| 研究生: |
陳冠瑜 Chen, Kuan-Yu |
|---|---|
| 論文名稱: |
無因次模型及機器學習之雷射疊焊參數優化研究 Optimization of Lap-joint Laser Welding Parameters Using Dimensionless Model and Machine Learning |
| 指導教授: |
陳元方
Chen, Yuan-Fang |
| 共同指導教授: |
羅裕龍
Lo, Yu-Lung |
| 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: | 英文 |
| 論文頁數: | 44 |
| 中文關鍵詞: | 雷射焊接 、疊焊 、人工神經網路 、優化 |
| 外文關鍵詞: | Laser welding, Welding-induced crack, Lap-joint, Dimensionless model |
| 相關次數: | 點閱:122 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 |
在雷射焊接過程中,通常會在較大的參數範圍內尋找可用參數(如:雷射功率、掃描速度),並進行反覆的試驗來獲得最佳的參數。本文著重於運用無因次模型所提供的預測來得到最佳的優化參數。研究中發現,在keyhole 焊接模式下的深度預測精度高於熔池的寬度預測。因此在熔池寬度預測中基於圓形包裝設計,使用 ANN 的模型對實驗數據進行訓練。透過 ANN 模型的訓練可以得到各雷射功率及掃描速度參數所對應的熔池縱橫比(深度/寬度)。其特別之處在於該方法是由無因次模型所提供的熔池深度和通過光學顯微鏡觀察及測量的熔池寬度進行組合。然後進一步以孔隙率、裂紋、熔池深度和縱橫比為標準建立雷射焊接優化區。最終得到無孔隙、無裂紋、熔池深度適中、縱橫比合適的最佳焊接區域,其中在無因次預測的平均誤差為 20%,而 ANN 學習的深寬比準確率高達 92%。此外,基於四個標準的最佳區域孔隙率皆得到驗證,而在搭接焊接過程中增加了 0.038mm 的介面間隙並不影響預測深度。
In a laser welding process, a weld sample of experiments is usually performed on a wide range of parameters (such as laser power, scanning speed...etc.) with trial-and-error in order to obtain optimized parameters. This paper is focused on the data based on the dimensionless model to extract optimized parameters for lap-join laser welding. It is found that a prediction map of melt pool depth in keyhole welding mode has much higher accuracy than that of melt pool width based a dimensionless model. Therefore, the predicted melt pool width is directly based upon the experimental data based upon the circular packaging design using the ANN model to train the experimental data. Accordingly, the corresponding aspect ratio (depth/width ratio) under various parameter sets of laser power and scanning speed was characterized. Uniquely, the melt pool depth simply extracted by the dimensionless model and the melt pool width easily measured by an observation using an optical microscope are designed in this proposed methodology. Subsequently, porosity, crack, penetration depth, and aspect ratio were set as criteria to establish optimized region, and finally, the optimal II zone with less porosity, less crack, proper penetration depth, and proper aspect ratio of lap-join welding can be achieved. As a result, the error of the predicted depth and the average accuracy rate with NN learning the depth width ratio is within 20% and 92%, respectively. Also, the optimal zone based upon four criteria is verified by the experiments in pore. It is noted the criteria introduced in this study are irresponsible for materials. Additionally, an interface gap of 0.038 mm for the lap-join laser welding is added to reduce the porosity, and it does not affect the predicted depth.
[1] A. Ruggiero, L. Tricarico, A. Olabi, and K. Benyounis, "Weld-bead profile and costs optimisation of the CO2 dissimilar laser welding process of low carbon steel and austenitic steel AISI316," Optics & Laser Technology, vol. 43, no. 1, pp. 82-90, 2011.
[2] S. Katayama, "Defect formation mechanisms and preventive procedures in laser welding," in Handbook of laser welding technologies: Elsevier, 2013, pp. 332-373.
[3] J. C. Ion, H. R. Shercliff, and M. F. Ashby, "Diagrams for laser materials processing," Acta metallurgica et materialia, vol. 40, no. 7, pp. 1539-1551, 1992.
[4] D. Hann, J. Iammi, and J. Folkes, "Keyholing or conduction–prediction of laser penetration depth," in Proceedings of the 36th International MATADOR Conference, 2010: Springer, pp. 275-278.
[5] D. Hann, J. Iammi, and J. Folkes, "A simple methodology for predicting laser-weld properties from material and laser parameters," Journal of Physics D: Applied Physics, vol. 44, no. 44, p. 445401, 2011.
[6] A. Robert and T. Debroy, "Geometry of laser spot welds from dimensionless numbers," Metallurgical and materials transactions B, vol. 32, pp. 941-947, 2001.
[7] X. Xie, J. Zhou, and J. Long, "Numerical study on molten pool dynamics and solute distribution in laser deep penetration welding of steel and aluminum," Optics & Laser Technology, vol. 140, p. 107085, 2021.
[8] T. Liu, R. Hu, X. Chen, S. Gong, and S. Pang, "Localized boiling-induced spatters in the high-power laser welding of stainless steel: three-dimensional visualization and physical understanding," Applied Physics A, vol. 124, pp. 1-14, 2018.
[9] A. Grossmann, J. Felger, T. Froelich, J. Gosmann, and C. Mittelstedt, "Melt pool controlled laser powder bed fusion for customised low-density lattice structures," Materials & Design, vol. 181, p. 108054, 2019.
[10] M. Thomas, G. J. Baxter, and I. Todd, "Normalised model-based processing diagrams for additive layer manufacture of engineering alloys," Acta Materialia, vol. 108, pp. 26-35, 2016.
[11] H.-C. Tran and Y.-L. Lo, "Systematic approach for determining optimal processing parameters to produce parts with high density in selective laser melting process," The International Journal of Advanced Manufacturing Technology, vol. 105, no. 10, pp. 4443-4460, 2019.
[12] Y.-A. Tsai et al., "Optimization of Lap-Joint Laser Welding Parameters Using High-Fidelity Simulations and Machine Learning Mode," Journal of Materials Research and Technology, 2023.
[13] L. Pellone, G. Inamke, K.-M. Hong, and Y. C. Shin, "Effects of interface gap and shielding gas on the quality of alloy AA6061 fiber laser lap weldings," Journal of Materials Processing Technology, vol. 268, pp. 201-212, 2019.
[14] C. Cross, "On the origin of weld solidification cracking," Hot cracking phenomena in welds, pp. 3-18, 2005.
[15] Y. Sun, H. Cui, X. Tang, and F. Lu, "Characterization and formation mechanism of periodic solidification defects in deep-penetration laser welding of NiCrMoV steel with heavy section," The International Journal of Advanced Manufacturing Technology, vol. 100, pp. 2857-2866, 2019.
[16] J. Mazumder, "Laser welding: state of the art review," Jom, vol. 34, pp. 16-24, 1982.
[17] A. El-Batahgy and M. Kutsuna, "Laser beam welding of AA5052, AA5083, and AA6061 aluminum alloys," Advances in Materials Science and Engineering, vol. 2009, 2009.
[18] J. Elmer, J. Vaja, R. Pong, T. Gooch, and H. Barth, "The Effect of Ar and N2 Shielding Gas on Laser Weld Porosity in Steel, Stainless Steel, and Nickel," Welding Journal, vol. 2015, no. LLNL-JRNL-663819, 2015.
[19] O. Ola and F. Doern, "Factors controlling keyhole-induced porosity in cold wire laser welded aluminum," Journal of Laser Applications, vol. 29, no. 1, p. 012008, 2017.
[20] M. Miyagi, H. Wang, R. Yoshida, Y. Kawahito, H. Kawakami, and T. Shoubu, "Effect of alloy element on weld pool dynamics in laser welding of aluminum alloys," Scientific reports, vol. 8, no. 1, p. 12944, 2018.
[21] C. Yuce, M. Tutar, F. Karpat, N. Yavuz, and G. Tekin, "The effect of process parameters on the microstructure and mechanical performance of fiber laser-welded AA5182 aluminium alloys," Strojniski Vestnik/Journal of Mechanical Engineering, vol. 63, no. 9, pp. 510-518, 2017.
校內:2028-08-15公開