旋窯是一種熱處理設備，是水泥廠的核心設備，基本組成中以金屬窯殼(kiln shell)最為重要，窯殼內層安裝耐火磚(refractory brick)，保護其免受熟化物料的腐蝕性影響，並與窯內的高溫隔離，為了維持與延長耐火磚的壽命，必須在耐火磚表面形成狀態良好的窯皮(coating)，如果窯皮崩塌(coating collapse)太嚴重，甚至會造成耐火磚損壞，停機損失約台幣三千萬元，現代化的水泥廠建置窯殼溫度監視系統(Kiln Shell Temperature Monitoring System, KSTMS)監視整個窯殼的溫度。
耐火磚故障的維修工作，除了更換已確定損壞的耐火磚外，只能以人工抽樣檢查其他未受損耐火磚，故增加了故障停機的風險。因此，本論文提出基於希爾伯特黃轉換(Hilbert-Huang transform, HHT)的KSTMS時間序列數據分析方法，找出窯磚脫落的時間點及其溫度頻率變化的特徵，可做為旋窯耐火磚健康程度的參考。此外，水泥廠依靠人工方式觀察旋窯驅動馬達電流的變化趨勢變化，判斷是否發生窯皮崩塌，本論文結合經驗模態分解法(Empirical Mode Decomposition, EMD)與計算智慧(Computational Intelligence, CI)發展智慧型窯皮崩塌偵測方法，可以很容易辨識出窯皮崩塌的發生，解決人員疏忽或遺漏造成的問題，並可自動搜尋窯皮崩塌位置。
本論文提出的方法，未來可應用於窯皮崩塌與窯磚脫落間的關聯性分析，以預測耐火磚的健康程度指標，耐火磚異常警報功能的改善，提升旋窯設備的可靠度與可用性，降低停機與維修造成的生產損失。

The rotary kiln is a pyroprocessing device that is the core piece of equipment in a cement plant. The most crucial component of the kiln is its steel shell. The interior of the kiln’s steel shell is protected by refractory bricks that insulate it from the high temperatures inside the kiln as well as protect it from the corrosive properties of the processed material. To maintain and extend the life of the refractory brick, forming an effective coating on its surface is necessary. If coating collapse occurs and is significantly large, the refractory bricks can be damaged, requiring a shutdown; the cost of production loss and maintenance for each shutdown is approximately NT$30 million. Modern cement plants install a kiln shell temperature monitoring system (KSTMS) to monitor the temperature of the entire rotary kiln shell.
Regarding the repair work required after refractory brick failure, in addition to replacing the damaged refractory bricks, only manual inspection of other undamaged refractory bricks increases the risk of downtime. Therefore, this dissertation proposes a Hilbert–Huang transform–based KSTMS time series data analysis method to determine the relationship between the time at which refractory bricks fall off and the frequency variation in temperature. The information provided can be used as a reference for the health level of rotary kiln refractory bricks. In addition, cement plants rely on manual observation of the variations in the current trend of the rotary kiln drive motor to determine whether coating collapse has occurred.
Therefore, this study combined empirical mode decomposition with computational intelligence to develop an intelligent coating collapse detection method. It easily identifies the occurrence of coating collapse, solves problems caused by personnel negligence or omission, and automatically searches for the location of the coating collapse.
The proposed method can be applied to analyzing the correlations between coating collapse and refractory brick falling in the future to predict the health index of refractory bricks, thereby improving the alarm function for abnormal refractory bricks and the reliability and availability of rotary kiln equipment as well as reducing the production losses caused by downtime and maintenance.

致謝 I
摘要 III
ABSTRACT V
CONTENTS VII
LIST OF FIGURES IX
LIST OF TABLES XIII
LIST OF ABBREVIATIONS XV
LIST OF SYMBOLS XVII
CHAPTER 1 INTRODUCTION 1
1.1 OVERVIEW 1
1.2 MOTIVATION OF THE STUDY 15
1.3 CONTRIBUTIONS AND FEATURES 16
1.4 ORGANIZATION OF THE DISSERTATION 18
CHAPTER 2 BACKGROUND KNOWLEDGE 21
2.1 KILN DRIVE SYSTEM 21
2.2 KSTMS 27
2.3 EMD AND HHT 34
2.4 FIS AND FNN 42
2.4.1 FIS 42
2.4.2 FNN 52
CHAPTER 3 HHT-BASED KSTMS TIME SERIES DATA ANALYSIS 57
3.1 ANALYSIS METHOD 58
3.2 DATA ANALYSIS RESULTS 59
CHAPTER 4 COATING COLLAPSE DETECTION 71
4.1 COATING COLLAPSE PHENOMENON DETECTION 71
4.1.1 Data observation and analysis 72
4.1.2 System modeling 73
4.1.3 FNN 76
4.1.4 EMD-based preprocessing with the trained FNN 78
4.2 DATA ANALYSIS RESULTS 79
4.2.2 Results for Step 1 80
4.2.3 Results for Step 2 81
4.2.4 Results for Step 3 87
4.2.5 Results for Step 4 89
4.3 COATING COLLAPSE LOCATION ESTIMATION 92
CHAPTER 5 CONCLUSION AND FUTURE WORK 99
5.1 CONCLUSION 99
5.2 FUTURE WORK 101
REFERENCE 103
VITA 109

[1] P. Hewlett, M. Liska, Lea's Chemistry of Cement and Concrete 5th Edition, Elsevier Science, 2018.
[2] J. Szczerba, E. Śnieżek, V. Antonovič, “Evolution of Refractory Materials for Rotary Cement Kiln Sintering Zone,” Refractories and Industrial Ceramics, Vol. 58, no.4, pp. 426–433, Nov. 2017.
[3] L. M. Aksel’rod, O. N. Pitsik, I. G. Maryasev, O. A. Maryaseva, V. A. Ustinov, “Development of New Refractory Materials for Cement Industry Rotary Kilns,” Refractories and Industrial Ceramics, vol. 58, no.1, pp. 5–9, May 2017.
[4] J. Szczerba, “Chemical corrosion of basic refractories by cement kiln materials,” Ceramics International, vol. 36, no. 6, pp. 1877-1885, Aug. 2010.
[5] G. Liu, N. Li, W. Yan, C. Gao, W. Zhou, Y. Li, “Composition and microstructure of a periclase–composite spinel brick used in the burning zone of a cement rotary kiln,” Ceramics International, vol. 40, no. 6, , pp. 8149-8155, Jul. 2014.
[6] J. Chen, M. Yan, J. Su, B. Li, J. Sun, “The kiln coating formation mechanism of MgO–FeAl2O4 brick,” Ceramics International, vol. 42, no. 1, pp. 569-575 , Jan. 2016.
[7] V. I. Shubin, “Mechanical Effects on the Lining of Rotary Cement Kilns,” Refractories and Industrial Ceramics, vol.42, no.5-6, pp. 245–250, May 2001.
[8] P. A. Alsop, H. Chen, H. H. Tseng, The Cement Plant Operations Handbook, 5th edn. International Cement Review, Surrey, UK, 2007.
[9] A. J. K. Al-Yasiri, M. A. Muhammed, “Estimating the thickness of coating in the burning zone of cement kilns including the aging factor.” The Iraqi Journal for Mechanical and Material Engineering, vol.12, no.3, pp.545-563, 2012.
[10] T. Honda, H. Kozuka, Y. Tsuchiya, Y. Mizuno, K. Tokunaga, S. Ota, “Damages of Refractory Bricks in Cement Rotary Kiln,” Journal of the Ceramic Society of Japan, vol. 112, S1272-S1277, 2004.
[11] G. Seggewiss, N. Schachter, G. Obermeyer, “Kiln drive application considerations,” in Proc. Cement Industry Technical Conference , pp. 52-76, Kansas City, MO, USA, May 2005.
[12] M. Järvensivu, K. Saari, SL. Jämsä-Jounela, “Intelligent control system of an industrial lime kiln process,” Control Engineering Practice, vol. 9, no. 6, pp. 589-606, Jun. 2001.
[13] S. M. Zanoli, C. Pepe, M. Rocchi, “Improving Performances of a Cement Rotary Kiln,: A Model Predictive Control Solution,” Journal of Automation and Control Engineering, vol. 4, no. 4, Aug. 2016.
[14] G. A. Christopoulos, A. N. Safacas, “Upgrade of a kiln drive system for cement industry,” in Proc. IEEE International Conference Electrical Machines(ICEM), Rome, Italy, 6-8 Sept. 2010.
[15] I. X. Bogiatzidis, A. N. Safacas, E. D. Mitronikas, “A Novel Control Strategy Applicable for a Dual AC Drive With Common Mechanical Load,” IEEE Transactions on Industry Applications, vol. 48, no.6, pp. 2022-2036, 2012.
[16] C. K. Carr, “From DC to AC [rotary lime kiln conversion],” IEEE Industry Applications Magazine, vol. 9, no.4, pp.13-21, 2003.
[17] B. J. Sauer, “Selection of AC Induction Motors for Cement Plant Applications,” in Proc. IEEE Cement Industry Technical Conference Record, Miami, FL, USA, 18-22 May 2008.
[18] J. H. Meyer, J. P. Puckett, “A rotary kiln shell temperature scanner,” Electrical Engineering, vol. 82, no.10, pp. 618–620, 1963.
[19] J. A. Hall, N. G. Bruun, G. Slot, “A Computer-Based Kiln Shell Temperature Monitoring (Control) System,” IEEE Transactions on Industry Applications, vol. IA-21, no.2, pp 333–343, 1985.
[20] J. P. Kemmerer, “Refractory management with a high speed kiln shell scanner,” in Proc. IEEE Technical Conference on cement Industry, pp. 237-251, Tarpon Springs, FL, USA ,1990.
[21] B. K. Chakrabarti, “Investigations on heat loss through the kiln shell in magnesite dead burning process: a case study,” Applied Thermal Engineering, vol.22, no.12, pp. 1339–1345, Aug. 2002.
[22] J. H. Potgieter, R. H. M. Godoi, R. Van Grieken, “A case study of high-temperature corrosion in rotary cement kilns,” Journal of the Southern African Institute of Mining and Metallurgy, vol. 104, no. 10, pp. 603-606, Nov. 2004.
[23] H. Knopfel, D. Schmidt, “Optimized Kiln Operation by Keeping Kiln Shell Temperatures at Set Point with Controlled Water Cooling,” in Proc. IEEE-IAS/PCA Cement Industry Conference, Nashville, TN, USA, May 2018.
[24] Raytek Corporation, “Kiln Shell Temperature Monitoring manual,” Rev J, 2001.
[25] N. E. Huang, Z. Shen, S. R. Long, M. C. Wu, H. H. Shih, Q. Zheng, N. Yen, C.C. Tung, H.H. Liu, “The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis,” Proceedings of the Royal Society A, vol. 454, no.1971, pp. 903-995, 1998.
[26] Y. Jing, D. Qiliang, “Multi-scale predictive modeling for achromic power in a Lithopone Calcination process based on EMD,” in Proc. 9th Asian Control Conference, Istanbul, Turkey, Jun. 2013.
[27] K. Zheng, Y. Zhang, C. Zhao, L. Liu, “Rotary Kiln Cylinder Deformation Measurement and Feature Extraction Based on EMD Method,” Engineering Letters 23(4), pp. 283-291, 2015.
[28] H. Z. Ding, Y. J. Sun, “Bearing Fault Diagnosis Application in Cement Vertical Mill based on Improved Empirical Mode Decomposition,” in Proc. 43rd Annual Conference of the IEEE Industrial Electronics Society, Beijing, China, Oct. 29-Nov. 1, 2017.
[29] S. Lu, X. Wang, H. Yu, H. Dong, Z. Yang, “Trend extraction and identification method of cement burning zone flame temperature based on EMD and least square,” Measurement, vol. 111, pp. 208-215, Dec. 2017.
[30] L. A. Zadeh, “Fuzzy sets,” informat. Control, 8: 338-353, 1965
[31] L. A. Zadeh, “Outline of a New Approach to the Analysis of Complex Systems and Decision Processes,” IEEE Transactions on Systems, Man, and Cybernetics, vol. SMC-3, no. 1, pp. 28-44, Jan. 1973.
[32] L. A. Zadeh, “The concept of a linguistic variable and its application to approximate reasoning—I,II,III”, Information Sciences, vol. 8, no.3, pp. 199-249, vol. 8, no.4, pp.301-357, vol. 9, no. 1, pp.43-80, 1975.
[33] L. X. Wang, “Fuzzy systems are universal approximators,” in Proc. IEEE International Conf. On Fuzzy Systems, San Diego, pp. 1163-1170, 1992.
[34] E. H. Mamdani, S. Assilian, “An experiment in linguistic synthesis with a fuzzy logic controller,” International Journal of Man-Machine Studies, vol. 7, no. 1, pp.1-13, Jan. 1975.
[35] T. Takagi, M. Sugeno, “Fuzzy identification of systems and its applications to modeling and control,” IEEE Transactions on SMC, vol.15, no. 1, pp. 116-132, 1985.
[36] C. C. Lee, “Fuzzy logic in control systems: fuzzy logic controller. II,” IEEE Transactions on SMC, vol.20, no. 2, pp. 419-435, 1990.
[37] E. H. Mamdani, “Application of Fuzzy Logic to Approximate Reasoning Using Linguistic Synthesis,” IEEE Transactions on Computers, vol. C-26, no. 12, pp. 1182-1191, Dec. 1977.
[38] J. Yen and R. Langari, Fuzzy logic: Intelligence, Control, and Information, Prentice Hall, New Jersey, 1999.
[39] S. Akkurt, G. Tayfur, S. Can, “Fuzzy logic model for the prediction of cement compressive strength,” Cement and Concrete Research, vol. 34, no. 8, pp. 1429-1433, Aug. 2004.
[40] A. K. Pani, H. K. Mohanta, “Online monitoring of cement clinker quality using multivariate statistics and Takagi-Sugeno fuzzy-inference technique,” Control Engineering Practice, vol. 57, pp. 1-17, Dec. 2016.
[41] A. K. Pani, H. K. Mohanta, “Soft sensing of particle size in a grinding process: Application of support vector regression, fuzzy inference and adaptive neuro fuzzy inference techniques for online monitoring of cement fineness,” Powder Technology, vol. 264, pp.484-497, Sep. 2014.
[42] C. Anjana, C. N. Dinakardas, “Adaptive Fuzzy Logic Controller for Rotary Kiln Control,” International Research Journal of Engineering and Technology, vol. 3, no. 10, pp. 231-235, Oct. 2016.
[43] H. Zermane, H. Mouss, “Internet and Fuzzy Based Control System for Rotary Kiln in Cement Manufacturing Plant,” International Journal of Computational Intelligence Systems, vol. 10, pp. 835-850, 2017.
[44] M. Sadeghian, A. Fatehi, Identification, prediction and detection of the process fault in a cement rotary kiln by locally linear neuro-fuzzy technique, Journal of Process Control, vol. 21, no. 2, pp. 302-308, Feb. 2011.
[45] M. Fallahpour, A. Fatehi, B. N. Araabi, M. Azizi, “A Neuro-Fuzzy Controller for Rotary Cement Kilns,” IFAC Proceedings, vol.41, no. 2, pp. 13259-13264, 2008.
[46] T. Y. Sun, M. C. Yang, S. J. Tsai, and J. S. He, “Adaptive neural network controller for the flush material belt weigh feeder,” in Proc. 2009 Inter. Joint Conf. Neural Networks, pp. 3018-3023, Atlanta, Georgia, USA, June 14-19, 2009.
[47] F. N. Koumboulis, N. D. Kouvakas, “Indirect adaptive neural control for precalcination in cement plants,” Mathematics and Computers in Simulation, vol. 60, no. 3-5, pp. 325-334, Sep. 2002.
[48] I. Makaremi, A. Fatehi, B. N. Araabi, M. Azizi, A. Cheloeian, “Abnormal condition detection in a cement rotary kiln with system identification methods,” Journal of Process Control, vol. 19, no. 9, pp. 1538-1545, Oct. 2009.
[49] C. T. Chao, Y. J. Chen, C. C. Teng, “Simplification of fuzzy-neural systems using similarity analysis,” IEEE Transaction Systems Man Cybernetic Part B: Cybernetics, vol.26, no. 2, pp. 344-354, 1996.
[50] T. Y. Sun, M. C. Yang, S. J. Tsai, and J. S. He, “An improved flush material belt weigh feeder system via fuzzy logic controller and adaptive neural networks,” in Proc. 2009 IEEE Inter. Conf. Fuzzy Systems, pp. 1259-1263, JeJu Island, Korea, Aug. 2009.