臺灣博碩士論文加值系統

本文主要在有關探討鎂合金在熱間擠製之製程開發及在半溶融狀態下之加工特性問題，運用田口實驗法、變異數分析、類神經網路等技術，來執行實驗並建構相關之分析模式。
本文內共有五項研究主題：(一) 採用田口實驗法針對AZ31及AZ61鎂合金進行薄板熱間擠製實驗，所使用之模具為直角模其擠製比為35.89及55.85，並藉由擠製實驗嘗試整理出薄板之最佳熱間擠製製程，同時由因子回應分析與變異數分析推論出擠製參數對擠製成品之關連性。(二) 根據以直角模進行鎂合金薄板之熱間擠製實驗所開發出之製程，推導出模具角度為30o鎂合金之薄板擠製時，並藉由類神經網路分析，預測出在不同胚料溫度下使用變速法之擠製加工時調降衝錘速度之時機。(三) 建立田口法之類神經之鎂合金擠製資料庫，資料庫中有關擠製之參數包含材料種類、擠製比、成品種類、擠製初速、胚料加熱溫度、盛錠筒溫度、潤滑劑、及保溫時間等，並針對擠製鎂合金棒材之成品特性進行分析預測。(四) 採用類神經網路分析法預測擠製鎂合金矩形管之抗拉強度分佈趨勢，探討胚料溫度與成品抗拉強度之關係，同時針對不同成形溫度之成品進行金相組織之觀察，確認溫度與強度之變化關係。(五) 尋找AZ61鎂合金之半溶融狀態之溫度範圍，並由此結果探討當AZ61處於半溶融狀態時經水冷卻及經壓縮與擠製成形後，其材料組織之變化狀況。
本研究中開發了可獲得健全鎂合金薄板之熱間擠製製程，並將影響擠製加工成品機械性質之各參數做一有系統之分析，建立一資料庫，並藉由類神經網路分析法，可預測一最佳擠製成品之加工條件，同時也探討了鎂合金半溶融區域及經半溶融加工後之各項特性。
希望上述之研究成果可提供給鎂合金之成形加工之產業界及學術界之相關研究人員參考。

This study mainly investigated the hot extrusion processes of magnesium alloy and the characteristics of this alloy under semi-solid state. Taguchi’s method, ANOVA analysis and Artificial Neural networks (ANNs) are applied to estalsish the analytical model for analyzing the experimental results.

The main topics discussed in this study are the fllowing five items.
1. Taguchi’s experimental method is used to carry out the experiments of AZ31 and AZ61 magnesium alloys on the hot extrusion, extrusion ratios of the die are 35.89 and 55.85. From the results of experiment we try to find out the optimal hot extrusion process of magnesium alloy sheet, furthermore, by the factor effect response analysis and ANOVA analysis, the relation between process parameters and products can be derived.
2. According to the process developed on hot extrusion of magnesium alloy sheet. We try to predict the timing of adjusting extrusion speed during extrusion of converging die (die angle is 60°). Moreover, ANNs analysis is applied to the developed multi-speed method, to predict the timing of adjustment of the initial speed of extrusion in the different billet temperatures.
3. For the prupose of establiship a database of hot extrusion process of magnesium alloy, process parameters considered in this study are material types, extrusion ratios, product types, initial extrusion speeds, billet heting temperatures, lubricants, and hold time at specified temperature. Moreover, the mechanical properties of magnesium alloy rod can be predicted.
4. Applying the ANNs analylical model to predict tensile strength of extruded rectangular tube, and to find the relationship between the billet temperatures and tensile strengths. Furthermore, the microstructures of products under different forming temperature were observed.
5. Some experiments are carried out to find the semi-solid temperature range of AZ61 magnesium alloy, and the changes of microstructures of AZ61 magnesium alloy from normal temperature to 400℃, in a hot chamber within the semi-solid state, and under the condition of semi-solid compression and extrusion.

In this study a called multi-speed method is developed in the extrusion of magnesium alloy sheet. The parameters that influence the mechanical properties of extruded parts are analyzed syematically, then the database of magnesium alloy extrusion in established. Applying ANNs analytical model, the optimal condition of acquring sound extruded products can be obtained. Finally, the semi-solid temperature range and the characteristics of semi-solid forming of AZ61 magnesium alloy are investigated. We hope the results derived from this study can be a useful reference to the industry of magnesium alloy forming, and researches who engage in the related metal forming field.

摘 要 I
ABSTRACT III
誌 謝 V
目 錄 VI
符號索引 XII
圖表索引 XIV
第一章 緒論 1
1.1 研究動機及目的 1
1.2 文獻回顧 5
1.2.1 有關擠製加工之文獻 5
1.2.2 有關田口實驗法之文獻 7
1.2.3 有關類神經網路分析之文獻 8
1.2.4 鎂合金相關製程之文獻 9
1.3 研究方法 10
1.3.1 高擠製比模具之田口擠製實驗 10
1.3.2 不同模具角度之高擠製比之鎂合金擠製 11
1.3.3 結合田口法之類神經網路模式 11
1.3.4 探討矩形管在設定加熱溫度內之強度分佈現象 12
1.3.5 建立鎂合金半溶融狀態之特性研究 13
第二章 鎂合金板材之熱間擠製實驗 14
2.1 熱間擠製原理 14
2.2 不同擠製比之模具及實驗規劃 15
2.3 直交表實驗規劃 17
2.4 試擠之實驗規劃 18
2.5 直交表實驗結果及田口回應分析 22
2.4.1 不同擠製比之擠製負荷分析 23
2.4.2 成品之抗拉強度分析 26
2.5實驗變異數分析結果 30
2.5.1 變異數分析理論 30
2.5.2擠製比為55.85之成品抗拉強度分析 33
2.5.3擠製比為39.5之成品抗拉強度分析 34
2.5.4 討論 35
2.6 田口加法模式訊噪比預測值與確認實驗 35
2.6.1 加法模式之訊噪比預測值 36
2.6.2 驗證實驗 37
2.6.3板材之金相組織之初步結果 38
2.7 結論 40
第三章 變速法運用於高擠製比之加工模式 42
3.1 錐度模具之熱間擠製實驗規劃 42
3.2變速法之選用時機點 47
3.2.1 擠製成品之品質評估 47
3.2.2 直角模與錐度模之調變時機點之差別 49
3.3運用類神經網路建立高擠製比之鎂合金擠製分析模式 53
3.3.1 網路訓練例輸入模組 54
3.3.2 網路架構規劃 55
3.3.3 網路訓練 56
3.4分析結果 57
3.4.1 不同潤滑劑與溫度對調變時機點之影響 57
3.4.2不同擠製速度與溫度對調變時機點之影響 61
3.4.3 不同調整時機點對擠製成形之影響 64
3.5結論 65
第四章 結合田口法之類神經網路建構模式 67
4.1 田口類神經網路模式 68
4.1.1 直交表實驗與因子回應分析 69
4.1.2 變異數分析與田口加模式預測 69
4.2 田口類神經網路訓練輸入模組 70
4.2.1 網路訓練例與追加實驗規劃 70
4.3 網路模式開發實例 76
4.3.1 擠製負荷及成品抗拉強度之預測結果與驗證實驗 76
4.3.2 擠製比25.44之分析結果與驗證實驗 82
4.4 結果討論 84
第五章 不同擠製比之成品抗拉強度及組織之變化 86
5.1 類神經網路訓練模式 86
5.1.1 網路訓練例與追加實驗 86
5.1.2 網路架構規劃與訓練 91
5.2 網路模式分析預測結果 91
5.2.1 不同擠製比及不同擠製速度之抗拉強度之變化 93
5.2.2 驗證實驗 96
5.3 微結構組織變化 98
5.3.1 不同胚料溫度之組織變化 99
5.3.2 不同擠製比之組織變化 101
5.4 結論 102
第六章 AZ61鎂合金之半溶融加工之探討 104
6.1 實驗方法及流程 104
6.2 AZ61鎂合金之半溶融凝固曲線之實驗 105
6.3 半溶融加工之成形特性 107
6.3.1 AZ61鎂合金之冷卻實驗 107
6.3.2 半溶融壓縮成形性 110
6.3.3 半溶融擠製成形特性 116
6.4 材料之顯微組織之觀察 122
6.4.1 半溶融壓縮實驗之組織變化 122
6.4.2 半溶融擠製品之組織觀察 123
6.5 SEM觀察結果 128
6.6 結論 131
第七章 結論與未來研究方向 133
7.1 結論 133
7.2 未來研究方向 135
參考文獻 136
附錄 A 網路訓練例之訓練樣本 145
附錄 B 訊噪比之計算 152
附錄C 倒傳遞類神經網路 154
作者簡介 158
受權書 159

[1] H. Friedrich and S. Schumann “Research for a new age of magnesium
in the automotive industry” Journal of Materials Processing Technology 117, pp.276-281 (2001).
[2] G.E. Dieter, “Mechanical Metallurgy”, McGraw-Hill Inc., New York. (1998)
[3] W.Y. Fowlkes and C.M. Creveling, “Engineering methods for robust product design : using Taguchi methods in technology and product development”, Addision-Wesley, Reading, Massachusetts (1995).
[4] J.A. Freeman and D.M. Skapura, “Neural networks : algoithms, applications, and programming techniques”, Addision-Wesley, Reading, Massachusetts (1991).
[5] Patrick Ulysse and Robert E. Johnson, “A die design model for thin section extrusions”, International Journal of Mechanical Sciences, Vol.41, pp.1067-1088 (1999).
[6] Y.T. Kim, K. Ikeda and T. Murakami, “Metal flow in porthole die extrusion of aluminium”, Journal of Materials Processing Technology, Vol.121, pp.107-115 (2002).
[7] Y.-H. Frank Su, C.-S. Sam Chiang and Chi Y.A. Taso, “Extrusion characteristics of spray-formed AC9A aluminum alloy”, Materials Science and Engineering A, Vol.364, pp.305-312 (2004).
[8] H.R. Darani and M. Ketabchi, “Simulation of L section extrusion using upper bound method”, Materials and Design, Vol.25, pp.535-540 (2004).
[9] M. Bakhshi-Jooybari, “A theoretical and experimental study of friction in metal forming by the use of the forward extrusion process”, Journal of Materials Processing Technology, Vol.125-126, pp.369-374 (2002).
[10] K.Y. Rhee, W.Y. Han, H.J. Park and S.S. Kim, “Fabrication of aluminum/copper clad composite using hot hydrostatic extrusion process and its materials characteristics”, Materials Science and Engineering A, Vol.384, pp.70-76 (2004).
[11] H.J. Park, K.H. Na, N.S. Cho, Y.S. Lee and S.W. Kim, “A study of the hydrostatic extrusion of copper-clad aluminum tube”, Journal of Materials processing Technology, Vol.67, pp.24-28 (1997).
[12] I.N. Chou and C.H. Hung, “The finite-element study on extrusion of powder/solid composite clad rods”, Journal of Materials Processing Technology, Vol.96, pp.124-132 (1999).
[13] C. Kaya and S. Blackburn, “Extrusion of ceramic tubes with complex structures of non-uniform curvatures made from nano-powders”, Journal of the European Ceramic Society, Vol.24, pp.3663-3670 (2004).
[14] Tsutomu Murai, Shin-ichi Matsuoka, Susumu Miyamoto and Yoshinari Oki, “Effects of extrusion conditions on microstructure and mechanical properties of AZ31B magnesium alloy extrusions”, Journal of Materials Processing Technology, Vol.141, pp.207-212 (2003).
[15] M. Chandrasekaran and Y.M.S. John, “Effect of materials and temperature on the forward extrusion of magnesium alloys”, Materials Science and Engineering A, Vol.381, pp.308-319 (2004).
[16] J.Y. Jeng, Y.S. Wong and C.T. Ho, “Over-curing reduction of parts fabricated by the solid laser-diode plotter rapid prototyping system using the Taguchi method”, International Journal of Advanced Manufacturing Technology, Vol.18, pp.683-691 (2001).
[17] G.P. Syrcos, “Die casting process optimization using Taguchi methods”, Journal of Materials Processing Technology, Vol.135, pp.68-74 (2003).
[18] D.C. Ko, D.H. Kim and B.M. Kim, “Application of artificial neural network and Taguchi method to perform design in metal forming considering workability”, International Journal of Machine Tools & Manufacture, Vol.39, pp.771-785 (1999).
[19] K.M. Tsai and P.J. Wang, “Semi-empirical model of surface finish on electrical discharge machining”, International Journal of Machine Tools & Manufacture, Vol.41, pp.1455-1477 (2001).
[20] J.P. Davim, “An experimental study of the tribological behaviour of the brass/stell pair”, Journal of Material Processing Technology, Vol.100, pp.273-277 (2000).
[21] S.S. Joshi, N. Ramakrishnan, H.E. Nagarwalla and P. Ramakrishnan, “Wear of rotary carbide tools in machining of Al/SiCp composites”, Wear, Vol.230, pp.124-132 (1999).
[22] J.P. Davim, N. Marques and A.M. Baptista, “Effect of carbon fiber reinforcement in the frictional behaviour of Peek in a water lubricated environment”, Wear, Vol.251, pp.1100-1104 (2001).
[23] T.Y. Lin and C.H. Tseng, “Optimum design for artificial neural networks: an example in a bicycle derailleur system”, Engineering Application of Artificial Intelligence, Vol.13, pp.3-14 (2000).
[24] R.S. Chen, H.C. Lin and C. Kung, “Optimal dimension of PQFP by using Taguchi method”, Composite Structures, Vol.49, pp.1-8 (2000).
[25] S.J. Dowey and A. Matthews, “Taguchi and TQM Quality issues for surface engineered applications”, Surface and Castings Techlology, Vol.110, pp.86-93 (1998).
[26] S. Yacout, “Assessment of quality activities using Taguchi’s loss function”, Computers industrial Engineering, Vol.35, pp.229-232 (1998).
[27] M. Bounou, S. Lefebvre and X. Dai Do, “Improving the quality of an optimal power flow solution by Taguchi method”, International Journal of Electrical Power & Energy Systems, Vol.17, pp.113-118 (1995).
[28] A.R. Khoei, I. Masters and D.T. Gethin, “Design optimisation of aluminium recycling processes using Taguchi technique”, Journal of Materials Processing Technology, Vol.127, pp.96-106 (2002).
[29] F.M. Ham, “Principles of Neurocomputing for Science and Engineering”, McGraw Hill, 2001.
[30] V. Tandon, H. E1-Mounayri and H. Kishawy, “NC end milling optimization using evolutionary computation”, International Journal of Machine Tools & Manufacture, Vol.42, pp.595-605 (2002).
[31] J.F. Briceno, H. E1-Mounayri and S. Mukhopadhyay, “Selecting and artificial neural network for efficient modeling and accurate simulation of the milling process”, International Journal Tools & Manufacture, Vol.42, pp.663-674 (2002).

[32] P.G. Benardos and G.C. Vosniakos, “Prediction of surface roughness in CNC face milling using neural networks and Taguchi’s design of experiments”, Robotics and Computer Integrated Manufacturing, Vol.18, pp.343-354 (2002).
[33] R.P. Cherian, L.N. Smith and P.S. Midha, “A neural network approach for selection of powder metallurgy materials and process parameters”, Journal of Artificial Intelligence Engineering, Vol.14, pp.39-44 (2000).
[34] D.H. Kim, D.J. Kim and B.M. Kim, “The application of neural networks and statistical methods to process design in metal forming processes”, The international Journal of Advanced Manufacturing Technology, Vol.15, pp.889-894 (1999).
[35] M. Inamdar, P.P. Date, K. Narasimhan, S.K. Maiti and U.P. Singh, “Development of an artificial neural network to predict springback in air vee benging”, The international Journal of Advanced Manufacturing Technology, Vol.16, pp.376-381 (2000).
[36] R. Pilani, K. Narasimhan, S.K. Maiti, U.P. Singh and P.P. Date, “A hybrid intelligent systems approach for die design in sheet metal forming”, The international Journal of Advanced Manufacturing Technology, Vol.16, pp.370-375 (2000).
[37] Y.Y. Li and J. Bridgwater, “Prediction of extrusion pressure using an artificial neural network”, Power Technology, Vol.108, pp.65-73 (2000).
[38] R.K. Jain, V.K. Jain and P.K. Kalra, “Modeling of abrasive flow machining process : a neural network approach”, Wear, Vol.213, pp.242-248 (1999).
[39] J.C. Lin, “Prediction of rolling force and deformation in three-dimensional cold rolling by using the finite-element method and a neural network”, International Journal of Advanced Manufacturing Technology, Vol.20, pp.799-806 (2002).
[40] J.S. Gunasekera, Z. Jia, J.C. Malas and L. Rabelo, “Development of a neural network model for a cold rolling”, Engineering Application of Artificial Intelligence, Vol.11, pp.597-603 (1998).
[41] J.Y. Jeng, T.F. Mau and S.M. Leu, “Prediction of laser butt joint welding parameters using back propagation and learning vector quantization networks”, Journal of Materials Processing Technology, Vol.99, pp.207-218 (2000).
[42] T.Y. Lin and C.H. Tseng, “Optimum design for artificial neural networks: an example in a bicycle derailleur system”, Engineering Application of Artificial Intelligence, Vol.13, pp.3-14 (2000).
[43] I.P. Moreno, T.K. Nandy, J.W. Jones, J.E. Allison, and T.M. Pollock, “Microstructural characterization of a die-cast magnesium-rare earth alloy”, Scripta Materialia, Vol.45, pp.1423-1429 (2001).
[44] Louis Braun, “Casting of magnesium alloys on hot and cold chamber die casting machines”, International Magnesium Conference in Taipei ,R.O.C., pp.153-169 (2001).
[45] G. song, A. Atrens and M. Dargusch, “Influence of microstructure on
the corrosion of diecast AZ91D”, Corrosion Science, Vol.41, pp.249-273 (1999).
[46] S. Kleiner, O. Beffort and P.J. Uggowitzer, “Microstructure evolution during reheating of an extruded Mg-Al-Zn alloy into the semisolid state”, Scripta Materialia, Vol.51, pp.405-410 (2004).
[47] F. Czerwinski, “Magnesium alloy particulates for Thixomolding applications manufactured by rapid solidification”, Materials Science and Engineering A, Vol.367, pp.261-271 (2004).
[48] Z. Koren, H. Rosenson, E.M. Gutman, Ya.B. Unigovski and A. Elieze, “Development of semisolid casting for AZ91 and AM50 magnesium alloys”, Journal of Light Metals, Vol.2, pp.81-87 (2002).
[49] M.T. Pérez-Prado, J.A. del Valle and O.A. Ruano, “Effect of shet thickness on the microstructural evolution of an Mg AZ61 alloy during large strain hot rolling”, Scripta Materialia, Vol.50, pp.667-671 (2004).
[50] T.C. Chang, J.Y. Wang, C.M. O and S. Lee, “Grain refining of magnesium alloy AZ31 by rolling”, Journal of Materials Processing Technology, Vol.140, pp.588-591 (2003).
[51] J. Bohlen, F. Chmelik, P. Dobroñ and F. Kaiser, “Orientation effects on acoustic emission during tensile deformation of hot rolled magnesium alloy AZ31”, Journal of Alloys and Compounds, Vol.378, pp.207-213 (2004).
[52] K. Matsubara, Y. Miyahara, Z. Horita and T.G. Langdon, “Developing superplasticity in a magnesium alloy through a combination of extrusion and ECAP”, Acta Materialia, Vol.51, pp.3073-3084 (2003).
[53] H. Watatnabe, T. Mukai, K. Ishikawa, and K. Higashi, “Low temperature superplasticity of a fine-grained ZK60 magnesium alloy processes by equal-channel-angular extrusion”, Scripta Materialia, Vol46, pp.851-856 (2002).
[54] Shoichiro Yoshihara, Hisashi Nishimura, Hirokuni Yamamoto and Ken-ichi Manabe, “Formability enhancement in magnesium alloy stamping using a local heating and cooling technique: circular cup deep drawing process”, Journal of Materials Processing Technology, Vol. 142, pp.609-613 (2003).
[55] F.K. Chen, T.B. Huang and C.K. Chang, “Deep drawing of square cups with magnesium alloy AZ31 sheets”, International Journal of Machine Tools & Manufacture, Vol.43, pp.1553-1559 (2003).
[56] M. Manoharan, S.C.V. Lim and M. Gupta, “Application of a model for the work hardening behavior to Mg/SiC composites synthesized using a fluxless casting process,” Materials Science and Engineering A, Vol.333, pp.243-249 (2002).
[57] H. Ferkel and B.L. Mordike, “Magnesium strengthened by SiC nanoparticles”, Materials Science and Engeering A, Vol.298, pp.193-199 (2001).
[58] Z.Y. Nan, S. Ishihara, T. Goshima and R. Nakanishi, “Scanning probe microscope observations of fatigue process in magnesium alloy AZ31 near the fatigue limit”, Scripta Materialia, Vol.50, pp.429-434 (2004).
[59] F. Czerwinski, “The oxidation behaviour of an AZ91D magnesium alloy at high temperatures”, Acta Materialis, Vol.50, pp.2639-2654 (2002).
[60] S.H. Yeo, K.A. Ngoi and C. Hang, “A cost-tolerance model for process sequence optimisation”, International Journal of Advanced Manufacturing Technology, Vol.12, pp.423-431 (1997).

[61] P.J. Ross, “Taguchi Techniques for quality engineering”, McGraw-Hill Inc., 2nd, New York. (1996)
[62] 李輝煌，”田口方法：品質設計的原理與實務”，高立圖書，台北市，民國八十九年。
[63] 中國國家標準金屬材料拉伸試驗法的規範
[64] Robert E.Reed-Hill and Reza Abbacshian, “Physical Metallurgy Principles”, THOMSON (1992).
[65] 黃忠良編著，”擠壓加工理論與工藝”，復漢出版社，民國八十三年。
[66] 蔣氏工業慈善基金工業技術課程系列，”鎂合金應用於3C產品之製程技術研習會”，2001。
[67] Z.C. Lin and D.Y. Chang, “Application of a neural network machine learning model in the selection system of sheet metal bending tooling”, International Journal of Artificial Intelligence in Engineering, Vol.10, pp.21-37 (1996).