本論文針對熔融紡絲直徑變異異常情況做研究，在機台加工參數偏離設定值時，可辨識出為何種加工參數。本實驗採用聚丙烯作為實驗材料，機台加工參數共有九個因子，分別為螺桿三段加熱溫度、齒輪幫浦溫度、模頭溫度、紡嘴溫度、螺桿轉速、齒輪幫浦轉速及捲取速度。藉由雙軸向雷射感測器即時量測紗線直徑，並透過田口實驗規劃法及變異數分析，可得到直徑變異最小之最佳參數組合，並得知各加工參數對直徑變異的影響程度。以最佳參數組合為參數設定值，再改變各因子的加工參數進行實驗以取得異常情況之訊號。特徵擷取的部分則使用兩種方法，首先透過小波轉換訊號之最小熵值作為特徵向量。再藉由分析統計製程管制圖選取兩個特徵，分別為偏態(Skewness)及峰值(Kurtosis)。利用這三種特徵值配合三層倒傳遞類神經網路，可有效且精準的區別出異常的加工參數及正常情況，在訓練樣本數足夠的條件下，其辨識成功率可達100 %。為了達到熔融紡絲機台異常診斷之完整性，運用單因子分類結果和倒傳遞類神經網路建構一套雙因子分類流程，實驗結果證明本研究所提出的方法可以成功且有效的辨識各種異常情況。

This thesis forces on diagnosing fault yarn diameter uniformity in the melting spinning process. When the process parameters in the machine deviate from the set values, we can identify which process parameters are faulty. In this study, we use polypropylene as the experimental materials and a total of nine machine process parameters, including screw three-stage heating temperatures, the metering pump temperature, die temperature, spinning temperature, rotation speed of an extruder, metering pump speed and take-up velocity. We use the biaxial laser sensor to measure yarn diameters, get the smallest diameter variance in the best combination of parameters, and know the influence of process parameters on diameter variance by using the Taguchi method and the analysis of variance (ANOVA). We operate the machine at the best combination of parameters, and then change each process parameters to do experiments for obtaining fault diameter signal. For feature extraction, the minimum entropy of signal is computed by wavelet transform as a feature and another two features, namely Skewness and Kurtosis, from statistical processing control charts are obtained. A three-layer back propagation neural network with three feature inputs is used to distinguish fault process parameters from normal situation effectively and accurately. In the situation that the training samples are sufficient, the recognition rate can reach 100%. In order to achieve the integrity of the fault diagnosis in the melting spinning process, we use the single-factor classification results and the back- propagation neural network to construct a two-factor classification process. The experimental results show that the proposed method can effectively and successfully identify various abnormal situations.

參考文獻
[1] 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.
[2] K. Slater, “Yarn evenness,” Textile Institute, Vol. 14, No. 3-4, pp. 1-90, 1986.
[3] H. Fu, H. Huang and H. Chen, “Progress in Modified Polypropylene Fiber,” China Synthetic Fiber Industry, Vol. 26, No. 6, pp. 39-41, 2003.
[4] 鄔國銘，“高分子材料加工工藝學”，中國紡織出版社，2000。
[5] K. Wilczynski, “A Computer Model for Single-Screw Plasticating Extrusion,” Polymer-Plastics Technology and Engineering, Vol. 35, No. 3, pp. 449-477, 1996.
[6] P. L. Shah, E. Steward and G. Yazbak, “A Study of the Effect of the Extrusion Variables and Screw Design on the Thermal and Rheological Characteristics of Acetal,” Polymer Engineering and Science, Vol. 34, No. 15, pp. 1196-1201, 1994.
[7] K. W. Hutchenson, D. D Edie and D. M. Riggs, “Radial Temperature Differences During the Melt Spinning of Fibers,” Journal of Applied Polymer Science, Vol. 29, No.11, pp. 3621-3640, 1984.

[8] V. B. Gupta, S. A. Mondal and Y. C. Bhuvanesh, “Spinning Speed-Throughput Rate Relationships for Polyester, Nylon, and Polypropylene Fibers,” Journal of Applied Polymer Science, Vol. 65, No. 9, pp. 1773-1788, 1997.
[9] 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, No. 3, pp. 663-674, 1995.
[10] V. V. Ginzburg, “Kinetic Model of Fiber Drawing,” Polymer, Vol. 34, No. 24, pp. 5123-5127, 1993.
[11] 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。
[12] 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.
[13] B. Ozcelik and T. Erzurumlu, “Comparison of the Warpage Optimization in the Plastic Injection Molding using ANOVA, Neural Network Model and Genetic Algorithm,” Journal of Materials Processing Technology, Vol. 171, No. 3, pp. 437-445, 2006.
[14] 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.
[15] J. Jin and J. Shi, “Automatic Feature Extraction of Waveform Signals for In-Process Diagnostic Performance Improvement,” Journal of Intelligent Manufacturing, Vol. 12, No. 3, pp. 257-268, 2001.
[16] 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.
[17] A. Elmitwally, S. Farghal, M. Kandil, S. Abdelkader and M. Elkateb, “Proposed Wavelet-Neurofuzzy Combined System for Power Quality Violations Detection and Diagnosis,” IEE Proceedings: Generation, Transmission and Distribution, Vol. 148, No. 1, pp. 15-20, 2001.
[18] 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.
[19] P. Zhuang, “A New Method of Fault Diagnosis for HV Circuit Breakers Based on Wavelet Packet,” IEEE Conference on Industrial Electronics and Applications, pp. 509-513 , 2011.
[20] 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.
[21] R. S. Guh, “Robustness of the Neural Network Based Control Chart Pattern Recognition System to Non-normality,” International Journal of Quality and Reliability Management, Vol. 19, No.1, pp. 97-112, 2002.
[22] D.T. Pham and M.A. Wani, “Feature-Based Control Chart Pattern Recognition,” International Journal of Production Research, Vol. 35, No. 7, pp. 1875-1890, 1997.
[23] A. Hassan, M. S. N. Baksh, A.M. Shaharoun and M. Jamaluddin, “Improved SPC Chart Pattern Recognition using Statistical Features,” International Journal of Production Research, Vol. 41, No. 7, pp. 1587- 1603, 2003.
[24] Y. Zhang, and M. Yang, “A Coordinate SPC Model for Assuring Designated Fit Quality via Quality-Oriented Statistical Tolerancing,” Computers and Industrial Engineering, Vol. 57, No. 1, pp.73-79, 2009.