永磁馬達之高功率密度特性使其應用領域相當廣泛，然而永磁馬達之轉子永久磁鐵於高溫及高逆向外加磁場下，易發生不可逆退磁故障，永磁馬達隨退磁嚴重程度上升，功率密度將因此下降。且退磁故障馬達須增加電流以達相同輸出轉矩，電流之增加將導致退磁情形惡化，形成惡性循環，因此退磁故障初期之偵測及預防相當重要。本文透過監督式機器學習建立故障分類系統，並以退磁故障數學模型輔助有效減少需訓練之資料數量。為增加訓練標籤數量及降低製造退磁故障之馬達成本，本文以硬體在環系統(HIL)作為馬達開關訊號接收及電流訊號輸出平台，並由ANSYS Maxwell等效電路提取(ECE)功能，匯入健康及退磁故障馬達模型至硬體在環系統，以進行馬達硬體驅動控制。而獲得之訓練資料，透過監督式學習演算法進行分類，完成系統建立。

The high power density characteristics of permanent magnet synchronous motors (PMSM) contribute to their wide applications. However, permanent magnets of PMSM are prone to irreversible demagnetization fault under high temperature and high reverse applied magnetic field. As the severity of demagnetization increases for PMSM, the power density will therefore decrease. The motor with demagnetization fault must increase the current to achieve the same output torque. Such increase in current will worsen the demagnetization situation and form a vicious circle. Therefore, the detection and prevention of the initial demagnetization fault is very important. This thesis establishes a fault classification system through supervised learning, and uses a mathematical model of demagnetization faults to effectively reduce the amount of data that needs to be trained. In order to increase the number of training labels and reduce the cost of motors that produce demagnetization faults, this thesis uses a hardware-in-the-loop (HIL) as a platform for motor switch signal reception and current signal output. The equivalent circuit extraction function of ANSYS Maxwell is used to import the healthy and demagnetized faulty motor models to the HIL for the motor hardware drive control. The training data obtained is classified through a supervised learning algorithm to complete the system establishment.

[1] C. Jaszczolt, (2017). “Understanding permanent magnet motors,” Yaskawa America, Inc.
[2] Y. V Sarvesha and S. Narula, "Analytical Modeling and Designing of SMPMSM for Electric Bike Application," 2019 4th International Conference on Electrical, Electronics, Communication, Computer Technologies and Optimization Techniques (ICEECCOT), Mysuru, India, 2019, pp. 118-124.
[3] E. Styvaktakis, M. H. J. Bollen and I. Y. H. Gu, "Expert system for classification and analysis of power system events," in IEEE Transactions on Power Delivery, vol. 17, no. 2, pp. 423-428.
[4] J. Zheng, Z. Wang, D. Wang, B. Wang and M. Li, "Diagnostic strategy and modeling of PMSM stator winding fault in electric vehicles," 2017 Chinese Automation Congress (CAC), Jinan, 2017, pp. 3870-3874.
[5] B. Vaseghi, N. Takorabet and F. Meibody-Tabar, "Analytical circuit-based model of PMSM under stator inter-turn short-circuit fault validated by time-stepping finite element analysis," The XIX International Conference on Electrical Machines - ICEM 2010, Rome, 2010, pp. 1-6.
[6] S. Hamidizadeh, N. Alatawneh, R. R. Chromik and D. A. Lowther, "Comparison of Different Demagnetization Models of Permanent Magnet in Machines for Electric Vehicle Application," in IEEE Transactions on Magnetics, vol. 52, no. 5, pp. 1-4.
[7] J. Hong et al., "Detection and Classification of Rotor Demagnetization and Eccentricity Faults for PM Synchronous Motors," in IEEE Transactions on Industry Applications, vol. 48, no. 3, pp. 923-932.
[8] S. Shehata, H. El-Goharey, M. Marei and A. Ibrahim, (2013). “Detection of Induction Motors Rotor/Stator Faults Using Electrical Signatures Analysis,” Renewable Energy and Power Quality Journal, pp. 382-387.
[9] D. T. L. Lee and A. Yamamoto, (1994). “Wavelet Analysis: Theory and Applications,” Hewlett-Packard Journal, December, 1994.
[10] J. Riba Ruiz, J. A. Rosero, A. Garcia Espinosa and L. Romeral, "Detection of Demagnetization Faults in Permanent-Magnet Synchronous Motors Under Nonstationary Conditions," in IEEE Transactions on Magnetics, vol. 45, no. 7, pp. 2961-2969.
[11] P. Kaur and P. Bahl, (2012). “Comparative Analysis between DWT and WPD Techniques of Speech Compression,” IOSR Journal of Engineering (IOSRJEN), Vol. 2, Issue 8 (August 2012), pp. 120-128.
[12] S. Wang, J. Bao, S. Li, H. Yan, T. Tang and D. Tang, "Research on Interturn Short Circuit Fault Identification Method of PMSM based on Deep Learning," 2019 22nd International Conference on Electrical Machines and Systems (ICEMS), Harbin, China, 2019, pp. 1-4.
[13] Y. Luo, J. Qiu and C. Shi, "Fault Detection of Permanent Magnet Synchronous Motor Based on Deep Learning Method," 2018 21st International Conference on Electrical Machines and Systems (ICEMS), Jeju, 2018, pp. 699-703.
[14] I. Kao, W. Wang, Y. Lai and J. Perng, "Analysis of Permanent Magnet Synchronous Motor Fault Diagnosis Based on Learning," in IEEE Transactions on Instrumentation and Measurement, vol. 68, no. 2, pp. 310-324.
[15] M. Si, X. Yang, S. Zhao, J. Si and W. Cao, "Modeling and analysis of the magnetic field of a surface-interior permanent magnet synchronous motor," 2015 IEEE International Magnetics Conference (INTERMAG), Beijing, 2015, pp. 1-1.
[16] M. Z. Islam, A. K. M. Arafat and S. Choi, "Determining the operating region for demagnetization-free fault tolerant control of multiphase PMa-SynRM," 2018 IEEE Applied Power Electronics Conference and Exposition (APEC), San Antonio, TX, 2018, pp. 198-204.
[17] M. Baranski, W. Szelag and C. Jedryczka, "Influence of temperature on partial demagnetization of the permanent magnets during starting process of line start permanent magnet synchronous motor," 2017 International Symposium on Electrical Machines (SME), Naleczow, 2017, pp. 1-6.
[18] Z. Ullah, S. Lee, M. R. Siddiqi and J. Hur, "Online Diagnosis and Severity Estimation of Partial and Uniform Irreversible Demagnetization Fault in Interior Permanent Magnet Synchronous Motor," 2019 IEEE Energy Conversion Congress and Exposition (ECCE), Baltimore, MD, USA, 2019, pp. 1682-1686.
[19] B. Guo, Y. Huang, F. Peng and J. Dong, "General Analytical Modeling for Magnet Demagnetization in Surface Mounted Permanent Magnet Machines," in IEEE Transactions on Industrial Electronics, vol. 66, no. 8, pp. 5830-5838.
[20] K. Kim, Y. Lee and J. Hur, "Transient Analysis of Irreversible Demagnetization of Permanent-Magnet Brushless DC Motor With Interturn Fault Under the Operating State," in IEEE Transactions on Industry Applications, vol. 50, no. 5, pp. 3357-3364.
[21] A. Singh, N. Thakur and A. Sharma, "A review of supervised machine learning algorithms," 2016 3rd International Conference on Computing for Sustainable Global Development (INDIACom), New Delhi, 2016, pp. 1310-1315.
[22] S. Wang, "Support Vector Machines Classification for High-Dimentional Dataset," 2012 Fourth International Conference on Multimedia Information Networking and Security, Nanjing, 2012, pp. 315-318.
[23] N. Tomasev and D. Mladenic, "Nearest Neighbor Voting in High-Dimensional Data: Learning from Past Occurrences," 2011 IEEE 11th International Conference on Data Mining Workshops, Vancouver, BC, 2011, pp. 1215-1218.
[24] M. E. Mavroforakis and S. Theodoridis, "Support Vector Machine (SVM) classification through geometry," 2005 13th European Signal Processing Conference, Antalya, 2005, pp. 1-4.
[25] S. Li and Y. Li, "An application of SVM to fingerprint image segmentation," 2011 International Conference on Electric Technology and Civil Engineering (ICETCE), Lushan, 2011, pp. 994-997.
[26] W. Tan, J. Cao and H. Li, "Algorithm of Shot Detection Based on SVM with Modified Kernel Function," 2009 International Conference on Artificial Intelligence and Computational Intelligence, Shanghai, 2009, pp. 11-14.
[27] N. Bajaj, G. T. -. Chiu and J. P. Allebach, "Reduction of memory footprint and computation time for embedded Support Vector Machine (SVM) by kernel expansion and consolidation," 2014 IEEE International Workshop on Machine Learning for Signal Processing (MLSP), Reims, 2014, pp. 1-6.
[28] S. Zhang, X. Li, M. Zong, X. Zhu and R. Wang, "Efficient kNN Classification With Different Numbers of Nearest Neighbors," in IEEE Transactions on Neural Networks and Learning Systems, vol. 29, no. 5, pp. 1774-1785.