10.57647/spre.2026.1001.01

Physics-Guided Temporal Transformer for CO2 Corrosion with SHAP Case study: Asaluyeh SPGC Complex, Iran

PDF
  • Afshin Mohammadi1,
  • Ebrahim Akbari*1,
  • Homayun Motameni2,
  • Faraein Aeini1,
  1. Department of Computer Engineering, Sar.C., Islamic Azad University, Sari, Iran
  2. Department of Industrial and Computer Engineering, Mazandaran University of Science and Technology, Babol, Iran

Received: 2025-11-11

Revised: 2025-12-20

Accepted: 2025-01-21

Published in Issue 2026-03-31

Copyright (c) 2026 Afshin Mohammadi, Ebrahim Akbari, Homayun Motameni, Faraein Aeini (Author)

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

Mohammadi, A., Akbari, E., Motameni, H., & Aeini, F. (2026). Physics-Guided Temporal Transformer for CO2 Corrosion with SHAP Case study: Asaluyeh SPGC Complex, Iran. Signal Processing and Renewable Energy (SPRE), 10(1). https://doi.org/10.57647/spre.2026.1001.01
Crossref
Scopus
Google Scholar
Europe PMC

PDF views: 27

Abstract

The structural integrity of the oil and gas industry is at high risk due to CO₂-induced corrosion; consequently, practical, explainable mechanisms are required to accurately predict this phenomenon. The available mechanistic models, like the de Waard–Milliams equation, though entrapping fundamental thermodynamics, do not withstand dynamic dosing and time-varying conditions. Although the conventional machine learning and deep learning models improve accuracy, though often lack physics consistency and interpretability. A Physics-Guided Temporal Transformer (PGTT), which integrates a Temporal Fusion Transformer backbone with a physics-informed loss (of de Waard–Milliams), and SHapley Additive exPlanations (SHAP) for feature attribution, is proposed here. By applying a comprehensive sequential dataset from 22 inhibitor dosing experiments (15,400 samples during 60+ hours), this PGTT achieves higher performance with an average , , and  of  over five independent runs, and outperforms Random Forest , , and MLP  baselines. According to the SHAP, Temperature (26.5%) and CO₂ pressure (17.7%) are identified as the dominant stimulants, consistent with corrosion science. This case study reveals a reduction in prediction error of less than 5%, supporting proactive inhibitor dosing and pipeline integrity management.

Keywords

  • Co2 corrosion,
  • physics-informed learning,
  • temporal transformer,
  • SHAP,
  • pipeline integrity
PDF

References

  1. C. d. W. D. E. Milliams, "Carbonic acid corrosion of steel," Corrosion, vol. 31, no. 5, May 1975 1975, DOI:https://doi.org/10.5006/0010-9312-31.5.177.
  2. M. M. H. J. Imran, S.; Mohamad Ayob, A. F, "A critical review of machine learning algorithms in maritime, offshore and oil & gas corrosion research: A comprehensive analysis of ANN and RF models," Ocean Engineering, vol. 295, March 2024 2024, DOI:https://doi.org/10.1016/j.oceaneng.2024.116796.
  3. J. C. Fang, X.; Gai, H.; Lin, S.; Lou, H. H, "Development of machine learning algorithms for predicting internal corrosion of crude oil and natural gas pipelines," Computers & Chemical Engineering, vol. 177, p. 108 358, July 2023 2023.DOI:https://doi.org/10.1016/j.compchemeng.2023.108358.
  4. L. B. Z. Coelho, Dawei; Van Ingelgem, Yves; Steckelmacher, Denis; Nowé, Ann; Terryn, Herman, "Reviewing machine learning of corrosion prediction in a data-oriented perspective," npj Materials Degradation, vol. 6, p. 8, 2022, DOI: https://doi.org/10.1038/s41529-022-00218-4.
  5. Z. Z. Dong, Min; Li, Weirong; Wen, Fenggang; Dong, Guoqing; Zou, Lu; Zhang, Yongqiang, "Development of a Predictive Model for Carbon Dioxide Corrosion Rate and Severity Based on Machine Learning Algorithms," Materials, vol. 17, no. 16, p. 4046, 14 August 2024 2024, DOI: https://doi.org/10.3390/ma17164046.
  6. A. S. Vaswani, N.; Parmar, N.; Uszkoreit, J.; Jones, Llion; Gomez, A. N.; Kaiser, Ł.; Polosukhin, I, "Attention Is All You Need," Advances in Neural Information Processing Systems (NeurIPS 2017), vol. 30, pp. 5998–6008, December 2017 2017, DOI:https://doi.org/10.48550/arXiv.1706.03762.
  7. J. W. Jiang, X.; Zhu, F.; Xiang, D.; Hu, Z.; Mu, S, "A deep learning framework integrating Transformer and LSTM architectures for pipeline corrosion rate forecasting," Computers & Chemical Engineering, vol. 190, p. Article No. 109365, 2025,DOI:https://doi.org/10.1016/j.compchemeng.2025.109365.
  8. M. P. Raissi, P.; Karniadakis, G. E, "Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations," Journal of Computational Physics, vol. 378, pp. 686–707, February 2019 2019, DOI: https://doi.org/10.1016/j.jcp.2018.10.045.
  9. A. V. Dourado, Felipe A. C, "Physics-Informed Neural Networks for Corrosion-Fatigue Prognosis," Proceedings of the Annual Conference of the PHM Society, vol. 11, 2019, DOI: https://doi.org/10.36001/phmconf.2019.v11i1.814.
  10. N. M. Chen, Rujin; Chen, Airong; Cui, Chuanjie, "PC-PINNs: Physics-informed neural networks for solving the phase-field model of pitting corrosion," Bridge Maintenance, Safety, Management, Digitalization and Innovation: CRC Press / Taylor & Francis, 2024, pp. 183–189. DOI:https://doi.org/10.1201/9781003483755-25
  11. X. L. Ma, Zhenzhi; Zhan, Jie; et al, "Prediction method of CO₂ plume distribution based on physics-informed neural networks," Petroleum Drilling & Production Technology, vol. 46, no. 5, pp. 90–98, 2024,DOI: https://doi.org/10.11911/syztjs.2024090.
  12. A. W. enkatraman, Mark A.; Montes de Oca Zapiain, David, "Accelerating charge estimation in molecular dynamics simulations using physics-informed neural networks: corrosion applications," npj Computational Materials, vol. 11, no. 1, 2025, DOI: https://doi.org/10.1038/s41524-024-01495-0.
  13. J. W. Jiang, X.; Zhu, F.; Xiang, D.; Hu, Z.; Mu, S, "A deep learning framework integrating Transformer and LSTM architectures for pipeline corrosion rate forecasting," Computers & Chemical Engineering, vol. 190, p. 109365, 2025, DOI:https://doi.org/10.1016/j.compchemeng.2025.109365.
  14. S. M. L. Lundberg, S.-I., "A unified approach to interpreting model predictions," presented at the Advances in Neural Information Processing Systems (NeurIPS 2017), 2017.DOI: https://doi.org/10.48550/arXiv.1705.07874
  15. Y. S. Q. W. X. Z. L. D. S. B. D. Z. Z. Z. H. Z. Y. Xi, "Interpretable machine learning for maximum corrosion depth and influence factor analysis," npj Materials Degradation, vol. 7, no. 1, 2023,DOI: https://doi.org/10.1038/s41529-023-00324-x.
  16. Q. H. Chen, W.; Wen, Y.; Wang, B.; Liu, G, "Degradation prediction and explainable analyses of the corroded subsea pipelines based on INSGAII-PGNN and SHAP algorithm," Ocean Engineering, vol. 334, p. Article no. 121495, 2025/08/01 (Aug 2025) 2025, DOI: https://doi.org/10.1016/j.oceaneng.2025.121495.
  17. B. A. Lim, S. Ö.; Loeff, N.; Pfister, T, "Temporal Fusion Transformers for interpretable multi-horizon time series forecasting," International Journal of Forecasting, vol. 37, no. 4, pp. 1748–1764, October 2021 2021,DOI: https://doi.org/10.1016/j.ijforecast.2021.03.012.
  18. L. Y. Tan, Y.; Zhang, K.; Liao, K.; He, G.; Tian, J.; Lu, X, "Prediction of internal corrosion rate for gas pipeline: A new method based on Transformer architecture," Computers & Chemical Engineering, vol. 198, p. Article No. 109084, 2025, DOI: https://doi.org/10.1016/j.compchemeng.2025.109084.
  19. I. M. S. Alqahtani, A.; Khan, M, "Experimental and theoretical aspects of crack assisted failures of metallic alloys in corrosive environments – A review," Materials Today: Proceedings, vol. 66, pp. 2530–2535, July 2022 2022, DOI: https://doi.org/10.1016/j.matpr.2022.07.075.
  20. M. M. Cid Montoya, A.; Verma, S.; Mures, O. A.; Rubio-Medrano, C. E, "Aeroelastic force prediction via temporal fusion transformers," Computer-Aided Civil and Infrastructure Engineering, vol. 39, pp. 319–344, January 2024 2024, DOI: https://doi.org/10.1111/mice.13381.
  21. R. W. U. Revie, H. H, Corrosion and Corrosion Control: An Introduction to Corrosion Science and Engineering (4th ed.). John Wiley & Sons, 2008.DOI: https://doi.org/10.1002/9780470277270
  22. G. T. He, J.; Sun, L.; Zhao, F.; Li, S.; Li, C.; Liao, K.; Chen, X.; Yang, W., "Research on natural gas pipeline corrosion prediction by integrating extreme gradient boosting and generative adversarial network," SSRN Electronic Journal, 2025,DOI: https://doi.org/10.2139/ssrn.5489826.