臺灣博碩士論文加值系統

感應馬達是產業界設備的原動力，為了防止感應馬達非預期的故障所帶來的損失，本研究即針對感應馬達的軸承做軸承狀態診斷，並經由支持向量機(SVM)訓練，建立一套軸承自動診斷系統。本論文研究的軸承狀態分別為正常軸承、外環損壞與內環損壞。首先將量測所得的振動信號分別以時域和頻域分析，並經由五種統計特徵值的運算辨識軸承狀態。此五種統計特徵值分別為波峰至波谷、標準差、波峰因數、偏度和峰度，再以支持向量機分類軸承狀態，將時域和頻域的統計特徵值訓練支持向量機以自動辨識軸承的狀態。依據支持向量機的軸承狀態診斷結果，發現頻域信號較時域信號更易辨識軸承狀態。

Induction motors are the "workhorse" of modern industry. In order to prevent the occurrence of faults of the induction motor, therefore, the primary aim of this research is to development of an automatic approach to condition monitor of motor bearing with support vector machine (SVM). The bearing conditions concerned here are the normal bearing, the faulty outer race bearing and faulty inner race bearing. The time series and frequency domain approaches are used to analyze the measured vibration signal generated by the bearings. At the same time, five statistical features calculated both from time and frequency domains are employed to identify the bearing conditions. There are peak-to-valley, standard deviation, crest factor, skewness and kurtosis. The support vector machine is used to automatic classification of bearing condition. In order to automate the bearing condition, the statistical features computed from both time and frequency domains are employed to train the SVMs. The experimental results obtained show that the performance of classification of the SVM using the five statistical features computed from the frequency domain is better than those computed from the time domain.

摘 要 I
英文摘要 II
誌 謝 III
目 錄 IV
圖 目 錄 V
表 目 錄 VII
第一章 前言 1
第一節 研究動機與目的 1
第二節 文獻回顧 1
第三節 研究方法 2
第二章 感應馬達軸承的故障與監視方法 4
第一節 感應馬達 4
第二節 感應馬達的構造 5
第三節 感應馬達的故障 6
第四節 軸承的損壞原因 6
第五節 感應馬達的損壞監測方法 7
第三章 功率頻譜及統計運算 8
第一節 頻譜估測理論 8
第二節 統計運算方法 12
第四章 支持向量機理論 17
第一節 線性支持向量機 17
第二節 非線性支持向量機 19
第三節 多類別支持向量機的分類法 20
第五章 實驗結果與討論 23
第一節 量測系統的架構 23
第二節 統計分析 26
第三節 建立軸承診斷器 32
第六章 結論 38
參考文獻 39

1.F. Briz et al., Online stator winding fault diagnosis in inverter-fed AC machines using high-frequency signal injection, IEEE Transactions on Industry Applications, Vol. 39, pp. 1109-1117, 2003.
2.H. Watanabe et al., Application of a fuzzy discrimination analysis for diagnosis of valvular heart disease, IEEE Transactions on Fuzzy Systems, Vol. 2, pp. 267-276, 1994.
3.J. A. Antonino-Daviu et al., Validation of a new method for the diagnosis of rotor bar failures via wavelet transform in industrial induction machines, IEEE Transactions on Industry Applications, Vol. 42, pp. 990-996, 2006.
4.J. C. Jung, Y. W. Chang and P. H. Seong, Application of an integrated power line signal analysis methodology to the detection of major rotating machine abnormalities in nuclear power plants, Nuclear Engineering and Design, Vol. 229, pp. 307-315, 2004.
5.R. B. W. Heng and M. J. M. Nor, Statistical Analysis of Sound and Vibration Signals for Monitoring Rolling Element Bearing Condition, Applied Acoustics, Vol. 53, No. 1-3, pp. 211-226, 1998.
6.S. F. Yuan and F. L. Chu, Fault diagnostics based on particle swarm optimisation and support vector machines, Mechanical Systems and Signal Processing, Vol. 21, pp. 1787-1798, May, 2007.
7.A. Sadoughi, M. Ebrahimi and E. Rezaei, A new approach for induction motor broken bar diagnosis by using vibration spectrum, SICE-ICASE International Joint Conference, pp. 4715-4720, Busan, Korea, 2006.
8.D. Guangbin and P. Peilin, Fault diagnosis method based on wavelet neural network for power system turbo-generator, Chinese Control Conference, pp. 259-263, Hunan, China, 2007.
9.J. Shi, Y. Zheng and Z. Yan, SVM for estimation of wrist angle from sonomyography and SEMG signals, 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 4806-4809, Lyon, France, Aug., 2007.
10.L. Peng et al., Gear intelligent fault diagnosis based on support vector machines, Proceedings of the 26th Chinese Control Conference, pp. 496-500, Hunan, China, 2007.
11.L. Na, L. Hong and F. Yanjun, Fault diagnosis of rotating system based ICA-SVM, The Sixth World Congress on Intelligent Control and Automation, pp. 5590-5594, Dalian, China, 2006.
12.M. Yoshikawa, M. Mikawa and K. Tanaka, A myoelectric interface for robotic hand control using support vector machine, IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 2723 – 2728, CA, USA, Oct. 29 -Nov. 2, 2007.
13.Q. Song et al., Prediction of human elbow torque from EMG using SVM based on AWR information acquisition platform, IEEE International Conference on Information Acquisition, pp. 1274-1278, Shandong, China, Aug., 2006.
14.S. Lu and M. Li, Bearing fault diagnosis based on PCA and SVM, IEEE International Conference on Mechatronics and Automation, pp. 3503-3507, Harbin, China, 2007.
15.X. Zhao, N. Jin and J. Zhang, Flow pattern identification of gas/water two phase flow based on SVM, The Sixth World Congress on Intelligent Control and Automation, Vol. 2, pp. 5657-5661, Dalian, China, June, 2006.
16.Y. Ping and W. Qing-miao, Fault diagnosis system for turbo-generator set based on fuzzy neural network, International Conference on Artificial Reality and Telexistence, pp. 228-231, Hangzhou, China, 2006.
17.Y. Guozheng et al., Pattern recognition using hybrid optimization for a robot controlled by human thoughts, 19th IEEE International Symposium on Computer-Based Medical Systems, pp. 396-400, Utah, USA, 2006.
18.Z. Sheng et al., Fault diagnosis system for rotary machines based on fuzzy neural networks, IEEE/ASME International Conference on Advanced Intelligent Mechatronics, Vol. 191, pp. 199-204, Kobe, Japan 2003.
19.Z. Hao and Z. Li-Qing, ECG analysis based on PCA and support vector machines, IEEE International Conference on Neural Networks and Brain, pp. 743-747, Beijing, China, 2005.
20.Motor Reliability Working Group, Report of Large Motor Reliability Survey of Industrial and Commercial Installations, Part I and Part II, IEEE Transactions on Industry Applications, Vol. 21, pp. 853-864, 1985.
21.O.V. Thorsen and M. Dalva, Failure identification and analysis for high-voltage induction motors in the petrochemical industry, IEEE Transactions on Industry Applications, Vol. 35, pp. 810-818, 1999.
22.L. Bottou et al., Comparison of classifier methods: A case study in handwritten digit recognition, Proceedings of the 12th IAPR International Conference on Pattern Recognition, Vol. 2, pp. 77-82, Los Alamitos, CA, 1994.
23.S. R. Gunn, Support vector machines for classification and regression, Technical Report, University of Southampton, May, 1998.
24.J. H. Friedmam, Another approach to polychotomous classification, Technical report, Department of Statistics, Stanford University, October, 1996.
25.J. C. Platt, N. Cristianini and J. Shawe-Taylor, Large margin DAGs for multiclass classification, In Advances in Neural Information Processing Systems, Vol. 12, pp. 547-553, MIT Press, 2000.
26.張英彬、曾相彬和曹登發，電機機械，新文京開發出版股份有限公司，台北縣，93年6月。
27.小野繁著，林子銘譯，滾動軸承之應用設計，正言出版社，台南市，1983。
28.SKF, Bearing failures and their causes, Product information 401, 1994.
29.S. Salivahanan, A. Vallavaraj and C. Gnanapriya著，博鴻工作室譯，數位訊號處理，美商麥格羅．希爾，台北市，2002。
30.杜德銘、洪崇嶺、黃柄森和陳德煒，水下音響信號頻譜特徵之萃取與識別，中正嶺學報，第三十二卷，第二期，民國93年5月。
31.W. L. Martinez and A. R. Marinez, Computational statistics handbook with Matlab, Chapman & Hall/CRC, London, New York, 2002.
32.江建良，統計學，龍騰出版股份有限公司，台北縣，86年2月。
33.L. Wang, Support vector machines: Theory and applications, Springer, Berlin, New York, January, 2005.
34.徐業良，工程最佳化設計，國立編譯館主編，宏明書局，台北市，84年10月。