在進行熔融紡絲製程加工時，為了得到所期望之品質特性，往往需先決定適當的加工參數，但因其加工參數眾多且複雜，一旦製程發生異常時，需花費相當大的成本在故障排除的工作上。本實驗以最常使用的聚丙稀作為實驗的材料，實驗機台的加工參數共十種，包括三段熱套筒溫度、齒輪幫浦溫度、模頭溫度、紡嘴溫度、螺桿轉速、齒輪幫浦轉速、冷卻風速與捲取速度。本論文以田口實驗計畫法配合變異數分析，可得到最佳的參數組合，並獲知各加工參數對張力變異的影響程度。實驗方面則進行改變機台加工參數，各取十組數據，並根據本實驗設定之判斷法則來辨識顯著與非顯著因子，如在十組的數據中，不超過三組數據落在最佳的參數組合張力變異區間外，則為非顯著因子；反之，則為顯著因子，如此一來即可明確得知張力變異異常之狀態。再透過小波包轉換擷取訊號五個的特徵值，此特徵值包括有最小熵值和小波包最佳基的四個節點，並建構一具準確預測能力的類神經網路架構，使其可以有效且精準區別出異常的加工參數。實驗結果證明了本研究所提出的辨識方法，對於改變單一因子或是同時改變兩因子機台參數下的數據，有著100%辨識成功率，可成功地辨識出異常的加工參數。

In the melt spinning processing, appropriate machine parameters are required to get expected qualities. Once the processing is abnormal, it’s essential to spend a lot of time in fault dignosis. In this experiment, we focus on polypropylene which is the commonly used material and the experiment machine have ten kinds of machine parameters, including three section extruder barrel temperatures, meteting pump temperature, die temperature, spinning temperature, rotation speeds of an extruder, metering pump speed, cooling air speed and take-up velocity. In this thesis, the Taguchi quality design method cooperated with analysis of variance is used to find the best combination of machine parameters to yield the smaller variance of tension and identify significant factors. The experiment goes on with change the machine parameters. Among these parameters, we choose ten groups of data and distinguish significants factor and nonsignificant factors according to the predefined judgment rule. If there are less than three groups of data outside the interval defined by the best combination of machine parameters, it is nonsignificant factor; On the contrary, it is significant factor. Therefore, we can clearly indicate the abnormal state of tension variance. Furthermore, the five feature values of the tension signal are computed from wavelet package transform, which includes the minimum entropy and four nodes of wavelet package best tree. Based on this, we establish a neural network model to distinguish abnormal machine parameters effectively and accurately. The experimental results have proved that either changing one factor or two factors simultaneously in machine parameters, the proposed method can locate abnormal machine parameters with 100% accuracy rate.

1 吳偉欽，“聚丙烯纖維染色性之探討”，絲織園地，第26期，(1998)。
2 A. Dutta and V. M. Nadkarni, “Identifying Critical Process Variable in Poly(ethylene terephthalate) Melt Spinning,” Textile Research Journal, Vol. 54, No. 1, pp. 35-42, (1984).
3 C. Jinan, T. Kikutani, A. Takaku and J. Shimizu, “Nonisothermal Orientation-induced Crystallization in Melt Spinning of Polypropylene,” Journal of Applied Polymer Science, Vol. 37, pp. 2686-2697, (1989).
4 J. S. Denton, J. A. Cuculo and P. A. Tucker, “Computer Simulation of High-speed Spinning of PET,” Journal of Applied Polymer Science, Vol. 57, pp. 939-951, (1995).
5 S. Chen, W. Yu and J. E. Spruiell, “On-line Studies of Structure Development during Melt Spinning of Poly(butylenes telephthalate),” Journal of Applied Polymer Science, Vol. 34, pp. 1477-1492, (1987).
6 K. F. Zieminski and J. Spruiell, “On-line Studies and Computer Simulation of the Melt Spinning of Nylon-66 Filaments,” Journal of Applied Polymer Science, Vol. 35, pp. 2223-2245, (1988).
7 Y. C. Bhuvanesh and V. B. Gupta, “Computer Simulation of Melt Spinning of Polypropylene Fibers Using a Steady-State Model,” Journal of Applied Polymer Science, Vol. 58, pp. 663-674, (1995).
8 鍾清章 校訂，“田口式品質工程導論”，中華民國品質管制學會發行， (1989)。
9 R. S. Chen, H. H. Lee and C. Y. Yu, “Application of Taguchi Method on the Optimal Process Design of an Injection Model PC/PBT Automobile Bumper,” Composite Structures, Vol. 39, No. 3-4, pp. 209-214, (1997).
10 Y. S. Zu and S. T. Lin, “Optimizing the Mechanical Properties of Injection Molded W - 4.9% Ni - 2.1% Fe in Debinding,” Journal of Materials Processing Technology, Vol. 71, pp. 337-342, (1997).
11 C. Chen, P. Jen and F. S. Lai, “Optimization of the Coathanger Manifold Via Computer Simulation and an Orthogonal Array method,” Polymer Engineering and Science, Vol. 37, No. 1, pp. 188-196, (1997).
12 S. J. Liu, C. C. Lai and S. T. Lin, “Optimizing the Impact Strength of Rotationally Molded Parts,” Polymer Engineering and Science, Vol. 40, No. 2, pp. 473-480, (2000).
13 W. M. Lin, C. D. Yang and J. H. Lin, “A Fault Classification Method by RBF Neural Network with OLS Learning Procedure,” IEEE Transactions on Power Delivery, Vol. 16 No. 4, pp. 473-477, (2001).
14 P. Pillay and A. Bhattacharjee, “Application of Wavelets to Model Short-Term Power System Disturbances,” IEEE Transactions on Power Delivery, Vol. 11 No. 4, pp. 2031-2037, (1996).
15 S. Mallat, “A Theory for Multiresolution Signal Decomposition: The Wavelet Representation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 11, No. 7, pp. 674-693, (1989).
16 S. Mallat, “Multiresolution Approximation and Wavelet,” Transactions of The American Mathematical Society, Vol. 135, No. 1, pp. 69-87, (1989).
17 A. Grossman and J. Morlet, “Decompositions of Hardy Functions into Square Integrable Wavelets of Constant Shape,” SIAM Journal of Mathematical Analysis, Vol. 15, No. 4, pp. 723-736, (1984).
18 C. Sidney Burrus, Ramesh A. Gopinath and Haitao Guo, “Wavelets and Wavelet Transforms,” Prentice-Hall, Inc., (1998)
19 S. Santoso, E. J Powers, W. M Grady and P. Hofmann, “Power Quality Assessment via Wavelet Transform Analysis,” IEEE Transactions on Power Delivery, Vol. 11, No. 2, pp. 924-930, (1996).
20B. Perunicic, M. Mallini, Z. Wang and Y. Lin, “Power Quality Disturbance Detection and Classification Using Wavelet and Artificial Neural Networks,” Proceedings of The 8th International Conference on Harmonics and Quality of Power, Vol. 1, pp. 14-16, (1998).
21 L. Angrisani, P. Daponte, D. Apuzzo and A. Testa “A Measurement Method Based on the Wavelet Transform for Power Quality Analysis,” IEEE Transactions on Power Delivery, Vol. 13, No. 4, pp. 990-998, (1998).
22 X. Lou and K. Loparo, “Bearing Fault Diagnosis Based on Wavelet Transform and Fuzzy Inference,” Mechanical Systems and Signal Processing, Vol. 18, No. 5, pp. 1077-1095, (2004).
23 Y. H. Peng, X. G. Xu and H. X. Zhao, “Application of Wavelet Packet Analysis in Turbine Fault Diagnosis,” Proceedings of the Fifth International Conference on Machine Learning and Cybernetics, pp. 13-16, (2006).
24 D. S. S. Lee, B. J. Lithgow and R. E. Morrison, “New Fault Diagnosis of Circuit Breakers,” IEEE Transactions on Power Delivery, Vol. 18, No. 2, pp. 454-459, (2003).
25 K. Yang, G. Shan and L. Zhao, “Application of Wavelet Packet Analysis and Probabilistic Neural Networks in Fault Diagnosis,” Proceedings of The 6th World Congress on Intelligent Control and Automation, pp.21-23, (2006).
26 Z. Wang, Y. Liu and P. J. Griffin, “Neural Net and Expert System Diagnose Transformer Faults,” IEEE Computer Applications in Power, Vol. 13, pp. 50-55, (2000).
27 S. Kase and T. Matsuo, “Studies on Melt Spinning. I. Fundamental Equation on the Dynamics of Melt Spinning,” Journal of Applied Polymer Science, Vol. 3, pp. 2541-2554, (1965).
28 S. Kase and T. Matsuo, “Studies on Melt Spinning. II. Steady-state and Transient Solutions of Fundamental Equations Compared with Experimental Results,” Journal of Applied Polymer Science, Vol. 11, pp. 251-287, (1967).
29 A. Ziabicki, “Fundamentals of Fiber Formation,” John Wiley, London (1976).
30 Z. Tadmor and C. G. Gogos, “Principle of Polymer Processing,” John Wiley, New York, (1979).
31 林朝蒼， “田口品質工程”，中山科學研究院品保中心，(1999)。
32 黎正中譯， “穩健設計之品質工程”，台北圖書有限公司，(1993)。
33 劉克琪， “實驗設計與田口式品質工程”，華泰書局，(1994)。
34 李輝煌， “田口方法—品質設計的原理與實務”，高立圖書有限公司，(2000)。
35 蘇朝墩，“產品穩健設計”，中華民國品質學會，(2002)。
36 胡昌柴、李國柴、劉濤、周志杰，“基於Matlab 6.x的系統分析與設計-小波分析”，西安電子科技大學出版社，(2004)。
37 MATLAB Wavelet Toolbox User’s Guide Version 2.
38 E. K. Diane, “Machine Learning,” Training and development Journal, Vol. 44, No. 12, pp. 24-29, (1990).
39 T. Henson, W. Huxhold and D. Bowman, “An Enhanced Neural Network Learning Algorithm with Simulated Annealing,” Third Workshop on Neural Networks, pp. 87-94, (1992).
40 R. J. Williams and D. Zipser, “A Learning Algorithm for Continually Running Fully Recurrent Neural Networks,” Neural Computation, Vol. 1, pp. 271-279, (1989)
41 S. Geman, G. Bienenstock and R. Doursat, “Neural Networks and the Bias/Variance Dilemma,” Neural Computation, Vol. 4, pp. 1-58, (1992).
42 J. K. Kruschke and J. R. Movellan, “Benefits of gain: Speeded Learning and Minimal Hidden Layers in Back-propagation networks,” IEEE Transactions on n Systems Man＆Cybernetics, Vol. 21, pp. 273-280, (1991).
43 E. E. Martinez, A. E. Smith and B. Idanda, “Reducing Waste in Casting with a Predictive Neural Model,” Journal of Intelligent Manufacturing, Vol. 5, No. 4, pp. 277-286, (1994).
44 D. R. Hush and B. G. Horne, “Progress in Supervised Neural Networks: What’s new since Lippmann,” IEEE Signal Processing Magazine, Vol. 10, No. 1, pp. 8-39, (1993).
45 S. Wang, “An Insight into the Standard Back-Propagation Neural Network Model for Regression Analysis,” International Journal of Management Science, Vol. 26, No. 1, pp. 133-140, (1998).
46 葉怡成，“類神經網路模式應用與實作”，儒林圖書有限公司，(2000)。
47 湯燦泰，“熔融紡絲系統參數最佳化之研究”，台灣科技大學高分子系碩士論文，(2001)。
48 溫耀銘，“應用小波封包轉換與類神經網路於針織物瑕疵辨識”，台灣科技大學高分子系碩士論文，(2002)。
49 劉政原，“熔融紡絲機台之異常診斷”，台灣科技大學高分子系碩士論文， (2007)。
50 A. K. Tiwari and K. K. Shukla, “A Comparative Study of Fuzzy Inferencing Systems Using Discrete Wavelet Transforms in Feature Extraction for the Detection and Classification of Power Quality,” Processdings of Fuzzy Set Theory and its Mathematical Aspects and Applicatons, Vol. 1, pp.220-227, (2002).
51 孫延奎，“小波分析及其應用”，機械工業出版社，(2005)