長久以來馬達一直圍繞在我們生活之中，其中感應馬達因為有價格低廉與安
裝簡易等優點，在工業中被廣泛使用。然而不論使用何種類型的馬達，一旦損壞
發生，將中斷生產或損壞設備，嚴重者則發生意外等，為此馬達的監測與診斷就
顯得格外重要。本研究透過量化馬達的電流派克圓輸出特徵，並搭配機器學習方
法協助診斷，其中也比較了馬達電流特徵分析與標準規範兩者之特徵，以驗證派
克圓特徵之效果。
　　由於某些特徵會隨著馬達負載改變，本研究透過線性回歸的方法建立基線，
以排除負載的影響，並在進行分類前使用正規化、標準化、變異數分析、主成分
分析與線性判別分析等特徵處理方法調整資料集。接著透過嵌套交叉驗證的方法
調校支持向量機的各項超參數，以確保每種組合的方法所訓練出來的模型都處於
最佳的狀態。從結果發現，在排除負載因素的影響後，所有實驗組的分類率都有
所提升，特別是使用了變異數分析之實驗組，特徵數量不僅由原本的29種降低至
10種外，分類率還達到了100%。這個結果除了證明使用基線的效果外，也間接驗
證了本研究量化派克圓之方法用於診斷馬達之成效。

Motors have played an important role in our lives for a long time. Because of the low prize and the easy installation, the induction motors are widely used in the industry. However, regardless of the type of the motor, once the fault occurs, it might interrupt the production, damage the equipment, or even worst, cause an accident. Therefore, motor monitoring and diagnosis become important. This research uses Park’s circle to generate the features and uses the machine learning to diagnose the fault. In order to verify the effect of the Park’s circle features, the features generated from the motor current signature analysis and the standard specification are also compared.
Because some features change according to the motor loading, this research builds the baseline through the linear regression to remove the loading influence. The feature processing methods such as normalization, standardization, analysis of variance, principal component analysis and linear discriminant analysis are applied before the classification. Then the hyperparameters of the support vector machine are tunned by using the nested cross validation to ensure that the model trained by each combination method is in the best state. As a result, the classification accuracies of all experiments are increased after removing the fault influence, especially for the experiment with analysis of variance. The number of the features not only reduces from 29 to 10 but also the classification rate has reached 100%. The result proofs the effectiveness of the proposed baseline method and Park’s circle method for the motor diagnosis.

[1] H. Gavrila, V. Manescu, G. Paltanea, G. Scutaru, and I. J. P. E. Peter, "New trends in energy efficient electrical machines," vol. 181, pp. 568-574, 2017.
[2] S. Bistak and S. Kim, "AC induction motors vs. permanent magnet synchronous motors, 2017," ed.
[3] P. Waide and C. U. Brunner, "Energy-efficiency policy opportunities for electric motor-driven systems," 2011.
[4] M. Garcia, P. A. Panagiotou, J. A. Antonino-Daviu, and K. N. J. I. T. o. I. E. Gyftakis, "Efficiency assessment of induction motors operating under different faulty conditions," vol. 66, no. 10, pp. 8072-8081, 2018.
[5] A. G. J. E. F. A. Yetgin, "Effects of induction motor end ring faults on motor performance. Experimental results," vol. 96, pp. 374-383, 2019.
[6] A. Choudhary, D. Goyal, S. L. Shimi, and A. J. A. o. C. M. i. E. Akula, "Condition monitoring and fault diagnosis of induction motors: A review," vol. 26, no. 4, pp. 1221-1238, 2019.
[7] X. Liang and K. Edomwandekhoe, "Condition monitoring techniques for induction motors," in 2017 IEEE Industry Applications Society Annual Meeting, 2017, pp. 1-10: IEEE.
[8] A. Glowacz and Z. J. M. Glowacz, "Diagnostics of stator faults of the single-phase induction motor using thermal images, MoASoS and selected classifiers," vol. 93, pp. 86-93, 2016.
[9] K. M. Siddiqui, K. Sahay, V. J. I. J. o. A. R. i. E. Giri, Electronics, and I. Engineering, "Health monitoring and fault diagnosis in induction motor-a review," vol. 3, no. 1, pp. 6549-6565, 2014.
[10] D. Lopez-Perez and J. J. I. T. o. I. A. Antonino-Daviu, "Application of infrared thermography to failure detection in industrial induction motors: case stories," vol. 53, no. 3, pp. 1901-1908, 2017.
[11] E. L. Bonaldi, L. E. d. L. de Oliveira, J. G. B. da Silva, G. Lambert-Torresm, and L. E. B. da Silva, "Predictive maintenance by electrical signature analysis to induction motors," in Induction Motors-Modelling and Control: IntechOpen, 2012.
[12] K. Raghavendra and B. J. I. J. S. R. Karabasanagouda, "Frequency response analysis of deep groove ball bearing," vol. 3, no. 8, pp. 1920-1926, 2014.
[13] E. Bechhoefer, R. Li, and D. J. J. o. I. M. He, "Quantification of condition indicator performance on a split torque gearbox," vol. 23, no. 2, pp. 213-220, 2012.
[14] T. W. Chow and S. J. I. T. o. I. E. Hai, "Induction machine fault diagnostic analysis with wavelet technique," vol. 51, no. 3, pp. 558-565, 2004.
[15] Ž. Kanović, D. Matić, Z. Jeličić, M. Rapaić, B. Jakovljević, and M. Kapetina, "Induction motor broken rotor bar detection using vibration analysis—A case study," in 2013 9th IEEE international symposium on diagnostics for electric machines, power electronics and drives (SDEMPED), 2013, pp. 64-68: IEEE.
[16] W. T. Thomson and M. J. I. I. A. M. Fenger, "Current signature analysis to detect induction motor faults," vol. 7, no. 4, pp. 26-34, 2001.
[17] M. Messaoudi, L. J. I. J. o. S. Sbita, and I. Processing, "Multiple faults diagnosis in induction motor using the MCSA method," vol. 1, no. 3, pp. 190-195, 2010.
[18] F. B. Batista, P. C. M. Lamim Filho, R. Pederiva, V. A. D. J. I. T. o. I. Silva, and Measurement, "An empirical demodulation for electrical fault detection in induction motors," vol. 65, no. 3, pp. 559-569, 2016.
[19] A. M. C. MA Cruz, S %J Electric Machines and P. Systems, "Rotor cage fault diagnosis in three-phase induction motors by extended Park's vector approach," vol. 28, no. 4, pp. 289-299, 2000.
[20] A. Sharma, S. Chatterji, and L. Mathew, "A novel Park’s vector approach for investigation of incipient stator fault using MCSA in three-phase induction motors," in 2017 International Conference on Innovations in Control, Communication and Information Systems (ICICCI), 2017, pp. 1-5: IEEE.
[21] N. Mehala, R. J. I. J. o. E. Dahiya, and Technology, "Detection of bearing faults of induction motor using Park’s vector approach," vol. 2, no. 4, pp. 263-266, 2010.
[22] J. L. Silva and A. M. Cardoso, "Bearing failures diagnosis in three-phase induction motors by extended Park's vector approach," in 31st Annual Conference of IEEE Industrial Electronics Society, 2005. IECON 2005., 2005, p. 6 pp.: IEEE.
[23] K. Yahia, A. Cardoso, A. Ghoggal, and S. J. I. t. Zouzou, "Induction motors airgap-eccentricity detection through the discrete wavelet transform of the apparent power signal under non-stationary operating conditions," vol. 53, no. 2, pp. 603-611, 2014.
[24] T. Ahmad and M. N. J. I. E. L. Aziz, "Data preprocessing and feature selection for machine learning intrusion detection systems," vol. 13, no. 2, pp. 93-101, 2019.
[25] V. Bolón-Canedo, N. Sánchez-Maroño, A. J. K. Alonso-Betanzos, and i. systems, "A review of feature selection methods on synthetic data," vol. 34, no. 3, pp. 483-519, 2013.
[26] Y. Saeys, I. Inza, and P. J. b. Larranaga, "A review of feature selection techniques in bioinformatics," vol. 23, no. 19, pp. 2507-2517, 2007.
[27] J.-h. Woo, J.-Y. Song, and Y.-J. Choi, "Performance enhancement of deep neural network using feature selection and preprocessing for intrusion detection," in 2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), 2019, pp. 415-417: IEEE.
[28] A. Malhi, R. X. J. I. t. o. i. Gao, and measurement, "PCA-based feature selection scheme for machine defect classification," vol. 53, no. 6, pp. 1517-1525, 2004.
[29] I. S. Thaseen, C. A. J. J. o. K. S. U.-C. Kumar, and I. Sciences, "Intrusion detection model using fusion of chi-square feature selection and multi class SVM," vol. 29, no. 4, pp. 462-472, 2017.
[30] N. O. F. Elssied, O. Ibrahim, A. H. J. R. J. o. A. S. Osman, Engineering, and Technology, "A novel feature selection based on one-way anova f-test for e-mail spam classification," vol. 7, no. 3, pp. 625-638, 2014.
[31] A. Mustaqeem, S. M. Anwar, M. Majid, and A. R. Khan, "Wrapper method for feature selection to classify cardiac arrhythmia," in 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2017, pp. 3656-3659: IEEE.
[32] I. Guyon, J. Weston, S. Barnhill, and V. J. M. l. Vapnik, "Gene selection for cancer classification using support vector machines," vol. 46, no. 1, pp. 389-422, 2002.
[33] S. Yadav and S. Shukla, "Analysis of k-fold cross-validation over hold-out validation on colossal datasets for quality classification," in 2016 IEEE 6th International conference on advanced computing (IACC), 2016, pp. 78-83: IEEE.
[34] R. Kohavi, "A study of cross-validation and bootstrap for accuracy estimation and model selection," in Ijcai, 1995, vol. 14, no. 2, pp. 1137-1145: Montreal, Canada.
[35] P. Liashchynskyi and P. J. a. p. a. Liashchynskyi, "Grid search, random search, genetic algorithm: A big comparison for nas," 2019.
[36] J. Wainer and G. J. E. S. w. A. Cawley, "Nested cross-validation when selecting classifiers is overzealous for most practical applications," vol. 182, p. 115222, 2021.
[37] E. Swana and W. J. E. Doorsamy, "An Unsupervised Learning Approach to Condition Assessment on a Wound-Rotor Induction Generator," vol. 14, no. 3, p. 602, 2021.
[38] R. Liu, B. Yang, E. Zio, X. J. M. S. Chen, and S. Processing, "Artificial intelligence for fault diagnosis of rotating machinery: A review," vol. 108, pp. 33-47, 2018.
[39] S. B. Kotsiantis, I. D. Zaharakis, and P. E. J. A. I. R. Pintelas, "Machine learning: a review of classification and combining techniques," vol. 26, no. 3, pp. 159-190, 2006.
[40] I. Martin-Diaz, D. Morinigo-Sotelo, O. Duque-Perez, and R. J. J. I. T. o. I. A. Romero-Troncoso, "An experimental comparative evaluation of machine learning techniques for motor fault diagnosis under various operating conditions," vol. 54, no. 3, pp. 2215-2224, 2018.
[41] S. Zolfaghari, S. B. M. Noor, M. Rezazadeh Mehrjou, M. H. Marhaban, and N. J. A. S. Mariun, "Broken rotor bar fault detection and classification using wavelet packet signature analysis based on fourier transform and multi-layer perceptron neural network," vol. 8, no. 1, p. 25, 2018.
[42] F. B. Abid, S. Zgarni, and A. Braham, "Bearing fault detection of induction motor using SWPT and DAG support vector machines," in IECON 2016-42nd Annual Conference of the IEEE Industrial Electronics Society, 2016, pp. 1476-1481: IEEE.
[43] A. Widodo, B.-S. Yang, and T. J. E. s. w. a. Han, "Combination of independent component analysis and support vector machines for intelligent faults diagnosis of induction motors," vol. 32, no. 2, pp. 299-312, 2007.
[44] P. K. Kankar, S. C. Sharma, and S. P. J. A. S. C. Harsha, "Fault diagnosis of ball bearings using continuous wavelet transform," vol. 11, no. 2, pp. 2300-2312, 2011.
[45] 劉孟昆教授, "非線性系統訊號分析," 上課講義.
[46] "ISO 20958-2013 Condition monitoring and diagnostics of machine systems - Electrical signature analysis of three-phase induction motors.," 2013.
[47] T. Ameid, A. Menacer, H. Talhaoui, and Y. J. I. t. Azzoug, "Discrete wavelet transform and energy eigen value for rotor bars fault detection in variable speed field-oriented control of induction motor drive," vol. 79, pp. 217-231, 2018.
[48] A. Cardoso and E. Saraiva, "The use of Park's vector in the detection of electrical failures on three-phase induction motors," in Proceed. Int. AMSE Confer Modelling & simulation, 1987, vol. 2, pp. 99-109.
[49] A. M. Cardoso and E. S. J. I. T. o. I. A. Saraiva, "Computer-aided detection of airgap eccentricity in operating three-phase induction motors by Park's vector approach," vol. 29, no. 5, pp. 897-901, 1993.
[50] L. Oliveira and A. J. I. e. p. a. Cardoso, "Extended Park's vector approach-based differential protection of three-phase power transformers," vol. 6, no. 8, pp. 463-472, 2012.
[51] S. M. Cruz and A. M. J. I. T. o. i. a. Cardoso, "Stator winding fault diagnosis in three-phase synchronous and asynchronous motors, by the extended Park's vector approach," vol. 37, no. 5, pp. 1227-1233, 2001.
[52] J. L. Crassidis and J. L. Junkins, Optimal estimation of dynamic systems. Chapman & Hall/CRC Boca Raton, FL, 2004.
[53] 呂金河, 變異數分析. 三民, 1993.
[54] 林惠玲 and 陳正倉, 統計學: 方法與應用. 雙葉書廊, 2010.
[55] 盧家峰教授, "醫學訊號分析原理與MATLAB程式應用實作課程," 線上課程.
[56] T. Huang. 機器/統計學習:主成分分析.
[57] 李政軒教授, "統計學習理論," 線上課程.
[58] 劉益宏教授, "實用機器智慧," 上課講義.
[59] M. J. M. NEMA and Generators, "MG 1-2009."
[60] I. IEEE Standards Association %J New York, "519-2014-IEEE Recommended Practices and Requirements for Harmonic Control in Electric Power Systems," 2014.
[61] Y.-J. Goh and O. J. A. S. Kim, "Linear method for diagnosis of inter-turn short circuits in 3-phase induction motors," vol. 9, no. 22, p. 4822, 2019.
[62] H. J. T. j. o. e. m. Akoglu, "User's guide to correlation coefficients," vol. 18, no. 3, pp. 91-93, 2018.
[63] D. V. Spyropoulos and E. D. Mitronikas, "Induction motor stator fault diagnosis technique using Park vector approach and complex wavelets," in 2012 XXth International Conference on Electrical Machines, 2012, pp. 1730-1734: IEEE.
[64] K. D. Kompella, V. G. R. Mannam, S. R. J. J. o. E. S. Rayapudi, and I. Technology, "DWT based bearing fault detection in induction motor using noise cancellation," vol. 3, no. 3, pp. 411-427, 2016.
[65] R. Senthil Kumar and I. J. I. J. o. E. Gerald Christopher Raj, "Broken Rotor Bar Fault Detection Using DWT and Energy Eigenvalue for DTC fed Induction Motor Drive," 2021