基于GRU神经网络的自相关过程残差控制图

doi:10.3969/j.issn.1007-7375.2022.01.013

摘要/Abstract

摘要： 为提升自相关过程监控的效率，提出基于门控循环单元 (gated recurrent unit，GRU)神经网络的自相关过程残差控制图。采用受控下的自相关过程数据对GRU网络进行离线训练与测试，对预测误差进行监控，形成控制用残差控制图。采用训练好的GRU网络预测当前过程波动，利用控制用残差控制图判定当前过程是否失控。运用蒙特卡洛仿真法，与基于一阶自回归模型、BP神经网络以及支持向量回归构建的残差控制图进行性能对比。研究表明，过程受控时，所提残差控制图与其他3种的稳态平均运行链长相差不大，即4者的性能表现相当；而在均值偏移异常过程中，所提残差控制图的平均运行链长远小于其他3种，对自相关过程均值偏移具有较好的监控性能。

Abstract: In order to further improve the efficiency of autocorrelation process monitoring, the residual control chart for autocorrelation process using gated recurrent unit (GRU) neural network is proposed. The GRU network is off-line trained and tested with the autocorrelation process data in control to monitor the prediction error and form the residual control chart for control. The trained GRU network is used to predict the current process variation and the residual control chart is used to determine whether the current process is out of control. Monte Carlo simulation method is used to compare the performance with the residual control chart based on first-order autoregressive model, BP neural network and support vector regression. The experiment results indicate that the difference of ARL between the proposed residual control chart and the other three kinds of control charts is small, that is, the performance of the four kinds of control charts is equivalent when the process is in control; while in the process of abnormal mean shift, the ARL of the proposed residual control chart is smaller than the other three kinds of control charts. The monitoring efficiency of the residual chart presented in this study has a remarkable improvement for mean shift of autocorrelation process.

Key words: autocorrelated process, deep learning, gated recurrent unit neural network, control chart, statistical process control

中图分类号:

O213.1
TP277

周昊飞. 基于GRU神经网络的自相关过程残差控制图[J]. 工业工程, 2022, 25(1): 108-113.

ZHOU Haofei. GRU Neural Network-based Residual Control Chart for Autocorrelated Processes[J]. Industrial Engineering Journal, 2022, 25(1): 108-113.

参考文献

[1] LEE P H, TORNG C C, LIAO L F. An economic design of combined double sampling and variable sampling interval X control chart[J]. International Journal of Production Economics, 2012, 138(6): 102–106.
[2] COSTA A F B, MACHADO M A G. Variable parameter and double sampling X charts in the presence of correlation: the Markov chain approach[J]. International Journal of Production Economics, 2011, 130(9): 224–229.
[3] 马义中, 田甜, 刘利平. 自相关过程协方差阵的残差MEWMA控制图[J]. 系统工程学报, 2012, 27(2): 279-286
MA Yizhong, TIAN Tian, LIU Liping. Residual-based MEWMA control chart for the covariance matrix of autocorrelated processes[J]. Journal of Systems Engineering, 2012, 27(2): 279-286
[4] WRIGHT C M, BOOTH D E, HU M Y. Joint estimation: SPC method for short–run autocorrelated data[J]. Journal of Quality Technology, 2001, 33(3): 365-378
[5] 张黎, 董晓阳. 可变抽样区间 ARMA 控制图经济设计[J]. 系统仿真学报, 2018, 30(10): 3869-3874
ZHANG Li, DONG Xiaoyang. Economic design of VSI ARMA charts[J]. Journal of System Simulation, 2018, 30(10): 3869-3874
[6] KALGONDA A A, KULKARNI S R. Multivariate quality process control for autocorrelated processes[J]. Journal of Applied Statistics, 2004, 31(3): 317-327
[7] GUH R S. A hybrid learning-based model for on-line detection and analysis of control chart patterns[J]. Computers and Industrial Engineering(S, 0360, -8352),2005,49(1): 35-62
[8] 孙静. 自相关过程的残差控制图[J]. 清华大学学报(自然科学版), 2002, 42(6): 735-738
SUN Jing. Residual charts for autocorrelated processes[J]. Journal of Tsinghua University (Science & Technology), 2002, 42(6): 735-738
[9] ALSHRAIDEH H, RUNGER G. Process monitoring using hidden markov models[J]. Quality and Reliability Engineering, 2014, 30(8): 1379-1387
[10] ALBARRACIN O Y E, ALENCAR A P, HO LL. CUSUM chart to monitor autocorrelated counts using negative binomial GARMA model[J]. Statistical Methods in Medical Research, 2017, 27(9): 2859-2871
[11] 肖艳, 李亚平, 俞少君. 残差控制图应用中的自相关过程模型研究[J]. 工业工程与管理, 2020, 25(4): 69-76
XIAO Yan, LI Yaping, YU Shaojun. Research on selection of autocorrelation process model in statistical process control[J]. Industrial Engineering and Management, 2020, 25(4): 69-76
[12] ZOBEL C W, COOK D F, NOTTINGHAM Q J. An augmented neural network classification approach to detecting mean shifts in correlated manufacturing process parameters[J]. International Journal of Production Research, 2004, 42(4): 758
[13] CHINNAM R B. Support vector machines for recognizing shifts in correlated and other manufacturing processes[J]. International Journal of Production Research, 2002, 40(17): 4449-4466
[14] WHITE A K, SAFI S K. The efficiency of artificial neural networks for forecasting in the presence of autocorrelated disturbances[J]. International Journal of Statistics and Probability, 2016, 5(2): 51-58
[15] ISSAM B K, MOHAMED L. Support vector regression based residual MCUSUM control chart for autocorrelated process[J]. Applied Mathematics and Computation, 2008, 201(1): 565-574
[16] CHEN S, YU J. Deep recurrent neural network–based residual control chart for autocorrelated processes[J]. Quality and Reliability Engineering International, 2019, 35(4): 1-22
[17] 欧阳红兵, 黄亢, 闫洪举. 基于LSTM神经网络的金融时间序列预测[J]. 中国管理科学, 2020, 28(4): 27-35
OUYANG Hongbing, HUANG Kang, YAN Hongju. Prediction of financial times series based on LSTM neural network[J]. Chinese Journal of Management Science, 2020, 28(4): 27-35
[18] XIANG Wang, LI Feng, WANG Jiaxu. Quantum weighted gated recurrent unit neural network and its application in performance degradation trend prediction of rotating machinery[J]. Neurocomputing, 2018, 313(3): 85-95
[19] FISCHER T, KRAUSS C. Deep learning with long short-term memory networks for financial market predictions[J]. European Journal of Operational Research, 2017, 270(2): 654-669.
[20] 刘明宇, 吴建平, 王钰博, 等. 基于深度学习的交通流量预测[J]. 系统仿真学报, 2018, 30(11): 4100-4105
LIU Mingyu, WU Jianping, WANG Yubo, et al. Traffic flow prediction based on deep learning[J]. Journal of System Simulation, 2018, 30(11): 4100-4105