簡易檢索 / 詳目顯示

回結果列表
研究生： 曹恩豪
Tsau, En-Hao
論文名稱： 運用雲端與非監督式深度學習方法智慧化診斷在改變加工條件下加工表面品質之研究
Intelligent Diagnosis for the Machined Surface Quality Using the Cloud System and the Unsupervised Deep Learning Method and under Different Cutting Conditions
指導教授： 林仁輝
Lin, Jen Fin
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： 2022
畢業學年度： 110
語文別： 中文
論文頁數： 166
中文關鍵詞： 非監督式深度學習 、卷積自編碼器 、鋸帶性能衰退分析 、雲端智慧化診斷
外文關鍵詞： Unsupervised deep learning, Convolutional autoencoder, Analysis of blade wear degradation, Intelligent diagnosis cloud system
相關次數： 點閱：72  下載：4 
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

    • 隨著機械加工產業的進步，精密度(Accuracy)為加工產品好壞之關鍵因素。帶鋸床(Band saw machine)為加工製程之前端設備，將金屬材料鋸切至合適的大小後，再進行後續加工。若在前端加工對於加工表面之精度有一定程度的控管，不僅可以減少材料的浪費，同時也能夠減少後續加工所需的時間與成本。隨著人工智慧興起，使得訊號處理有大幅度的發展，影響加工表面品質之主要原因為鋸帶性能衰退，透過振動訊號之分析，即時的診斷當下鋸帶之運作狀況，在發生輕微異常現象時馬上發出警報，通知現場操作人員處理，既能保持產品品質也能提高生產效率。
    本研究有兩種實驗方式：變切削實驗與定切削實驗，兩種實驗皆使用帶鋸床鋸切S45C圓棒，各鋸切1000片工件。在定切削實驗過程中，加工條件固定不變；在變切削實驗過程中，會改變鋸切之材料直徑。接著利用非監督式卷積自編碼器(Convolutional autoencoder)，將帶鋸床運作之振動訊號代入模型中進行分析，得到振動訊號健康指數(〖HI〗_vib)。發現鋸帶之磨耗與衰退行為主要會反應在側向振動(Lateral vibration)，由於側向振動增大，使得鋸帶無法平整的將材料移除，導致開始有材料殘留於加工表面上形成小凸丘(Hills)，加工表面品質下降。分析結果顯示當加工表面開始有小凸丘形成時，〖HI〗_vib 能夠即時的反應出來。因此，本研究將 〖HI〗_vib 之分析結果傳輸至雲端，建立雲端智慧化診斷系統，利用局部斜率作為判斷依據，當局部斜率有急遽上升之現象時，代表此時的 〖HI〗_vib 有明顯的下降趨勢，雲端智慧化診斷系統便會即時發出警報。結果顯示本研究之雲端智慧化診斷系統可以在鋸帶有輕微異常，加工表面開始有小凸丘產生時，即時的發出警報，能應用於實際加工條件變動的加工場域。相較於文獻[25]使用之分析方法與結果，本研究使用之分析方法與模型具有訓練模型時間很短、振動訊號健康指數(〖HI〗_vib)之分析結果較準確與穩定、小凸丘面積比例(A_hills)分析效率大幅提升以及能應用於加工條件變動的實際加工場域之優勢。

    Bandsaw machine is the front equipment of the machining process to cut the material into required dimensions. Controlling the machined surface accuracy at this process stage could reduce the waste of material and the cost required for subsequent processing. The main reason affecting the machined surface quality is the deterioration of the blade. Therefore, this study focuses on the wear degradation assessment of bandsaw blade and establishes an intelligent diagnosis cloud system. We used an unsupervised deep learning model named convolutional autoencoder to analyze two types of experiments. By substituting the vibration signal into the model, we could obtain the healthy index of vibration (〖HI〗_vib). The analysis result shows that 〖HI〗_vib could respond immediately when there is residual material on the machined surface at the beginning. Finally, we established an intelligent diagnosis cloud system and quantified the decline of 〖HI〗_vib with a local slope. With the result of this study, the intelligent diagnosis cloud system could issue an alarm immediately when the blade has slight abnormality. Comparing with the literature [25], the analysis methods and model used in this study had several advantages. The training time of the model is shorter, and the analysis result of 〖HI〗_vib is more accurate and stable. The analysis efficiency of the area percentage with hills (A_hills) is greatly improved. It could be applied to the actual processing field with different cutting conditions.

    摘要 I Extended Abstract III 誌謝 VI 目錄 VII 表目錄 XII 圖目錄 XIII 第一章 緒論 1 1-1 前言 1 1-2 文獻回顧 2 1-3 研究目的 7 1-4 論文架構 10 第二章 基本理論 12 2-1 振動訊號處理 12 2-1-1 快速傅立葉轉換(Fast Fourier Transform, FFT) 12 2-1-2 短時距傅立葉變換(Short-Time Fourier Transform, STFT) 17 2-1-3 特徵頻率(Characteristic Frequency) 18 2-1-3-1 側向頻率(Lateral frequency) 18 2-2 影像處理 19 2-2-1 顏色空間RGB、HSV以及灰階(Gray Scale) 19 2-2-2 結構相似性指標(Structural Similarity Index, SSIM Index) 22 2-2-3 影像成像 23 2-3 機器學習 24 2-3-1 機器學習理論 24 2-3-2 人工類神經網路(Artificial Neural Network, ANN) 27 2-3-2-1 神經元(Neurons) 27 2-3-2-2 激活函數(Activation function) 29 2-3-2-3 梯度下降法(Gradient Descent) 30 2-3-2-4 反向傳播法(Back Propagation) 32 2-3-3 深度學習(Deep Learning) 36 2-3-3-1 自編碼器(Autoencoder) 37 2-3-3-2 卷積神經網路(Convolutional Neural Network, CNN) 39 2-3-3-3 卷積自編碼器（Convolutional Autoencoder） 41 2-4 皮爾森相關係數(Pearson Correlation Coefficient) 43 第三章 試驗方法與資料擷取 61 3-1 實驗設備 61 3-1-1 實驗機台簡介 61 3-1-2 切削刀具簡介 61 3-1-3 量測儀器簡介 62 3-1-3-1 加速規 62 3-1-3-2 應變規 62 3-1-3-3 相機 63 3-1-3-4 表面粗度儀 63 3-2 試驗內容 64 3-2-1 試驗條件與參數設置 64 3-2-2 加速規訊號之擷取 66 3-2-3 應變規訊號之擷取 67 3-2-4 加工表面形貌之影像擷取 68 3-2-5 表面粗糙度之量測 68 3-3 Microsoft Azure 簡介與服務介紹 68 3-3-1 IOT Hub 69 3-3-2 Data lake storage 69 3-3-3 Machine learning service 70 3-3-4 Container 70 3-3-5 Power BI 71 3-3-6 雲端服務平台之架構 72 第四章 結果與討論 86 4-1 變切削實驗之分析 87 4-1-1 變切削實驗之振動訊號分析 87 4-1-1-1 變切削實驗之卷積自編碼器之超參數設置 87 4-1-1-2 變切削實驗之振動訊號前處理結果 89 4-1-1-3 變切削實驗之卷積自編碼器分析結果 91 4-1-1-4 變切削實驗之振幅分析 98 4-1-1-5 變切削實驗之頻譜分析 98 4-1-2 變切削實驗之加工表面影像分析 100 4-1-2-1 變切削實驗之影像前處理結果 101 4-1-2-2 變切削實驗之分析表面缺陷結果 101 4-1-2-3 變切削實驗之表面粗糙度結果 104 4-2 定切削實驗之分析 105 4-2-1 定切削實驗之振動訊號分析 105 4-2-1-1 定切削實驗之振動訊號前處理結果 106 4-2-1-2 定切削實驗之卷積自編碼器之分析結果 106 4-2-1-3 定切削實驗之振幅分析 108 4-2-1-4 定切削實驗之頻譜分析 108 4-2-1-5 與文獻[25]比較定切削實驗之振動訊號健康指數分析結果 109 4-2-2 定切削實驗之加工表面影像分析 110 4-2-2-1 定切削實驗之影像前處理結果 110 4-2-2-2 定切削實驗之分析表面缺陷之結果 111 4-3 雲端智慧化診斷系統之判斷方法與結果 112 第五章 結論與未來展望 157 5-1 結論 157 5-2 未來展望 160 參考文獻 161

    [1] S. Söderberg, L. Åhman, and M. Svenzon, "A metallurgical study of the wear of band-saw blades," Wear, vol. 85, no. 1, pp. 11-27, 1983.
    [2] M. Sarwar, M. Persson, and H. Hellbergh, "Wear and failure modes in the bandsawing operation when cutting ball-bearing steel," Wear, vol. 259, no. 7-12, pp. 1144-1150, 2005.
    [3] N. Zhu, C. Tanaka, T. Ohtani, and H. Usuki, "Automatic detection of washboarding in bandsaws," Journal of wood science, vol. 47, no. 2, pp. 102-108, 2001.
    [4] C. Tanaka, Y. Shiota, A. Takahashi, and M. Nakamura, "Experimental studies on band saw blade vibration," Wood Science and Technology, vol. 15, no. 2, pp. 145-159, 1981.
    [5] P. Gendraud, J.-C. Roux, and J.-M. Bergheau, "Vibrations and stresses in band saws: A review of literature for application to the case of aluminium-cutting high-speed band saws," Journal of materials processing technology, vol. 135, no. 1, pp. 109-116, 2003.
    [6] M. Sarwar, M. Persson, H. Hellbergh, and J. Haider, "Measurement of specific cutting energy for evaluating the efficiency of bandsawing different workpiece materials," International Journal of Machine Tools and Manufacture, vol. 49, no. 12-13, pp. 958-965, 2009.
    [7] X. Li, A. Djordjevich, and P. K. Venuvinod, "Current-sensor-based feed cutting force intelligent estimation and tool wear condition monitoring," IEEE Transactions on Industrial Electronics, vol. 47, no. 3, pp. 697-702, 2000.
    [8] C. Madhusudana, H. Kumar, and S. Narendranath, "Face milling tool condition monitoring using sound signal," International Journal of System Assurance Engineering and Management, vol. 8, no. 2, pp. 1643-1653, 2017.
    [9] A. Azmi, "Monitoring of tool wear using measured machining forces and neuro-fuzzy modelling approaches during machining of GFRP composites," Advances in Engineering Software, vol. 82, pp. 53-64, 2015.
    [10] M. Nouri, B. K. Fussell, B. L. Ziniti, and E. Linder, "Real-time tool wear monitoring in milling using a cutting condition independent method," International Journal of Machine Tools and Manufacture, vol. 89, pp. 1-13, 2015.
    [11] H. Zhang, J. Zhao, F. Wang, J. Zhao, and A. Li, "Cutting forces and tool failure in high-speed milling of titanium alloy TC21 with coated carbide tools," Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, vol. 229, no. 1, pp. 20-27, 2015.
    [12] M. He and D. He, "A new hybrid deep signal processing approach for bearing fault diagnosis using vibration signals," Neurocomputing, vol. 396, pp. 542-555, 2020/07/05/ 2020.
    [13] E. Kayahan, H. Oktem, F. Hacizade, H. Nasibov, and O. Gundogdu, "Measurement of surface roughness of metals using binary speckle image analysis," Tribology International, vol. 43, no. 1-2, pp. 307-311, 2010.
    [14] S. Dutta, A. Kanwat, S. Pal, and R. Sen, "Correlation study of tool flank wear with machined surface texture in end milling," Measurement, vol. 46, no. 10, pp. 4249-4260, 2013.
    [15] S. Dutta, S. K. Pal, S. Mukhopadhyay, and R. Sen, "Application of digital image processing in tool condition monitoring: A review," CIRP Journal of Manufacturing Science and Technology, vol. 6, no. 3, pp. 212-232, 2013/01/01/ 2013.
    [16] G. E. Hinton and R. Zemel, "Autoencoders, minimum description length and Helmholtz free energy," Advances in neural information processing systems, vol. 6, 1993.
    [17] K. Fukushima, "Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position," Biological Cybernetics, vol. 36, no. 4, pp. 193-202, 1980/04/01 1980, doi: 10.1007/BF00344251.
    [18] J. Masci, U. Meier, D. Cireşan, and J. Schmidhuber, "Stacked convolutional auto-encoders for hierarchical feature extraction," in International conference on Artificial Neural Networks, 2011: Springer, pp. 52-59.
    [19] R. Zhao, R. Yan, Z. Chen, K. Mao, P. Wang, and R. X. Gao, "Deep learning and its applications to machine health monitoring," Mechanical Systems and Signal Processing, vol. 115, pp. 213-237, 2019.
    [20] S. Khan and T. Yairi, "A review on the application of deep learning in system health management," Mechanical Systems and Signal Processing, vol. 107, pp. 241-265, 2018.
    [21] 陳冠宇, "運用非監督式深度學習方法分析滾珠螺桿傳動系統性能之智慧化診斷," 成功大學機械工程學系學位論文, pp. 1-144, 2020.
    [22] 謝祐禾, "運用非監督式學習診斷刀具磨耗階段之研究," 成功大學機械工程學系學位論文, pp. 1-186, 2020.
    [23] J. Ma, D. Luo, X. Liao, Z. Zhang, Y. Huang, and J. Lu, "Tool wear mechanism and prediction in milling TC18 titanium alloy using deep learning," Measurement, vol. 173, p. 108554, 2021/03/01/ 2021.
    [24] Q. Qian, Y. Qin, Y. Wang, and F. Liu, "A new deep transfer learning network based on convolutional auto-encoder for mechanical fault diagnosis," Measurement, vol. 178, p. 109352, 2021/06/01/ 2021.
    [25] 魏新偉, "運用雲端深度學習智慧化診斷系統分析鋸床鋸帶磨耗衰退性能之研究," 成功大學機械工程學系學位論文, pp. 1-149, 2021.
    [26] J. W. Cooley and J. W. Tukey, "An algorithm for the machine calculation of complex Fourier series," Mathematics of computation, vol. 19, no. 90, pp. 297-301, 1965.
    [27] 李天龍, "以 FFT 為架構建立之諧波參數建立方法," 撰者, 2000.
    [28] H. Tao, P. Wang, Y. Chen, V. Stojanovic, and H. Yang, "An unsupervised fault diagnosis method for rolling bearing using STFT and generative neural networks," Journal of the Franklin Institute, vol. 357, no. 11, pp. 7286-7307, 2020/07/01/ 2020.
    [29] A. Eschler, "Stresses and vibrations in bandsaw blades," University of British Columbia, 1982.
    [30] Wikiwand. "三原色光模式." https://reurl.cc/A75eme, 2022/04/18 (accessed.
    [31] ITREAD. "顏色空間RGB與HSV(HSL)的轉換." https://www.itread01.com/content/1549111148.html, 2022/04/18 (accessed.
    [32] W. Wang, Z. Chen, X. Yuan, and X. Wu, "Adaptive image enhancement method for correcting low-illumination images," Information Sciences, vol. 496, pp. 25-41, 2019.
    [33] C. Saravanan, "Color image to grayscale image conversion," in 2010 Second International Conference on Computer Engineering and Applications, 2010, vol. 2: IEEE, pp. 196-199.
    [34] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, "Image quality assessment: from error visibility to structural similarity," IEEE Transactions on Image Processing, vol. 13, no. 4, pp. 600-612, 2004.
    [35] S. Albawi, T. A. Mohammed, and S. Al-Zawi, "Understanding of a convolutional neural network," in 2017 international conference on engineering and technology (ICET), 2017: Ieee, pp. 1-6.
    [36] W. S. Noble, "What is a support vector machine?," Nature biotechnology, vol. 24, no. 12, pp. 1565-1567, 2006.
    [37] N. Zhang, S. Ding, J. Zhang, and Y. Xue, "An overview on restricted Boltzmann machines," Neurocomputing, vol. 275, pp. 1186-1199, 2018.
    [38] J. An and S. Cho, "Variational autoencoder based anomaly detection using reconstruction probability," Special Lecture on IE, vol. 2, no. 1, pp. 1-18, 2015.
    [39] S. C. Johnson, "Hierarchical clustering schemes," Psychometrika, vol. 32, no. 3, pp. 241-254, 1967.
    [40] W. S. McCulloch and W. Pitts, "A logical calculus of the ideas immanent in nervous activity," The bulletin of mathematical biophysics, vol. 5, no. 4, pp. 115-133, 1943.
    [41] D. O. Hebb, "The organization of behavior; a neuropsycholocigal theory," A Wiley Book in Clinical Psychology, vol. 62, p. 78, 1949.
    [42] S. Sharma, S. Sharma, and A. Athaiya, "Activation functions in neural networks," Towards Data Science, vol. 6, no. 12, pp. 310-316, 2017.
    [43] K. Hara, D. Saito, and H. Shouno, "Analysis of function of rectified linear unit used in deep learning," in 2015 international joint conference on neural networks (IJCNN), 2015: IEEE, pp. 1-8.
    [44] S. Ruder, "An overview of gradient descent optimization algorithms," arXiv preprint arXiv:1609.04747, 2016.
    [45] V. K. Ojha, A. Abraham, and V. Snášel, "Metaheuristic design of feedforward neural networks: A review of two decades of research," Engineering Applications of Artificial Intelligence, vol. 60, pp. 97-116, 2017.
    [46] S.-L. Chang, L.-S. Chen, Y.-C. Chung, and S.-W. Chen, "Automatic license plate recognition," IEEE transactions on intelligent transportation systems, vol. 5, no. 1, pp. 42-53, 2004.
    [47] H. Singh, Practical Machine Learning and Image Processing: For Facial Recognition, Object Detection, and Pattern Recognition Using Python. Springer, 2019.
    [48] K. Greff, R. K. Srivastava, J. Koutník, B. R. Steunebrink, and J. Schmidhuber, "LSTM: A search space odyssey," IEEE transactions on neural networks and learning systems, vol. 28, no. 10, pp. 2222-2232, 2016.
    [49] P. Baldi, "Autoencoders, unsupervised learning, and deep architectures," in Proceedings of ICML Workshop on Unsupervised and Transfer Learning, 2012: JMLR Workshop and Conference Proceedings, pp. 37-49.
    [50] K. O'Shea and R. Nash, "An introduction to convolutional neural networks," arXiv preprint arXiv:1511.08458, 2015.
    [51] Y. Zhang, "A better autoencoder for image: Convolutional autoencoder," in ICONIP17-DCEC. Available online: http://users.cecs.anu.edu.au/Tom.Gedeon/conf/ABCs2018/paper/ABCs2018_paper_58.pdf (accessed on 23 March 2017), 2018.
    [52] H. Akoglu, "User's guide to correlation coefficients," Turkish journal of emergency medicine, vol. 18, no. 3, pp. 91-93, 2018.
    [53] Cosen. "Cosen Saws." https://www.cosen.com/product/band-saws/snc%EF%BC%8Fnc--programmable-automatic-mass-production/c-320gnc, 2022/04/18 (accessed.
    [54] Amada. "SGLB Saw Blades." https://amadamca.com/?dispatch=products.view&product_id=16, 2022/04/18 (accessed.
    [55] PicoCoulomB. "PCB Model 356A33." https://www.pcb.com/products?model=356a33, 2022/04/18 (accessed.
    [56] 智盛生物科技. "PolyChrome 20MP." https://www.pentagontek.com/, 2022/04/18 (accessed.
    [57] 三朋儀器股份有限公司. "ET-4000." https://www.sanpany.com.tw/, 2022/04/18 (accessed.
    [58] MEADinfo. "JIS S45C-Mild Steel-An Overview." https://www.meadinfo.org/2010/03/s45c-jis-mechanical-properties.html, 2022/04/18 (accessed.
    [59] M.Sarwar, M.Persson, and H.Hellbergh, Proceedings of the 34th International MATADOR Conference. University of Manch-ester Institute of Science avd Technology UMIST, 2004, pp. 103-110.
    [60] P. Li et al., "A novel scalable method for machine degradation assessment using deep convolutional neural network," Measurement, vol. 151, p. 107106, 2020/02/01/ 2020.

    下載圖示 校內：立即公開
    校外：2025-09-01公開
    QR CODE