Deep learning approaches to complex chemical process control manipulating strategies

doi:10.11949/0438-1157.20210357

Abstract

Abstract:

Production operations of modern chemical processes record a large amount of process control manipulating temporal data. How to extract valuable manipulating experiences and rules is of great significance to enhance process operation intelligent level. Previous research results have shown that time series clustering is an effective method for mining historical control manipulating sequences. However, practical working conditions often deviate from the historical data, making it difficult to reconstruct accurate process control manipulating strategies. In response to this problem, this paper proposes a process control manipulation extraction method based on deep learning of data fragmentation, using agglomerative hierarchical temporal clustering based on Levenshtein distance to obtain different process disturbance state classes, extracting corresponding effective manipulating sequences for fragmentation, and employing convolutional neural networks for deep learning and reconstructions of manipulating strategies. The approach is demonstrated by industrial heat exchanger processes, in which, the experiment shows that the proposed approach is able to overcome the poor adaptability and strong dependence on data sources of conventional control manipulating sequence mining methods in practical applications, achieving satisfactory results.

Key words: manipulation strategies, hierarchical clustering, Levenshtein distance, SAX symbolizations, convolutional neural networks

摘要：

现代复杂化工过程生产运行记录了大量的过程调控时序数据，如何提取其中有价值的调控操纵经验和规则，对于提升过程运行智能化水平具有重要意义。时间序列聚类是一种挖掘历史调控操纵序列的有效方法，然而由于实际工况经常与历史数据出现偏差，使得重构准确的过程调控操纵策略出现困难。为此，本文提出了一种基于数据碎片化深度学习的过程调控操纵提取方法，采用基于Levenshtein距离的凝聚层次时序聚类获取不同过程扰动状态类别，提取对应的有效操纵序列进行碎片化处理，采用卷积神经网络对调控操纵策略进行深度学习和重构。将此方法在工业换热器过程上进行了应用，获得了满意的结果，表明所提出的方法能够克服常规操纵序列挖掘的工程应用适应性差、对数据源依赖性强等缺点。

关键词: 操纵策略, 层次聚类, Levenshtein距离, SAX符号化, 卷积神经网络

CLC Number:

TQ 028.8

Xiaojie TANG, Bo YANG, Hongguang LI. Deep learning approaches to complex chemical process control manipulating strategies[J]. CIESC Journal, 2021, 72(9): 4830-4837.

唐晓婕, 杨博, 李宏光. 复杂化工过程调控操纵策略的深度学习方法[J]. 化工学报, 2021, 72(9): 4830-4837.

Figures/Tables 14

Fig.1 Diagrams of complex chemical process manipulations

Fig.2 Aggregations of hierarchical clustering trees

Fig.3 Diagram of the symbol aggregation approximation method

Fig.4 Symbolizations of process control manipulating time series

Fig.5 Flow charts of manipulating strategy deep learning for complex chemical processes

Fig.6 CNN structures for deep learning of industrial process time series data

Fig.7 Flow diagram of the heat exchanger

Fig.8 Hierarchical agglomerative clustering results of T1 and F1

Fig.9 Disturbance categories of T1 and F1

Fig.10 Disturbance states of the process disturbances variables

Fig.11 SAX symbolizations of manipulating variables

Fig.12 Examples of a multi-category confusion matrix

Table 1 Key evaluation indicators of the confusion matrix

重构类别	精确率	准确率	召回率	F₁-Score
1	95.33%	89.77%	94.05%	91.99%
2	92.98%	92.86%	92.86%	92.86%
3	99.33%	94.44%	94.44%	94.44%
4	94.33%	95.83%	87.62%	91.54%
5	94.1%	87.14%	87.14%	87.14%

Table 2 IAE contrasts of the key variable corresponding to the disturbance states

过程扰动状态	关键变量IAE(监督控制)	关键变量IAE(学习策略)
1	9989.62	9129.6
2	10003.09	9061.3
3	9949.4	9201.5
4	9960.45	10381
5	9909.4	9733
6	10019.84	9061.3
7	—	10144

References 31

1	Lines J, Davis L M, Hills J, et al. A shapelet transform for time series classification[C]∥Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD'12. August12-16, 2012.
	Beijing, China. New York: ACM Press, 2012: 289-297.
2	LeCun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
3	Lee K B, Cheon S, Kim C O. A convolutional neural network for fault classification and diagnosis in semiconductor manufacturing processes[J]. IEEE Transactions on Semiconductor Manufacturing, 2017, 30(2): 135-142.
4	Janssens O, Slavkovikj V, Vervisch B, et al. Convolutional neural network based fault detection for rotating machinery[J]. Journal of Sound and Vibration, 2016, 377: 331-345.
5	Yang B, Li H G. A novel convolutional neural network based approach to predictions of process dynamic time delay sequences[J]. Chemometrics and Intelligent Laboratory Systems, 2018, 174: 56-61.
6	张浩, 刘振娟, 李宏光, 等. 基于关联变量时滞分析卷积神经网络的生产过程时间序列预测方法[J]. 化工学报, 2017, 68(9): 3501-3510.
	Zhang H, Liu Z J, Li H G, et al. Process time series prediction based on application of correlated process variables to CNN time delayed analyses[J]. CIESC Journal, 2017, 68(9): 3501-3510.
7	易令, 吕忠元, 丁进良, 等. 面向原油总氢物性预测的数据扩增预处理方法[J]. 控制与决策, 2018, 33(12): 2153-2160.
	Yi L, Lyu Z Y, Ding J L, et al. Data pretreatment approach for crude oil hydrogen properties prediction[J]. Control and Decision, 2018, 33(12): 2153-2160.
8	Wang Y J, Li H G. Industrial process time-series modeling based on adapted receptive field temporal convolution networks concerning multi-region operations[J]. Computers & Chemical Engineering, 2020, 139: 106877.
9	Wang Y J, Zhang Y C, Li H G. Adapted receptive field temporal convolutional networks with bar-shaped structures tailored to industrial process operation models[J]. Industrial & Engineering Chemistry Research, 2020, 59(13): 5482-5490.
10	Keogh E, Lin J. Clustering of time-series subsequences is meaningless: implications for previous and future research[J]. Knowledge and Information Systems, 2005, 8(2): 154-177.
11	Kumar M, Patel N R, Woo J. Clustering seasonality patterns in the presence of errors[C]∥Proceedings of the Eighth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '02. July23-26, 2002.
	Edmonton, Alberta, Canada. New York: ACM Press, 2002: 557-563.
12	Begum N, Ulanova L, Wang J, et al. Accelerating dynamic time warping clustering with a novel admissible pruning strategy[C]∥Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Sydney NSW Australia. New York, NY, USA: ACM, 2015: 49-58.
13	Lin Y P, Yang Y W. Stock markets forecasting based on fuzzy time series model[C]∥2009 IEEE International Conference on Intelligent Computing and Intelligent Systems. November 20-22, 2009, Shanghai, China. IEEE, 2009: 782-786.
14	Fu T, Chung F, Ng V, et al. Pattern discovery from stock time series using self-organizing maps[EB/OL]. [2021-01-27].
15	Akatsuka S, Noda M, Sugimoto K. Similarity analysis of sequential alarms in plant operation data by using levenshtein distance[J]. Kagaku Kogaku Ronbunshu, 2013, 39(4): 352-358.
16	Chen G C, Liu Y, Ge Z Q. K-means Bayes algorithm for imbalanced fault classification and big data application[J]. Journal of Process Control, 2019, 81: 54-64.
17	Luo F L. An improved K-means algorithm and its application in customer classification of network enterprises[J]. Applied Mechanics and Materials, 2014, 543/544/545/546/547: 2124-2127.
18	Schölkopf B, Platt J. A local learning approach for clustering[M].Advances in Neural Information Processing Systems 19: Conference. MIT Press, 2007: 1529-1536.
19	Li J F, Li J S, He H Q. A simple and accurate approach to hierarchical clustering[J]. Journal of Computational Information Systems, 2011, 7(7): 2577-2584.
20	朱坚, 杨博, 王永健, 等. 一种新型的基于Levenshtein距离层次聚类的时序操作优化方法[J]. 化工学报, 2019, 70(2): 581-589.
	Zhu J, Yang B, Wang Y J, et al. New operation optimization method with time series based on Levenshtein distance hierarchical clustering[J]. CIESC Journal, 2019, 70(2): 581-589.
21	Xu D K, Tian Y J. A comprehensive survey of clustering algorithms[J]. Annals of Data Science, 2015, 2(2): 165-193.
22	Bouguettaya A, Yu Q, Liu X M, et al. Efficient agglomerative hierarchical clustering[J]. Expert Systems with Applications, 2015, 42(5): 2785-2797.
23	Putera Utama Siahaan A, Aryza S, Hariyanto E, et al. Combination of levenshtein distance and Rabin-karp to improve the accuracy of document equivalence level[J]. International Journal of Engineering & Technology, 2018, 7(2.27): 17.
24	Levenshtein V. Binary Codes Capable of Correcting Deletions, Insertions, and Reversals[J]. Soviet Physics Doklady, 1966,10(8): 707-710.
25	Ho T, Oh S R, Kim H. A parallel approximate string matching under Levenshtein distance on graphics processing units using warp-shuffle operations[J]. PLoS One, 2017, 12(10): e0186251.
26	Lin J, Keogh E, Lonardi S, et al. A symbolic representation of time series, with implications for streaming algorithms[C]∥Proceedings of the 8th ACM SIGMOD workshop on Research issues in data mining and knowledge discovery - DMKD '03. June13, 2003.
	San Diego, California. New York: ACM Press, 2003:2.
27	Lin J, Keogh E, Wei L, et al. Experiencing SAX: a novel symbolic representation of time series[J]. Data Mining and Knowledge Discovery, 2007, 15(2): 107-144.
28	Keogh E, Kasetty S. On the need for time series data mining benchmarks: a survey and empirical demonstration[J]. Data Mining and Knowledge Discovery, 2003, 7(4): 349-371.
29	Keogh E, Chakrabarti K, Pazzani M, et al. Dimensionality reduction for fast similarity search in large time series databases[J]. Knowledge and Information Systems, 2001, 3(3): 263-286.
30	李海林, 郭崇慧. 基于云模型的时间序列分段聚合近似方法[J]. 控制与决策, 2011, 26(10): 1525-1529.
	Li H L, Guo C H. Piecewise aggregate approximation method based on cloud model for time series[J]. Control and Decision, 2011, 26(10): 1525-1529.
31	Larsen R J, Marx M L. An Introduction to Mathematical Statistics and Its Applications[M]. Upper Saddle River:Prentice-Hall, 1981.

[1]	Zhuang YUAN, Yiqun LING, Zhe YANG, Chuankun LI. Critical parameters prediction based on TA-ConvBiLSTM for chemical process [J]. CIESC Journal, 2022, 73(1): 342-351.
[2]	Jian ZHU, Bo YANG, Yongjian WANG, Xiaojie TANG, Hongguang LI. New operation optimization method with time series based on Levenshtein distance hierarchical clustering [J]. CIESC Journal, 2019, 70(2): 581-589.