臺灣博碩士論文加值系統

摘要

本論文利用小波理論、奇異值分解、類神經網路與網路服務等技術發展了一個「智慧型診斷與預兆偵測系統」。首先，我們利用小波理論與奇異值分解技術設計了一套訊號前處理機制，以有效降低資料維度與相關之運算量，包括資料擷取、消除雜訊、資料壓縮與特徵擷取等步驟。其次，我們利用倒傳遞類神經網路技術建構智慧型診斷與預兆偵測功能，以進行設備失效的診斷與失效前剩餘時間的預測。另ㄧ方面，本系統採用三層式網路架構來部署所須的功能元件。第一層為前端控制器負責即時診斷工作；中間層為安全暨預兆偵測伺服器負責預測設備失效與用戶之認證授權；第三層為用戶端，因本系統利用新一代系統整合技術-網路服務來建構通訊的基礎建設，因此使用者可在任何時間，從任何地點利用任何可上網的裝置來操作系統之各項功能。最後，我們實際建置了一套PLC-Based的馬達驅動平台，也設計了齒輪的失效樣本及進行Run-to-Failure試驗，以驗證本論文所提「智慧型診斷與預兆偵測系統」的有效性及實用性。經過系統整合測試與效能分析，結果顯示本系統可讓使用者從遠端即時地監測機械設備的運作狀態，同時也可準確地分辨齒輪故障模式及預測設備的失效趨勢。

ABSTRACT
In this thesis, wavelet, singular value decomposition (SVD), neural network, and Web Services are applied to develop an intelligent diagnostics and prognostics system (IDPS). First, we use wavelet and SVD technologies to design a data preprocess mechanism for effectively reducing data dimensions and the associated computational volumes. Specifically, the data preprocess mechanism includes four steps: data acquisition, noise elimination, data compression, and feature extraction. Next, the back propagation neural network is used to construct the intelligent diagnostics and prognostics functions that are able to diagnose equipment failures and predict the time to failure. On the other hand, the proposed system adopts a three-tiers network architecture to deploy the desired functional components. The first tier is the front-end controller, which is responsible for real-time diagnostic tasks of equipment failures. The middle tier is the security and prognosis server, which is in charge of predicting equipment failures, as well as authenticating and authorizing the clients. The third tier is the remote client side. Since the Web Services technology is utilized to establish the communication infrastructure of the system, the users can then employ any Web-enable device to operate the system functions at any time from anywhere. Finally, we construct a PLC-based motor-driven platform, design failure samples of gears, and conduct run-to-failure tests to verify the effectiveness and practicability of the proposed IDPS. The testing results show that the IDPS can allow the users to monitor the operational statuses of equipment in real time from remote sites. It can also accurately identify the failure modes of gears and predict the failure trend of equipment.

目錄

誌謝 i
摘要 ii
ABSTRACT iii
目錄 iv
表目錄 viii
圖目錄 viiii
1. 緒論 1
1.1研究背景 1
1.2研究動機與目的 5
1.3 論文結構 7
2. 應用技術 9
2.1小波理論 9
2.1.1小波轉換 9
2.1.2小波函數 9
2.1.3離散小波轉換 11
2.1.4 小波轉換多重解析度分析 12
2.2 奇異值分解法(SVD, Singular Value Decomposed) 14
2.3 類神經網路 15
2.3.1 處理單元(Processing Element) 15
2.3.2 層(layer) 16
2.3.3 網路(network) 17
2.3.4 倒傳遞類神經網路 18
2.4 網路服務技術 19
2.4.1 HTTP (Hyper Text Transfer Protocol) 20
2.4.2 SOAP 20
2.4.3 WSDL 21
2.4.4 可延伸參數語言技術 21
2.5 網路資訊安全技術 21
2.5.1 SSL(Secure Sockets Layer)] 21
2.5.2 XML簽章 22
2.5.3 XML加密 23
2.5.4 單一認證(SSO, Single Sign-On)服務 24
2.6 統一塑模語言 25
3. 智慧型診斷與預兆偵測系統技術之開發 29
3.1資料前處理(Data Pre-Processing) 30
3.1.1小波轉換(Wavelet Transformation) 30
3.1.2 門檻化(Thresholding) 35
3.1.3特徵值擷取(Feature Extraction) 36
3.2智慧型診斷及預兆偵測(Intelligent Diagnostics / Prognostics) 38
4. 智慧型診斷與預兆偵測系統架構設計 48
4.1系統需求與架構設計 48
4.2 系統運作流程 49
4.2.1區域伺服端訊號擷取及智慧型診斷與預兆偵測流程 51
4.2.2 遠方用戶端監測智慧型診斷與預兆偵測結果流程 52
4.3 軟體元件之開發 55
4.3.1區域伺服端訊號擷取及智慧型診斷與預兆偵測 55
4.3.2 遠方用戶端監測智慧型診斷與預兆偵測結果 64
5.系統硬體元件 72
5.1加速規(Accelerometer) 73
5.2資料擷取卡 74
5.3 可程式邏輯控制馬達平台 76
6.系統實作與整合測試 82
6.1 開發環境：軟體部分 82
6.2 開發環境：硬體部分 85
6.3實作測試結果 85
6.3.1小波特徵擷取結果 85
6.3.2奇異值分解法測試結果 87
6.3.3倒傳遞類神經網路學習測試 88
6.3.4類神經網路推論引擎測試 89
6.3.5 預兆偵測實驗結果 90
6.4 系統整合測試 94
6.4.1 前端控制器診斷功能 94
6.4.2安全暨預兆偵測伺服器預兆偵測功能 95
6.4.3遠方用戶端監測診斷及預兆偵測結果功能 98
7. 結論 102
7.1 論文總結 102
7.2 研究成果 102
7.3 未來研究方向 103
參考文獻 104
自傳 109

[1]Devaney M. J. and Eren, L., “Detecting Motor Bearing Faults,” IEEE Instrumentation & Measurement Magazine, Volume 7, Issue 4, pp. 30-50, December, 2004.
[2]Benko, U., Pentrovcic, J., Juricic, D., Tavcar, J.,and Rejec, J., “An Approach to Fault Diagnosis of Vacuum Clean Motor Based on Sound Analysis,” Mechanical System and Signal Processing, Vol.19, pp.427-445, 2005.
[3]Wang, W., “Some New Signal Processing Approaches for Gear Fault Diagnosis,” Proceedings of the Fifth International Symposium on Signal Processing and Its Applications, Brisbane, Australia, Vol. 2, pp. 587-590, August 22-25, 1999.
[4]Hochmann, D. and Bechhoefer, E., “Gear Tooth Crack Signal and their Detection via the FM4 Measure in Application for a Helicopter (HUMS) Health Usage and Management System,” Proceedings of IEEE Aerospace Conference ,Vol. 7, pp. 3313-3326, Montana, USA, March, 2003.
[5]Dempsey, J. and Afjeh, A., “Decision Fusion Analysis Detects Spiral Bevel Gear Damage,” IEEE AES System Magazine, Vol. 18, Issue 9, pp. 11-17, September, 2003.
[6]Samanta, B. and Balushi, K. R., “Artificial Neural Network Based Fault Diagnostics of Rolling Element Bearings Using Time-Domain Features,” Journal of Mechanical System and Signal Processing, Vol. 17, pp. 317-328, March, 2003.
[7]Wang, W., “Linear Model Identification for Gear Fault Detection Using Higher Order Statistics and Inverse Filter Criteria,” Proceedings of the Sixth International Symposium on Signal and its Applications, Kuala Lumpur, Malaysia, Vol. 1, pp. 371-374, August 13-16, 2001.
[8]Bai, M., Huang, J., Hong, M., and Su, F., “Fault Diagnosis of Rotating Machinery Using an Intelligent Order Tracking System,” Journal of Sound and Vibration, Vol. 280, pp. 699-718, February, 2005.
[9]Zhang, D., Du, R., and Shi, T., “Extracting Gear Fault Features Using Integrated Bispectrum,” Proceedings of the 2003 IEEE International Conference on Robotics, Changsha, China, Vol. 1, pp. 548-553, October, 2003.
[10]Singh, G.K. and Ahmed, S.A.K.S., “Vibration Signal Analysis Using Wavelet Transform for Isolation and Identification of Electrical Faults in Induction Machine,” Electrical Power Systems Research, Vo.68, pp. 119-136, May, 2004
[11]Wu, S. and Chow, W. S., “Induction Machine Fault Detection Using SOM-Based RBF Neural Networks,” IEEE Transaction on Industrial Electronics, Vol. 51, No.1, pp. 183-194, February, 2004.
[12]Betta, G., Ligouori, C., Paolillo, A., and Pietrosanto, A., “A DSP-Based FFT-Analyzer for the Fault Diagnosis of Rotating Machine Based on Vibration Analysis,” IEEE Transaction on Instrumentation and Measurement, Vol. 51, No. 6, pp. 1316-1322, December, 2002.
[13]Wang, W., Ismail, F., and Golnaraghi, F., “A Neuro-Fuzzy Approach to Gear System Monitor,” IEEE Transaction on Fuzzy Systems, Vol. 12, No. 5, pp. 710-7253, October, 2004.
[14]Lou, X. and Loparo, K. A., “Bearing Fault Diagnosis Based on Wavelet Transform and Fuzzy Inference,” Mechanical System and Signal Processing, Vol. 18, Issue 5, pp. 1077-1095, September, 2004.
[15]Dalpiaz, G., Rivola, A., and Rubini, R, “Effectiveness and Sensitivity of Vibration Processing Technique for Local Fault Detection in Gears,” Mechanical Systems and Signal Processing, Vol. 14, Issue 3, pp. 387-412, May, 2000.
[16]Donoho, D. L., “De-Noising by Soft-Thresholding,” IEEE Transactions on Information Theory, Vol. 41, No. 3, pp. 183-194, May, 1995.
[17]Lin, J. and Qu, L., “Feature Extraction Based on Morlet Wavelet and Its Application for Mechanical Fault Diagnosis,” Journal of Sound and Vibration, Vol. 234, Issue 1, pp. 135-148, June, 2006.
[18]Qiu, H., Lee, J., Lin, J. and Yu, G., “Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics,” Journal of Sound and Vibration, Vol. 289, Issues 4-5, pp. 1066-1090, February, 2006.
[19]Zhang, X., Xu, R., Kwan, C., Liang, S. Y., Xie, Q., and Haynes, L., “An Integrated Approach to Bearing Fault Diagnostics and Prognostics,” Proceedings of the 2005 American Control Conference, Portland, USA, pp. 2750-2755, June 8-15, 2005.
[20]Yen, G. and Lin, K. C., “Wavelet Packet Feature Extraction for Vibration Monitoring,” IEEE Transactions on Instrumentation and Measurement, Vol. 47, No. 3, pp. 650-666, June, 2000.
[21]Goumas, K., Zervakis E., and Stavarakis, S., “Classification of Washing Machines Vibration Signals Using Discrete Wavelet Analysis for Feature Extraction,” IEEE Transactions on Instrumentation and Measurement, Vol. 51, No. 3, pp. 497-508, June, 2000.
[22]楊大明，“智慧型感應馬達狀況診斷預測維護系統”，專題研究計畫，行政院國家科學委員會，2004。
[23]Liu, X, Zhang, H., Liu, J., and Yang, J., “Fault Detection and Diagnosis of Permanent-Magnet DC Motor Based on parameter Estimation and Neural Network,” IEEE Transactions on Instrumentation and Measurements, Vol. 47, No. 5, October, 2000.
[24]Automotive e-Safety Web-Site. http:// www.dtic.mil/ndia/2003NET3/dudzik.ppt
[25]Tang, K. Z., “A DSP-Based FFT-Analyzer for the Fault Diagnosis of Rotating Machine Based on Vibration Analysis,” IEEE Transaction on Instrumentation and Measurement, Vol. 51, No. 6, pp. 1316-1322, December, 2002.
[26]Campos, F. T., “A Reference Architecture for Remote Diagnostics and Prognostics Applications,” Proceedings of the Systems Readiness Technology Conference, Utah, USA, pp. 842-853, October, 2002.
[27]Li, P. S., “Prognostics Framework” Proceedings of the Systems Readiness Technology Conference, Utah, USA, pp. 497-504, Sep. 18-21, 2000.
[28]Goswami, J. and Chan, J, Fundamentals of Wavelets, John Wiley and Sons, Inc., USA, pp.73, 2000.
[29]單維彰，凌波初步，全華科技圖書中心，第10-69頁台北，1999。
[30]Yu, J., Singular Value Decomposition With Applications to IR and Text Clustering http:// icl.pku.edu.cn/yujs/papers/pdf/SVDcluster.pdf
[31]Haykin, S. S., Adaptive Filter Theory, Prentice-Hall , USA, 2001.
[32]葉怡成，類神經網路模式應用與實作，儒林出版社，台北，2003。
[33]An Introduction to Neural Networks. http://www.cs.stir.ac.uk/~lss/NNIntro/InvSlides.html
[34]Roy, J. and Ramanujan, A., “Understanding Web Services,” IT Professional, Vol. 3, pp. 69-73, November, 2001.
[35]Microsoft’s Web Services site. http://msdn.microsoft.com/webservices
[36]IBM’s Web Services site. http://www.alphaworks.ibm.com/webservices
[37]HTTP Specification. http://www.w3c.org/Protocols/Specs.html
[38]SOAP Specification. http://www.w3c.org/tr/soap
[39]WSDL Specification. http://www.w3c.org/tr/wsdl
[40]XML Specification. http://www.w3c.org/XML
[41]SSL, Netscape Communications Corporation.
http://developer.netscape.com/tech/security/ssl/howitworks.html
[42]XML-Signature Syntax and Processing. http://www.w3.org/Signature
[43]XML-Encryption Syntax and Processing. http://www.w3.org/Signature
[44]Web Services Enhancements for Microsoft .NET, Microsoft Corporation.
http://msdn.microsoft.com/webservices/building/wse/default.aspx
[45]Understanding Single Sign-On, Jupitermedia Corporation.
http://www.intranetjournal.com/articles/200205/se_05_28_02a.html
[46].NET Passport, Microsoft Corporation.
http://www.microsoft.com/net/services/passport/
[47]Securing Web Services with Single Sign-On, Systinet Inc.
http://www.theserverside.com/articles/
[48]What is UML. http://pigseye.kennesaw.edu/~dbraun/csis4650/A&D/UML_tutorial/what_is_uml.htm
[49]Korbicz, J., Koscielny, J. M., Kowalczuk, Z., and Cholewa, W., Fault Diagnosis, Springer, New York, 2004
[50]羅華強，類神經網路-MATLAB的應用，高立圖書，台北，第5-50頁，2005。
[51]姚孟華、吳英偉、蔡源成和高小茜，“新世代武器系統自主式後勤支援管理架構概述與實例研究”，海軍學術月刊，第39卷第02期，第40-45頁，七月，2005。
[52]Schach, S.R., Introduction to Object-Oriented Analysis and Design with UML and the United Process, McGraw Hill, USA, pp. 118, 2003.
[53]Booch, G., Rumbaugh, J., and Jacobson, I., The Unified Modeling Language User Guide, Addison Wesley, pp. 4, USA, 1998.
[54]Quatek Corporation Web-Site. http://www.quatek.com.tw
[55]Bruel & Kjar Corporation Web-Site. http://www.bksv.com/
[56]IEEE P1451.4 Web-Site. http://grouper.ieee.org/groups/1451/4/
[57]ADLINK Technology Corporation. Web-Site. http://www.adlinktech.com/PD/web/productline.php
[58]National Instruments Web-Site. http://www.ni.com/
[59]Two Way Trading Corporation. Web-Site. http://www.two-way.com.tw/