本研究工作為應用本團隊開發之高頻寬、高性價比之模組化壓電式微機電三軸加速計於工具機馬達滾動軸承健康診斷預測系統之開發。在工業4.0年代，工具機的智慧控制與維護逐漸受到重視，其中轉動軸承為眾多工具機械不可或缺的一部分，軸承損壞造成的故障在機械故障原因統計中占比高達80%，故本論文將研究目標聚焦在馬達轉動軸承損害的分析診斷上。
本研究主要希望開發一套結合神經網路的預處理流程來達成最佳軸承健康辨識效果，我們將三種不同健康狀態的軸承依500、1,000、1,500 rpm轉速的設計收集資料，應用總體型經驗模態分解法對數據進行本質模態函數的分離，最終選擇IMF3作為包絡譜分析的目標，並且搭配類神經網路對該包絡譜特徵進行學習預測。經過模型驗證後，該方法在軸承的Z軸方向上的辨識成功率達到99.61%，一方面對本團隊之高性價比三軸加速計模組進行可應用性驗證，另一方面也驗證了本方法對滾動軸承的健康診斷可行性。

In this research work, a cost-effective and high-bandwidth piezoelectric micro-electromechanical three-axis accelerometer module developed by our group is implemented to establish a health diagnosis and prediction system for motor rolling bearings. In Industry 4.0 era, intelligent control and predicted maintenance of machine tools play a vital role. Among the smart manufacturing applications (Industry 4.0), rolling bearings are an indispensable part of many machine tools. Failures caused by bearing damage account for up to 80% of the total mechanical failures. Therefore, this research will focus on the rolling bearing health analysis and diagnosis.
This research aims to achieve the best rolling bearing health diagnosis accuracy with a few steps of preprocessing. This thesis will start from explaining the method of analyzing the vibration data of rolling bearings. Next, the vibration data will be collected from various health-status bearings at three different rotating speeds of 500, 1,000, 1,500 rpm, respectively. Then, the three-axis acceleration vibration data of each experiment will be applied with the Ensemble Empirical Mode Decomposition (EEMD) method to separate the intrinsic mode function. The envelope analysis will be adopted on IMF3 to acquire frequency features. These features will be used to perform learning and prediction of the health states for rolling bearings through an Artificial Neural Network (ANN). The identification rate of this method in Z-axis direction of the bearing reached 99.61% after the neural network model validation and testing. On one hand, the applicability of the cost-effective accelerometer module was verified; on the other hand, the feasibility of the proposed method in this study was also verified in terms of rolling bearing health diagnosis.

[1] R. J. Alfredson and J. Mathew, “The condition monitoring of rolling element bearings using vibration analysis,” Journal of Vibration, Acoustics, Stress and Reliability in Design, vol.106, no. 3, pp. 447-453, 1984.
[2] A. Jerome: Fast Computing of the Kurtogram for the detection of transient faults. Mechanical Systems and Signal Processing, vol. 21, no. 1, pp. 108-124, 2007.
[3] B. Li, M. Y. Chow, Y. Tipsuwan, and J. C. Hung, “Neural-network-based motor rolling bearing fault diagnosis,” IEEE Transactions on Instrumentation Measurement, vol. 47, no. 5, pp. 1060-1069, 2000.
[4] G. Zurita, V. Sánchez, and D. Cabrera, “A review of vibration machine diagnostics by using artificial intelligence methods,” Investigación & Desarrollo, vol. 1, no. 16, pp. 102-114, 2016.
[5] R. B. Randall, J. Antoni, and S. Chobsaard, “The relationship between spectral correlation and envelope analysis in the diagnostics of bearing faults and other,” Mechanical Systems and Signal Processing, vol. 15, no. 5, pp. 945-962, 2001.
[6] W. C. Tsao, “Fault Diagnosis of Ball Bearings Using Appropriate Intrinsic Mode Functions,” National Central University, Doctoral Dissertation, 2013.
[7] P. D. McFadden and J. D. Smith, “The vibration monitoring of rolling element bearings by the high-frequency resonance technique - a review,” Tribology International, vol. 17, no. 1, pp. 3-10, Feb. 1984.
[8] D. N. Brown, “Envelope analysis detects bearing faults before major damage occurs,” Pulp and Paper, vol. 63, no. 13, pp. 113-117. 1989.
[9] D. Ho and R.B. Randall, “Optimisation of bearing diagnostic techniques using simulated and actual bearing fault signals,” Mechanical Systems and Signal Processing, vol. 14, no. 5, pp. 763-788. Sep. 2000.
[10] J. R. Stack, R. G. Harley, and T. G. Habetler, “An amplitude modulation detector for fault diagnosis in rolling element bearings”, IEEE Transactions on Industrial Electronics, vol. 51, no. 5, pp. 1097-1102. Oct. 2004
[11] K. Ramachandran and M. Lokesha, “Fault diagnosis in gear using wavelet envelope power spectrum,” Int. J. Eng. Sci. Technol., vol. 3, no. 8, Aug. 2012.
[12] M. He and D. He, “Deep learning based approach for bearing fault diagnosis,” IEEE Transactions on Industry Applications, vol. 53, no. 3, pp. 3057-3065, Jan. 2017.
[13] Y. Lei, J. Lin, Z. He, and M. J. Zuo, “A review on empirical mode decomposition in fault diagnosis of rotating machinery,” Mechanical Systems and Signal Processing, vol. 35, no. 1-2, pp. 108-126, 2013.
[14] D. J. Yu, J. S. Cheng, and Y. Yang, “Application of EMD method and Hilbert spectrum to the fault diagnosis of roller bearings,” Mechanical Systems and Signal Processing, vol. 19, no. 2, pp. 259-270, 2005.
[15] H. Li and H. Q. Zheng, “Bearing fault detection using envelope spectrum based on EMD and TKEO,” Proceedings, the Fifth International Conference on Fuzzy Systems and Knowledge Discovery, Shandong, China, pp. 142-146, 2008.
[16] H. Li, H. Q. Zheng, and L. W. Tang, “Wigner-Ville distribution based on EMD for faults diagnosis of bearing,” Fuzzy Syst. Knowl. Discovery, vol. 4223, pp. 803-812, 2006.
[17] A. R. Mohanty, and V. K. Rai, “Bearing fault diagnosis using FFT of intrinsic mode functions in Hilbert-Huang transform,” Mechanical Systems and Signal Processing, vol. 21, no. 6, pp. 2607-2615, 2007.
[18] Z. K. Peng, P. W. Tse, and F. L. Chu, “A comparison study of improved Hilbert-Huang transform and wavelet transform: application to fault diagnosis for rolling bearing,” Mechanical Systems and Signal Processing, vol. 19, no. 5 pp. 974-988, 2005.
[19] J. E. Shahan and G. Kamperman, “Handbook of Industrial Noise Control,” Industrial Press, 1976.