簡易檢索 / 詳目顯示

回結果列表
研究生： 江坤諦
Jiang, Kun-Di
論文名稱： 透過機器學習最佳化選擇性雷射熔融積層製造應用於軟磁複合材料產品特性
Selective laser melting of FeSiCr alloy: Parameter optimization, magnetic and mechanical properties
指導教授： 蔡明祺
Tsai, Mi-Ching
學位類別： 碩士
Master
系所名稱： 工學院 - 機械工程學系
Department of Mechanical Engineering
論文出版年： 2023
畢業學年度： 111
語文別： 中文
論文頁數： 92
中文關鍵詞： 金屬3D列印 、積層製造 、軟磁複合材料 、機器學習算法
外文關鍵詞： selective laser melting (SLM), soft magnetic composites (SMC), process parameter optimization, additive manufacturing, Taguchi method, machine learning
相關次數： 點閱：126  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 軟磁複合材料由於其極低的鐵損值在電磁應用中被廣泛使用，為了確保透過選擇性雷射熔融製造的元件滿足電磁應用的要求，並幫助製造工程師選擇最佳的製程參數，本研究開發了一種基於機器學習的優化方法，將四個關鍵製程參數(氧濃度、雷射功率、掃描速度以及線間距)與製造電磁元件相關的三個目標產品特性(磁導率、鐵損以及最大拉伸強度)相關聯，其中機器學習模型通常需要大量的數據，本研究使用田口方法建議之L9直交表以減少所需的實驗量，根據收集的實驗數據，測試了五個機器學習模型，並在不同的應用情境下選擇最佳的機器學習模型來預測成品的每個特性，製造參數建議系統整合了機器學習模型和非凌越排序基因演算法，幫助用戶選擇最佳的製程參數，以製造具有要求特性的元件。

    Soft magnetic composite products have been widely used in electromagnetic applications, owing to their unique properties of very low eddy current loss, relatively low total core loss at medium and high frequencies. To ensure that the part fabricated by SLM meets the requirements for electromagnetic applications and help the manufacturing engineers choose optimal process parameters, an optimization methodology based on machine learning was developed to relate four key process parameters (oxygen concentration, laser power, scanning speed, and hatch distance) and three target properties of the fabricated electromagnetic parts (permeability, core loss, and ultimate strength). Machine learning models usually require a large amount of experimental data, and L9 orthogonal array design was used to reduce the required amount of experiments in this study. Based on the collected experimental data, five machine learning models were developed, and the better machine learning models were adopted to predict each property of the part. A manufacture parameter suggestion system integrated the machine learning models and multi-objective optimal algorithm NSGA-II, helping users to select the optimal process parameters to fabricat products with the required properties.

    摘要 I 誌謝 XXII 目錄 XXIV 表目錄 XXVIII 圖目錄 XXXI 符號表 XXXIV 第一章 緒論 1 1.1 研究動機 1 1.2 文獻回顧 2 1.3 研究目的 4 1.4 本文架構 4 第二章 金屬3D列印積層製造與田口方法實驗設計 6 2.1 選擇性雷射熔融 6 2.1.1 選擇性雷射熔融介紹 6 2.1.2 實驗環境 7 2.1.3 試片幾何 9 2.2 田口方法品質設計及因子反應圖分析 9 2.2.1 田口方法與直交表 10 2.2.2 田口式直交表實驗設計 11 2.3 資料蒐集及因子反應圖分析 12 2.3.1 量測原理 13 2.3.2 實驗數據及因子反應圖分析 15 2.3.3 磁導率及鐵損值分析 22 2.3.4 最大拉伸應力分析 30 2.3.5 孔隙率分析 32 2.3.6 結果討論 35 2.3.7 田口式品質計量法 36 2.3.8 預測最佳設計下的品質特性 37 第三章 特性預測模型系統架構及模型選用 41 3.1 機器學習演算法選擇 41 3.1.1 機器學習模型分類 41 3.1.2 K Nearest Neighbor(KNN) 演算法 42 3.1.3 Support Vector Regression(SVR) 演算法 43 3.1.4 XGBoost 演算法 44 3.1.5 LightGBM 演算法 45 3.1.6 CatBoost 演算法 45 3.2 模型評估與最佳化超參數 46 3.2.1 指標函數選用 46 3.2.2 隨機搜索與網格搜索比較 47 3.2.3 產品特性預測 48 3.2.4 模型訓練結果 49 3.2.5 超參數最佳化 50 3.2.6 機器學習以及田口方法預測結果比較 53 第四章 製程參數建議系統建立 57 4.1 非凌越排序基因演算法 57 4.1.1 基因演算法 57 4.1.2 何謂凌越 59 4.1.3 擁擠距離 61 4.1.4 非凌越排序基因演算法選擇機制 62 4.2 製程參數建議系統開發流程 63 4.2.1 開發環境及軟硬體配置 63 4.2.2 機器學習及基因演算法函式庫 64 4.2.3 演算法之間的互動 66 4.2.4 產品特性權重分配 68 4.2.5 參數建議及產品特性預估案例 70 4.3 產品特性預測及製程參數建議系統實現 73 4.3.1 使用者介面設計 74 4.3.2 系統結果呈現 76 結論與未來建議 79 4.4 結論 79 4.5 未來建議 80 參考文獻 82 附錄 88 附件一、磁特性(磁導率、鐵損)訓練集以及測試集資料格式(部分) 88 附件二、機械性質訓練集以及測試集資料格式(部分) 89 附件三、模型訓練程式碼 (XGBoost) 90 附件四、隨機搜索最佳化超參數程式碼(部分) 91 附件五、C# 與 Python 互動程式碼 (部分) 92

    [1] D. T. Pham, S. S. Dimov, "Technical characteristics and technological capabilities of rapid prototyping systems.", Rapid Manufacturing: The Technologies and Applications of Rapid Prototyping and Rapid Tooling, pp. 43-69, 2001.
    [2] S. L. Sing, W. Y. Yeong "Laser powder bed fusion for metal additive manufacturing: perspectives on recent developments.", Virtual Phys Prototyping, pp. 359–370, 2020.
    [3] H. Shokrollahi, K. Janghorban, "Soft magnetic composite materials (SMCs)", Journal of Materials Processing Technology, pp. 1-12, 2007.
    [4] S. T. Wu, P. W. Huang, T. W. Chang, I. H. Jiang, M. C. Tsai "Application of magnetic metal 3-D printing on the integration of axial-flow impeller fan motor design.", IEEE Transactions on Magnetics, pp. 1-5, 2020.
    [5] P. W. Huang, M. C. Tsai, I. H. Jiang, "3-D Structure Line-Start Synchronous Reluctance Motor Design Based on Selective Laser Melting of 3-D Printing.", IEEE Transactions on Magnetics, pp. 1-4, 2018.
    [6] P. W. Huang, I. H. Jiang, M. C. Tsai, G. M. Chen, "New Hybrid Stator Design for High-Speed PMSMS Based on Selective Laser Melting of 3-D Printing.", IEEE Transactions on Magnetics, pp. 1-4, 2019.
    [7] G. M. Tseng, K. J. Jhong, M. C. Tsai, P. W. Huang, W. H. Lee, "Application of additive manufacturing for low torque ripple of 6/4 switched reluctance motor.", International Conference on Electrical Machines and Systems (ICEMS), pp. 1-4, 2019.
    [8] I. Tolosa, F. Garciandía, F. Zubiri, F. Zapirain, A. Esnaola "Study of mechanical properties of AISI 316 stainless steel processed by “selective laser melting”, following different manufacturing strategies.", The International Journal of Advanced Manufacturing Technology, pp. 639-647, 2010.
    [9] K. J. Jhong, W. C. Huang, W. H. Lee "Microstructure and magnetic properties of magnetic material fabricated by selective laser melting. ", Physics Procedia, pp. 818-824, 2016.
    [10] K. J. Jhong, T.W. Chang, W.H. Lee, M. C. Tsai "Characteristic of high frequency Fe–Si–Cr material for motor application by selective laser melting.", AIP Advances, pp. 1-4, 2019.
    [11] C. Kamath "Data mining and statistical inference in selective laser melting.", The International Journal of Advanced Manufacturing Technology, pp. 1659-1677, 2016.
    [12] M. Gor, A. Dobriyal, V. Wankhede, P. Sahlot, K. Grzelak, J. Kluczyński, J. Łuszczek "Density prediction in powder bed fusion additive manufacturing: machine learning-based techniques.", Applied Sciences, pp. 12-14, 2022.
    [13] D. S. Nguyen, H. S. Park, C. M. Lee "Optimization of selective laser melting process parameters for Ti-6Al-4V alloy manufacturing using deep learning.", Journal of Manufacturing Processes, pp. 230-235, 2020.
    [14] B. Yuan, G. M. Guss, A. C. Wilson, S. P. Hau‐Riege, P. I. DePond, S. McMains, B. Giera "Machine‐learning‐based monitoring of laser powder bed fusion.", Advanced Materials Technologies, pp. 3-12, 2018.
    [15] A. C. Lee, R. Y. Huang, T. D. Nguyen, C. W. Cheng, M. C. Tsai "Laser powder bed fusion of multilayer thin-walled structures based on data-driven model.", Journal of Laser Micro Nanoengineering, pp. 1-7, 2020.
    [16] T. W. Chang, K. W. Liao, C. C. Lin, M. C. Tsai, C. W. Cheng "Predicting magnetic characteristics of additive manufactured soft magnetic composites by machine learning.", The International Journal of Advanced Manufacturing Technology, pp. 3177-3184, 2021.
    [17] Q. Gao, X. Wang, Y. Zhang "Multi–physical-field characteristics modeling and structure optimization for kW-level ultra-high-speed PM motors with integrated support system.", Chinese Journal of Aeronautics, 2022.
    [18] S. Wu, X. Huang, C. Tian, P. Zhang, "Multi-Physical Field Optimization Analysis of High-Speed Permanent Magnet Synchronous Motor Based on NSGA-II Algorithm.", International Conference on Electrical Machines and Systems (ICEMS), pp. 1-6, 2019.
    [19] 東台精積AMP-160, http://www.tongtai.com.tw/tw/product-detail.php?id=315, 2023.
    [20] M. Solanki, P. Bhatt, A. Jain "Integration of value stream mapping with Taguchi’s orthogonal array to improve productivity and quality: A case study of SSE.", Materials Today: Proceedings, pp. 1304-1310, 2021.
    [21] T. R. Bement "Taguchi Techniques for Quality Engineering.", Quality Resources, pp. 253-255, 1989.
    [22] M. Van Elsen, F. Al-Bender, J. P. Kruth, “Application of dimensional analysis to selective laser melting.”, Rapid Prototyping Journal, pp. 15–22, 2008.
    [23] B-H Analyzer SY-8218, https://www.iti.iwatsu.co.jp/en/products/sy/sy8218_top_e.html, 2023.
    [24] SHIMADZU MWG-100KNA, https://www.shimadzu.com/an/test/tmc/find/material/00/3/1/868, 2023.
    [25] M. Fousová, D. Vojtěch, J. Jablonská "Promising characteristics of gradient porosity Ti-6Al-4V alloy prepared by SLM process.", Journal of the mechanical behavior of biomedical materials, pp. 368-376, 2017.
    [26] S. K. Karna, R. Sahai "An overview on Taguchi method." International journal of engineering and mathematical sciences, pp. 1-7, 2012.
    [27] J. G. Carbonell, R. S. Michalski, T. M. Mitchell "An overview of machine learning." Machine learning pp. 3-23, 1983.
    [28] Kaggle: Your Home for Data Science, https://www.kaggle.com/, 2023.
    [29] J. Bergstra, Y. Bengio "Random search for hyper-parameter optimization.", Journal of machine learning research, pp. 13, 2012.
    [30] K. Deb, A. Pratap, S. Agarwal, T. Meyarivan "A fast and elitist multiobjective genetic algorithm: NSGA-II. Evolutionary Computation.", IEEE transactions on evolutionary computation, pp. 182-197, 2002.
    [31] J. Blank, K. Deb "Pymoo: Multi-objective optimization in python.", IEEE Access, pp. 89497-89509, 2020.
    [32] A. B. Ruiz, R. Saborido, M. Luque "A preference-based evolutionary algorithm for multiobjective optimization: the weighting achievement scalarizing function genetic algorithm.", Journal of Global Optimization, pp. 101-129, 2015.
    [33] K. Deb, K. Sindhya, T. Okabe "Self-adaptive simulated binary crossover for real-parameter optimization.", Proceedings of the 9th annual conference on genetic and evolutionary computation, pp. 1187-1194, 2007.
    [34] Y. Song, J. Liang, J. Lu, X. Zhao "An efficient instance selection algorithm for k nearest neighbor regression.", Neurocomputing, pp. 26-34, 2017.
    [35] A. J. Smola, B. Schölkopf "A tutorial on support vector regression. ", Statistics and computing, pp. 199-222, 2004.
    [36] T. Chen, C. Guestrin "Xgboost: A scalable tree boosting system.", Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, pp. 785-794, 2016.
    [37] L. Prokhorenkova, G. Gusev, A. Vorobev, A. V. Dorogush, A. Gulin "CatBoost: unbiased boosting with categorical features.", Advances in neural information processing systems, pp. 31, 2018.
    [38] G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, T. Y. Liu "Lightgbm: A highly efficient gradient boosting decision tree.", Advances in neural information processing systems, pp. 30, 2017.
    [39] S. Noori-Darvish, R. Tavakkoli-Moghaddam "Minimizing the total tardiness and makespan in an open shop scheduling problem with sequence-dependent setup times.", Journal of industrial engineering international, pp. 1-13, 2012.
    [40] H. S. Park, D. S. Nguyen, T. Le-Hong, X. Van Tran "Machine learning-based optimization of process parameters in selective laser melting for biomedical applications.", Journal of Intelligent Manufacturing, pp. 1843-1858, 2022.
    [41] H. S. Park, T. T. Nguyen, X. P Dang "Multi-objective optimization of turning process of hardened material for energy efficiency.", International Journal of Precision Engineering and Manufacturing, pp. 1623-1631, 2016.
    [42] F. Miao, H. S. Park, C. Kim, S. Ahn "Swarm intelligence based on modified PSO algorithm for the optimization of axial-flow pump impeller.", Journal of Mechanical Science and Technology, pp. 4867-4876, 2015.

    無法下載圖示 校內：2028-07-26公開
    校外：2028-07-26公開
    電子論文尚未授權公開，紙本請查館藏目錄
    QR CODE