![]() |
|
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:47 、訪客IP:18.119.105.27
姓名 林靖宇(Ching-Yu Lin) 查詢紙本館藏 畢業系所 資訊工程學系在職專班 論文名稱
(Defective Wafer Detection Using Sensed Numerical Features)相關論文 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]
[檢視]
[下載]
- 本電子論文使用權限為同意立即開放。
- 已達開放權限電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。
- 請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
摘要(中) 半導體製造的基本過程之一是切片,這意味著將晶棒切成許多晶片。
在切片過程中,可能會產生有缺陷的晶圓。
不幸的是,識別缺陷晶片的檢查既浪費時間又檢查困難。為了解決這個問
題,我們建立了一個系統,該系統在切片及表面檢查過程中會使用傳感器
收集晶圓特性(例如溫度,厚度,晶片表面上的圖案等)以檢測晶片在製
造過程中是否有缺陷。
我們在此系統中應用兩種不同模型-GRU 神經網絡和XGBoost。此兩種模型
經過微調後,根據實際數據分析的實驗結果顯示在晶圓的缺陷檢測方面,
GRU 神經網絡的預測準確度和模型訓練時間均優於XGBoost摘要(英) One of the fundamental processes in semiconductor manufacturing is slicing,
which means cutting an ingot into many wafers. During the slicing process, it
is possible to produce defective wafers. Unfortunately, the inspection to identify
defective wafers is time-consuming and dicult. To solve this problem, we build a
system, which uses sensors to collect features (e.g., temperature, thickness, pattern
on wafer surface, etc.) during the slicing process to detect if the wafers are defective
in the manufacturing process. Two dierent models, the GRU neural network and
XGBoost, are implemented in the proposed system. After ne-tuning both models,
experimental results based on real dataset indicate that the GRU neural network
outperforms XGBoost for wafer defective detection in both the prediction accuracy
and model training time.
ii關鍵字(中) ★ GRU神經網絡
★ XGBoost
★ 深度學習
★ 缺陷晶圓檢測關鍵字(英) ★ GRU neural networks
★ XGBoost
★ Deep learning
★ Defective wafer detection論文目次 Contents
1 Introduction 1
2 RelatedWork 3
2.1 Unsupervised learning . . . . . . . . . . . . . . . . . . . . . . 4
2.2 Supervised learning . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2.1 Image-based prediction . . . . . . . . . . . . . . . . . . 4
2.2.2 Value-based prediction . . . . . . . . . . . . . . . . . . 5
3 Preliminary 7
3.1 Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.1.1 Supervised Learning . . . . . . . . . . . . . . . . . . . 7
3.1.2 Neural Networks . . . . . . . . . . . . . . . . . . . . . 9
3.1.3 Deep Neural Networks . . . . . . . . . . . . . . . . . . 9
3.1.4 Recurrent Neural Networks . . . . . . . . . . . . . . . 9
3.1.5 Long Short-Term Memory Cell . . . . . . . . . . . . . 10
3.1.6 Gated Recurrent Unit Cell . . . . . . . . . . . . . . . . 13
3.2 Overtting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.2.1 Data Regularization . . . . . . . . . . . . . . . . . . . 14
3.2.2 Early stopping . . . . . . . . . . . . . . . . . . . . . . 15
3.2.3 Dropout . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4 Design 17
4.1 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.2 Internal Calculation . . . . . . . . . . . . . . . . . . . . . . . . 20
4.3 Defective Wafer Detection . . . . . . . . . . . . . . . . . . . . 20
4.3.1 Feature scaling . . . . . . . . . . . . . . . . . . . . . . 21
4.3.2 RNN model . . . . . . . . . . . . . . . . . . . . . . . . 22
5 Performance 24
5.1 Data Description . . . . . . . . . . . . . . . . . . . . . . . . . 24
5.2 Performance Metrics . . . . . . . . . . . . . . . . . . . . . . . 25
5.3 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . 26
6 Conclusion and Future Work 31參考文獻 [1] Y. T. H. Yoda, Y. Ohuchi, M. Ejiri, An automatic wafer inspection system using
pipelined image processing techniques, 1988, pp. vol. 10,no. 1, pp. 4{16,.
[2] G. P. C. R. Fairley, T.-Y. Fu, B. m. B. Tsai, High throughput brightfield/darkfield
wafer inspection system using advanced optical techniques, 2007, pp. 164,475.
[3] R. M. M. C. Sun, S. Jansen, O. D. Patterson, Semiconductor integrated test struc-
tures for electron beam inspection of semiconductor wafers, 2010, pp. 679,083.
[4] F. W. Mohamed Baker Alawieh, X. Li, Identifying systematic spatial failure patterns
through wafer clustering, 2016.
[5] A. Holzinger, Biomedical informatics: Discovering knowledge in big data, 2014.
[6] A. Joshi, Proceedings of the 9th international joint conference on artificial intelligence
- volume 1, 1985.
[7] J. Y. Hwang, W. Kuo, Model-based clustering for integrated circuit yield enhance-
ment, in: Eur. J. Oper. Res., 2007, pp. 143{153.
[8] D. V. K. Takeshi Nakazawa, Wafer map defect pattern classification and image re-
trieval using convolutional neural network, Vol. 31, 2018, pp. 309{314.
[9] S. S. Mohamed El Mohadab, Belaid Bouikhalene, Predicting rank for scientific re-
search papers using supervised learning, 2018.
[10] C.-T. S. . T. Y. . C.-M. Ke, A neural-network approach for semiconductor wafer
post-sawing inspection, 2002, pp. 260 { 266.
[11] A. K. I. S. N. Srivastava, G. Hinton, R.Salakhutdinov, Dropout: A simple way to
prevent neural networks from overfitting, Vol. 15, 2014.
[12] G. H. D. Rumelhart, R. Williams, Learning representations by back-propagating
error, 1986.
[13] D. Soutner, L. Muller, Application of lstm neural networks in language modelling,
2013, pp. 105{112.
[14] O. K. A. P. J. Kazybek Adam, Kamilya Smagulova, Wafer quality inspection using
memristive lstm, ann, dnn and htm, 2018.
[15] A. Geron, Hands-On Machine Learning with Scikit-Learn TensorFlow, 2017.
[16] M. Z. Wang Guilan, Zhao Hongshan, Application of xgboost algorithm in fault pre-
diction of main bearing of fans, 2019, p. vol. 39.
[17] L. W. Zhao Tianao, Zheng Shanhong, Research on credit risk analysis based on
xgboost, 2018, pp. vol. 21,pp.33{35.
[18] M. Z. S. He, G. A. Wang, D. F. Cook, Multivariate process monitoring and fault
identification using multiple decision tree classifiers, 2013, pp. vol. 51, no. 11, pp.
3355{3371.
[19] Z. W. F. L. B. Xu, X. Liu, J. Liang, Fusion decision model for vehicle lane change
with gradient boosting decision tree, 2019, pp. vol. 53, no. 6, pp. 1{11.
[20] S.-S. K. J. Park, I.-H. Kwon, J.-G. Baek, Spline regression based feature extraction
for semiconductor process fault detection using support vector machine, 2011, pp.
vol. 38, no. 5, pp. 5711{5718.
[21] M.Avriel, Nonlinear programming: Analysis and methods, Dover Publications, 2003.
[22] Y. K. Hoyeop Lee, C. O. Kim, A deep learning model for robust wafer fault moni-
toring with sensor measurement noise, 2017.
[23] J. J. A. Karpathy, L. Fei-Fei, Visualizing and understanding recurrent network, 2015.
[24] E. Lewis, Control of body segment differentiation in drosophila by the bithorax gene
complex, 1982, pp. pages 239{253.指導教授 孫敏德(Ming-Te Sun) 審核日期 2020-7-30 推文 plurk
funp
live
udn
HD
myshare
netvibes
friend
youpush
delicious
baidu
網路書籤 Google bookmarks
del.icio.us
hemidemi
myshare