任何發動機的性能都一定隨著操作時間而受到磨損的影響。其中有若干機制導致燃氣渦輪機的退化和潛在故障：如積聚污垢，腐蝕，氧化，異物損壞，磨損的軸承或密封，過大的葉片端部間隙，燃燒或翹曲的渦輪葉片或葉片，堵塞的燃料噴嘴，裂紋和翹曲的燃燒器，或裂化的轉子盤及葉片等等。
過去以數學模型為主要依據的氣路分析法(Model-based Gas-path Analysis Method)，一直是傳統監控發動機健康狀態與故障診斷的主要方法。然而，現有使用此方法的監控診斷軟體卻因發動機性能參數的缺乏、建模手段的侷限性等種種限制而無法有效在日常修護中被普及使用。為此，本研究首先利用資料驅動(Data driven)的方式：由發動機實際飛行或地面試車之數據，反向對發動機健康狀態建模，來進行故障診斷技術的改善，並證實此演算法模型的確提供了快速且有效的解決方法。其中用於網路訓練、測試的資料乃由NASA所公開之九萬磅推力商用引擎資料集之數據進行正規化並做建模，發展出一套具有健康分類能力之監測診斷系統，並期能將此方法論推廣應用於各式渦輪扇發動機。

In general, engine wear is inevitable, even though it gets worse in engine performance over time. There are several mechanisms caused the degradation and potential failures of gas turbine engine, such as accumulation of dirt, oxidation, foreign object damage, worn bearings or seals, excessive blade end clearance, burning or warped turbine blades or blades, blocked fuel nozzles, cracked and warped burners, or cracked rotor disks and blades.
In the past, the Gas-path Analysis Method based on mathematical models has been the main method for traditional monitoring engine health and fault diagnosis. However, the existing monitoring and diagnostic software cannot be effectively used in daily maintenance due to various limitations of engine performance parameters and modeling methods. Consequently, this study used data driven method through engine's actual flight or ground data, to reverse modeling of the engine health state and to improve the diagnosis technology, and to confirm that the algorithm model provides a quicker and more effective solution. The data used for model training and testing were NASA's public 90,000-pound thrust commercial engine data set, and a diagnostic system with health classification capabilities was developed. Eventually, this methodology is expected to promote and apply to other turbo machines.

摘要 I
Abstract II
致謝 III
CONTENTS IV
List of Tables VI
List of Figure VII
CHAPTER I INTRODUCTION 1
1-1. Research background 1
1-1.1. Aviation Maintenance Management 1
1-1.2. Development of Engine Monitoring Diagnostic Technology 3
1-1.3. Concept of PHM 4
1-1.4. Aeroengine PHM System Constitution 6
1-1.5. PHM System Architecture 11
1-1.6. PHM System Standards 14
1-2. Research Niche 15
1-3. Research Purpose 17
1-4. Research Value 17
1-5. Research Objectives 18
CHAPTER II Experimental design and method 19
2-1. Experimental Data 19
2-1.1. Linear System Gas-path Analysis Method 19
2-1.2. Damage Propagation Model 21
2-2. Data Analysis 23
2-3. Research Design 24
2-4. Methodology 26
2-4.1. Descriptive statistics 26
2-4.2. Principal components analysis (PCA) 29
2-4.3. Normalization (PCA whitening) 32
2-4.4. Linear Regression 33
CHAPTER III Results and discussion 36
3-1. Data Description 37
3-2. View the signal of engine unit number 1 37
3-3. Dimensionality reduction 39
3-4. Outliers 43
3-5. Linear Regression 46
3-6. RUL Prediction 49
CHAPTER IV Conclusion 56
4-1. Review of the Important Research Findings 56
4-2. Applications of the Study 59
4-3. Limitations of the Study 60
4-4. Recommendations for Future Research 61

[1]S. Shappell, C. Detwiler, K. Holcomb, C. Hackworth, A. Boquet, and D. A. Wiegmann, Human error and commercial aviation accidents: an analysis using the human factors analysis and classification system, in Human Error in Aviation: Routledge, 2007, pp. 73-88.
[2]M. Andrenucci and R. Lazzeretti, Problems in fault diagnostics and prognostics for engine condition monitoring, AGARD Diagnostics and Engine Condition Monitoring 10 p(SEE N 75-31083 22-07), 1975.
[3]A. K. Jardine, D. Lin, and D. Banjevic, A review on machinery diagnostics and prognostics implementing condition-based maintenance, Mechanical systems and signal processing, vol. 20, no. 7, pp. 1483-1510, 2006.
[4]J. Banks and B. Walter. (2015). RCM to CBM from Platform To Enterprise. Available: http://slideplayer.com/slide/6241722/
[5]J. K. Beale and A. Hess, Experiences with A-7E and the AV-8B Engine Monitoring Systems: the Good and the Ugly, in Aerospace Conference Proceedings, 2000 IEEE, 2000, vol. 6, pp. 221-227: IEEE.
[6]G. J. Vachtsevanos, F. Lewis, A. Hess, and B. Wu, Intelligent fault diagnosis and prognosis for engineering systems. Wiley Online Library, 2006.
[7]尉询楷, 冯悦, 刘芳, 杨立, and 战立光, 军用航空发动机 PHM 发展策略及关键技术, 航空动力学报, vol. 26, no. 9, pp. 2107-2115, 2011.
[8]Aircraft Gas Turbine Engine Health Management System Guide ARP1587B, 2007.
[9]N. H. W. Eklund. (2009). Prognostics ＆ Health Management. Available: https://www.phmsociety.org/sites/phmsociety.org/files/Eklund_Diagnostics_TutorialPHM09.pdf
[10]Y. Zhou, J. Bo, and T. Wei, A review of current prognostics and health management system related standards, Chemical Engineering, vol. 33, 2013.
[11]A. Saxena and K. Goebel, Turbofan engine degradation simulation data set, NASA Ames Prognostics Data Repository, 2008.
[12]A. Saxena, K. Goebel, D. Simon, and N. Eklund, Damage propagation modeling for aircraft engine run-to-failure simulation, in Prognostics and Health Management, 2008. PHM 2008. International Conference on, 2008, pp. 1-9: IEEE.
[13]L. A. Urban, Gas Path Analysis Applied to Turbine Engine Condition Monitoring, Journal of Aircraft, vol. 10, no. 7, pp. 400-406, 1973.
[14]L. A. Urban and A. J. Volponi, Mathematical methods of relative engine performance diagnostics, SAE technical paper0148-7191, 1992.
[15]D. K. Frederick, J. A. DeCastro, and J. S. Litt, User's guide for the commercial modular aero-propulsion system simulation (C-MAPSS), 2007.
[16]Y. Liu, D. K. Frederick, J. A. DeCastro, J. S. Litt, and W. W. Chan, User's Guide for the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS): Version 2, 2012.
[17]M. Kurosaki, T. Morioka, K. Ebina, M. Maruyama, T. Yasuda, and M. Endoh, Fault Detection and Identification in an IM270 Gas Turbine Using Measurements for Engine Control, Journal of Engineering for Gas Turbines and Power, vol. 126, no. 4, pp. 726-732, 2004.
[18]T. Gale. Distribution, Normal [Online]. Available: https://www.encyclopedia.com/social-sciences/applied-and-social-sciences-magazines/distribution-normal
[19]S.-G. Miaou and J.-S. Chou, Probability density function diagram, in Fundamentals of probability and statistics: Gau Lih Books Co., Ltd., 2012, p. 147.
[20]K. Pearson, LIII. On lines and planes of closest fit to systems of points in space, The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, vol. 2, no. 11, pp. 559-572, 1901.
[21]A. M. Legendre, Nouvelles méthodes pour la détermination des orbites des comètes. F. Didot, 1805.
[22]C. F. Gauss, Theoria motus corporum coelestium in sectionibus conicis solem ambientium. FA Perthes, 1877.
[23]Wyatts. (2009). Bathtub curve.svg. Available: https://commons.wikimedia.org/wiki/File:Bathtub_curve.svg
[24]J. Reason, The contribution of latent human failures to the breakdown of complex systems, Phil. Trans. R. Soc. Lond. B, vol. 327, no. 1241, pp. 475-484, 1990.