臺灣博碩士論文加值系統

本研究乃針對原棉之各項特性與羅陀式紡紗(Rotor Spinning)之各項品質特性進行研究分析。為求一適當之原棉配置，首先以田口式實驗計畫進行實驗設計，再以High Volume Instrument (HVI)原棉檢測儀所量測之各項原棉性質，應用適當之直交表設計一實驗計畫。為求實驗符合現今業界所需，故實際以工廠之Rieter OE 機台進行Rotor紗之紡製，紡出之成紗再以Uster Tester 3量測其各項成紗特性。
在特性分析方面，應用智慧型控制理論進行要因分析判定。結果顯示在類神經網路(Neural Network)的訓練與測試下，網路學習的誤差均方根(Mean Square Error)會收斂到0.1以下。再以基因演算法(Genetic Algorithms)搜尋最佳解，則可於原棉品質條件內找出一組最佳解。依據智慧型控制系統所得到的結果，建構原棉性質與OE Rotor成紗特性預測模式，亦即在已知的原棉配置條件下、得出最終的成紗特性為何。
原棉性質與OE Rotor成紗特性預測模式採用演化式感知機網路為架構，結合了類神經網路與遺傳基因演算法的特性。與單應用倒傳遞網路相較，其擁有較佳的預測準確性以及較快的收斂速度，故本研究應用於各項成紗特性預測模式系統之建立。
經驗證實驗證明，所建立的各項成紗特性預測模式的平均誤差均可收斂在11%以下，此即證明了該預測系統能有效且正確的預測出最終成紗的品質特性，以及原棉性質與OE Rotor成紗特性間的關係。

In this paper, the quality specificity about the cotton and the Open-End (OE) Rotor Spun yarn is studied. First, the Taguchi method is applied for experiment design to find the fittest cotton collocation. Second, the cotton specificity is measured by using High Volume Instrument system and the suitable orthogonal array is applied to design the experimental plan. Then, the Rieter OE Rotor Spinning Machine M 1/1 is used for spinning the OE Rotor yarn. Finally, the Uster Tester-3 is employed to measure the specificity of yarn.
For quality consideration, the intelligent control is applied for factorial analysis. After training and testing the back-propagation neural network, the mean square error can be converged to 0.1. Then, the genetic algorithm is employed to find the optimal solution, and the best of corresponding solution in the cotton specificity conditions can be found. In this study, the predictive model of the correlation about characteristics of OE yarn and cotton is proposed. In other words, the condition of OE yarn characteristics can be obtained under the fitness condition of cotton collocation.
The predictive model of the correlation about characteristics of cotton and OE yarn is implemented by the evolve perceptron neural network. It combined with genetic algorithm and the back-propagation neural network. It has the better predictive accuracy and the faster converge velocity than that use of the back-propagation neural network.
The predictive model is supported by the experiment. The average error can be converged to 11%. It is proved that the model can predict the condition of OE yarn characteristics effectively and exactly. At the same time, the condition of OE yarn characteristics can also be obtained under the fitness condition of cotton collocation.

中文摘要I
ABSTRACTIII
誌謝V
目錄VI
圖索引XI
表索引XIII
CHAPTER 1. 緒論1
1.1.前言1
1.2.研究動機與目的1
1.3.文獻回顧2
1.4.研究步驟3
CHAPTER 2. 開端式羅陀紡紗之紡紗原理6
2.1.OE紗之成紗概論6
2.1.1.開纖羅拉(Combing Roller)機構6
2.1.2.羅陀(Rotor)紡紗室機構7
2.2.羅陀式紡紗與環錠式紡紗之技術比較10
2.3.羅陀式紡紗與環錠式紡紗之成紗品質特性比較11
CHAPTER 3. 田口式品質工程12
3.1.田口式品質工程概述12
3.2.田口式品質工程的特點13
3.3.參數設計15
3.4.品質特性的分類15
3.5.控制因素的效應分析18
3.6.實例說明19
CHAPTER 4. 類神經網路23
4.1.類神經網路簡述23
4.2.類神經網路的特性23
4.3.生物神經網路簡述25
4.4.類神經網路基本架構26
4.5.類神經網路的運作28
4.6.學習演算法29
4.7.倒傳遞類神經網路30
4.8.網路架構與轉換函數31
4.9.網路運算流程33
4.9.1.前授計算33
4.9.2.回傳調整加權值34
4.9.3.修正加權值36
4.10.運算流程圖37
4.11.網路測試38
CHAPTER 5. 遺傳基因演算法39
5.1.遺傳基因演算法簡介39
5.2.遺傳基因演算法特點40
5.3.遺傳基因演算法基本流程41
5.4.遺傳基因演算法工作原理42
5.4.1.染色體複製43
5.4.2.基因交配44
5.4.3.基因突變46
5.5.實例說明47
5.6.基因演算法與傳統最佳法演算法之比較50
CHAPTER 6. 實驗52
6.1.實驗步驟與方法52
6.2.實驗流程54
6.3.實驗原料55
6.4.實驗設備56
6.4.1.原棉檢驗(HVI 4000 型原棉電腦檢驗儀器)56
6.4.2.棉條紡製設備(USTER MDTA-3)62
6.4.3.OE Rotor Spinning成紗設備 (Rieter OE Rotor Spinning Machine M 1/1)63
6.4.4.成紗檢驗儀器PART - 1 (USTER TESTER 3型非均勻度電腦檢驗儀器)63
6.4.5.成紗檢驗儀器PART — 2 (拉力實驗部分)66
6.5.田口式實驗計劃66
CHAPTER 7. 結果與討論69
7.1.實驗結果整理69
7.2.以SN比回應表進行分析70
7.3.應用類神經網路分析結果84
7.3.1.整體特性架構84
7.3.2.單項特性架構90
7.4.以基因演算法尋找最佳解96
7.5.建構品質改善預測模式104
7.5.1.成紗變異率(CVm(%))預測模式105
7.5.2.I.P.I值預測模式107
7.5.3.毛羽(Hairiness)指數值預測模式109
7.5.4.單紗強力(Strength)預測模式111
7.5.5.成紗延伸性(Elongation)預測模式113
CHAPTER 8. 驗證115
CHAPTER 9. 總結與後續研究之建議117
9.1.總結117
9.2.後續研究之建議118
參考文獻119
附錄A123
附錄B125

[1]M. D. Ethridge, J. D. Towery, J. F. Hembree, “Estimating functional relationships between fiber properties and the strength of Open-End spun yarns”, Textile Research Journal, Vol 52, No 1, p35-44, Jan. 1982.
[2]R. Zhu, M. D. Ethridge “Predicting hairiness for ring and rotor spun yarns and analyzing the impact of fiber properties”, Textile Research Journal, Vol 67, No 9, p694-698, Sep. 1997.
[3]原棉電腦檢驗及配棉技術之開發推廣計劃, 經濟部工業局八十一年度專案計劃執行成果報告(1990)
[4]W. Schlafhorst AG & Co. Autocoro Technology Tour 1998.
[5]楊滄洲, 棉系紡紗技術, 永大書局(1990)
[6]楊先然, 現代棉紡工業技術與管制, 曉園出版社(1996)
[7]莊景文, 棉紡品質及作業管理, 師友工業圖書股份有限公司(1982)
[8]田口玄一, 田口實驗計劃法, 滄海出版社(1997)
[9]Diane M. Byrne and Shin Taguchi, “The Taguehi Approach to Parameter Design”, Quality Progress, P19~26, Dec. 1987.
[10]白賜清, 工業實驗計劃法, 中華民國品質管制學會(1991)
[11]林秀雄, 田口方法與品質工程, 新知企管(1991)
[12]鄭燕琴,田口品質工程技術理論與實務,中華民國品質管制學會(1993)
[13]Thomas P. Ryan, “Taguchi’s Approach to Experimental Design: Some Conerns”, Quality Progress, P34~36, Mar. 1988.
[14]張季娜; 羅仕勇; 宋振昌; 蔡彰文; 陳世璉; 莊泰旭; 邱鎮宏; 高述崙,田口式品質工程導論,中華民國品質管制學會(1993)
[15]Glen Stuart Peace, Taugchi methods: a hands-on approach to quality engineer, Addison - Wesley, P113~337(1993)
[16]葉怡成, 類神經網路-模式應用與實作, 儒林圖書有限公司(1994)
[17]Jacek M. Zurada, Introduction to Artifical Neural Systems, West(1992)
[18]Simon Haykin, Neural Networks-A Comprehensive Foundation, Macmillan(1994)
[19]Chin-Teng Lin, C. S. George Lee, Neural Fuzzy Systems, PHPTR(1996)
[20]林昇甫; 洪成安, 神經網路入門與圖樣辨識, 全華科技圖書股份有限公司(1993)
[21]Pei-Wen Chen, Tsair-Chun Liang, Hon-Fai Yau, Wan-Li Sun, Nai-Chueh Wang, Horng Chyi Lin, Rong Cherng Lien, “Classifying Textile Faults with a Back-Propagation Neural Network Using Power Spectra”, Textile Research Journal, Vol 68, No 2, p121-126, Feb. 1998.
[22]L. Vangheluwe, S. Sette, P. Kiekens, “Modelling Relaxation Behaviour of Yarns Part II: Back Propagation Neural Network Model”, Journal of the Textile Institute Part 1: Fibre Science and Textile Technology, Vol 87, No 2, p305-310, 1996.
[23]Sun-Bae Cho, “Neural-Network Classifiers for Recognizing Totally Unconstrained Handwritten Numerals”, IEEE Transactions on Neural Networks, Vol. 8, No. 1, Jan. 1997.
[24]G. R. Barrett, T. G. Clapp, K. J. Titus, “An On-Line Fabric Classification Technique Using a Wavelet-Based Neural Network Approach”, Textile Research Journal, Vol 66, No 8, p521-528, Aug. 1996.
[25]Kuo-Chin Fan, Yuan-Kai Wang, Bih-Lan Chang, Tzu-Po Wang, Chi-Hsiung Jou, I-Feng Kao, “Fabric Classification Based on Recognition Using a Neural Network and Dimensionality Reduction”, Textile Research Journal, Vol 68, No 3, p179-185, Mar. 1998.
[26]D. E. Goldberg, Genetic Algorithms in Search, Optimization, and Machine Learning, Addison-Wesley(1989)
[27]L. Davis, Handbook of Genetic Algorithms, Van Nostrand Reinhold(1991)
[28]Geoffrey E. Hinton, Peter Dayan, Michael Revow, “Modeling the Manifolds of Images of Handwritten Digits”, IEEE Transactions on Neural Networks, Vol. 8, No. 1, Jan. 1997.
[29]S. Rajasekaran, “Training-Free Counter Propagation Neural Network for Pattern Recognition of Fabric Defects”, Textile Research Journal, Vol 67, No 6, p401-405, Jun. 1997.
[30]M. C. Ramesh, R. Rajamanickam, S. Jayaraman, “The Prediction of Yarn Tensile Properties by Using Artificial Neural Networks”, Journal of the Textile Institute, Vol 86, No 3, p459-469, 1995.
[31]R. Zhu, M. D. Ethridge, “The Prediction of Cotton Yarn Irregularity Based on thr ‘AFIS’ Measurement”, Journal of the Textile Institute, Vol 87, No 3, p509-512, 1996.
[32]Chang Kyu Park, Tae Jen Kang, “Objective Rating of Seam Pucker Using Neural Networks”, Textile Research Journal, Vol 67, No 7, p494-502, Jul. 1997.
[33]Rangaswamy Rajamanickam, Steven M. Hansen, Sundaresan Jayaraman, “Analysis of the Modeling Methodologies for Predicting the Strength of Air-Jet Spun Yarns”, Textile Research Journal, Vol 67, No 1, p39-44, Jan. 1997.
[34]K. Nara, et al., “Implementation of Genetic Algorithm for Distribution Systems Loss Minimum Re-Configuration”, IEEE Transactions on Power Systems, Vol 7, No 3, p1044-1050,Aug. 1992.
[35]J. E. Lansberry, L. Wozniak, “Adaptive Hydrogenerator Governor Tuning with A Genetic Algorithm”, IEEE Transactions on Energy Conversion, Vol 9, No 1, p179-183, Mar. 1994.
[36]P. H. Chen, H. C. Chang, “Genetic Aided Scheduling of Hydraulically Coupled Plants in Hydro-Thermal Coordination”, IEEE Transactions on Power Systems, Vol 11, No 2, p975-981, May. 1996.
[37]王全福, 演化式計算在提昇水泥研磨製程效率之應用, 私立元智工學院管理研究所(1997)
[38]S. Sette, L. Boullart, L. Van Langenhove, “Optimising a Production Process by a Neural Network/Genetic Algorithm Approach”, Engineering Applications of Artificial Intelligence, Vol 9, No 6, p681-689, Dec. 1996.
[39]S. Sette, L. Boullart, L. Van Langenhove, P. Kiekens, “Optimizing the Fiber-to-Yarn Production Process with a Combined Neural Network/Genetic Algorithm Approach”, Textile Research Journal, Vol 67, No 2, p84-92, Feb. 1997.
[40]Vittorio Maniezzo, “Genetic Evolution of the Topology and Weight Distribution of Neural Networks”, IEEE Transactions on Neural Networks, Vol. 5, No. 1, Jan. 1994.
[41]謝政琪, 靜電開纖絲棉複合紗斷面均勻度改善之研究, 國立台灣科技大學纖維暨高分子所(1998)