臺灣博碩士論文加值系統

軸承是機械設備中重要的關鍵零組件，其複雜的工作環境，容易使軸承發生故障，當故障發生時不僅會造成機械性能下降，更可能造成其他元件之損壞，致使更多成本的增加，因此對軸承的機械故障進行診斷具有相當重要的意義。過去許多研究因無量化故障的標準，故不利於自動化之診斷。而本研究主要的目的在於能夠建立一套基於雲端架構之軸承故障診斷系統，該系統能夠提早得知軸承的健康狀況，並降低維修頻率，以減少停機之損失。本研究提出了一種故障定量診斷的新方法，此方法是基於接收者操作特徵曲線對訊號的碎形維度特徵進行量化，以此界定故障閾值，達到自動化診斷的目的。將軸承運轉時所量測到的振動訊號，計算出廣義碎形維度的大小，透過廣義碎形維度隨時間變化的趨勢，判斷軸承是否異常。為得到較為科學且客觀的診斷依據，本文利用接收者操作特徵曲線界定出最佳的故障閾值，作為判斷是否異常的標準。經研究結果發現，本文所提出之診斷方法所建立的最佳分類模型，在相同的測試條件下，針對不同位置的軸承進行診斷，也可以得到良好的診斷結果。為了達到現今產業對機械設備監控的需求，基於上述方法之研究成果，建立了一套軸承故障診斷的雲端系統，除了能夠即時得知軸承的健康狀況外，也能將診斷結果傳送至其它裝置，同時建立儲存軸承訊號的資料庫，方便於將龐大的數據進行備份，以利後續進行檢驗及模型修正，對於軸承故障診斷之雲端系統的建立，已具有相當的成效。

The bearing is a crucial component in mechanical equipment. Complex operating environments easily cause bearing faults, which may not only compromise mechanical performance, but also damage other components and consequently increase the cost. Therefore, diagnosis of mechanical faults of bearings is of great importance. The lack of quantitative standards for faults in the previous studies has hampered the development of automatic diagnosis. The main objective of the present study lies in establishing a cloud-based system for diagnosing bearing faults. The system can provide early notification of the health of bearings and reduce maintenance frequency to avoid losses arising from machine downtime. This study proposed a new quantitative diagnosis method for bearing faults; in this method, the fractal dimension features of signals were quantified based on the receiver operating characteristic curves. These features were then used to determine the fault threshold in order to achieve automatic diagnosis. The vibration signals measured during bearing operation were used to calculate the size of the generalized fractal dimension, whose variation over time was used to determine whether the bearing exhibited abnormality. To enhance the scientific and objective basis of the diagnosis, this study used receiver operating characteristic curves to set the optimal fault threshold for determining the presence of abnormality. According to the results, an optimal classification model established through the proposed diagnosis method performed well, under the same test conditions, in diagnosing faults of bearings even if they were at different positions. To meet the current industrial needs for monitoring mechanical equipment, this study established a cloud system for diagnosing bearing faults based on the study results derived from the aforementioned method. The system not only enables real-time detection of bearing health conditions but also allows for transmission of the diagnosis results to other devices and establishment of a database that stores bearing signals. The database is convenient for backing up a great amount of data to facilitate subsequent inspection and model correction. The proposed cloud system for bearing fault diagnosis was of satisfactory effectiveness.

摘 要......i
Abstract......ii
誌謝......iii
目錄......iv
表目錄......vii
圖目錄......viii
第一章 緒論......1
1.1 前言......1
1.2 文獻回顧......2
1.2.1 軸承故障診斷的發展進程......2
1.2.2 故障診斷的方法與流程......3
1.3 論文主要內容......5
1.4 論文架構......8
第二章 研究方法......9
2.1 碎形維度......9
2.1.1 單一碎形......9
2.1.2 多重碎形......11
2.1.3 振動訊號計算GFDs之方法......12
2.1.4 GFDs計算流程......13
2.2 接收者操作特徵曲線......14
2.2.1 混淆矩陣......14
2.2.2 ROC曲線繪製原理......14
2.2.3 評估檢測準確率指標......16
2.2.4 選擇最佳分類模型......17
2.2.5 最佳閾值求解......19
2.3 診斷系統開發技術......21
2.3.1 伺服器環境......21
2.3.2 資料庫管理系統與工具......23
2.3.3 資料處理......24
2.3.4 動態網頁技術......24
2.3.5 資料傳輸方法......26
2.3.6 警報通知......26
第三章 實例診斷與模型檢驗......29
3.1 診斷分析流程......29
3.2 實驗案例量測說明......30
3.3 時域與頻域分析......32
3.3.1 時域分析......32
3.3.2 頻域分析......33
3.4 建立GFDs與判斷標準......35
3.4.1 第一組實驗......35
3.4.2 第二組實驗......37
3.4.3 建立判斷標準......39
3.5 第一組實驗Bearing 4 模型建立......42
3.5.1 建立最佳分類模型......42
3.5.2 驗證模型......44
3.6 第二組實驗Bearing 1 模型建立......46
3.6.1 建立最佳分類模型......46
3.6.2 驗證模型......48
3.7 結果討論......50
3.8 與先前一些研究比較......51
第四章 診斷系統設計與實現......52
4.1 系統分析與架構規劃......52
4.1.1 系統需求分析......52
4.1.2 系統使用技術與架構......53
4.1.3 系統結構設計......54
4.1.4 使用情境......55
4.2 系統資料庫設計開發......57
4.2.1 資料庫需求分析......57
4.2.2 資料庫的結構設計......58
4.3 機械設備管理系統實現......63
4.3.1 登入模組......63
4.3.2 參數設置模組......63
4.3.3 自動上傳模組......66
4.3.4 診斷結果模組......66
4.4 軸承雲端診斷系統實現......68
4.4.1 登入模組......68
4.4.2 即時診斷結果模組......68
4.4.3 線上診斷模組......70
4.4.4 通報編號查詢模組......71
4.4.5 設備查詢模組......72
4.5 診斷結果通知與查詢......73
4.5.1 Email......73
4.5.2 LINE BOT......73
第五章 結論與未來展望......77
5.1 結論......77
5.2 未來展望......77
參考文獻......79
Extended Abstract......84

[1]Huang, W., Sun, H., Luo, J., & Wang, W. (2019). Periodic feature oriented adapted dictionary free OMP for rolling element bearing incipient fault diagnosis. Mechanical Systems and Signal Processing, 126, 137-160.
[2]Razavi-Far, R., & Saif, M. (2016, December). Ensemble of extreme learning machines for diagnosing bearing defects in non-stationary environments under class imbalance condition. In 2016 IEEE Symposium Series on Computational Intelligence (SSCI) (pp. 1-6). IEEE.
[3]Chesser, R. K., & Rodgers, B. E. (2014). Chernobyl. In Encyclopedia of Toxicology: Third Edition (pp. 822-829). Elsevier.
[4]Marples, D. R. (1988). The social impact of the Chernobyl disaster. Springer.
[5]Jan D. (2015). Cracked shaft possible cause of Repower turbine collapse. Windpower Monthly.
[6]Nandi, S., Toliyat, H. A., & Li, X. (2005). Condition monitoring and fault diagnosis of electrical motors—A review. IEEE transactions on energy conversion, 20(4), 719-729.
[7]Zhao, X., Jia, M., & Lin, M. (2020). Deep Laplacian Auto-encoder and its application into imbalanced fault diagnosis of rotating machinery. Measurement, 152, 107320.
[8]Kim, N.-H., An, D., & Choi, J.-H. (2016). Introduction. Prognostics and Health Management of Engineering Systems, 1–24.
[9]Vale, C., Bonifácio, C., Seabra, J., Calçada, R., Mazzino, N., Elisa, M., ... & Grimes, D. (2016). Novel efficient technologies in Europe for axle bearing condition monitoring–the MAXBE project. Transportation Research Procedia, 14, 635-644.
[10]Stancu, A. G., Zhou, L., & Soua, S. (2019). Acoustic Emission-Based Similarity Analysis: A Baseline Convergence Algorithm. Procedia Structural Integrity, 17, 238-245.
[11]Peng, Z., Kessissoglou, N. J., & Cox, M. (2005). A study of the effect of contaminant particles in lubricants using wear debris and vibration condition monitoring techniques. Wear, 258(11-12), 1651-1662.
[12]Wang, Y., Peter, W. T., Tang, B., Qin, Y., Deng, L., Huang, T., & Xu, G. (2019). Order spectrogram visualization for rolling bearing fault detection under speed variation conditions. Mechanical Systems and Signal Processing, 122, 580-596.
[13]Alian, H., Konforty, S., Ben-Simon, U., Klein, R., Tur, M., & Bortman, J. (2019). Bearing fault detection and fault size estimation using fiber-optic sensors. Mechanical Systems and Signal Processing, 120, 392-407.
[14]Kompella, K. D., Mannam, V. G. R., & Rayapudi, S. R. (2016). DWT based bearing fault detection in induction motor using noise cancellation. Journal of Electrical Systems and Information Technology, 3(3), 411-427.
[15]Kim, Y. H., Tan, A. C., Mathew, J., & Yang, B. S. (2006). Condition monitoring of low speed bearings: A comparative study of the ultrasound technique versus vibration measurements. In Engineering asset management (pp. 182-191). Springer, London.
[16]Li, J., Wang, H., Wang, X., & Zhang, Y. (2020). Rolling bearing fault diagnosis based on improved adaptive parameterless empirical wavelet transform and sparse denoising. Measurement, 152, 107392.
[17]Osman, S., & Wang, W. (2016). A morphological Hilbert-Huang transform technique for bearing fault detection. IEEE Transactions on Instrumentation and Measurement, 65(11), 2646-2656.
[18]Quinde, I. R., Sumba, J. C., Ochoa, L. E., & Morales-Menendez, R. (2019). Bearing fault diagnosis based on optimal time-frequency representation method. IFAC-PapersOnLine, 52(11), 194-199.
[19]Liu, X., Bo, L., & Luo, H. (2015). Bearing faults diagnostics based on hybrid LS-SVM and EMD method. Measurement, 59, 145-166.
[20]Lv, Y., Yuan, R., & Song, G. (2016). Multivariate empirical mode decomposition and its application to fault diagnosis of rolling bearing. Mechanical Systems and Signal Processing, 81, 219-234.
[21]Chen, Z., Mauricio, A., Li, W., & Gryllias, K. (2020). A deep learning method for bearing fault diagnosis based on cyclic spectral coherence and convolutional neural networks. Mechanical Systems and Signal Processing, 140, 106683.
[22]Dhamande, L. S., & Chaudhari, M. B. (2016). Bearing fault diagnosis based on statistical feature extraction in time and frequency domain and neural network. International Journal of Vehicle Structures & Systems, 8(4), 229.
[23]Saari, J., Strömbergsson, D., Lundberg, J., & Thomson, A. (2019). Detection and identification of windmill bearing faults using a one-class support vector machine (SVM). Measurement, 137, 287-301.
[24]Lopes, R., & Betrouni, N. (2009). Fractal and multifractal analysis: a review. Medical image analysis, 13(4), 634-649.
[25]Zhao, J., Chen, S., Zuo, R., & Carranza, E. J. M. (2011). Mapping complexity of spatial distribution of faults using fractal and multifractal models: vectoring towards exploration targets. Computers & Geosciences, 37(12), 1958-1966.
[26]Lin, J., & Chen, Q. (2013). Fault diagnosis of rolling bearings based on multifractal detrended fluctuation analysis and Mahalanobis distance criterion. Mechanical Systems and Signal Processing, 38(2), 515-533.
[27]Liu, H., Wang, X., & Lu, C. (2015). Rolling bearing fault diagnosis based on LCD–TEO and multifractal detrended fluctuation analysis. Mechanical Systems and Signal Processing, 60, 273-288.
[28]Shi, J., Liang, M., & Guan, Y. (2016). Bearing fault diagnosis under variable rotational speed via the joint application of windowed fractal dimension transform and generalized demodulation: A method free from prefiltering and resampling. Mechanical Systems and Signal Processing, 68, 15-33.
[29]Du, W., Tao, J., Li, Y., & Liu, C. (2014). Wavelet leaders multifractal features based fault diagnosis of rotating mechanism. Mechanical Systems and Signal Processing, 43(1-2), 57-75.
[30]Lu, S., Wang, J., & Xue, Y. (2016). Study on multi-fractal fault diagnosis based on EMD fusion in hydraulic engineering. Applied Thermal Engineering, 103, 798-806.
[31]Wang, B., Hu, X., & Li, H. (2017). Rolling bearing performance degradation condition recognition based on mathematical morphological fractal dimension and fuzzy C-means. Measurement, 109, 1-8.
[32]Zhang, Y., Xing, K., Bai, R., Sun, D., & Meng, Z. (2020). An enhanced convolutional neural network for bearing fault diagnosis based on time–frequency image. Measurement, 157, 107667.
[33]Sun, R. B., Yang, Z. B., Zhai, Z., & Chen, X. F. (2019). Sparse representation based on parametric impulsive dictionary design for bearing fault diagnosis. Mechanical Systems and Signal Processing, 122, 737-753.
[34]Nandi, S., Toliyat, H. A., & Li, X. (2005). Condition monitoring and fault diagnosis of electrical motors—A review. IEEE transactions on energy conversion, 20(4), 719-729.
[35]Cao, S., Xu, F., & Ma, T. (2021). Fault diagnosis of rolling bearing based on multiscale one-dimensional hybrid binary pattern. Measurement, 109552.
[36]Wang, Z., Yao, L., Chen, G., & Ding, J. (2021). Modified multiscale weighted permutation entropy and optimized support vector machine method for rolling bearing fault diagnosis with complex signals. ISA transactions.
[37]Antoni, J., & Borghesani, P. (2019). A statistical methodology for the design of condition indicators. Mechanical Systems and Signal Processing, 114, 290-327.
[38]Zhong, J., Wang, D., & Li, C. (2021). A nonparametric health index and its statistical threshold for machine condition monitoring. Measurement, 167, 108290.
[39]Deng, F., Qiang, Y., Liu, Y., Yang, S., & Hao, R. (2020). Adaptive parametric dictionary design of sparse representation based on fault impulse matching for rotating machinery weak fault detection. Measurement Science and Technology, 31(6), 065101.
[40]呂健合（2018）。軸承運轉狀態偵測系統。崑山科技大學電子工程研究所碩士論文，台南市。
[41]Ewert, P., Kowalski, C. T., & Orlowska-Kowalska, T. (2020). Low-cost monitoring and diagnosis system for rolling bearing faults of the induction motor based on neural network approach. Electronics, 9(9), 1334.
[42]洪詩堯（2017）。智能化高速主軸即時異常監控暨熱誤差預測建模。國立中興大學機械工程學系所碩士論文，台中市。
[43]Edgar, G. (2007). Measure, topology, and fractal geometry. Springer Science & Business Media.
[44]Mandelbrot, B. B., & Mandelbrot, B. B. (1982). The fractal geometry of nature (Vol. 1). New York: WH freeman.
[45]Barunik, J., Aste, T., Di Matteo, T., & Liu, R. (2012). Understanding the source of multifractality in financial markets. Physica A: Statistical Mechanics and its Applications, 391(17), 4234-4251.
[46]Yang, D., Zhang, C., & Liu, Y. (2015). Multifractal characteristic analysis of near-fault earthquake ground motions. Soil Dynamics and Earthquake Engineering, 72, 12-23.
[47]Baoliang, Y. Y. W. T. L., & Shixuan, Y. (2010). A New Signal Feature Extraction Method-Fractal dimensions of Time-Frequency Domain. International Journal of Computer Information Systems and Industrial Management Applications, Vol. 2, pp. 155-162.
[48]C. C. Lee and B. W. Tsai. (2005). The Relationship between the Spatial Pattern of Built-up Area and Landform-- An Application of Fractal Theory. Journal of Taiwan Geographic Information Science(3): pp.43-55.
[49]Zhang, S. Q., He, Y. Z., Zhang, J. M., & Zhao, Y. C. (2012). Multi-fractal based fault diagnosis method of rotating machinery. In Applied Mechanics and Materials (Vol. 130, pp. 571-574). Trans Tech Publications Ltd.
[50]Zheng, Z., Jiang, W., Wang, Z., Zhu, Y., & Yang, K. (2015). Gear fault diagnosis method based on local mean decomposition and generalized morphological fractal dimensions. Mechanism and Machine Theory, 91, 151-167.
[51]秦少凡（2015）。台灣股票市場多重碎形去趨勢波動分析。國立東華大學物理學系碩士論文，花蓮縣。
[52]Liu, Z., Shang, P., & Wang, Y. (2019). Multifractal weighted permutation analysis based on Rényi entropy for financial time series. Physica A: Statistical Mechanics and its Applications, 536, 120994.
[53]湯耀期、李國豪（2019）。振動訊號在軸承損傷診斷之應用。台灣聲學學會 108 年會員大會暨第 32 屆學術研討會，國立臺灣科技大學。
[54]Green, D. M., & Swets, J. A. (1966). Signal detection theory and psychophysics (Vol. 1). New York: Wiley.
[55]Netcraft. (2021, May 31). May 2021 Web Server Survey.
https://news.netcraft.com/archives/2021/05/31/may-2021-web-server-survey.html
[56]林怡臻（2021）。重大疫情之功能面與情感面聊天機器人實作：以新型冠狀病毒為例。國立臺北科技大學資訊與財金管理系碩士論文，台北市。
[57]Datareportal. (2021, Feb 11). Digital in Taiwan: All the Statistics You Need in 2021 — DataReportal – Global Digital Insights. https://datareportal.com/reports/digital-2021-taiwan
[58]J. Lee, H. Qiu, G. Yu, J. Lin, and Rexnord Technical Services (2007). IMS, University of Cincinnati. "Bearing Data Set", NASA Ames Prognostics Data Repository , NASA Ames Research Center, Moffett Field, CA
[59]Qiu, H., Lee, J., Lin, J., & Yu, G. (2006). Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics. Journal of sound and vibration, 289(4-5), 1066-1090.
[60]PCB. (2021, Feb 15). Model 353B33 Installation and Operating Manual.
https://www.pcbpiezotronics.fr/wp-content/uploads/353B33.pdf
[61]Arslan, A. T., & Yayan, U. (2019, April). Convolutional Auto-Encoder Based Degradation Point Forecasting for Bearing Data Set. In The International Conference on Artificial Intelligence and Applied Mathematics in Engineering (pp. 817-829). Springer, Cham.
[62]Cavalaglio Camargo Molano, J., Strozzi, M., Rubini, R., & Cocconcelli, M. (2019). Analysis of NASA Bearing Dataset of the University of Cincinnati by Means of Hjorth’s Parameters. In International Conference on Structural Engineering Dynamics ICEDyn 2019.
[63]Ambika, P. S., Rajendrakumar, P. K., & Ramchand, R. (2019, February). An Approach to Rolling Bearing Fault Diagnosis using Fractal Descriptors and Regularized Least Squares. In 2019 Second International Conference on Advanced Computational and Communication Paradigms (ICACCP) (pp. 1-6). IEEE.
[64]Tang, Y., & Li, K. (2021). Quantitative diagnosis of mechanical faults based on generalized fractal dimensions. Measurement Science and Technology, 32(6), 065013.
[65]Mi, L., Tan, W., & Chen, R. (2013). Multi-steps degradation process prediction for bearing based on improved back propagation neural network. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 227(7), 1544-1553.
[66]Lu, W., Li, Y., Cheng, Y., Meng, D., Liang, B., & Zhou, P. (2018). Early fault detection approach with deep architectures. IEEE Transactions on Instrumentation and Measurement, 67(7), 1679-1689.
[67]Liu, C., & Gryllias, K. (2020). A semi-supervised Support Vector Data Description-based fault detection method for rolling element bearings based on cyclic spectral analysis. Mechanical Systems and Signal Processing, 140, 106682.
[68]Yu, J. (2012). Health condition monitoring of machines based on hidden Markov model and contribution analysis. IEEE Transactions on Instrumentation and Measurement, 61(8), 2200-2211.