Please use this identifier to cite or link to this item:
標題: 應用類神經網路與振動訊號之微銑刀具狀態偵測系統開發
Development of tool wear monitoring system in the micro milling using vibration signal and Neural Network
作者: 謝萬澔
Hsieh, Wan-Hao
關鍵字: Micro cutting;微細加工;Neural Network;Tool monitoring system;Wavelet;類神經網路;刀具偵測系統;小波轉換
出版社: 機械工程學系所
引用: [1] Kwak, J.S., 2006, “Application of Wavelet Transform Technique Tool Failure in Turning Operations,” Int J Adv Manuf Technol 28,1078-1083 [2] Tarng, Y. S., and Lee, B. Y., 1999, “Amplitude Demodulation of The Induction Motor Current for the Tool Breakage Detection in Drilling Operations,” Robotics and Computer Integrated Manufacturing 15,313-318. [3]Li, X., 1999 ”On-line Detection of the Breakage of Small Diameter Drills using Current Signature Wavelet Transform,” International Journal of Machine Tools & Manufacture 39,157-164 [4]Li, P.Y., Fang, Y.W., Wang, Y., Yang, M.S., Yuan, Q.L. and Li,Y., 2006, “Time-Frequency Analysis For Cutting Tools Wear Characteristics, ”Proceeding of the Fifth International Conference on Machine Learning and Cybernetics,Dalian,13-16 [5]Elijah, K.A. and Erdal, E., 1987,”Linear Discriminant Function Analysis of Acoustic Emission Signals for Cutting Tool Monitoring,” Mechanical Systems and Signal Processing 1, 333-347. [6]Cus, F. and Zuperl, U., 2005, “Approach to Optimization of Cutting Conditions by using Artificial Neural Networks.” Journal of Materials Processing Technology 173, 281-290. [7]Issam, A.M., 2003,” Drilling Wear Detection and Classification using Vibration Signals and Artificial Neural Network,” International Journal of Machine Tools & Manufacture 43 , 707-720. [8]Patra, K., Pal, S.K. and Bhattacharyya, K., 2006,”Drill Wear Monitoring through Current Signature Analysis using Wavelet Packet Transform and Artifical Neural Network,” Department of Mechanical Engineering, Indian Institute of Technology Kharagpur, Kharagpur-721302, India. [9]Yan, J. and Lee, J., 2007, “A Hybrid Method for On-line Performance Assessment and Life Prediction in Drilling Operation,” Proceedings of the IEEE,International Conference on Automation and Logistics August 18-21,Jinan,China. [10]Pai, P.S., Nagabhushana T.N. and Rao, P.K.R., 2002, ”Flank wear Estimation in Face Milling Based on Radial Basis Function Neural Networks,” Int J Adv Manuf Technol 20,241-247 [11] Tansel, I. N., Arkan, T. T., Bao, W. Y., Mahendrakar, N., Shisler, B., Smith, D. and McCool, M., 1999 “Tool Wear Estimation in Micro-Machining. Part I: Tool Usage-Cutting Force Relationship.” International Journal of Machine Tools & Manufacture 40, 599-608. [12] Tansel, I. N., Arkan, T. T., Bao, W. Y., Mahendrakar, N., Shisler, B., Smith, D. and McCool, M., 1999 “Tool Wear Estimation in Micro-Machining. Part II: Neural-Network-Based Periodic Inspector.” International Journal of Machine Tools & Manufacture 40, 609-620. [13] Obikawa, T. and Shinozuka, J. , 2004, “Monitoring of Flank Wear of Coated Tools in High Speed Machining with a Neural Network ART2,”International Journal of Machine Tools & Manufacture 44,1311-1318 [14] Matlab, 2008,Wavelet Toolbox User’s Guide [15]楊福生,小波變換的工程分析與應用,科學出版社,北京 [16] Daubechies, I. ,1990,” The Trandform, Time-Frequency Localization and Signal Analysis.”IEEE Transation on Information Theory ,vot.36,NO.5 [17] 顏嘉良, 2008,應用類神經網路於微細切削刀具狀態偵測之研究,碩士論文,中興大學機械工程研究所 [18] 陳柏元,2005,應用小波轉換及人工智慧進行配電系統電容切換暫態位置之判斷,碩士論文,中原大學電機工程研究所 [19] 羅明哲,2001,小波轉換於轉子故障與切削顫振偵測之應用,碩士論文,中原大學機械工程研究所
系統開發之實驗刀具為700 m直徑之微銑刀,工件材料為SK2高碳鋼;刀具斷裂實驗結果顯示,振動訊號利用小波轉換與小波包轉換對於系統判斷刀具斷裂成功率皆有相同之水準,因小波轉換所需之運算量較少,故小波轉換應用於判斷振動與刀具狀態之關係為最佳之訊號轉換方式。磨耗實驗中,在相同頻寬30Hz,振動訊號經群組分離準則取5項可使系統準確判斷刀具狀態,擷取4項與3項特徵無法使系統準確判斷刀具狀態,在相同頻寬60Hz,振動訊號經群組分離準則取5項與4項特徵皆保持系統相同判斷水準,其擷取3項特徵可提高系統判斷刀具狀態之成功率,最後輸入雙方向振動訊號之特徵可改善系統準確判斷刀具狀態。

The development of sensing and monitoring system plays an important role in improving the accuracy and stability for machine tool. The micro tool condition monitoring system integrated by sensor system, signal transformation, feature selection, and classifier was developed in this study. A three axis accelerometer was installed on the sensor plate, which fixed on the spindle housing, to collect the vibration signal in the cutting processes. The Wavelet and FFT transformation were used for transforming the time domain signal to the other domains to modify and select the features for tool breakage and tool wear monitoring, respectively. For classification, the back propagation neural network was designed to classify the tool condition.
In order collect the date for training and verifying the system, an experiment was implemented along with 700 m diameter micro mill and SK2 workpiece. The results show that the transformation of vibration signal both by Digital Wavelet and Wavelet Package Transformation provides the same performance for classifying the tool breakage condition. In tool wear monitoring test, selecting 5 features for classification provides a better classification rate than selecting 4 and 3 features in the 30Hz bandwidth feature case. However, selecting 5 and 4 features all provide the better classification rate than 3 features with 60Hz bandwidth features selected. Finally, the multi sensor system was observed to show the better classification performance than only considering signal from a signal sensor.
其他識別: U0005-2008200915190100
Appears in Collections:機械工程學系所

Show full item record

Google ScholarTM


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