled structures based on data-driven model. Journal of Laser Micro Nanoengineering , 15 (1), (pp. 38-44).
10. Gao, Q., Wang, X., & Zhang, Y. (2023). Multi–physical-field characteristics modeling and structure optimization for kW-level ultra-high-speed PM motors with integrated support system. Chinese Journal of Aeronautics , 36 (4), (pp. 455-467).
11. Wu, S., Huang, X., Tian, C., & Zhang, P. (2019). Multi-Physical Field Optimization Analysis of High-Speed Permanent Magnet Synchronous Motor Based on NSGA-II Algorithm. In 2019 22nd International Conference on Electrical Machines and Systems (ICEMS) (pp. 1-6). Harbin, China.
12. TONGTAI. (2023). 東台精積AMP-160. Retrieved from http://www.tongtai.com.tw/tw/product-detail.php?id=315
13. 陳日昇，"積層製造成品之三維影像建模與品質預測"，碩士論文，國立成功大學機械工程研究所，2023。
14. Cannizzaro, D., Varrella, A. G., Paradiso, S., Sampieri, R., Macii, E., Patti, E., & Di Cataldo, S. (2021). Image analytics and machine learning for in-situ defects detection in Additive Manufacturing. In 2021 Design, Automation & Test in Europe Conference & Exhibition (DATE) (pp. 603-608). Grenoble, France.
15. Haralick, R. M., Shanmugam, K., & Dinstein, I. (1973). Textural Features for Image Classification. IEEE Transactions on Systems, Man, and Cybernetics, SMC-3 (6), (pp. 610-621).
16. Prasad, G., Vijay, G. S., & Kamath, R. C. (2022). Comparative study on classification of machined surfaces using ML techniques applied to GLCM based image features. Materials Today: Proceedings, 62 (3), (pp. 1440-1445).
17. Unni, A., Eg, N., Vinod, S., & Nair, L. S. (2018). Tumour detection in double threshold segmented mammograms using optimized GLCM features fed SVM. In Proceedings of the 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI) (pp. 554-559). Bangalore, India.
18. Raut, A., Patil, M. A., Dhondrikar, C. P., & Kamble, S. D. (2016). Texture Parameters Extraction of Satellite Image. International Journal of Science Technology & Engineering, 2 (11), (pp. 13-18).
19. Qiu, Q., Duan, Y., Ma, K., Tao, L., & Xie, Z. (2023). Information extraction and knowledge linkage of geological profiles and related contextual texts from mineral exploration reports for geological knowledge graphs construction. Ore Geology Reviews, 163 , 105739.
20. Sonam, & Jain, V. (2018). Geomorphic effectiveness of a long profile shape and the role of inherent geological controls in the Himalayan hinterland area of the Ganga River basin,India. Geomorphology,304 , (pp. 15-29).
21. IWATSU ELECTRIC CO., LTD. (2023). Instruction Manual B-H Analyzer SY-8218/SY-8219(pp.7–10). https://www.iti.iwatsu.co.jp/en/products/sy/sy8218_top_e.html,
22. SHIMADZUMWG-100KNA, (2023)
https://www.shimadzu.com/an/test/tmc/find/material/00/3/1/868,
23. Devaraj, S. J. (2019). Chapter 2 - Emerging Paradigms in Transform-Based Medical Image Compression for Telemedicine Environment. In H. D. Jude & V. E. Balas (Eds.), Telemedicine Technologies (pp. 15-29). Academic Press.
24. Haralick, R. M., Shanmugam, K., & Dinstein, I. (1973). Textural Features for Image Classification. IEEE Transactions on Systems, Man, and Cybernetics, SMC-3 (6), (pp. 610-621).
25. Haralick, R. M. (1979). Statistical and structural approaches to texture. Proceedings of the IEEE, 67 (5), (pp. 786-804).
26. Cover, T. M., & Thomas, J. A. (2006). Elements of Information Theory (2nd ed.). John Wiley & Sons.
27. Kraskov, A., Stögbauer, H., & Grassberger, P. (2004). Estimating mutual information. Physical Review E: Statistical, Nonlinear, and Soft Matter Physics, 69 (6 Part 2), 066138.
28. Bement, T. R. (1989). Taguchi Techniques for Quality Engineering., Technometrics, 31 (2), (pp. 253–255).
29. Elsen, M., Al-Bender, F., & Kruth, J.-P. (2008). Application of dimensional analysis to selective laser melting. Rapid Prototyping Journal, 14 (1), (pp. 15-22).
30. 許育銘，"選擇性雷射熔融鎳鐵軟磁材料之磁與機械特性最佳化研究 "，碩士論文，國立成功大學機械工程研究所，2023。
31. Kıvak, T. (2014). Optimization of surface roughness and flank wear using the Taguchi method in milling of Hadfield steel with PVD and CVD coated inserts. Measurement, 50 , (pp. 19-28).
32. Cord, M., & Cunningham, P. (Eds.). (2008). Machine Learning Techniques for Multimedia: Case Studies on Organization and Retrieval (pp. 21–45).
33. Su, X., Yan, X., & Tsai, C.-L. (2012). Linear regression. WIREs Computational Statistics, 4 , (pp. 275-294).
34. LaValley, M. P. (2008). Logistic regression. Circulation, 117 (18), (pp. 2395-2399).
35. Pan, J.-X., & Fang, K.-T. (2002). Growth Curve Models and Statistical Diagnostics (pp. 77-158).
36. Awad, M., & Khanna, R. (2015). Support Vector Regression. In Efficient Learning Machines (pp. 67-80).
37. Suthaharan, S. (2015). Machine Learning Models and Algorithms for Big Data Classification: Thinking with Examples for Effective Learning (pp. 237-250).
38. Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785-794).
39. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T.-Y. (2017). Lightgbm: A highly efficient gradient boosting decision tree. In Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS 2017).