臺灣博碩士論文加值系統

積層製造技術具高度客製化與實現複雜幾何的特性，應用快速發展如汽車、航太與生醫產業。選擇性雷射熔融金屬積層製造技術於磁性材料的製程試作，參數選擇流程相當繁複，一般須經由主觀地觀察及錯誤嘗試，導致效率不高且可重複性低，特別於磁性材料列印時，除了考慮材料結構強度外，還需要考慮成品的導磁特性，使得參數選用更加複雜。本研究導入機器學習於列印參數之選擇設定，採用資料導向(Data-driven)的方式，以減少錯誤嘗試的耗時，進而提高效率。本文主要在探討應用機器學習在磁性材料之積層製造時的參數選用，利用機器學習的演算法XGBoost，建立一套有效率、可重複性高的軟磁材料積層製造參數選用流程。

Selective laser melting (SLM) is one of the widely used metal additive manufacturing (AM) techniques. While SLM is able to produce high quality products, the parameter selection process can be very complicated, especially for magnetic materials. The mechanical properties need to be concerned as well as the magnetic characteristics, which makes the parameter selecting process more complicated. In this research, the parameter selection process of magnetic material for SLM will be explored. The process that integrates machine algorithm and evolutionary algorithm, comparing with the existing approach, is not only repeatable but timesaving.

摘要 I
ABSTRACT II
致謝 XIII
目錄 XIV
表目錄 XVII
圖目錄 XVIII
符號表 XX
第一章 緒論 1
1.1研究動機 1
1.2研究背景與文獻回顧 2
1.3研究目的 3
1.4本文架構 4
第二章 製程參數選擇方法與資料收集 6
2.1選擇性雷射熔融 6
2.1.1選擇性雷射熔融原理 6
2.1.2實驗環境 7
2.2 選擇性雷射熔融材料選擇方法 9
2.2.1現行製程選擇流程 9
2.2.2 機器學習製程選擇方法 11
2.3 學習模型所需資料搜集 12
2.3.1 機器學習流程 13
2.3.2模型特徵與標籤 14
2.3.3 資料搜集 15
第三章 磁特性預測模型系統架構 19
3.1 演算法介紹與選擇 19
3.1.1類神經網絡 20
3.1.2 XGBoost 22
3.1.3演算法比較 23
3.1.4 機器學習函式庫 24
3.1.5 R2 Score 決定係數 25
3.2 磁特性預測系統 26
3.2.1 磁特性預測系統架構 26
3.2.2 進化演算法 27
第四章 機器學習模型設計與結果分析 29
4.1 開發流程及背景說明 29
4.1.1 開發流程 29
4.1.2 開發使用軟硬體 30
4.1.3 實驗背景 31
4.2 機器學習模型選擇 31
4.2.1 類神經網路模型結果 31
4.2.2 XGBoost結果 33
4.2.3 最終模型選擇 35
4.3訓練及測試資料預處理 35
4.4 訓練模型優化 37
4.4.1以網格式搜索找出最佳模型參數 37
4.5 磁特性預測與製程參數建議系統實現 39
4.5.1 製程參數建議系統應用情境 39
4.5.2 製程參數建議系統結果驗證 41
第五章 結論與未來建議 43
5.1 結論 43
5.2 未來建議 44
參考文獻 45
附錄 47
附件一、訓練及測試資料CSV檔案樣本(部分) 47
附件二、PCV訓練模型PYTHON程式樣本(部分) 48
附件三、ΜA訓練模型PYTHON程式樣本(部分) 49
附件四、參數建議系統PYTHON程式樣本(部分) 50

[1]K.-J. Jhong, W.-C. Huang, and W.-H. Lee, Microstructure and Magnetic Properties of Magnetic Material Fabricated by Selective Laser Melting, Physics Procedia, 2016.
[2]C. Y. Yap, C. K. Chua, Z. L. Dong, Z. H. Liu, D. Q. Zhang, L. E. Loh, and S. L. Sing, Review of selective laser melting: Materials and applications, Applied Physics Reviews, 2015.
[3]H. Shokrollahi and K. Janghorban, Soft magnetic composite materials (SMCs), Journal of Materials Processing Technology, 2007.
[4]K.-J. Jhong, T.-W. Chang, W.-H. Lee, M.-C. Tsai, and I.-H. Jiang, Characteristic of high frequency Fe-Si-Cr material for motor application by selective laser melting, AIP Advence, 2019.
[5]R. Ramprasad, R. Batra, G. Pilania, A. Mannodi-Kanakkithodi, and C. Kim, Machine learning in materials informatics: recent applications and prospects, Computational Materials, 2017.
[6]C. Kamath, Data mining and statistical inference in selective laser melting, The International Journal of Advanced Manufacturing Technology, 2016.
[7]B. Kappes, S. Moorthy, D. Drake, H. Geerlings, and A. Stebner, Machine Learning to Optimize Additive Manufacturing Parameters for Laser Powder Bed Fusion of Inconel 718, in Proceedings of the 9th International Symposium on Superalloy 718 ＆ Derivatives: Energy, Aerospace, and Industrial Applications, 2018.
[8]B. Yuan, G. M. Guss, A. C. Wilson, S. P. Hau‐Riege, P. J. DePond, S. McMains, M. J. Matthews, and B. Giera, Machine‐Learning‐Based Monitoring of Laser Powder Bed Fusion, Advanced Materials Technologies, 2018.
[9]東台精積AMP-160. Available: http://www.tongtai.com.tw/tw/product-detail.php?id=315
[10]嘉鋼精密工業股份有限公司. Available: http://www.cysteel.com.tw/
[11]J. P. Kruth, L. Froyen, J. Van Vaerenbergh, P. Mercelis, M. Rombouts, and B. Lauwers, Selective laser melting of iron-based powder, Journal of Materials Processing Technology, 2004.
[12]KEYENCE VHX-5000. Available: https://www.keyence.com.tw/products/microscope/digital-microscope/vhx-5000/index.jsp
[13]B-H Analyzer SY-8218 Available: https://www.iti.iwatsu.co.jp/en/products/sy/sy8218_top_e.html
[14]Kaggle: Your Home for Data Science. Available: https://www.kaggle.com/
[15]A. K. Jain, J. C. Mao, and K. M. Mohiuddin, Artificial neural networks: A tutorial, (in English), Computer, 1996.
[16]T. Chen and C. Guestrin, Xgboost: A scalable tree boosting system, Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 2016.
[17]Scikit-learn algorithm cheat-sheet. Available: https://scikit-learn.org/stable/tutorial/machine_learning_map/index.html
[18]T. Bäck, Evolutionary algorithms in theory and practice: evolution strategies, evolutionary programming, genetic algorithms, Oxford University Press, Inc., 1996.
[19]Python API Reference. Available: https://xgboost.readthedocs.io/en/latest/python/python_api.html#module-xgboost.sklearn
[20]K. Deb, A. Pratap, S. Agarwal, and T. Meyarivan, A fast and elitist multiobjective genetic algorithm: NSGA-II, IEEE Transactions on Evolutionary Computation, 2002.