基于GSA-LSTM动态结构特征提取的间歇过程监测方法

doi:10.11949/0438-1157.20220665

摘要/Abstract

摘要：

间歇过程监测对于保证批次生产过程的稳定运行具有重要意义。传统过程监测方法难以提取间歇过程数据特有的非线性结构和动态时变特征。为此，提出了一种融合图采样聚合网络和长短期记忆网络(GSA-LSTM)的典型相关分析方法用于间歇过程在线监测。首先，利用K近邻方法将批次过程数据转化为图结构形式，利用图采样聚合网络(GraphSAGE)提取数据内部的结构特征，然后利用长短期记忆网络(LSTM)提取数据的非线性动态特征，通过权重系数将结构特征和动态特征融合得到更具有代表性的间歇过程数据特征。进一步地，利用典型相关分析方法对残差建立监测模型。最后将所提方法应用于数值例子和注塑过程监测，结果分析验证了所提方法的有效性。

关键词: 过程监测, 间歇性, 图神经网络, 长短期记忆网络, 典型相关分析, 算法, 过程系统

Abstract:

Intermittent process monitoring is of great importance to ensure the stable operation during the process running. However, it is difficult for traditional process monitoring methods to extract the structure-related and nonlinear dynamic time-varying features of batch process data. To solve this problem, this paper proposes a canonical correlation analysis method combining graph sample aggregate network and long-short term memory network (GSA-LSTM) for batch process monitoring. First, the K-nearest neighbor method is utilized to convert batch data into graph-structured form, and graph sample aggregate network (GraphSAGE) is used to extract the structure features of process data. Then, long-short term memory (LSTM) is used to extract the nonlinear time-varying feature of data at the same time. By integrating them with structure features through weight coefficients, the more representative batch process data features are obtained. After that, canonical correlation analysis method is utilized to carry out process monitoring. Finally, the proposed method is applied to numerical examples and injection molding process monitoring, and the results are analyzed to verify the effectiveness of the proposed method.

Key words: process monitoring, intermittent, graph neural network, long-short term memory, canonical correlation analysis, algorithm, process systems

中图分类号:

TQ 021.8

王雅琳, 潘雨晴, 刘晨亮. 基于GSA-LSTM动态结构特征提取的间歇过程监测方法[J]. 化工学报, 2022, 73(9): 3994-4002.

Yalin WANG, Yuqing PAN, Chenliang LIU. Intermittent process monitoring based on GSA-LSTM dynamic structure feature extraction[J]. CIESC Journal, 2022, 73(9): 3994-4002.

图/表 10

图1 GraphSAGE示意图

Fig.1 Schematic of GraphSAGE

图2 LSTM细胞单元

Fig.2 LSTM cell unit

图3 GSA-LSTM-CCA模型总体框架

Fig.3 The framework of GSA-LSTM-CCA

图4 GSA-LSTM-CCA过程监测流程图

Fig.4 Flow chart of GSA-LSTM-CCA based process monitoring

表1 数值例子故障设置

Table 1 Fault setting for numerical example

故障编号	故障描述
F1	第3000个样本以后，对变量 $y_{1}$ 设置幅值为6的阶跃故障
F2	第3000个样本以后，对 $y_{2}$ 设置斜率为6.0的斜坡故障
F3	第4500个样本以后，对 $y_{1}$ 设置斜率为1.5的斜坡故障

表1 数值例子故障设置

Table 1 Fault setting for numerical example

故障编号	故障描述
F1	第3000个样本以后，对变量 $y_{1}$ 设置幅值为6的阶跃故障
F2	第3000个样本以后，对 $y_{2}$ 设置斜率为6.0的斜坡故障
F3	第4500个样本以后，对 $y_{1}$ 设置斜率为1.5的斜坡故障

表2 数值例子实验结果对比

Table 2 Comparison of experimental results on numerical examples

方法	故障类型	FAR	FDR
PCA	F1	4.97%	100%
	F2	5.63%	26.20%
	F3	4.98%	99.80%
KPCA	F1	5.03%	99.40%
	F2	4.17%	0.40%
	F3	5.04%	99.40%
DPCA	F1	4.00%	91.09%
	F2	4.13%	89.40%
	F3	3.58%	72.60%
LSTM-CCA	F1	4.63%	97.41%
	F2	4.83%	87.76%
	F3	4.97%	71.78%
GSA-LSTM-CCA	F1	1.50%	100%
	F2	1.98%	99.60%
	F3	1.90%	100%

表3 注塑过程故障变量描述

Table 3 Description of fault variables during injection molding

变量分组	变量编号	故障描述
Ⅰ	1	模内压力
	2	模内温度
	3	模温机水流实际流量
Ⅱ	4	实际螺杆位置
Ⅱ	5	喷嘴头射出压力

表4 注塑过程故障设置

Table 4 Fault setting during injection molding

故障编号	故障描述
F1	第3000个样本以后，对模温机水流实际流量设置幅值为10的阶跃故障
F2	第3000个样本以后，对实际螺杆位置设置幅值为10的阶跃故障
F3	第4500个样本以后，对实际螺杆位置设置斜率为0.04的斜坡故障

表5 注塑过程的实验结果对比

Table 5 Comparison of experimental results on injection molding process

方法	故障类型	FAR	FDR
PCA	F1	5.00%	2.45%
	F2	5.00%	2.30%
	F3	5.02%	100%
KPCA	F1	4.97%	2.80%
	F2	4.97%	58.15%
	F3	4.98%	100%
DPCA	F1	14.23%	11.25%
	F2	14.23%	5.88%
	F3	10.57%	72.46%
LSTM-CCA	F1	4.90%	60.44%
	F2	5.03%	80.99%
	F3	5.07%	86.31%
GSA-LSTM-CCA	F1	4.88%	96.30%
	F2	4.87%	100%
	F3	4.93%	100%

图5 五种方法对故障2的监测结果对比

Fig.5 Comparison of monitoring results of fault 2 by five methods

参考文献 30

1	Jiang Q C, Gao F R, Yan X F, et al. Multiobjective two-dimensional CCA-based monitoring for successive batch processes with industrial injection molding application[J]. IEEE Transactions on Industrial Electronics, 2019, 66(5): 3825-3834.
2	薛峰, 李欣铜, 周琨, 等. 基于GLSAFIS的氟化工过程操作单元可靠性监测[J]. 化工学报, 2021, 72(11): 5696-5706.
	Xue F, Li X T, Zhou K, et al. Reliability monitoring of fluorochemical process operation unit based on GLSAFIS[J]. CIESC Journal, 2021, 72(11): 5696-5706.
3	唐俊苗, 俞海珍, 史旭华, 等. 基于潜变量自回归算法的化工过程动态监测方法[J]. 化工学报, 2019, 70(3): 987-994.
	Tang J M, Yu H Z, Shi X H, et al. Dynamic monitoring of chemical processes based on latent variable auto-regressive algorithm[J]. CIESC Journal, 2019, 70(3): 987-994.
4	Nomikos P, MacGregor J F. Monitoring batch processes using multiway principal component analysis[J]. AIChE Journal, 1994, 40(8): 1361-1375.
5	Nomikos P, MacGregor J F. Multi-way partial least squares in monitoring batch processes[J]. Chemometrics and Intelligent Laboratory Systems, 1995, 30(1): 97-108.
6	Yoo C K, Lee J M, Vanrolleghem P A, et al. On-line monitoring of batch processes using multiway independent component analysis[J]. Chemometrics and Intelligent Laboratory Systems, 2004, 71(2): 151-163.
7	Hu K L, Yuan J Q. Multivariate statistical process control based on multiway locality preserving projections[J]. Journal of Process Control, 2008, 18(7/8): 797-807.
8	Chen Z W, Ding S X, Zhang K, et al. Canonical correlation analysis-based fault detection methods with application to alumina evaporation process[J]. Control Engineering Practice, 2016, 46: 51-58.
9	Yao Y Q, Li Y, Jiang B B, et al. Multiple kernel k-means clustering by selecting representative kernels[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(11): 4983-4996.
10	Kurkova V, Coufal D. Translation-invariant kernels for multivariable approximation[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(11): 5072-5081.
11	Pilario K E S, Cao Y, Shafiee M. A kernel design approach to improve kernel subspace identification[J]. IEEE Transactions on Industrial Electronics, 2021, 68(7): 6171-6180.
12	Samuel R T, Cao Y. Kernel canonical variate analysis for nonlinear dynamic process monitoring[J]. IFAC-PapersOnLine, 2015, 48(8): 605-610.
13	Jiang Q C, Yan X F. Locally weighted canonical correlation analysis for nonlinear process monitoring[J]. Industrial & Engineering Chemistry Research, 2018, 57(41): 13783-13792.
14	高学金, 刘腾飞, 徐子东, 等. 基于循环自动编码器的间歇过程故障监测[J]. 化工学报, 2020, 71(7): 3172-3179.
	Gao X J, Liu T F, Xu Z D, et al. Intermittent process fault monitoring based on recurrent autoencoder[J]. CIESC Journal, 2020, 71(7): 3172-3179.
15	Pal A, Hsieh S H. Deep-learning-based visual data analytics for smart construction management[J]. Automation in Construction, 2021, 131: 103892.
16	Yan X Q, Hu S Z, Mao Y Q, et al. Deep multi-view learning methods: a review[J]. Neurocomputing, 2021, 448: 106-129.
17	Cosgriff C V, Celi L A. Deep learning for risk assessment: all about automatic feature extraction[J]. British Journal of Anaesthesia, 2020, 124(2): 131-133.
18	Jiang Q C, Yan X F. Learning deep correlated representations for nonlinear process monitoring[J]. IEEE Transactions on Industrial Informatics, 2019, 15(12): 6200-6209.
19	Zhan Z H, Li J Y, Zhang J. Evolutionary deep learning: a survey[J]. Neurocomputing, 2022, 483: 42-58.
20	Liu X L. Feature recognition of English based on deep belief neural network and big data analysis[J]. Computational Intelligence and Neuroscience, 2021: 5609885.
21	Chen Z, Liang K, Ding S X, et al. A comparative study of deep neural network-aided canonical correlation analysis-based process monitoring and fault detection methods[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021: 1-15.
22	Bai L, Cui L X, Jiao Y H, et al. Learning backtrackless aligned-spatial graph convolutional networks for graph classification[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(2): 783-798.
23	Ruiz L, Gama F, Ribeiro A. Graph neural networks: architectures, stability, and transferability[J]. Proceedings of the IEEE, 2021, 109(5): 660-682.
24	Zou D M, Lerman G. Graph convolutional neural networks via scattering[J]. Applied and Computational Harmonic Analysis, 2020, 49(3): 1046-1074.
25	Chen K J, Hu J, Zhang Y, et al. Fault location in power distribution systems via deep graph convolutional networks[J]. IEEE Journal on Selected Areas in Communications, 2020, 38(1): 119-131.
26	Zhang D C, Stewart E, Entezami M, et al. Intelligent acoustic-based fault diagnosis of roller bearings using a deep graph convolutional network[J]. Measurement, 2020, 156: 107585.
27	Chen Z, Xu J, Peng T, et al. Graph convolutional network-based method for fault diagnosis using a hybrid of measurement and prior knowledge[J]. IEEE Transactions on Cybernetics, 2021: 1-13.
28	Rossi R A, Zhou R, Ahmed N K. Deep inductive graph representation learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2018, 32(3): 438-452.
29	Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780.
30	Chen Z W, Cao Y, Ding S X, et al. A distributed canonical correlation analysis-based fault detection method for plant-wide process monitoring[J]. IEEE Transactions on Industrial Informatics, 2019, 15(5): 2710-2720.

[1]	陈哲文, 魏俊杰, 张玉明. 超临界水煤气化耦合SOFC发电系统集成及其能量转化机制[J]. 化工学报, 2023, 74(9): 3888-3902.
[2]	曹跃, 余冲, 李智, 杨明磊. 工业数据驱动的加氢裂化装置多工况切换过渡状态检测[J]. 化工学报, 2023, 74(9): 3841-3854.
[3]	诸程瑛, 王振雷. 基于改进深度强化学习的乙烯裂解炉操作优化[J]. 化工学报, 2023, 74(8): 3429-3437.
[4]	尹刚, 李伊惠, 何飞, 曹文琦, 王民, 颜非亚, 向禹, 卢剑, 罗斌, 卢润廷. 基于KPCA和SVM的铝电解槽漏槽事故预警方法[J]. 化工学报, 2023, 74(8): 3419-3428.
[5]	郑玉圆, 葛志伟, 韩翔宇, 王亮, 陈海生. 中高温钙基材料热化学储热的研究进展与展望[J]. 化工学报, 2023, 74(8): 3171-3192.
[6]	李贵贤, 曹阿波, 孟文亮, 王东亮, 杨勇, 周怀荣. 耦合固体氧化物电解槽的CO₂制甲醇过程设计与评价研究[J]. 化工学报, 2023, 74(7): 2999-3009.
[7]	邵远哲, 赵忠盖, 刘飞. 基于共同趋势模型的非平稳过程质量相关故障检测方法[J]. 化工学报, 2023, 74(6): 2522-2537.
[8]	高学金, 姚玉卓, 韩华云, 齐咏生. 基于注意力动态卷积自编码器的发酵过程故障监测[J]. 化工学报, 2023, 74(6): 2503-2521.
[9]	周继鹏, 何文军, 李涛. 异形催化剂上乙烯催化氧化失活动力学反应工程计算[J]. 化工学报, 2023, 74(6): 2416-2426.
[10]	郑书闽, 郭鹏程, 颜建国, 王帅, 李文博, 周淇. 微小通道内过冷流动沸腾阻力特性实验及预测研究[J]. 化工学报, 2023, 74(4): 1549-1560.
[11]	贠程, 王倩琳, 陈锋, 张鑫, 窦站, 颜廷俊. 基于社团结构的化工过程风险演化路径深度挖掘[J]. 化工学报, 2023, 74(4): 1639-1650.
[12]	宋冰, 郑城风, 侍洪波, 陶阳, 谭帅. 基于VAE-OCCA的质量相关故障检测方法研究[J]. 化工学报, 2023, 74(4): 1630-1638.
[13]	王子宗, 索寒生, 赵学良. 数字孪生智能乙烯工厂研究与构建[J]. 化工学报, 2023, 74(3): 1175-1186.
[14]	张生安, 刘桂莲. 高效太阳能电解水制氢系统及其性能的多目标优化[J]. 化工学报, 2023, 74(3): 1260-1274.
[15]	顾学荣, 刘硕士, 杨思宇. 基于并行EGO和代理模型辅助的多参数优化方法研究[J]. 化工学报, 2023, 74(3): 1205-1215.