A Compact CNN-Transformer Model for Robust Fault Diagnosis in Large-Scale Chemical Processes
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
Ensuring the safe and stable operation of large-scale chemical processes requires accurate fault detection and diagnosis under nonlinear dynamics and strong variable interactions. This study investigates deep learning-based fault diagnosis for the Tennessee Eastman Process (TEP), focusing on whether performance improvements beyond near-saturation Long Short-Term Memory (LSTM) baselines can be achieved. A standardised TEP dataset with 52 measured and manipulated variables is used, excluding the non-detectable fault cases (IDV 3, 9, and 15), resulting in a 19-class classification problem.
A hybrid Convolutional Neural Network–Transformer (CNN-Transformer) architecture is proposed in which one-dimensional convolutional layers capture local cross-variable correlations, while a Transformer encoder models long-range temporal dependencies through self-attention. To ensure fair comparison, both the proposed model and a strong LSTM baseline are trained and evaluated under identical preprocessing, optimization, and evaluation protocols.
The CNN-Transformer achieves an overall classification accuracy of 99.92%, marginally outperforming the LSTM baseline (99.86%). Although the numerical improvement is slight, the proposed model consistently yields higher macro-averaged F1-scores and reduced fault-wise misclassification, indicating enhanced robustness in challenging fault scenarios.
The key contribution of this work is demonstrating that combining convolutional feature extraction with attention-based temporal modelling provides consistent class-level robustness beyond near-saturated recurrent architectures, while maintaining a compact structure suitable for practical deployment.
##plugins.themes.bootstrap3.article.details##
Tennessee Eastman Process, Deep learning, Fault Diagnosis, Long short term memory, CNN, Transformer
Agarwal, P., Tamer, M., Budman, H., 2021, Explainability: relevance based dynamic deep learning algorithm for fault detection and diagnosis in chemical processes. Comput. Chem. Eng. 154, 107467. https://doi.org/10.1016/j.compchemeng.2021.107467
Agarwal, P., Tamer, M., Budman, H., 2020. Assessing observability using supervised autoencoders with application to Tennessee eastman process, in: IFAC-PapersOnLine. Elsevier B.V., pp. 206–211. https://doi.org/10.1016/j.ifacol.2020.12.122
Chadha, G.S., Schwung, A., 2017. Comparison of deep neural network architectures for fault detection in Tennessee Eastman process, in: 2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA). Presented at the 2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA), IEEE, Limassol, pp. 1–8. https://doi.org/10.1109/ETFA.2017.8247619
Chen, B., Zhang, H., He, F., Zhang, C., Chen, Y., Liao, H., Zheng, S., 2023. Modified Q-sigma Rule for Incipient Fault Detection in Industrial Processes on Analysis of Intermittent Process Variation. SSRN Electronic Journal null, null. https://doi.org/10.2139/ssrn.4375935
Chen, H., Cen, J., Yang, Z., Si, W., Cheng, H., 2022. Fault Diagnosis of the Dynamic Chemical Process Based on the Optimized CNN-LSTM Network. ACS Omega 7, 34389–34400. https://doi.org/10.1021/acsomega.2c04017
Chiang, L.H., Russell, E.L., Braatz, R.D., 2001. Fault Detection and Diagnosis in Industrial Systems, Advanced Textbooks in Control and Signal Processing. Springer London, London. https://doi.org/10.1007/978-1-4471-0347-9
Downs, J.J., Vogel, E.F., 1993 A plant-wide industrial process control problem Comput. Chem. Eng. 17 (3), 245–255 https://doi.org/10.1016/0098-1354(93)80018-I
Han, Y., Ding, N., Geng, Z., Wang, Z., Chu, C., 2020. An optimized long short-term memory network based fault diagnosis model for chemical processes. Journal of Process Control 92, 161–168. https://doi.org/10.1016/j.jprocont.2020.06.005
Heo, S., Lee, J.H., 2019. Statistical process monitoring of the Tennessee Eastman process using parallel autoassociative neural networks and a large dataset. Processes 7. https://doi.org/10.3390/pr7070411
Heo, S., Lee, J.H., 2018. Fault detection and classification using artificial neural networks. IFAC-PapersOnLine 51, 470–475. https://doi.org/10.1016/j.ifacol.2018.09.380
Labbaf-Khaniki, M.A., Manthouri, M., Ajami, H., 2024 Twin Transformer using Gated Dynamic Learnable Attention mechanism for Fault Detection and Diagnosis in the Tennessee Eastman Process. https://doi.org/10.48550/arXiv.2403.10842
Li, M., Shao, Y., 2021. Deep Compression of Neural Networks for Fault Detection on Tennessee Eastman Chemical Processes.
Lomov, I., Lyubimov, M., Makarov, I., Zhukov, L.E., 2021. Fault detection in Tennessee Eastman process with temporal deep learning models. Journal of Industrial Information Integration 23, 100216. https://doi.org/10.1016/j.jii.2021.100216
Onel, M., Kieslich, C.A., Pistikopoulos, E.N., 2019. A nonlinear support vector machine-based feature selection approach for fault detection and diagnosis: Application to the Tennessee Eastman process. AIChE Journal 65, 992–1005. https://doi.org/10.1002/aic.16497
Pal, P.K., Hens, A., Behera, N., Lahiri, S.K., 2025. Digital twins: Transforming the chemical process industry—A review. Can J Chem Eng 103, 3611–3636. https://doi.org/10.1002/cjce.25611
Pozdnyakov, V., Kovalenko, A., Makarov, I., Drobyshevskiy, M., Lukyanov, K., 2024. Adversarial Attacks and Defenses in Fault Detection and Diagnosis: A Comprehensive Benchmark on the Tennessee Eastman Process. IEEE Open J. Ind. Electron. Soc. 5, 428–440. https://doi.org/10.1109/OJIES.2024.3401396
Reinartz, C., Kulahci, M., Ravn, O., 2021. An extended Tennessee Eastman simulation dataset for fault-detection and decision support systems. Computers and Chemical Engineering 149. https://doi.org/10.1016/j.compchemeng.2021.107281
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, Ł., Polosukhin, I., 2017 Attention is All you Need.
Verma, R., Yerolla, R., Besta, C.S., 2022. Deep Learning-based Fault Detection in the Tennessee Eastman Process, in: Proceedings of the 2nd International Conference on Artificial Intelligence and Smart Energy, ICAIS 2022. Institute of Electrical and Electronics Engineers Inc., pp. 228–233. https://doi.org/10.1109/ICAIS53314.2022.9743021
Yin, S., Ding, S.X., Haghani, A., Hao, H., Zhang, P., 2012. A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark Tennessee Eastman process. Journal of Process Control 22, 1567–1581. https://doi.org/10.1016/j.jprocont.2012.06.009
Yin, S., Gao, X., Karimi, H.R., Zhu, X., 2014. Study on Support Vector Machine-Based Fault Detection in Tennessee Eastman Process. Abstract and Applied Analysis 2014, 1–8. https://doi.org/10.1155/2014/836895
Zhang, C., Zheng, X., Li, Y., 2021. A novel fault detection and diagnosis scheme based on independent component analysis-statistical characteristics: Application on the tennessee eastman benchmark process. Journal of Chemical Engineering of Japan 54, 304–312. https://doi.org/10.1252/jcej.20we045
Zhang, L., Song, Z., Zhang, Q., Peng, Z., 2022. Generalized transformer in fault diagnosis of Tennessee Eastman process. Neural Comput & Applic 34, 8575–8585. https://doi.org/10.1007/s00521-021-06711-2