Process fault root cause diagnosis through state evolution mapping based on temporal unit shapelets

doi:10.1016/j.cjche.2025.04.011

Chinese Journal of Chemical Engineering ›› 2025, Vol. 84 ›› Issue (8): 96-106.DOI: 10.1016/j.cjche.2025.04.011

Previous Articles Next Articles

Process fault root cause diagnosis through state evolution mapping based on temporal unit shapelets

Zhenhua Yu¹, Guan Wang², Qingchao Jiang¹, Xuefeng Yan¹

1. Key Laboratory of Smart Manufacturing in Energy Chemical Process, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China;
2. State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai 200237, China

Received:2024-09-03 Revised:2025-04-18 Accepted:2025-04-20 Online:2025-05-14 Published:2025-08-28
Contact: Qingchao Jiang,E-mail:qchjiang@ecust.edu.cn
Supported by:
The authors gratefully acknowledge the support from the following foundations: the National Natural Science Foundation of China (62322309, 62433004), Shanghai Science and Technology Innovation Action Plan (23S41900500), and Shanghai Pilot Program for Basic Research (22TQ1400100-16).

Process fault root cause diagnosis through state evolution mapping based on temporal unit shapelets

Zhenhua Yu¹, Guan Wang², Qingchao Jiang¹, Xuefeng Yan¹

1. Key Laboratory of Smart Manufacturing in Energy Chemical Process, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China;
2. State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai 200237, China

通讯作者: Qingchao Jiang,E-mail:qchjiang@ecust.edu.cn
基金资助:
The authors gratefully acknowledge the support from the following foundations: the National Natural Science Foundation of China (62322309, 62433004), Shanghai Science and Technology Innovation Action Plan (23S41900500), and Shanghai Pilot Program for Basic Research (22TQ1400100-16).

Abstract

Abstract: Accurate fault root cause diagnosis is essential for ensuring stable industrial production. Traditional methods, which typically rely on the entire time series and overlook critical local features, can lead to biased inferences about causal relationships, thus hindering the accurate identification of root cause variables. This study proposed a shapelet-based state evolution graph for fault root cause diagnosis (SEG-RCD), which enables causal inference through the analysis of the important local features. First, the regularized autoencoder and fault contribution plot are used to identify the fault onset time and candidate root cause variables, respectively. Then, the most representative shapelets were extracted to construct a state evolution graph. Finally, the propagation path was extracted based on fault unit shapelets to pinpoint the fault root cause variable. The SEG-RCD can reduce the interference of noncausal information, enhancing the accuracy and interpretability of fault root cause diagnosis. The superiority of the proposed SEG-RCD was verified through experiments on a simulated penicillin fermentation process and an actual one.

Key words: Root cause diagnosis, Neural networks, Shapelet, Fermentation, Bioprocess

摘要： Accurate fault root cause diagnosis is essential for ensuring stable industrial production. Traditional methods, which typically rely on the entire time series and overlook critical local features, can lead to biased inferences about causal relationships, thus hindering the accurate identification of root cause variables. This study proposed a shapelet-based state evolution graph for fault root cause diagnosis (SEG-RCD), which enables causal inference through the analysis of the important local features. First, the regularized autoencoder and fault contribution plot are used to identify the fault onset time and candidate root cause variables, respectively. Then, the most representative shapelets were extracted to construct a state evolution graph. Finally, the propagation path was extracted based on fault unit shapelets to pinpoint the fault root cause variable. The SEG-RCD can reduce the interference of noncausal information, enhancing the accuracy and interpretability of fault root cause diagnosis. The superiority of the proposed SEG-RCD was verified through experiments on a simulated penicillin fermentation process and an actual one.

关键词: Root cause diagnosis, Neural networks, Shapelet, Fermentation, Bioprocess

Zhenhua Yu, Guan Wang, Qingchao Jiang, Xuefeng Yan. Process fault root cause diagnosis through state evolution mapping based on temporal unit shapelets[J]. Chinese Journal of Chemical Engineering, 2025, 84(8): 96-106.

Zhenhua Yu, Guan Wang, Qingchao Jiang, Xuefeng Yan. Process fault root cause diagnosis through state evolution mapping based on temporal unit shapelets[J]. 中国化学工程学报, 2025, 84(8): 96-106.

References

[1] H. Safaeipour, M. Forouzanfar, A. Casavola, A survey and classification of incipient fault diagnosis approaches, J. Process. Contr. 97 (2021) 1-16.
[2] L.M. Elshenawy, T.A. Mahmoud, Fault diagnosis of time-varying processes using modified reconstruction-based contributions, J. Process. Contr. 70 (2018) 12-23.
[3] J.J. Luo, Z.H. Jin, H.P. Jin, Q. Li, X. Ji, Y.Y. Dai, Causal temporal graph attention network for fault diagnosis of chemical processes, Chin. J. Chem. Eng. 70 (2024) 20-32.
[4] J.X. Zhang, W.J. Luo, Y.Y. Dai, Y.M. Yao, Cycle temporal algorithm-based multivariate statistical methods for fault diagnosis in chemical processes, Chin. J. Chem. Eng. 47 (2022) 54-70.
[5] L.Y. Jiang, L. Xie, S.Q. Wang, Fault diagnosis for batch processes by improved multi-model fisher discriminant analysis, Chin. J. Chem. Eng. 14 (3) (2006) 343-348.
[6] X. Zhao, W.W. Yan, H.H. Shao, Monitoring and fault diagnosis for batch process based on feature extract in fisher subspace, Chin. J. Chem. Eng. 14 (6) (2006) 759-764.
[7] M. Ram Maurya, R. Rengaswamy, V. Venkatasubramanian, Application of signed digraphs-based analysis for fault diagnosis of chemical process flowsheets, Eng. Appl. Artif. Intell. 17 (5) (2004) 501-518.
[8] D. Gao, C.G. Wu, B.K. Zhang, X. Ma, Signed directed graph and qualitative trend analysis based fault diagnosis in chemical industry, Chin. J. Chem. Eng. 18 (2) (2010) 265-276.
[9] H.L. Jiang, R. Patwardhan, S.L. Shah, Root cause diagnosis of plant-wide oscillations using the concept of adjacency matrix, J. Process. Contr. 19 (8) (2009) 1347-1354.
[10] S. Kabir, An overview of fault tree analysis and its application in model based dependability analysis, Expert Syst. Appl. 77 (2017) 114-135.
[11] Y.M. Wan, F. Yang, N. Lv, H.P. Xu, H. Ye, W.C. Li, P. Xu, L.M. Song, A.K. Usadi, Statistical root cause analysis of novel faults based on digraph models, Chem. Eng. Res. Des. 91 (1) (2013) 87-99.
[12] L.H. Chiang, B.B. Jiang, X.X. Zhu, D.X. Huang, R.D. Braatz, Diagnosis of multiple and unknown faults using the causal map and multivariate statistics, J. Process. Contr. 28 (2015) 27-39.
[13] J. Dong, K.R. Cao, K.X. Peng, Hierarchical causal graph-based fault root cause diagnosis and propagation path identification for complex industrial process monitoring, IEEE Trans. Instrum. Meas. 72 (2023) 3513911.
[14] C.W.J. Granger, Investigating causal relations by econometric models and cross-spectral methods, Econometrica 37 (3) (1969) 424.
[15] J.F. Geweke, Measures of conditional linear dependence and feedback between time series, J. Am. Stat. Assoc. 79 (388) (1984) 907-915.
[16] L. Barnett, A.K. Seth, The MVGC multivariate Granger causality toolbox: a new approach to Granger-causal inference, J. Neurosci. Methods 223 (2014) 50-68.
[17] T. Yuan, S.J. Qin, Root cause diagnosis of plant-wide oscillations using Granger causality, IFAC Proc. Vol. 45 (15) (2012) 160-165.
[18] H.S. Chen, Z.B. Yan, Y. Yao, T.B. Huang, Y.S. Wong, Systematic procedure for granger-causality-based root cause diagnosis of chemical process faults, Ind. Eng. Chem. Res. 57 (29) (2018) 9500-9512.
[19] G. Shen, P.L. Wang, K.L. Hu, Q.Y. Ye, Fault root cause diagnosis method based on recurrent neural network and granger causality, 2021 CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes (SAFEPROCESS). December 17-18, 2021, Chengdu, China. IEEE, (2021) 1-6.
[20] Y.X. Liu, B. Jafarpour, Graph attention network with Granger causality map for fault detection and root cause diagnosis, Comput. Chem. Eng. 180 (2024) 108453.
[21] P.Y. Song, C.H. Zhao, B. Huang, Addressing heterogeneous time-frequency causality: source consistency exploring for industrial root cause alignment and diagnosis, IEEE Trans. Cybern. 55 (3) (2025) 1107-1120.
[22] P.Y. Song, C.H. Zhao, B. Huang, MPGE and RootRank: a sufficient root cause characterization and quantification framework for industrial process faults, Neural Netw. 161 (2023) 397-417.
[23] Y.D. Shu, J.S. Zhao, Data-driven causal inference based on a modified transfer entropy, Comput. Chem. Eng. 57 (2013) 173-180.
[24] P. Duan, F. Yang, T.W. Chen, S.L. Shah, Direct causality detection via the transfer entropy approach, IEEE Trans. Contr. Syst. Technol. 21 (6) (2013) 2052-2066.
[25] G. Li, S.J. Qin, T. Yuan, Data-driven root cause diagnosis of faults in process industries, Chemom. Intell. Lab. Syst. 159 (2016) 1-11.
[26] B. Rashidi, D.S. Singh, Q. Zhao, Data-driven root-cause fault diagnosis for multivariate non-linear processes, Contr. Eng. Pract. 70 (2018) 134-147.
[27] P. Duan, T.W. Chen, S.L. Shah, F. Yang, Methods for root cause diagnosis of plant-wide oscillations, AlChE. J. 60 (6) (2014) 2019-2034.
[28] Y. Liu, H.S. Chen, H.B. Wu, Y. Dai, Y. Yao, Z.B. Yan, Simplified Granger causality map for data-driven root cause diagnosis of process disturbances, J. Process. Contr. 95 (2020) 45-54.
[29] Q.C. Jiang, W.J. Wang, S.T. Chen, C.J. Pan, W.M. Zhong, Hierarchical fault root cause identification in plant-wide processes using distributed direct causality analysis, IEEE Trans. Ind. Inform. 20 (3) (2024) 3232-3241.
[30] Q.C. Jiang, J.S. Jiang, W.J. Wang, C.J. Pan, W.M. Zhong, Partial cross mapping based on sparse variable selection for direct fault root cause diagnosis for industrial processes, IEEE Trans. Neural Netw. Learn. Syst. 35 (5) (2024) 6218-6230.
[31] J.G. Wang, R. Chen, X.Y. Ye, Y. Yao, Z.T. Xie, S.W. Ma, L.L. Liu, Data-driven root cause diagnosis of process disturbances by exploring causality change among variables, J. Process. Contr. 129 (2023) 103062.
[32] P.Y. Song, C.H. Zhao, B. Huang, M. Wu, Sparse and time-varying predictive relation extraction for root cause quantification of nonstationary process faults, IEEE Trans. Instrum. Meas. 71 (2022) 3525013.
[33] S. Wang, Q. Zhao, Y.H. Han, J.K. Wang, Root cause diagnosis for process faults based on multisensor time-series causality discovery, J. Process. Contr. 122 (2023) 27-40.
[34] Y.J. Zhou, K. Xu, F. He, Root cause diagnosis in multivariate time series based on modified temporal convolution and multi-head self-attention, J. Process. Contr. 117 (2022) 14-25.
[35] Z.Q. Cheng, Y. Yang, W. Wang, W.J. Hu, Y.T. Zhuang, G.J. Song, Time2Graph: revisiting time series modeling with dynamic shapelets, Proc. AAAI Conf. Artif. Intell. 34 (4) (2020) 3617-3624.
[36] L.X. Ye, E. Keogh, Time series shapelets: a novel technique that allows accurate, interpretable and fast classification, Data Min. Knowl. Discov. 22 (1) (2011) 149-182.
[37] W.W. Yan, P.J. Guo, L. gong, Z.K. Li, Nonlinear and robust statistical process monitoring based on variant autoencoders, Chemom. Intell. Lab. Syst. 158 (2016) 31-40.
[38] E. Parzen, On estimation of a probability density function and mode, Ann. Math. Statist. 33 (3) (1962) 1065-1076.
[39] M. Cuturi, M. Blondel, Soft-DTW: a differentiable loss function for time-series, (2017): 1703.01541.
[40] G. Birol, U. Cenk, C. Ali, A modular simulation package for fed-batch fermentation: penicillin production, Comput. Chem. Eng. 26 (11) (2002) 1553-1565.
[41] P.Y. Song, C.H. Zhao, B. Huang, J.L. Ding, Explicit representation and customized fault isolation framework for learning temporal and spatial dependencies in industrial processes, IEEE Trans. Neural Netw. Learn. Syst. 35 (3) (2024) 2997-3011.

Process fault root cause diagnosis through state evolution mapping based on temporal unit shapelets

Process fault root cause diagnosis through state evolution mapping based on temporal unit shapelets

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics

Comments

[1]	Qingyu Wang, Duiping Liu, Yong Lu, Jibin Zhou, Xiangang Ma, Mao Ye. Deep learning approach for morphology classification and particle sizing of industrial methanol-to-olefins (MTO) catalyst [J]. Chinese Journal of Chemical Engineering, 2025, 84(8): 1-10.
[2]	Nan Liu, Chunmeng Zhu, Zeng Li, Yunpeng Zhao, Xiaogang Shi, Xingying Lan. Hierarchical framework for predictive maintenance of coking risk in fluid catalytic cracking units: A data and knowledge-driven method [J]. Chinese Journal of Chemical Engineering, 2025, 84(8): 35-46.
[3]	Ali Tarik Karagoz, Omar Alqusair, Chao Liu, Jie Li. Advances in conceptual process design: From conventional strategies to AI-assisted methods [J]. Chinese Journal of Chemical Engineering, 2025, 84(8): 60-76.
[4]	Tiantong Zhang, Haolin Cheng, Yao Nian, Jinli Zhang, Qingbiao Li, You Han. Application of generative artificial intelligence in catalysis [J]. Chinese Journal of Chemical Engineering, 2025, 84(8): 86-95.
[5]	Zheyuan Pang, Siying Liu, Yiting Lin, Xiangchen Fang, Honglai Liu, Chong Peng, Cheng Lian. Large language model-based multi-objective modeling framework for vacuum gas oil hydrotreating [J]. Chinese Journal of Chemical Engineering, 2025, 84(8): 133-145.
[6]	Zhenhua Yu, Zongyu Yao, Weijun Wang, Qingchao Jiang, Zhixing Cao. SmdaNet: A hierarchical hard sample mining and domain adaptation neural network for fault diagnosis in industrial process [J]. Chinese Journal of Chemical Engineering, 2025, 84(8): 146-157.
[7]	Xuedan Hou, Pengfei Zhao, Xiaohui Lin, Yunxing Gao, Huidong Chen, Di Cai, Peiyong Qin. Current advances in distillation processes for fermentative acetone-butanol-ethanol purification [J]. Chinese Journal of Chemical Engineering, 2025, 79(3): 91-108.
[8]	Zongyu Yao, Qingchao Jiang, Xingsheng Gu. Distributed process monitoring based on Kantorovich distance-multiblock variational autoencoder and Bayesian inference [J]. Chinese Journal of Chemical Engineering, 2024, 73(9): 311-323.
[9]	Zhiwei Zhu, Minglei Yang, Wangli He, Renchu He, Yunmeng Zhao, Feng Qian. A deep reinforcement learning approach to gasoline blending real-time optimization under uncertainty [J]. Chinese Journal of Chemical Engineering, 2024, 71(7): 183-192.
[10]	Cheng Yang, Chao Jiang, Guo Yu, Jun Li, Cuimei Bo. Transferable adversarial slow feature extraction network for few-shot quality prediction in coal-to-ethylene glycol process [J]. Chinese Journal of Chemical Engineering, 2024, 71(7): 258-271.
[11]	Xiaoxiao Dong, Zhuohong He, Xu Yan, Dong Gao, Jingyu Jiao, Yan Sun, Haibin Wang, Haibin Qu. Real-time model correction using Kalman filter for Raman-controlled cell culture processes [J]. Chinese Journal of Chemical Engineering, 2024, 70(6): 251-260.
[12]	Qianlin Wang, Jiaqi Han, Feng Chen, Xin Zhang, Cheng Yun, Zhan Dou, Tingjun Yan, Guoan Yang. Real-time risk prediction of chemical processes based on attention-based Bi-LSTM [J]. Chinese Journal of Chemical Engineering, 2024, 75(11): 131-141.
[13]	Peng Xu, Yuwei Song, Jingbo Du, Feilong Zhang. Town gas daily load forecasting based on machine learning combinatorial algorithms: A case study in North China [J]. Chinese Journal of Chemical Engineering, 2024, 75(11): 239-252.
[14]	Chenyang Xu, Lijuan Guo, Kang Zhou, Hai Yu, Chaoliang Wei, Fengqi Fan, Lei Zhang. RSscore: Reaction superiority learned from reaction mapping hypergraph [J]. Chinese Journal of Chemical Engineering, 2024, 74(10): 203-215.
[15]	Zeeshan Ikram Butt, Iftikhar Ahmad, Muhammad Shoaib, Syed Ibrar Hussain, Hira Ilyas, Muhammad Asif Zahoor Raja. Stochastic neuro-swarming intelligence paradigm for the analysis of magneto-hydrodynamic Prandtl-Eyring fluid flow with diffusive magnetic layers effect over an elongated surface [J]. Chinese Journal of Chemical Engineering, 2024, 74(10): 295-311.