臺灣博碩士論文加值系統

由於機械技術的高度發展，對產品品質的需求逐漸增加，一般而言，精密零組件大多經過多次加工製程才能滿足需求，而其所須的精密表面是靠磨削而獲得。然而，為了達成磨削自動化的實現與維持工件品質，則必須能即時判斷砂輪磨損之情形。因此，若能將磨削狀況以多層級的方法分類，確實掌控磨粒磨損的程度，提早告知砂輪是否即將進入不適合磨削狀態而提出警告，必能有效降低時間及成本上的浪費。
本研究的主要目的是對 CBN 磨削製程分類分級，以確實掌控磨削作業中磨粒的變化情形，進而維持產品品質。藉由選擇能類比磨粒在磨削過程變化的聲射訊號及功率訊號，同時對應於磨後工件表面特性，並以支持向量機的分析技術進行砂輪磨削狀態的分類，以探討製程資訊與磨削表面間的關聯。
從研究結果的內容來看，將實驗中所記錄的數據透過相關處理，再依所設定的級別進行分類訓練後，可得到能判別砂輪於磨削進行中其磨削狀態等級之分類模型，而利用此模型辨別於後續加工作業其分類的預測上，能有效的判斷出砂輪當下之磨削狀況；除此之外，同時利用支持向量機的回歸功能可計算出當下砂輪所磨出之工件表面粗糙度值，將有助於提供磨削作業時之參考依據。

Grinding is a manufacturing processes utilized rotary abrasive wheel to generate quality surface. Wheel wears as grinding proceed. A dressing operation is needed to restore the wheel grinding ability as the abrasive particle of wheel worn out. In order to perform automation while keeping the surface quality, wheel status needs to be detected white workpiece is being ground, in order to confirm the guy quality, and decide the timing for dressing operation.
The main objective of this research is to classify the status of CBN grinding process, while the process is being monitored. Acoustic emission signals and grinding power signals are selected for study, which correspond to the variation of abrasives in grinding process. As the signals were collected during the process, it was then classified with support vector machines based ground surface integrity. The relationship between grinding signals and ground surface were then build accuracy experimental result.
The utilization of support vector machines method can efficiently obtain the grinding status model to correctly character grinding status. The model is then used to with regressive model for the result of support vector machines classification, the grinding surface roughness can then predicted with little error, this would be beneficial for grinding process monitor.

目錄
中文摘要 ……………………………………………………………i
英文摘要 ……………………………………………………………ii
致謝 …………………………………………………………………iv
目錄 …………………………………………………………………v
圖目錄 ………………………………………………………………vii
表目錄 ………………………………………………………………ix
符號說明 ……………………………………………………………x
一、緒論 ……………………………………………………………1
1.1 前言 ……………………………………………………………1
1.2 研究動機與目的 ………………………………………………3
1.3 論文架構 ………………………………………………………3
二、文獻探討 ………………………………………………………5
2.1 CBN磨削 …………………………………………………………5
2.1.1 CBN 砂輪和傳統砂輪之分別 ……………………………6
2.1.2 CBN 砂輪之損耗與修整 …………………………………7
2.2 製程監測 ………………………………………………………8
2.3 製程訊號處理與分析 …………………………………………11
2.3.1 訊號處理技術 ……………………………………………11
2.3.2 數據分析方法 ……………………………………………12
三、監測裝置的設計與磨削實驗之規劃 …………………………15
3.1 CBN 磨削監測裝置 ……………………………………………15
3.2 工件表面品質 …………………………………………………17
3.3 感測器 …………………………………………………………21
3.3.1功率訊號 ……………………………………………………22
3.3.2 聲射感測器 ………………………………………………23
3.4特徵資訊的萃取 …………………………………………………26
3.4.1小波分析 …………………………………………………27
3.4.2 AE 特徵資訊 ………………………………………………29
3.5 支持向量機 ……………………………………………………31
3.6 研究方法 ………………………………………………………38
3.7實驗規劃與設備 …………………………………………………41
四、結果與討論 ……………………………………………………49
4.1 工件表面粗糙度的變化 ………………………………………49
4.2 磨削訊號數據 …………………………………………………59
4.3 分類結果 ………………………………………………………61
4.4 分類錯誤點的探討 ……………………………………………72
五、結論與未來展望 ………………………………………………77
5.1 結論 ……………………………………………………………77
5.2 未來展望 ………………………………………………………79
六、參考文獻 ………………………………………………………81
附錄一 分割式分群法 ………………………………………………87
附錄二 表面粗糙度趨勢圖 …………………………………………88

圖目錄
圖3.1 磨削監測裝置概念圖 ………………………………………16
圖3.2 熱傳導率 ……………………………………………………18
圖3.3 Ra 量測示意圖 ……………………………………………19
圖3.4 Rz 量測示意圖 ……………………………………………20
圖3.5 功率訊號與聲射訊號反應在磨削狀態示意圖 ……………22
圖3.6 聲射訊號簡示圖 ……………………………………………23
圖3.7 聲射能量示意圖 ……………………………………………24
圖3.8 一維小波轉換示意圖 ………………………………………27
圖3.9 多層小波轉換示意圖 ………………………………………28
圖3.10 實際利用小波轉換作多層濾波示意圖 ……………………28
圖3.11 磨削之小波分解之能量示意圖 ……………………………30
圖3.12 磨削之AE訊號頻譜圖 ………………………………………30
圖3.13 最佳化超平面示意圖 ………………………………………33
圖3.14 核函數轉換示意圖 …………………………………………36
圖3.15 研究方法及流程 ……………………………………………40
圖3.16 實驗規劃流程圖 ……………………………………………41
圖3.17 平面磨床 ……………………………………………………43

圖3.18 樹脂法 CBN 砂輪 …………………………………………44
圖3.19 實驗用工件材料 (SKD11) …………………………………44
圖3.20 聲射感測器 …………………………………………………46
圖3.21 表面粗度儀 …………………………………………………47
圖4.1 第一組加工條件表面粗糙度趨勢圖 ………………………50
圖4.2 第二組加工條件表面粗糙度趨勢圖 ………………………51
圖4.3 第三組加工條件表面粗糙度趨勢圖 ………………………51
圖4.4 第四組加工條件表面粗糙度趨勢圖 ………………………52
圖4.5 第五組加工條件表面粗糙度趨勢圖 ………………………52
圖4.6 第六組加工條件表面粗糙度趨勢圖 ………………………53
圖4.7 第七組加工條件表面粗糙度趨勢圖 ………………………53
圖4.8 第八組加工條件表面粗糙度趨勢圖 ………………………54
圖4.9 第一組其體積移除相對表面粗糙度之分級 ………………55
圖4.10 第二組其體積移除相對表面粗糙度之分級 ………………55
圖4.11 第三組其體積移除相對表面粗糙度之分級 ………………56
圖4.12 第四組其體積移除相對表面粗糙度之分級 ………………56
圖4.13 第五組其體積移除相對表面粗糙度之分級 ………………57
圖4.14 第六組其體積移除相對表面粗糙度之分級 ………………57
圖4.15 第七組其體積移除相對表面粗糙度之分級 ………………58
圖4.16 第八組其體積移除相對表面粗糙度之分級 ………………58
圖4.17 功率電壓訊號原始圖(vs:31m/s、vw:8m/min、a:10μm) …59
圖4.18 聲射訊號原始圖(vs:31m/s、vw:8m/min、a:10μm) ………60
圖4.19 SVM 分類結果(第一組加工條件) …………………………62
圖4.20 粗糙度值的預測(第一組加工條件)…………………………63
圖4.21 SVM 分類結果(第二組加工條件) …………………………64
圖4.22 粗糙度值的預測(第二組加工條件)…………………………64
圖4.23 SVM 分類結果(第三組加工條件) …………………………65
圖4.24 粗糙度值的預測(第三組加工條件)…………………………65
圖4.25 SVM 分類結果(第四組加工條件) …………………………66
圖4.26 粗糙度值的預測(第四組加工條件)…………………………67
圖4.27 SVM 分類結果(第五組加工條件) …………………………68
圖4.28 粗糙度值的預測(第五組加工條件)…………………………68
圖4.29 SVM 分類結果(第六組加工條件) …………………………69
圖4.30 粗糙度值的預測(第六組加工條件)…………………………69
圖4.31 SVM 分類結果(第七組加工條件) …………………………70
圖4.32 粗糙度值的預測(第七組加工條件)…………………………70
圖4.33 SVM分類結果(第八組加工條件) …………………………71
圖4.34 粗糙度值的預測(第八組加工條件)…………………………71
圖4.35 第三組參數在45000mm3磨削量之粗糙度值的預測 ………73
圖4.36 第六組參數在15000mm3磨削量之粗糙度值的預測 ………74
圖4.37 第六組參數在55000mm3磨削量之粗糙度值的預測 ………75

表目錄
表3.1 磨削加工條件組合 …………………………………………42
表3.2 磨床規格 ……………………………………………………43
表3.3 工件材料之物理特性(20℃) ………………………………45
表4.1 磨削加工條件組別 …………………………………………50
表4.2 磨削數據資料 ………………………………………………60
表4.3 第三組參數在45000mm3磨削量之分類結果 ………………73
表4.4 第六組參數在15000mm3磨削量之分類結果 ………………74
表4.5 第六組參數在55000mm3磨削量之分類結果 ………………74

[1]林維新，1993，CBN磨輪研削加工技術，全華總經銷初版，機械技術出版。
[2]賴耿陽，2001，精密加工新技術全集，復漢出版社印行。
[3]K. Subramanian, R. P. Lindsay, 1992, “A systems approach for the use of vitrified bounded superabrasive wheels for precision production grinding”, Journal of Engineering for Industry，Vol.114, pp.41-52.
[4]蕭瑞聖，1999，“CBN磨輪高速磨輪技術”，機械工業雜誌，5月。
[5]E. Gotou, M. Touge, 1996, “Monitoring of wear of Abrasive Grains”, Journal of Materials Processing Technology, Vol.62, pp.408-414.
[6]M. Younis, M. M. Sadek, T. EI-Wardani, 1987, “A New Approach to Development of a Grinding Force Model”, ASME Journal of Engineering for Industry, Vol.109, pp.306-313.
[7]徐世穹，2002，瓷質CBN砂輪磨削工具鋼性能之分析，華梵大學，碩士論文。
[8]E. S. Lee, N. H. Kim, 2001, “A study on the machining characteristics in the external plunge grinding using the current signal of the spindle motor”, International Journal of Machine Tools and Manufacture, Vol.41, pp.937-951.
[9]X. Li, A. Djordjevich, P. K. Venuvinod, 2000, “Current-sensor-based feed cutting force intelligent estimation and tool wear condition monitoring”, IEEE Transactions on Industrial Electronics, Vol.47, pp.697-702.
[10]陳鴻榮，2002，磨削能量對磨後工件表面品質之探討，國立高雄第一科技大學，碩士論文。
[11]馮昌碩，2003，判斷砂輪再削銳時機之監控系統的建構，國立高雄第一科技大學，碩士論文。
[12]K. Iwata, T. Moriwaki, 1977, “An Application of Acoustic Emission Measurement to In-Process Sensing of Tool Wear”, Annals of CIRP, Vol.26, No.1, pp.21-26.
[13]D. Dornfled, 1980, “Acoustic Emission and Metal Working - Survey of Potential and Examples of Applications”, Proc. 8th North American Metal Working Research Conference, pp.207-213.
[14]E. Kannatey-Asibu, D. A. Dornfeld, 1981, “Quantitative Relationship for Acoustic Emission from Orthogonal Metal Cutting”, ASME Journal of Engineering for Industry, Vol.103, pp.330-340.
[15]D. Dornfled, H. G. Cai, 1984, “An Investigation of Grinding and Wheel Loading Using Acoustic Emission”, Journal of Engineering for Industry, Vol.106, pp.23-28.
[16]T.J. Holroyd, 1986, “Practical Applications of Acoustic Emission Technology”, British Journal of NDT, Vol.28, pp.224-227.
[17]T.W. Hwang, E.P. Whitenton, N.N. Hsu, G.V. Blessing, C.J. Evans, 2000, “Acoustic emission monitoring of high speed grinding of silicon nitride”, Ultransonics, Vol.38, pp.614-619.
[18]H. K. Tönshoff, M. Jung, S. Männel, W. Rietz, 2000, “Using acoustic emission signals for monitoring of production processes”, Ultrasonics, Vol.37, pp.681-686.
[19]P. Srinivasa Pai, P. K. Ramakrishna Rao, 2002, “Acoustic Emission Analysis for Tool Wear Monitoring in Face Milling”, International Journal of Production Research, Vol.40, pp.1081-1093.
[20]S.N.A. Saravananriyan, L. Vijayaraghavan, R. Krishnamurthy, 2001, “On-line Detection of Grinding Burn by Integrated Sensing”, Proceedings of the First ISA/IEEE Conference, pp.89-94.
[21]S. Gu, J. Ni, J. Yuan, 2002, “Non-stationary signal analysis and transient machining process condition monitoring”, International Journal of Machine Tools and Manufacture, Vol.42, pp.41-51.
[22]L. Xiaoli, Y. Zhejun, 1998, “Tool wear monitoring with wavelet packet transform-fuzzy clustering method”, Wear, Vol.219, pp.145-154.
[23]T. Lundholm, 1991, “A flexible real-time adaptive control system for turning”, Annals of CIRP, Vol.40, No.1, pp.441-444.
[24]D. A. Dornfeld, 1990, “Neural Network Sensor Fusion for Tool Condition Monitoring”, Annals of the CIRP, Vol. 39, pp.101-105.
[25]S. Spiewak, 1991, “A predictive monitoring and diagnosis system for manufacturing”, Annals of the CIRP, Vol.40, No.1, pp.401-404.
[26]許家旗，2002，智慧型球銑刀銑削過程刀具磨耗之偵測與補償系統，大同大學，碩士論文。
[27]M. Misra, H. H. Yue, S. J. Qin, C. Ling, 2002, “Multivariate process monitoring and fault diagnosis by multi-scale PCA”, Computers and Chemical Engineering, Vol.26, pp.1281-1293.
[28]J. Chen, C. M. Liao, 2002, “Dynamic process fault monitoring based on neural network and PCA”, Journal of Process Control, Vol.12, pp.277-289.
[29]J. Sun, M. Rahman, Y.S. Wong, G.S. Hong, 2004, “Multiclassification of tool wear with support vector Machine by manufacturing loss consideration”, International Journal of Machine Tools and Manufacture, Vol.44, pp.1179-1187.
[30]X. K. Wei, B. LU, C. Wang, J. M. LU, Y. H. LI, 2004, “Applications of support vector machines in aeroengine fault diagnosis”, Journal of Aerospace Power, Vol.19, pp.844-848.
[31]X. K. Wei, Y. H. LI, S. Wang, J. M. LU, C. Wang, 2004, “Aeroengine lubrication monitoring analysis via support vector machines”, Journal of Aerospace Power, Vol.19, pp.392-397.
[32]J. Yang, Y. Zhang, R. Zhao, 2005, “Application of Support Vector Machines in Trend Prediction of Vibration Signal of Mechanical Equipment”, Journal of XI’AN Jiaotong University, Vol.39, No.9, pp.950-953.
[33]M. Field, J. F. Kahles, 1971, “Review of Surface Integrity of Machined Components”, Annals of the CIRP, Vol.20, pp.153-163.
[34]廖運炫，1990，磨削加工之表面整合，機械工業雜誌，5月。
[35]R.S. Hahn, 1976, “On the Loss of Surface Integrity and Surface Form due to Thermoplastic Stress in Grinding Operation”, Annals of the CIRP, Vol.25, No.1, pp.203-207.
[36]K. Takazawa1, 1964, “The Ratio of Heat Entering the Work to Total Grinding Heat–Theoretical Analysis on the Grinding Temperature”, J. Japan Society of Precision Engineering, Vol.30, No.12, p.914.
[37]K. Yokoyama, R. Ichimiya, 1977, “Thermal Deformation of Workpiece in Surface Grinding”, Bull. Japan Soc. Precis. Eng., Vol.11, No.4, pp.195-201.
[38]C. Guo, 1993, Investigation of Fluid Flow and Heat Transfer in Grinding, University of Massachusetts at Amherst.
[39]D.J. Whitehouse, 2001, “Fractal or fiction”, Wear, Vol.249, pp.345-353.
[40]吳杉堯，1993，精密量具及機械檢驗實習，大揚出版社。
[41]陳鴻榮，2002，磨削能量對磨後工件表面品質之探討，國立高雄第一科技大學，碩士論文。
[42]B. Karpuschewski, M. Wehmeier, I. Inasaki, 2000, “Grinding Monitoring System Based on Power and Acoustic Emission Sensors”, Annals of the CIRP, Vol.49, No.1, pp.235-240.。
[43]E. S. Lee, N.H. Kim, 2001, “A study on the machining characteristics in the external plunge grinding using the current signal of the spindle motor”，International Journal of Machine Tools and Manufacture, Vol.41, pp.937-951.
[44]R. K. Miller, P. Mclntire, 1987, “Acoustic Emission Testing”, American Society for Nondestructive Testing, Vol.5.
[45]楊明輝，1992，“第六屆非破壞性檢測技術研討會論文集”，中華民國非破壞檢測協會，頁36-39頁。
[46]劉時風，1996，焊接缺陷聲發射檢測信號譜估計及人工神經網路模式識別研究，清華大學機械工程系，博士論文，北京。
[47]M. Misiti, 2000, Wavelet Toolbox For Use with MATLAB, Matlab user’s guide, USA.
[48]X. Li, 2002, “A brief review：acoustic emission method applied for tool wear monitoring during turning”, International Journal of Machine Tools and Manufacture, Vol.42, pp.157-165.
[49]R. Y. Chiou, S. Y. Liang, 2000, “Dynamic modeling of cutting acoustic emission via piezo-electric actuator wave control”, International Journal of Machine Tools and Manufacture, Vol.40, pp.641-659.
[50]徐美珍，2004，企業財務危機之預測，國立政治大學，碩士論文。
[51]V. Vapnik, 1995, “The Nature of Statistical Learning Theory”, Springer.
[52]劉乃揚，田英傑，2004，數據挖掘中的新方法-支持向量機，科學出版社，北京。
[53]C. Nello, S. T. John, 2000, An Introduction to Support Vector Machines and Other Kernel-based Learning Methods, ISBN 7-5053-9336-7.
[54]M. A. Aizerman, E. M. Braverman, L. Rozonoer, 1964, “Theoretical foundations of the potential function method in pattern recognition learning”, Automation and Remote Control, Vol.25, pp.821-837.
[55]E. Forgy, 1965, “Cluster analysis of multivariate data : Efficiency versus interpretability of classifications”, Biometerics, Vol.21, pp.768-169.