Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/2087
標題: 應用類神經網路於微細切削刀具狀態偵測之研究
Study on the Application of Neural Network for Tool Wear Monitoring in Micro Cutting
作者: 顏嘉良
Yen, Chia-Liang
關鍵字: micro cutting tool;微細刀具;SOFM(SOM);LVQ;自組性特徵映射;學習向量量化
出版社: 機械工程學系所
引用: [1] Teti, R., 1995, ” A review of tool condition monitoring literature database,” Annals of the CIRP, Vol.44(2), pp.659-666. [2] Dan, L. and Mathew, J., 1990, “Tool wear and failure monitoring techniques for turning-a review,” International Journal of Machine Tools and Manufacture, Vol.30(4), pp.579-598. [3] ISO, 3685, 1993, “Tool-Life Testing with Single-point Turning Tools,” ISO 3685:1993(E), International Standard, Second Edition, 1993-11-15. [4] Chao, P. Y. and Hwang Y. D., 1997, “An improved neural network model for the prediction of cutting tool life. Journal of Intelligent Manufacturing,” Vol.8, pp.107-115. [5] Ezugwu, E. O., Arthur, S. J. and Hines, E. L., 1995, “Tool-wear prediction using artificial neural networks,” Journal of Materials Processing Technology, Vol.49, pp.255-264. [6] Li, X and Nee, A. Y. C., 1996, “Monitoring cutting conditions for tool scheduling in CNC machining,” Manufacturing Systems, Vol.25, pp. 377-383. [7] Prasad, K. N. and Ramamoorthy, B., 2001, “Tool wear evaluation by stereo vision and prediction by artificial neural network,” Journal of Materials Processing Technology, Vol.112, pp.43-52. [8] Byrne, G., Dornfeld, D., Inasaki, I., Ketteler, G., König, W. and Teti, R., 1995, “Tool condition monitoring (TCM)-the status of research and industrial application,” Annals of CIRP ,Vol.44(2), pp.541-567. [9] Dimla, E. D., 2000, “Sensor signals for tool-wear monitoring in metal cutting operations-a review of methods,” International Journal of Machine Tools and Manufacture, Vol.40(8), pp.1073-1098. [10] Dornfeld, D.A., 1992, “Application of acoustic emission techniques in manufacturing,” NDT&E International, Vol.25(6), pp.259-269. [11] Dornfeld, D.A., 1994, “In process recognition of cutting states,” JSME International Journal Series C: Dynamics Control, Vol.37(4), pp. 638-650. [12] Jemielniak, K. and Otman, O., 1998, ”Tool failure detection based on analysis of accoustic emission signals,” Journal of Material Processing Technology, Vol.76, pp.192–197. [13] Kakade, S., Vijayaraghavan, L. and Krishnamurthy, R., 1994, “In-process tool wear and chip-form monitoring in face milling operation using acoustic emission,” Journal of Material Processing Technology, Vol.44, pp.207-214. [14] König, W., Kutzner, K. and Schehl, U., 1992, “Tool monitoring of small drills with acoustic emission,” International Journal of Machine Tools and Manufacture, Vol.32(4), pp.487-493. [15] Blum, T. and Inasaki, I., 1990, “A study on acoustic emission from the orthogonal cutting process,” ASME Trans. Journal of Engineering for Industry, Vol.112(3), pp.203-211. [16] Moriwaki, T. and Tobito, M., 1990, “A new approach to automatic detection of life of coated tool based on acoustic emission measurement,” ASME Trans. Journal of Engineering for Industry, Vol. 112(3), pp.212-218. [17] Ravindra, H. V., Srinivasa, Y. G. and Krishnamurthy, R., 1993, ” Modelling of tool wear based on cutting forces in turning,” Wear, Vol.169, pp.25-32. [18] Purushothaman, S. and Srinivasa, Y.G., 1994, “A back-propagation algorithm applied to tool wear monitoring,” International Journal of Machine Tools and Manufacture, Vol.34(5), pp.625-631. [19] Dornfeld, D.A.,1990, “Neural network sensor fusion for tool condition monitoring,” Annals of the CIRP, Vol.39(1), pp.101-105. [20] Kim, J.S. and Lee, B.H., 1991, “An analytical model of dynamic cutting forces in chatter vibration,” International Journal of Machine Tools and Manufacture, Vol.31(3), pp.371-381. [21] Oraby, S.E. and Hayhurst, D.R., 1991, “Development of models for tool wear force relationships in metal cutting,” International Journal of Machine Tools and Manufacture, Vol.332, pp.125-138. [22] Yao, Y., Fang, X.D. and Arndt, G.,1990, “Comprehensive tool wear estimation in finish-machining via multivariate time-series analysis of 3-D cutting forces,” Annals of the CIRP Vol.391, pp.57-60. [23] Yao, Y. and Fang, X.D., 1992, ”Modelling of multivariate time series for tool wear estimation in finish-turning,” International Journal of Machine Tools and Manufacture, Vol.324, pp.495-508. [24] Shi, T. and Ramalingam, S.,1990, “Real-time flank wear sensing,” Winter Annual Meeting of the ASME, PED, Vol.43, pp. 57-170. [25] El-Wardany, T. I., Gao, D. and Elbestawi, M. A.,1996, “Tool condition monitoring in drilling using vibration signature analysis,” International Journal of Machine Tools and Manufacture, Vol.36(6), pp.687-711. [26] Rotberg, J., Braun, S. and Lenz, E., 1987, “Mechanical signature analysis in interrupted cutting,” Annals of the CIRP, Vol.36(1), pp. 249-252. [27] Rotberg, J., Braun, S. and Lenz, E., 1989, ”Vibration generation models for cutting tool monitoring,” Diagnostics, Vehicle Dynamics and Special Topics-ASME Design Engineering Division ,Vol.18(5), pp.1-6. [28] Sadat, A. B., and Raman, S., 1987, ‘‘Detection of Tool Flank Wear Using Acoustic Signature Analysis,’’ Wear, Vol.115, pp.265-272. [29] Delio, T., Tlusty, J., and Smith, S., 1992, ‘‘Use of Audio Signals for Chatter Detection and Control,’’ ASME Trans. Journal of Engineering for Industry, Vol.114, pp.146-157. [30] Trabelsi, H. and Kannatey-Asibu, E. Jr., 1991, ”Pattern-Recognition Analysis of Sound Radiation in Metal Cutting," The International Journal of Advanced Manufacturing Technology, Vol.6, pp.220-231. [31] Lu, M. C. and Kannatey-Asibu, E. Jr., 2002, ” Analysis of sound signal generation due to flank wear in turning,” Journal of Manufacturing Science and Engineering, Transactions of the ASME, Vol.124(4), pp. 799-808. [32] Lu, M. C. and Kannatey-Asibu, E. Jr., 2004, ” Flank wear and process characteristic effect on system dynamics in turning,” Journal of Manufacturing Science and Engineering, Transactions of the ASME, Vol. 126(1), pp.131-140. [33] Anderson, D., 1988, ‘‘Method for Monitoring Cutting Tool Wear During a Machining Operation,’’ The Boeing Company, USA, USP. 04744242. [34] Sarwar, M., Li, J., Penlington, R. and Ahmed, W., 1996, “Development of thermal imaging systems for metal cutting applications,” Proceedings Advanced Manufacturing Processes Systems and Technologies Conference, pp.361-369. [35] Lin, J., 1995, ”Inverse estimation of the tool-work interface temperature in end milling,” International Journal of Machine Tools and Manufacture , Vol.355, pp.751-760. [36] Raman, S., Shaikh, A. and Cohen, P. H., 1992, “A mathematical model for tool temperature sensing,” ASME Computational Methods in Materials Processing, PED, Vol.61, pp.181-193. [37] Noori-Khajavi, A. and Komanduri, R., 1995, “Frequency and time domain analyses of sensor signals in drilling - I. Correlation with drill wear,” International Journal of Machine Tools and Manufacture, Vol.356, pp. 775-793. [38] Zhou, J. M., Andersson, M. and Ståhl, J. E., 1995, ”A system for monitoring cutting tool spontaneous failure based on stress estimation,” Journal of Materials Processing Technology, Vol.48, pp.231. [39] Zhang, D., Dai, S., Han, Y. and Chen, D., 1994, ”On-line monitoring of tool breakage using spindle current in milling,” International Conference Progress of Cutting and Grinding, pp.270-276. [40] Constantinides, N. and Bennett, S., 1987, ”An investigation of methods for on-line estimation of tool wear,” International Journal of Machine Tools and Manufacture, Vol.27(2), pp.225-237. [41] Rangwala, S. and Dornfeld, D., 1987, ”Integration of sensors via neural networks for detection of tool wear states,” Winter Annual Meetings of the ASME, PED, Vol.25, pp.109-120. [42] Sick, B., 2002, ”On-line and indirect tool wear monitoring in turning with artificial neural networks: A review of more than a decade of research,” Mechanical Systems and Signal Processing, Vol.16(4), pp.487-546. [43] Emel, E., 1991, ”Tool Wear Detection by Neural Network Based Acoustic Emission Sensing,” American Society of Mechanical Engineers, Dynamic Systems and Control Division, Vol.28, pp.77-84. [44] Zhou, Q., Hong, G. S. and Rahman, M. A., 1995, “A New Tool Wear Criterion for Tool Condition Monitoring Using Neural Network,” Engineering Applications of Artificial Intelligence, Vol.8(5), pp.579-588. [45] Hong, G. S., Rahman, M. A. and Zhou, Q., 1996, ”Using neural network for tool condition monitoring based on wavelet decomposition,” International Journal of Machine Tools & Manufacture, Vol.36(5), pp. 551-566. [46] ELanayar, S. V. T. and Shin, Y. C., 1995, “Robust tool wear estimation with radial basis function neural networks,” ASME Trans. Journal of Dynamic Systems, Measurement, and Control, Vol.117, pp.459-467. [47] ELanayar, S. V. T. and Shin, Y. C., 1995, “Design and implementation of tool wear monitoring with radial basis function neural networks,” Proceedings of the American Control Conference, pp.1722-1726. [48] Scheffer, C. and Heyns, P.S., 2001, “Development of an adaptable tool condition monitoring system,” Mechanical Systems and Signal Processing, Vol.15(6), pp.1185-1202. [49] Scheffer, C., Kratz, H., Heyns, P.S., Klocke, F., 2003, “Development of a tool wear-monitoring system for hard turning,” International Journal of Machine Tools and Manufacture, Vol.43(10), pp. 973-985. [50] Niu, Y., Wong, Y., and Hong, G., 1998, “Intelligent sensor system approach for reliable tool flank wear recognition,” The International Journal of Advanced Manufacturing Technology, Vol.14(2), pp.77-84. [51] Emel, E., 1991, “Tool wear detection by neural network based acoustic emission sensing”, American Society of Mechanical Engineers, Dynamic Systems and Control Division, Vol.28, pp.79-85. [52] Emel, E. and Kannatey-Asibu, E. Jr., 1989, ”Acoustic Emission and Force sensor fusion for monitoring the cutting process,” International Journal of Mechanical Sciences, Vol.31, pp.795-809. [53] Emel, E. and Kannatey-Asibu, E. Jr., 1988, ”Tool failure monitoring in turning by pattern recognition analysis of AE signals,” ASME Trans. Journal of Engineering for Industry, Vol.110, pp.137-145. [54] Damodarasamy, S. and Raman, S., 1993, “Inexpensive system for classifying tool wear states using pattern recognition,” Wear, Vol.170(2), pp.149-160. [55] Li, X. and Yuan, Z., 1998, ”Tool wear monitoring with wavelet packet transform-fuzzy clustering method,” Wear, Vol.219(2), pp.145-154. [56] Yao, Y., Li, X. and Yuan, Z., 1999, ”Tool wear detection with fuzzy classification and wavelet fuzzy neural network,” International Journal of Machine Tools and Manufacture, Vol.39(10), pp.1525-1538. [57] Wavelet Toolbox User''s Guide, 2008, http://www.mathworks.com/access/helpdesk/help/pdf_doc/wavelet/wavelet_ug.pdf. [58] Cooley, J. W., and Tukey, J. W., 1965, “An algorithm for the machine calculation of complex Fourier series“, Mathematics of Computation, Vol.19(90), pp.297-301. [59] Schilling, R. J. and Harris, S. L., 2005, Fundamentals of Digital Signal Processing Using MATLAB, Thomson. [60] Kohonen, T.,1982, “Self-Organized formation of topologically correct feature maps,” Biological Cybernetics, Vol.43, pp.59-69. [61] 張斐章、張麗秋,2005,類神經網路,東華書局。 [62] Neural Network Toolbox User''s Guide, 2008, http://www.mathworks.com/access/helpdesk/help/pdf_doc/nnet/nnet.pdf [63] 葉怡成,1993,類神經網路模式應用與實作,儒林圖書公司。 [64] Michael Negnevitsky, 2002, Artificial Intelligence: A Guide to Intelligence Systems, Addison Wesley. [65] Hagan, M. T. and Demuth, H. B., 1996, Neural Network Design, Thomson. [66] Kohonen, T., 1986, Learning Vector Quantization for Pattern Recognition, Technical Report TKK-F-A601, Helsinki University of Technology, Finland.
摘要: 
隨著光、機、電、生醫、通訊等產業的蓬勃發展,零件的微型化或是較高的加工的精度要求是加工技術發展的趨勢,而一般傳統加工技術已漸無法滿足此需求,因此微細切削加工之發展漸漸有其必要性,但於微細切削加工過程中,受限材料的選擇刀具極易摩耗,而刀具摩耗對產品精度之影響也較傳統尺寸大,因此刀具狀態偵測系統之發展更形重要,但由於微細切削所產生的訊號能量遠較傳統切削低,因此系統的發展挑戰更高。因此本研究發展微細刀具狀態偵測系統,藉由切削過程中訊號的變化與所建立之模型來辨識所對應之刀具狀態。
本研究發展之微細刀具狀態偵測系統,係以音洩感測器來量測切削過程中材料晶格能量之變化為基本之判別訊號。為了探討不同方法之模型建立對系統辨識之影響度,本研究分別採用兩種方法:群組分離準則以及自組性特徵映射神經網路,來對特徵資料作處理。在分類器的設計方面,則以學習向量量化神經網路建構所需之分類器。研究訊號由音洩感測器在加工過程中取得,實驗刀具為700 直徑之微細銑刀,工件材料為SK2高碳鋼,結果顯示透過群組分離準則可找出與刀具狀態變化最具密切相關之頻帶能量,且經由特徵萃取後再輸入至分類器,對尖刀之辨識成功率為95-99%,對鈍刀則有74-99%不等之辨識成功率,在利用自組性特徵映射神經網路模型處理訊號特徵後再輸入至分類器,則對尖刀之辨識率皆能達到100%,對鈍刀之辨識亦有92%以上之辨識成功率,且由辨識結果可知,利用不同特徵處理方法會造成分類器隨著頻帶寬度設定大小不同而有所影響。

With the fast development in the industries such as the optical, mechatronic, micro-electronic, biomedical, communication, etc., the demand of the miniaturization and the high accuracy processing in manufacturing increases dramatically. However, the conventional machining technology can not meet the demands properly. Therefore, the development of micro-cutting technology draws a much more attention. In the micro-cutting process, the effect of tool wear on the product quality is much more serious than that in the conventional cutting because the limited material choice for the tool which leads to the higher tool wear rate and the usually high accuracy demand for processing quality. Therefore, although it is much more challengeable to develop the tool wear monitoring system than that in the conventional cutting due to low signal energy, it is much more important to establish this kind of system for the micro cutting process.
In this research, the tool wear monitoring system developed for the micro milling process is based on the Acoustic Emission (AE) signal obtained by the AE sensor in the cutting process. Two different ways of feature signal processing (1) class scatter criterion (2) self-organization feature map neural network were investigated for their effect on the classification performance. For the classifier design, Learning Vector Quantification (LVQ) network is used to classify the tool wear condition. The experiment was setup with Sk2 workpiece milled by the micro mill of 700 μm in diameter. The results show that the feature closest to the tool wear can be obtained after calculating the class scatter index for each feature. After putting the chosen features into the LVQ classifier, the classification rate for sharp tool test is from 95% to 99%, as well as from 74% to 99% for worn tool test. With the feature processing by the SOFM and classified by the LVQ algorithms, the classification rates for sharp and worn tool test are 100% and 92%, respectively. Moreover, the effect of bandwidth size of features on classification rate is observed clearly, but it varies for the different feature processing methods.
URI: http://hdl.handle.net/11455/2087
其他識別: U0005-2208200807463600
Appears in Collections:機械工程學系所

Show full item record
 
TAIR Related Article

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.