Keywords
causal mechanism discovery
multivariate time series
fine-grained fault localization
causal residuals
fault propagation
process anomaly detection
Abstract
Continuous chemical production systems generate multivariate time series from reactor temperature, feed flow, pressure, pH, catalyst activity, cooling-water rate, reflux ratio, product concentration, and valve opening signals. These variables interact through reaction kinetics, heat transfer, material balance, and feedback control loops. A small abnormal change in feed composition or cooling efficiency may propagate through the process and appear as delayed deviations in several downstream variables. This study proposes a causal mechanism discovery method for fine-grained fault localization in chemical process time series. The method first constructs a time-lagged causal graph by combining conditional independence testing, Granger constraints, and process-control knowledge. A causal residual estimator is then used to compare observed trajectories with expected variable behavior under normal causal mechanisms. Finally, abnormal variables are ranked by their causal contribution to system-level deviation, separating primary fault variables from secondary responses. Experiments are conducted on a continuous chemical process dataset containing 36 production units, 128 process variables, and 5-second records collected over 164 operating days. The dataset contains 362 million timestamped observations and 1,870 engineer-confirmed abnormal episodes, including cooling-loop degradation, feed-pump instability, reactor-pressure oscillation, catalyst deactivation, reflux-control failure, and product concentration drift. The proposed method reduces median root-cause localization time from 42.6 minutes to 12.8 minutes compared with a correlation-based temporal baseline. The mean reciprocal rank for initiating-variable localization reaches 0.836. Causal path analysis assigns 1,430 abnormal episodes to interpretable fault-propagation chains, while unnecessary unit inspection tickets decrease from 1,120 to 418. The full plant-level assessment is completed in 10.6 minutes. The results show that causal mechanism discovery can improve fine-grained anomaly localization in highly coupled chemical process monitoring.
References
1. Gao, G., Gao, R., Lu, C., Gao, R., & Kuang, Y. (2026, March). Security Governance Methods and Quantitative Evaluation for Enterprise SMS and Verification Code Systems. In 2026 International Conference on Generative Artificial Intelligence and Information Security (GAIIS) (pp. 455-458). IEEE.
2. Nawaz, M., Maulud, A. S., Zabiri, H., & Suleman, H. (2022). Review of multiscale methods for process monitoring, with an emphasis on applications in chemical process systems. IEEE Access, 10, 49708-49724.
3. Xu, T., Zhu, W., & Zhang, J. (2026). A Study on the Application of Alternative Data in Credit Assessment for the Unbanked Population.
4. Farber, J. A., & Al Rashdan, A. Y. (2025). Unsupervised Process Anomaly Detection and Identification Using the Leave-One-Variable-Out Approach. Sensors, 25(7), 2098.
5. Chen, X., Xiao, H., Zeng, Z., Zhang, S., & Xiao, R. (2025). Fine-Grained Multivariate Time Series Anomaly Detection via Causal Inference. Knowledge-Based Systems, 114765.
6. Habiba, M., Pearlmutter, B. A., & Maleki, M. (2026). Neural Network Techniques in Continuous Time Series Prediction. Recent Trends in Modelling the Continuous Time Series Using Deep Learning, 95-143.
7. Fu, Y., Gui, H., Li, W., & Wang, Z. (2020, August). Virtual Material Modeling and Vibration Reduction Design of Electron Beam Imaging System. In 2020 IEEE International Conference on Advances in Electrical Engineering and Computer Applications (AEECA) (pp. 1063-1070). IEEE.
8. Chen, F., Yue, L., Xu, P., Liang, H., & Li, S. (2025). Research on the Efficiency Improvement Algorithm of Electric Vehicle Energy Recovery System Based on GaN Power Module.
9. Chakraborti, A. S. (2024). Graph-based Model Reduction of Machine System Digital Twins (Doctoral dissertation, University of Tampere, Finland).
10. Liang, R., Feifan, F. N. U., Liang, Y., & Ye, Z. (2025). Emotion-Aware Interface Adaptation in Mobile Applications Based on Color Psychology and Multimodal User State Recognition. Frontiers in Artificial Intelligence Research, 2(1), 51-57.
11. Yang, M., Wang, Y., Shi, J., & Tong, L. (2025). Reinforcement Learning Based Multi-Stage Ad Sorting and Personalized Recommendation System Design.
12. Mattera, G., Cao, Y., Zhang, Y., & Nele, L. (2026). Vision-based online quality tracking in wire arc additive manufacturing via hybrid unsupervised deep learning–statistical process monitoring. Welding in the World, 70(4), 1261-1281.
13. Zhu, W., & Yang, J. (2025). Causal Assessment of Cross-Border Project Risk Governance and Financial Compliance: A Hierarchical Panel and Survival Analysis Approach Based on H Company's Overseas Projects.
14. Yin, J., Huang, Y., & Rao, H. (2026). A Study on the Dynamic Evolution of Learning Behavior and Outcome Prediction in Digital Educational Environments. Available at SSRN 6607139.
15. Rodriguez Dominguez, A. (2026). Order-Constrained Spectral Causality in Multivariate Time Series. Available at SSRN 6035875.
16. Yang, J. (2026). Computational Analysis of How Digital Non-Clinical Communication Tools Influence Social Participation Among Older Adults in Community-Based Elderly Care. Available at SSRN 6682838.
17. Zhang, Z. (2026). A Study on Multimodal Product Understanding and Assembly Guidance Optimization in e-commerce for Complex Consumer Goods. Available at SSRN 6734559.
18. Kim, J., Kim, H., Kim, H., Lee, D., & Yoon, S. (2025). A comprehensive survey of deep learning for time series forecasting: architectural diversity and open challenges. Artificial Intelligence Review, 58(7), 216.
19. Su, D., & Wu, Y. (2026). Multilevel Growth and Social Network Modeling for Functional Communication Enhancement in Students with Intellectual Disabilities.
20. Shao, W. (2026, March). Interpretable Ensemble Learning for Network Traffic Anomaly Detection: A SHAP-based Explainable AI Framework for Embedded Systems Security. In 2026 14th International Symposium on Digital Forensics and Security (ISDFS) (pp. 1-4). IEEE.
21. Souza, B., Castro, M., Esmin, A., Machado, L., Ferreira, A., & Rocha, A. (2025). Causality-driven feature representation for connectivity prediction. Frontiers in Artificial Intelligence, 8, 1686750.
22. Gao, G., Gao, R., Gao, R., Zhou, H., & Lu, C. (2026, March). Performance Study of Enterprise SMS Communication Scheduling Mechanisms in Triple-Play Convergence Environments. In 2026 IEEE 8th International Conference on Communications, Information System and Computer Engineering (CISCE) (pp. 82-86). IEEE.
23. Khoo, T. L., Lee, T. S., Bee, S. T., Ma, C., & Zhang, Y. Y. (2025). A comparative review of large language models in engineering with emphasis on chemical engineering applications. Processes, 13(9), 2680.
24. Zhang, Z., Gao, Y., & Tong, Y. (2026). CausalML4CX: A Causal Inference and Machine Learning Framework for Customer Experience Optimization with Application in Banking.
25. Banerjee, D., Hassanali, A., & Laio, A. (2026). Investigating causality between principal components in protein dynamics. arXiv preprint arXiv:2606.24283.
26. Qi, C., & Qiao, X. (2026, May). Design Patterns for Multi-Agent Systems in Production. In 2026 International Conference on Intelligent Systems, Automation and Control (ISAC) (pp. 198-202). IEEE.
27. Ishii, K., Jia, Z. X., & Yamamoto, I. (2025). Intergenerational persistence of subjective well-being: Evidence from the Japanese Household Panel Survey on parents and children. Review of Economics of the Household, 23(3), 1047-1074.
28. Xu, T., Zhang, J., & Zhu, W. (2026). Fairness-Accuracy Trade-offs and Calibration Mechanisms in Machine Learning-Based Subprime Credit Scoring. Available at SSRN 6893759.
29. Xie, Z., Ren, X., Zheng, T., Bai, J., Fan, W., Xu, B., ... & Song, Y. (2026). A Survey on AI Agent Harness. ResearchGate Preprint. DOI, 10.
30. Iriany, I., Alexander, V., Taslim, T., Zuhri, R. R. S., Benguerba, Y., & Burmana, A. D. (2026). Energy investigation of pressure pulsation and flow instability in centrifugal pumps used in biodiesel plants. Array, 100803.