基于误差注意力的晶圆制造数据异常检测

doi:10.3969/j.issn.1004-132X.2020.14.007

摘要/Abstract

摘要： 针对晶圆制造数据异常检测过程中异常特征提取难度大且检测效率不高的问题，提出了一种基于误差注意力的晶圆制造数据异常检测方法。在保持数据分布不变的前提下，将晶圆制造数据转化成灰度图像，根据与正常样本的误差对灰度图像生成基于位置的柔性注意力图，增加误差特征的显性表达并略去冗余特征；利用深度学习神经网络LeNet-5模型将注意力图进行卷积训练，得到异常检测的最优化模型。采用晶圆制造数据集与现有方法进行对比，所提方法耗时缩短160%、F2-Score提高3%，证明了所提方法的有效性。

关键词: 注意力机制, 异常检测, 晶圆制造, 特征提取, 卷积神经网络

Abstract: In wafer manufacturing processes, it was difficult and inefficient to extract abnormal data in the anomaly detection. Therefore, an anomaly detection method was proposed based on error attention mechanism. Firstly, the wafer manufacturing data was converted into grayscale images while keeping the data distribution unchanged. Then, based on the errors with the normal sample, a position-based flexible attention map was generated for the grayscale images. The dominant expressions of the error features were developed while the redundant features areomitted. In addition, deep learning neural network model LeNet-5 was used to perform convolution training to obtain an optimal model for anomaly detection. Finally, the wafer fabrication data set was used to compare the performance between the proposed method and the existing methods. The proposed method reduces the time consuming by 160% and increases the F2-Score by 3%, which verifies the effectiveness of the method.

Key words: attention mechanism, anomaly detection, wafer fabrication, feature extraction, convolutional neural network

中图分类号:

TP183

余石龙, 鲍劲松, 李婕, 张启华. 基于误差注意力的晶圆制造数据异常检测[J]. 中国机械工程, 2020, 31(14): 1686-1692.

YU Shilong, BAO Jinsong, LI Jie, ZHANG Qihua. Wafer Manufacturing Data Anomaly Detection Based on Error Attention[J]. China Mechanical Engineering, 2020, 31(14): 1686-1692.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.cmemo.org.cn/CN/10.3969/j.issn.1004-132X.2020.14.007

http://www.cmemo.org.cn/CN/Y2020/V31/I14/1686

参考文献

[1] 马羽,王志宽,崔伟.SiGe集成电路工艺技术现状及发展趋势[J].微电子学,2018,48(4):508-514. MA Yu,WANG Zhikuan,CUI Wei.Current Status and Development Trend of SiGe Integrated Circuit Technology[J].Microelectronics,2018,48(4):508-514.
[2] 杨振良,汪俊亮,张洁,等.数据驱动的晶圆图缺陷模式识别方法[J].中国机械工程,2019,30(2):230-236. YANG Zhenliang, WANG Junliang, ZHANG Jie,et al. Data-driven Wafer Pattern Defect Pattern Recognition Method[J]. China Mechanical Engineering, 2019, 30(2):230-236.
[3] 许鸿伟,张洁,吕佑龙,等.基于改进的连续型深度信念网络的晶圆良率预测方法[J/OL].计算机集成制造系统:1-13[2020-03-18].http://kns.cnki.net/kcms/detail/11.5946.tp.20190315.0936.028.html. XU Hongwei, ZHANG Jie, LYU Youlong, et al. Wafer Yield Prediction Method Based on Improved Continuous Deep Belief Network[J/OL].[2020-03-18].http://kns.cnki.net/kcms/detail/11.5946.tp.20190315.0936.028.html.
[4] KOSTROS M, JAKAB F, JANITOR J. Overview of Big Data Analysis for Root Cause Determination and Problem Predictions[C]//2014 IEEE 12th International Conference on Emerging E-Learning Technologies and Applications. Stary Smokovec, Slovakia,2014:15127623.
[5] 韩敏,李宇,韩冰.基于改进结构保持数据降维方法的故障诊断研究[J/OL].自动化学报:1-11[2019-04-12].https://doi.org/10.16383/j.aas.c180138. HAN Min, LI Yu, HAN Bing. Research on Fault Diagnosis Based on Improved Structure to Maintain Data Dimensionality Reduction Method[J/OL].Acta AutomaticaSinica:1-11[2019-04-12].https://doi.org/10.16383/j.aas.c180138.
[6] 韩敏,张占奎.基于改进核主成分分析的故障检测与诊断方法[J].化工学报,2015,66(6):2139-2149. HAN Min,ZHANG Zhankui.Fault Detection and Diagnosis Based on Improved Kernel Principal Component Analysis[J].Journal of Chemical Industry and Engineering,2015,66(6):2139-2149.
[7] GARCIA-ALVAREZ D, FUENTE M J, SAINZ G I. Fault Detection and Isolation in Transient States Using Principal Component Analysis[J]. Journal of Process Control, 2012, 22(3):551-563.
[8] GONG Ruikun, YUAN Kui, NIAN Shanpo, et al. Real-time Reading Recognition of Digital Display Instrument Based on BP Neural Network[C]//20108th IEEE International Conference on Control and Automation. Xiamen,2010:11446536.
[9] WANG J, JIN Z, WAN S. Faults Diagnosis of Induction Motors Based on Artificial Neural Network[C]//International Conference on Control & Automation. Xiamen,2019:167.
[10] YU X, WANG X. Uncorrelated Discriminant Locality Preserving Projections[J]. IEEE Signal Processing Letters, 2008, 15:361-364.
[11] GUO W, YOU X, ZHU Z, et al. Locally Linear Embedding Based Dynamic Texture Synthesis[M]. Berlin:Springer, 2015.
[12] 廖文雄,曾碧,梁天恺,等. 面向高维数据的个人信贷风险评估方法[J]. 计算机工程与应用, 2020, 56(4):219-224. LIAO Wenxiong, ZENG Bi, LIANG Tiankai, et al. Personal Credit Risk Assessment Method for High-Dimensional Data[J]. Computer Engineering and Applications, 2020, 56(4):219-224.
[13] 车建国,赵赛. 基于数据深度的过程工业故障检测方法[J]. 计算机工程与应用, 2020, 56(1):265-271. CHE Jianguo, ZHAO Sai. Fault Detection Method Based on Data Depth for Process Industry[J].Computer Engineering and Application, 2020, 56(1):265-271.
[14] 林荫.基于KNN-SVM的垃圾邮件过滤模型[J].现代电子技术,2016,39(23):90-92. LIN Yin.Spam Mail Filtering Model Based on KNN-SVM[J].Modern Electronic Technique,2016,39(23):90-92.
[15] LI L S, HUANG D, ZHENG C, et al. Image Feature Extraction Based on an Extended Non-negative Sparse Coding Neural Network Model[C]//Advances in Neural Networks,Second International Symposium on Neural Networks. Chongqing, 2005:807-812.
[16] 王勇,周慧怡,俸皓,等.基于深度卷积神经网络的网络流量分类方法[J].通信学报,2018,39(1):14-23. WANG Yong,ZHOU Huiyi,FENG Hao,et al.A Network Traffic Classification Method Based on Deep Convolution Neural Network[J].Journal on Communications,2018,39(1):14-23.
[17] 孙萍,胡旭东,张永军. 结合注意力机制的深度学习图像目标检测[J]. 计算机工程与应用, 2019, 55(17):180-184. SUN Ping, HU Xudong, ZHANG Yongjun. Object Detection Based on Deep Learning and Attention Mechanism[J].Computer Engineering and Application, 2019, 55(17):180-184.
[18] 王培森,宋彦,戴礼荣.基于多通道视觉注意力的细粒度图像分类[J].数据采集与处理,2019,34(1):157-166. WANG Peisen,SONG Yan,DAI Lirong.Classification of Fine-grained Image Based on Multi-channel Visual Attention[J].Data Acquisition and Processing,2019,34(1):157-166.
[19] 杨燕,胡小鹏,吴思宁,等.基于注意力选择的局部特征匹配方法[J].大连理工大学学报,2019,59(2):186-193. YANG Yan, HU Xiaopeng, WU Sining,et al.Local Feature Matching Method Based on Attention Selection[J].Journal of Dalian University of Technology,2019,59(2):186-193.
[20] WANG Xiaoming, XIONG Jiulong, WANG Zhihu, et al. Approach for Image Segmentation Based on Improved Visual Attention Mechanism[C]//IEEE International Conference on Electronic Measurement & Instruments. Harbin,2013:14116726.
[21] 刘金利,张培玲. 改进LeNet-5网络在图像分类中的应用[J]. 计算机工程与应用, 2019, 55(15):32-37. LIU Jinli, ZHANG Peiling. Application of LeNet-5 Neural Network in Image Classification[J].Computer Engineering and Application, 2019, 55(15):32-37.
[22] OLIVEROS G A,WANG R,SEETHARAMAN S,et al. Modeling and Laboratory Scale Proof of Concept of the Horizontal Ribbon Growth Process:Application to Silicon Wafer Manufacturing[C]//38th IEEE Photovoltaic Specialists Conference.Austin,2012:13055682.
[24] 王文冠,沈建冰,贾云得.视觉注意力检测综述[J].软件学报,2019,30(2):416-439. WANG Wenguan, SHEN Jianbing, JIA Yunde.A Review of Visual Attention Detection[J].Journal of Software,2019,30(2):416-439.
[25] HAMKER F H. Predictions of a Model of Spatial Attention Using Sum-and Max-pooling Functions[J]. Neurocomputing, 2018, 56:329-343.