應用均勻實驗設計法優化倒傳遞網路之超參數以預測射出成型成品之比容

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：105

、訪客IP：3.145.155.58

姓名

鄧修奇(Justin H.C. Teng) 查詢紙本館藏

畢業系所

機械工程學系

論文名稱

應用均勻實驗設計法優化倒傳遞網路之超參數以預測射出成型成品之比容
(Optimizing BPNN hyperparameters via uniform design to predict specific volume of injection molded part)

相關論文

★ 田口分析法驗證射出參數對光碟機面板翹曲變形量之研究	★ 聚丙烯射出成型品表面具抗沾黏特性之研究
★ 光學鏡片之有限元素網格品質探討暨模仁全方位體積收縮補償法之研究	★ 從模流到結構的集成分析光學鏡片之模仁變形研究
★ 應用反應曲面法進行鏡筒真圓度之射出成型參數優化	★ 冠狀動脈三維重建之初步架構
★ Zienkiewicz動態多孔彈性力學模型之穩定性探討	★ 外加磁場輔助射出成型對於導電高分子複合材料的磁性纖維配向與導電度之實驗與模擬
★ 骨板與骨釘之參數模型應用於股骨骨折術前規劃	★ 光學鏡片模具之異型水路最佳化設計
★ 傳統骨板與解剖骨板對於固定Sanders II-B型跟骨骨折力學分析	★ 以線性迴歸分析驗證射出成型縫合角與抗拉強度呈正相關
★ 異形水路模具設計對於金屬粉末射出成型槍機卡榫影響之研究	★ 槍機卡榫模流分析參數最佳化之研究
★ 聚碳酸酯與碳纖維複合材料之射出參數對於縫合線強度之研究	★ 運用田口方法分析ABS塑膠材料之射出成型參數對拉伸強度的影響

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

至系統瀏覽論文 (2026-8-25以後開放)

摘要(中)

本研究利用模內感測器收集並記錄射出成型之溫度、壓力值，並
利用壓力、溫度與比容之關聯性以公式量化成品上三點位置比容的不
均勻性。利用 python 程式語言撰寫 Tensorflow 框架之倒傳遞神經網
路模型，將射出成型製程參數與相應比容數據分成訓練組與驗證組進
行訓練，實際以驗證組預測後得到 2.8802%的誤差。為了要降低預測
誤差提升準確度，利用實驗設計法探討超參數之學習率、迭代次數、
批量與預測誤差之關係，分別使用田口方法規劃 L25(53)實驗表與均勻實驗法規劃 U20(203)實驗表，兩種方法經過迴歸望小分析後皆得到最
佳參數組為學習率 0.01、迭代次數 4800、批量 29 之組合，以此組合
實際訓練模型過後得到 1.1015%之預測誤差，相較初始設定之預測誤
差 2.8802%優化了 61.76%，顯示實驗設計法於超參數優化之可行性。

摘要(英)

In this research, in-mold sensors were used to record pressure and
temperature during the injection molding process. A relationship between
pressure, temperature and specific volume was established to quantify the
nonuniformity of three spots on the molded part. The measured data was
then divided into training set and validation set for a back propagation
neural network model based on TensorFlow framework to make
prediction on specific volume of the part, with a prediction error of
2.8802% on validation set. In order to improve prediction accuracy,
applying design of experiment methods to make observation on the
relationship among hyperparameters of learning rate, epoch, batch size
and prediction error. The aforementioned design of experiment methods
were a L25(53
) table of Taguchi method and a U20(203
) table of uniform
design, respectively. The two methods obtained the same outcome of best
parameters set with learning rate 0.01, epoch 4800, and batch size 29
from smaller-the-better regression analyzation. The best parameters set
leads to a lower prediction error of 1.1015%. The improvement of
61.76% compared to initial hyperparameters setting shows that tuning
hyperparameters via design of experiment methods is feasible.

關鍵字(中)

★ 倒傳遞神經網路
★ 實驗設計法
★ 超參數
★ 均勻實驗法

關鍵字(英)

★ Back propagation neural network (BPNN)
★ Design of experiment (DOE)
★ Hyperparameters
★ Uniform design (UD)

論文目次

摘要............................................................................................................II
Abstract.....................................................................................................III
致謝.......................................................................................................... IV
目錄............................................................................................................V
圖目錄.....................................................................................................VII
表目錄...................................................................................................... IX
第 1 章、緒論...........................................................................................1
1-1 前言..................................................................................................1
1-2 文獻回顧..........................................................................................2
1-3 研究目的..........................................................................................5
第 2 章、研究方法 ..................................................................................6
2-1 量化產品收縮程度..........................................................................6
2-2 射出成型數據組..............................................................................8
2-3 類神經網路....................................................................................11
2-4 田口方法........................................................................................15
2-5 均勻實驗法....................................................................................17
第 3 章、研究設備 ................................................................................21
3-1 射出成型設備................................................................................21
3-2 實驗材料........................................................................................27
3-3 測量設備........................................................................................29
第 4 章、結果與討論 ............................................................................33
4-1 BPNN 之目標值量測結果.............................................................33
4-2 BPNN 訓練與預測結果.................................................................36
4-3 均勻實驗法優化結果....................................................................38
4-4 田口方法優化結果........................................................................41
4-5 均勻實驗法與田口方法優化結果比較........................................43
第 5 章、結論與未來展望 ....................................................................45
5-1 結論................................................................................................45
5-2 未來展望........................................................................................46
參考文獻...................................................................................................47

參考文獻

[1] T. Ageyeva, S. Horváth, and J.G. Kovács, "Mold sensors for injection
molding: on the way to industry 4.0", Sensors, Vol. 19(16), 3551, 2019.
[2]科盛科技股份有限公司.Moldex3D 模流分析軟體. Available:
https://ch.moldex3d.com/products/software/moldex3d/
[3] Y. Chang, R.Y. Chang, C.H. Hsu, C.W. Chang, C.C. Chien, H.S.
Chiu, "Method for operating a molding machine with a predicted in-mold
PVT waveform of a molding resin", United States Patent No.9,555,571
B1,2017.
[4] Y. Chang, C.W. Chang, R.Y. Chang, C.H. Hsu, C.C. Chien, H.S.
Chiu, "Molding system and method for operating the same", United
States Patent No.9,684,295 B2.2017.
[5] F. Yin, H. Mao, L. Hua, W. Guo, and M. Shu, "Back propagation
neural network modeling for warpage prediction and optimization of
plastic products during injection molding," Materials & design, vol. 32,
no. 4, pp. 1844-1850, 2011.
[6] W.-C. Chen, P.-H. Tai, M.-W. Wang, W.-J. Deng, and C.-T. Chen,
"A neural network-based approach for dynamic quality prediction in a
plastic injection molding process," Expert systems with Applications, vol.
35, no. 3, pp. 843-849, 2008.
[7] G. A. Lujan-Moreno, P. R. Howard, O. G. Rojas, and D. C.
Montgomery, "Design of experiments and response surface methodology
to tune machine learning hyperparameters, with a random forest
case-study," Expert Systems with Applications, vol. 109, pp. 195-205,
2018.
[8] Z. Yang and A. Zhang, "Hyperparameter Optimization via Sequential
Uniform Designs," Journal of Machine Learning Research, vol. 22, no.
149, pp. 1-47, 2021.
[9] W.-R. Jong, Y.-M. Huang, Y.-Z. Lin, S.-C. Chen, and Y.-W. Chen,
"Integrating Taguchi method and artificial neural network to explore
machine learning of computer aided engineering," Journal of the Chinese
Institute of Engineers, vol. 43, no. 4, pp. 346-356, 2020.
[10]K.-M. Tsai and H.-J. Luo, "An inverse model for injection molding
of optical lens using artificial neural network coupled with genetic
48
algorithm," Journal of Intelligent Manufacturing, vol. 28, no. 2, pp.
473-487, 2017.
[11]M. Altan, "Reducing shrinkage in injection moldings via the Taguchi,
ANOVA and neural network methods," Materials & Design, vol. 31, no.
1, pp. 599-604, 2010.
[12]R. Chang, C. Chen, and K. Su, "Modifying the tait equation with
cooling‐rate effects to predict the pressure–volume–temperature
behaviors of amorphous polymers: Modeling and experiments," Polymer
Engineering & Science, vol. 36, no. 13, pp. 1789-1795, 1996.
[13]W. Hao, Z. Hongtao, G. Qianjian, W. Xiushan, and Y. Jianguo,
"Thermal error optimization modeling and real-time compensation on a
CNC turning center," Journal of materials processing technology, vol.
207, no. 1-3, pp. 172-179, 2008.
[14] A. Aylin Tokuç, "Underfitting and Overfitting in Machine
Learning," baeldung.cs,2022. Available:
https://www.baeldung.com/cs/ml-underfitting-overfitting
[15]Will Koehrsen, "Overtting vs. Undertting: A Complete Example,"
Towards Data Science,2019.
[16]Kai-Tai Fang, Dennis K. J. Lin, Peter Winker and Yong Zhang,
"Uniform Design: Theory and Application, " Technometrics 0040-1706,
42, pp. 237-248,2000.
[17] K.-T. Fang and D. K. Lin, "Uniform experimental designs and their
applications in industry," Handbook of statistics, vol. 22, pp. 131-170,
2003.
[18]台中精機.台中精機射出成型機規格資訊.Available:
https://www.victortaichung.com/injection-machines/tw/vsp-60-e.htm
[19]潤輝科技有限公司. AO系列油式溫度控制機規格表. Available:
http://www.a1-max.com.tw/t/pro_ao.htm#spec
[20]晏邦電機工業有限公司. 料斗乾燥機 (HD/IHD/DHD). Available:
https://www.yannbang.com/hopper-dryer-tw
[21] 双葉電子工業株式会社. 圧力センサボタン形 SSBシリーズ.
Available:
https://premium.ipros.jp/futaba/product/detail/2000410951/
49
[22] 双葉電子工業株式会社. 樹脂温度センサフラッシュマウン
ト形 EPSSZT. Available:
https://premium.ipros.jp/futaba/product/detail/2000411098/

指導教授

鍾禎元(Chen-Yuan Chung)

審核日期

2022-9-7

推文