滾珠軸承作為機械設備中重要的旋轉零件之一，也時常是機械故障的問題之一。軸承的工作狀態正常與否直接影響了機械設備的性能，因此軸承故障檢測技術的研究具有其發展意義與價值。本文主要討論滾動軸承在實際使用和安裝場景中可能出現的異常和故障。本論文采用自行設計的平台結合高、低成本兩個加速度規來採集滾動軸承正常和異常狀態的振動信號。採用主成分分析（PCA）和小波包分解（WPD）作為特徵提取方法，對原始軸承震動訊號進行提取、分解和重構，得到小波包能量特徵。根據特徵提取方法提取的特徵作為長短期記憶（LSTM）網絡的輸入特徵，對軸承的狀態（如正常、偏心、不平衡、撞擊）進行分類訓練。通過提出的兩種特徵提取方法與不同規格的加速度規進行比較，驗證我們的模型精度達到了 97%。達到在不同狀態的訊號特徵情況下能夠分辨軸承問題是屬於哪種異常，提早進行故障排除，以利工廠機械設備能有最大程度的產能與效能。

Rolling bearing is one of the essential parts of mechanical equipment and is often the cause of mechanical failure. And the bearing’s working condition affects the performance of mechanical equipment. Therefore, the research on bearing anomaly detection techniques has excellent value. This thesis primarily discusses the abnormalities and failures of rolling bearings during actual use and installation scenarios. This thesis uses a self-designed platform combined with two high- and low-cost accelerometers to collect vibration signals of both normal and abnormal states of rolling bearings. Principal components analysis (PCA) and wavelet packet decomposition (WPD) are used as feature extraction methods to extract, decompose and reconstruct the bearing signal and calculate wavelet packet energy feature. According to the features extracted by the feature extraction method as the input features of the long short-term memory (LSTM) network, the state of the bearing (e.g., normal, off-center, unbalanced, impact) is classified and trained. By comparing the two proposed feature extraction methods with accelerometers of different specifications, verify our model accuracy reaches 97%. Knowing the signal characteristics of different states, bearing problems can be distinguished by abnormality type, and troubleshooting can thus be performed in advance so that machines can achieve their most tremendous capacity and efficiency.

[1] I. Howard, “A review of rolling element bearing vibration׳detection, diagnosis and prognosis׳,” Defence Science and Technology Organization Canberra (Australia), 1994.
[2] B. Zhang, G. Georgoulas, M. Orchard, A. Saxena, D. Brown, G. Vachtsevanos, and S. Liang, "Rolling element bearing feature extraction and anomaly detection based on vibration monitoring," 2008 16th Mediterranean Conference on Control and Automation, Ajaccio, pp. 1792-1797, 2008.
[3] S.P. Mogal, and D.I. Lalwani, “Experimental investigation of unbalance and misalignment in rotor bearing system using order analysis,” Journal of Measurements in Engineering, vol. 3, Issue 4, pp. 114-122, 2015.
[4] SKF Group, “Bearing damage and failure analysis,” 2017.
[5] X. Jin, Y. Sun, Z. Que, Y. Wang, and T. W. S. Chow, "Anomaly Detection and Fault Prognosis for Bearings," in IEEE Transactions on Instrumentation and Measurement, vol. 65, no. 9, pp. 2046-2054, Sept. 2016.
[6] D. Fernández-Francos, D. Martínez-Rego, O. Fontenla-Romero, A. Alonso-Betanzos, “Automatic bearing fault diagnosis based on one-class ν-SVM,” in Computers & Industrial Engineering, vol. 64, Issue 1, Pages 357-365, 2013.
[7] C. Sun, Z. Zhang, and Z. He. "Research on bearing life prediction based on support vector machine and its application." Journal of Physics: Conference Series. vol. 305, no. 1, 2011.
[8] M. Saddam, H. Tjandrasa, and D.A. Navastara, "Classification of alcoholic EEG using wavelet packet decomposition, principal component analysis, and combination of genetic algorithm and neural network," 2017 11th International Conference on Information & Communication Technology and System (ICTS), Surabaya, Indonesia, pp. 19-24, 2017.
[9] B. Li, M.Y. Chow, Y. Tipsuwan, and J.C. Hung, "Neural-network-based motor rolling bearing fault diagnosis," IEEE Transactions on Industrial Electronics, vol. 47, no. 5, pp. 1060-1069, Oct. 2000.
[10] K.F. Al-Raheem, and W. Abdul-Karem, “Rolling bearing fault diagnostics using artificial neural networks based on Laplace wavelet analysis,” International Journal of Engineering, Science and Technology, vol. 2, no. 6, pp. 278-290, 2010.
[11] F.A. Gers, D. Eck, and J. Schmidhuber, “Applying LSTM to time series predictable through time-window approaches.” Neural Nets WIRN Vietri-01. Springer, London, pp. 193-200, 2002.
[12] L. Eren, “Bearing Fault Detection by One-Dimensional Convolutional Neural Networks,” Mathematical Problems in Engineering, vol. 2017, 2017.
[13] R. Zhang, Z. Peng, L. Wu, B. Yao, and Y. Guan, “Fault Diagnosis from Raw Sensor Data Using Deep Neural Networks Considering Temporal Coherence,” Sensors, 17, no. 3: 549, 2017.
[14] K. Lee, J. Kim, J. Kim, K. Hur, and H. Kim, “CNN and GRU combination scheme for Bearing Anomaly Detection in Rotating Machinery Health Monitoring,” 2018 1st IEEE International Conference on Knowledge Innovation and Invention (ICKII), pp. 102-105, 2018.
[15] H. Pan, X. Tang, S. Meng, X. He, and F. Meng, “An Improved Bearing Fault Diagnosis Method using One-Dimensional CNN and LSTM,” Strojniški vestnik - Journal of Mechanical Engineering [Online], vol. 64, no. 7-8, pp. 443-452, 2018.
[16] X. Xu, H. Zhao, H. Liu, and H. Sun, “LSTM-GAN-XGBOOST Based Anomaly Detection Algorithm for Time Series Data,” 2020 11th International Conference on Prognostics and System Health Management (PHM-2020 Jinan), pp. 334-339, 2020.
[17] C. Lessmeier, J.K. Kimotho, D. Zimmer, and W. Sextro, “Condition monitoring of bearing damage in electromechanical drive systems by using motor current signals of electric motors: A benchmark data set for data-driven classification,” in Proceedings of the European conference of the prognostics and health management society, pp. 5-8, 2016.
[18] Case Western Reserve University (CWRU) Bearing Data Center, [Online]. https://csegroups.case.edu/bearingdatacenter/pages/welcome-casewestern-reserve-university-bearing-data-center-website, Dec. 2018.
[19] H. Qiu, J. Lee, J. Lin, and G. Yu, “Robust performance degradation assessment methods for enhanced rolling element bearing prognostics,” Advanced Engineering Informatics, vol. 17, no. 3-4, pp. 127-140, 2003.
[20] P. Nectoux, R. Gouriveau, K. Medjaher, E. Ramasso, B. Chebel-Morello, N. Zerhouni, C. Varnier, “Pronostia: An experimental platform for bearings accelerated life test,” In IEEE International Conference on Prognostics and Health Management, PHM'12, pp. 1-8, 2012.
[21] B. Wang, Y. Lei, N. Li, and N. Li, "A Hybrid Prognostics Approach for Estimating Remaining Useful Life of Rolling Element Bearings," in IEEE Transactions on Reliability, vol. 69, no. 1, pp. 401-412, March 2020.
[22] ANALOG DEVICES, ADcmXL3021 Datasheet and Product Info, https://www.analog.com/en/products/adcmxl3021.html#product-overview
[23] HIPNUC ELECTRONIC, https://hipnuc.com/HI226_229.html
[24] A. Rai, and S.H. Upadhyay, “A review on signal processing techniques utilized in the fault diagnosis of rolling element bearings,” Tribology International, vol. 96, pp. 289-306, 2016.
[25] M.D. Felder, J.C. Mason, and B.L. Evans, “Efficient dual-tone multifrequency detection using the nonuniform discrete Fourier transform,” in IEEE Signal Processing Letters, vol. 5, no. 7, pp. 160-163, July 1998.
[26] A. Grossman, and J. Morlet, “Decomposition of hardy functions into square integrable wavelets of constant shape,” SIAM J. Math., vol. 15, pp. 723-736, 1984.
[27] T. Chang, and C.C.J. Kuo, “Texture analysis and classification with tree-structured wavelet transform,” in IEEE Transactions on Image Processing, vol. 2, no. 4, pp. 429-441, Oct. 1993.
[28] J. Luo, H. Li, M. Zhong, and C. Zhang, “Wavelet Packet Energy Feature Extraction Method of Recoil Mechanism Wear Signal,” 2015 2nd International Conference on Electrical, Computer Engineering and Electronics. Atlantis Press, 2015.
[29] I.T. Chen, “Optimization of SVM Parameters Based on Artificial Fish Swarm Algorithm for Fault Diagnosis of Ball Bearings,” 2018.
[30] S. Hochreiter, and J. Schmidhuber. “Long Short-Term Memory,” Neural computation, 9 (8), pp. 1735-1780, 1997.
[31] S. Hochreiter, “The vanishing gradient problem during learning recurrent neural nets and problem solutions,” International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 6(02), pp. 107-116, 1998.
[32] P. Malhotra, L. Vig, G. Shroff, and P. Agarwal. “Long Short Term Memory Networks for Anomaly Detection in Time Series,” In Proceedings, Presses universitaires de Louvain, vol. 89, pp. 89-94, 2015.
[33] K. Greff, R.K. Srivastava, J. Koutník, B.R. Steunebrink, and J. Schmidhuber, “LSTM: A search space odyssey,” IEEE transactions on neural networks and learning systems, vol. 28, no. 10, pp. 2222-2232, Oct. 2017
[34] J. Pan, Y. Zi, J. Chen, Z. Zhou, and B. Wang, “LiftingNet: A Novel Deep Learning Network With Layerwise Feature Learning From Noisy Mechanical Data for Fault Classification,” in IEEE Transactions on Industrial Electronics, vol. 65, no. 6, pp. 4973-4982, June 2018.