本篇論文提出一個讓永磁同步馬達在額定轉速下之任意轉速皆能有效診斷正確多種故障問題以及嚴重程度的演算法流程。故障診斷分析使用霍爾元件量測電流訊號，並透過小波轉換處理、時域與頻域特徵作為有效參數，最後使用支持向量積分類器成功正確分類出故障的馬達。目前文獻中的故障診斷方法常將馬達的運轉狀態設定於一至兩種轉速，並將訊號處理方法針對該轉速下進行調整。不過，本論文提出的演算法精神之一，透過小波轉換設計出適合處理任意轉速下諧波分佈的參數，同時能夠將故障特徵數值化利於監控。由於同性質的故障種類也會有程度上的不同，若能分辨出不同程度上的故障即可發揮監控的精神。因此，本論文提出的演算法精神之二，能處理兩個不同程度以及兩個不同型態的故障馬達共四個，使用目前文獻中沒用過的特徵參數，藉此提高同性質不同程度故障的診斷正確率。由於在全域轉速的設計下，有別於文獻中以單一轉速設計而侷限使用的範圍，本論文提出的演算法精神之三，使用四個支持向量機預測模型處理頻譜分佈廣泛的全域轉速，以平均分配轉速範圍為設計，使得馬達在任意轉速下皆有高正確率的診斷效果。同時，當有已知故障種類馬達之任何轉速下的電流資料皆能成為資料庫的訓練樣本。在此三種設計精神下，能夠實現在全域轉速下診斷四種故障型態的馬達，有相當好的正確率，同時任意轉速下的已知故障皆能成為訓練資料，讓系統有工業價值的可行性。

This paper presents the study of the effectively diagnosis algorithm for permanent magnet synchronous motors (PMSMs) running with a variety of fault and severity at any speed. The fault diagnosis is based on the current signature analysis by the Hall sensor, and processing through discrete wavelet transform, the features from time-domain and frequency domain as effective parameters, and finally classifying what type the motors filed correctly using the support vector machines. According to the journal papers, the fault detection methods supposed the state of the motor operating at one or two different speeds, and the adjustment of the parameters for the speed by the signal processing. However, on the basis of the spirit of this paper proposed, in order to detect the motors at any speed, the parameters of discrete wavelet transform are designed to be deal with the problem of the characteristic harmonics distribution shifting with different speeds and diffident faults. According to the second and third spirit of this paper proposed, in order to improve diagnostic accuracy, the algorithm comes up with a solution by two features eliminating the disadvantages of energy coefficient normalized. On the other hand, it provides an idea for predicting motors under different types of faults at different velocity by using four predictive models of the support vector machines. In addition to prediction, it could collect the current data from the known failure motors at any speed for training the predict models. With the innovation of the fault diagnosis system, the system not only has excellent accuracy against four types of faults but also the ability to monitor the motor at any speed. Overall, the system is valuable in industrial feasibility.

目 錄
論文審定書 i
誌 謝 ii
摘 要 iii
Abstract iv
目 錄 v
圖 次 viii
表 次 x
第一章 緒論 1
1-1 前言 1
1-2 文獻回顧 2
1-3 研究動機與範疇 6
1-4 組織章節 9
第二章 訊號分析基礎 10
2-1 前言 10
2-2 奈奎斯特取樣定理 10
2-3 快速傅立葉轉換 12
2-3-1 簡介 12
2-3-2 快速傅立葉轉換基本原理 13
2-4 小波轉換 15
2-4-1 前言 15
2-4-2 小波轉換數學定義 15
2-4-3 離散小波轉換 17
2-5 訊號能量 21
第三章 支持向量機分類器 23
3-1 前言 23
3-2 支持向量機的基本概念 23
3-2-1 基本概念 23
3-2-2 支持超平面 26
3-2-3 支持向量 27
3-3 核函數 28
3-3-1 核函數的概念 28
3-3-2 核函數的種類 29
3-4 C-支持向量分類法 31
第四章 故障馬達之時頻觀測 33
4-1 前言 33
4-2 永磁同步馬達基本規格 34
4-2-1 介紹 34
4-2-2 馬達規格 34
4-2-3 電氣角頻率與機械角頻率關係 38
4-3 退磁故障馬達 39
4-3-1 退磁故障現象與原因 39
4-3-2 退磁故障馬達製造 39
4-3-3 退磁故障時頻特徵 41
4-4 軸承故障馬達 44
4-4-1 軸承故障現象與原因 44
4-4-2 軸承故障馬達製造 44
4-4-3 軸承故障時頻特徵 45
第五章 故障診斷演算法與實驗 49
5-1 實驗硬體與規劃 49
5-1-1 故障診斷系統架構 49
5-1-2 硬體與規格 50
5-2 故障診斷演算法與實驗安排 53
5-2-1 實驗理念與特色 53
5-2-2 故障診斷演算法流程圖 54
5-3 分析故障特徵 56
5-3-1 訊號擷取與特徵頻帶分佈 56
5-3-2 小波母函數規格選定 58
5-3-3 離散小波轉換與多層解析度分析 59
5-4 特徵萃取 62
5-4-1 小波能量係數正規化 62
5-4-2 時頻能量強度特徵 63
5-4-3 故障馬達特徵參數 65
5-5 判定故障種類 67
5-5-1 分類器使用 67
5-5-2 支持向量機分類器 67
5-6 實驗結果 68
第六章 結論與未來展望 73
6-1 結論 73
6-2 未來展望 74
參考文獻 75
簡 歷 xi

[1] 朱效賢、潘敏俊，“包絡譜分析於軸承故障診斷之探討暨工程應用”，碩士論文，國立中央大學機械工程研究所，民國九十四年九月。
[2] 康淵、趙安民，“馬達故障診斷之模糊類神經網路”，碩士論文，中央大學機械工程研究所，民國九時三年七月。
[3] M. S. K. Khan and M. A. Rahman, “Development and implementation of a novel fault diagnostic and protection technique for IPM motor drives,” IEEE Trans. Ind. Electron., vol. 56, no. 1, pp. 85-92, Jan. 2009.
[4] E. G. Strangas, S. Aviyente, and S. S. H. Zaidi, “Time-frequency analysis for efficient fault diagnosis and failure prognosis for interior permanent-magnet AC motors,” IEEE Trans. Ind. Electron., vol. 55, no.12, pp. 4191-4199, Dec. 2008.
[5] A. Gandhi, T. Corrigan, and L. Parsa, “Recent advances in modeling and online detection of stator interturn faults in electrical motors,” IEEE Trans. Ind. Electron., vol. 58, no. 5, pp. 1564-1575, May 2011.
[6] J. Faiz, B. M. Ebrahimi, B. Akin, and H. A. Toliyat, “Comprehensive eccentricity fault diagnosis in induction motors using finite element method,” IEEE Trans. Magn., vol. 45, no. 3, pp. 1564-1575, Mar. 2009.
[7] W. le Roux, R. G. Harly, and T. G. Habetler, “Detection rotor faults in low power permanent magnet synchronous machines,” IEEE Trans. Power Electron., vol. 22, no. 1, pp. 322-328, Jan. 2007.
[8] B. M. Ebrahimi, J. Faiz, and M. J. Roshtkhari, “Static, dynamic and mixed eccentricity fault diagnosis in permanent magnet synchronous motors,” IEEE Trans. Ind. Electron., vol. 56, no. 1, pp. 85-92, Jan. 2009.
[9] Jordi-Roger R. Ruiz, J. A. Rosero, A. G. Espinosa, and L. Romeral, “Detection of demagnetization faults in permanent-magnet synchronous motors under nonstationary conditions,” IEEE Trans. Magn., vol. 45, no. 7, pp. 2961-2968, Jul. 2009.
[10] Randy R. Schoen, Thomas G. Habetler, Farrukh Kamran, and Robert G. Bartheld, “Motor bearing damage detection using stator current monitoring,” IEEE Trans. Ind. Applicat., vol. 31, no. 6, pp. 1274-1279, Nov./Dec. 1995.
[11] J. Rosero, L. Romeral, J. Cusido, and J. A. Ortega, “Fault detection by means of wavelet transform in a PMSM under demagnetization,” in the 33nd Annual Conference of the IEEE Industrial Electronics Society, Nov. 2007, Taipei, Taiwan.
[12] Bashir M. Ebrahimi, and J. Faiz, “Feature extraction for short-circuit fault detection in permanent-magnet synchronous motors using stator-current monitoring,” IEEE Trans. Power Electron., vol. 25, no. 10, pp. 2673-2682, Oct. 2010.
[13] 劉昌換、吳添保、黃仲欽、彭榮芳、楊宗銘、詹前茂、劉添華、賴炎生，“電機機械”，東華書局，民國九十一年
[14] Zhifu Wang, Jingzhe Yang, Huiping Ye, and Wei Zhou, “A review of permanent magnet synchronous motor fault diagnosis,” in the Conference and Exposition of the IEEE Transportation Electrification Asia-Pacific , Sep. 2014, Beijing, China.
[15] J. Rosero, L. Romeral, E. Rosero, and J. Urresty, “Fault detection in dynamic conditions by means of discrete wavelet decomposition for PMSM running under bearing damage,” in the 24 Annual Conference and Exposition of the IEEE Applied Power Electronics, Feb. 2009, Washington, DC.
[16] 劉俊麟，“A Tutorial of the Wavelet Transform”，數位影像處理及訊號處理實驗室，國立台灣大學電信所，民國九十九年。
[17] Bashir M. Ebrahimi, Jawad Faiz, M. Javan-Roshtkhari, and A. Zargham Nejhad, “Static eccentricity fault diagnosis in permanent magnet synchronous motor using time stepping finite element method,” IEEE Trans. Magn., vol. 44, no. 11, pp. 4297-4300, Nov. 2008.
[18] B. P. Lathi, Signal Processing and Linear Systems, Oxford University Press, pp. 319-321, 1998.
[19] Walker, James S., “Fast Fourier transforms,” London CRC Press, 1991.
[20] B. P. Lathi, Signal Processing and Linear Systems, Oxford University Press, pp. 352-356, 1998.
[21] Paul S. Addison, “The Illustrated Wavelet Transform Handbook, Institute of Physics,” 2002.
[22] J. Rosero, A. G. Espinosa, J. Cusido, J. A. Ortega, and L. Romeral, “Simulation and fault detection of short circuit winding in permanent magnet synchronous machine (PMSM) by means of Fourier and wavelet transform,” in the Conference of the IEEE International Instrumentation and Measurement Technology, May 2008, Canada.
[23] 葛哲學、沙威，“小波分析理論與MATLAB R2007實現”，電子工業出版社，民國九十六年。
[24] Mallat, S., "A theory for multiresolution signal decomposition: the wavelet representation," IEEE Trans. Pattern Anal. and Machine Intell., vol. 11, no. 7, pp 674–693, 2002.
[25] Mallat and S. Zhong, “Characterization of signals from multiscale edges,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 14, no. 7, pp. 710– 732, Jul. 1992.
[26] B. P. Lathi, Signal Processing and Linear Systems, Oxford University Press, pp. 52, 1998.
[27] Boser, B. E., Guyon, I. M., and Vapnik, V. N., “A training algorithm for optimal margin classifiers” Proceedings of the 55 annual workshop on Computational learning theory, 1992.
[28] William H., Teukolsky, Saul A., Vetterling, William T., Flannery, B. P, “Section 16.5 Support Vector Machines,” Numerical Recipes: The Art of Scientific Computing, New York: Cambridge University Press, 2007.
[29] 李根逸，“支持向量機教學文件”，通訊與多媒體實驗室，國立台灣大學資訊工程研究所，民國九十九年。
[30] Kuhn, H. W.; and Tucker A. W., “Nonlinear programming”, Proceedings of 2nd Berkeley Symposium. Berkeley: University of California Press. pp. 481–492, 1951.
[31] Chih-Wei Hsu, Chih-Chung Chang, and Chih-Jen Lin, “A Practical Guide to Support Vector Classification,” Department of Computer Science National Taiwan University, Taipei, Taiwan
[32] 王建昌，“高速永磁同步馬達驅動技術”，財團法人工業技術研究院，能源知識庫，民國一百零二年。
[33] 陳賜恭、王醴，”單相感應發電機與單相感應馬達之特性分析”，碩士論文，國立成功大學電機工程研究所，民國九十一年。
[34] Yongchun Liang, “Diagnosis of inter-turn short-circuit stator winding fault in PMSM based on stator current and noise,” in the Conference of the IEEE Ind. Technol., Feb. 2014, Busan, Korea.
[35] S. Nandi, H. A. Toliyat, and X. Li, “Condition monitoring and fault diagnosis of electrical motors – a review,” IEEE Trans. Energy Convers., vol.20, no. 4, pp. 719-729, 2005.
[36] M. El Hachemi Benouzid, “A review of induction motors signature analysis as a medium for faults detection,” IEEE Trans. Ind. Electron., vol. 40, no. 3, pp. 735-739, 2004.
[37] J. R. Stack, R. G. Harley, and T. G. Habetler, “An amplitude modulation detector for fault diagnosis in rolling element bearings,” IEEE Trans. Ind. Electron., vol. 51, no. 5, pp. 1097-1102, 2004.
[38] 張宏展、鄭強、鄭育明、郭政謙、黃立彬，”馬達運轉頻譜監測系統之研製”，中華民國第三十五屆電力工程研討會，台灣科技大學電機工程系，民國一百零三年。
[39] 郭紘森、陳遵立，”高性能永磁式同步馬達驅動器之研製”，碩士論文，國立中山大學電機工程研究所，民國一百零二年。
[40] Bashir M. Ebrahimi, Mehrsan J. Roshtkhari, and Jawad Faiz, “Advanced eccentricity fault recognitions in permanent magnet synchronous motors using stator current signature analysis,” IEEE Trans. Ind. Electron., vol. 61, no. 4, Apr. 2013.