Volume 15, Issue 8 · IJARST · ISSN 2319-1783 / 2320-1126

Modeling of the Vacuum Distillation Unit of a Refinery Using Attention-Based and Conventional Neural Network Approaches

Modeling of the Vacuum Distillation Unit of a Refinery Using Attention-Based and Conventional Neural Network Approaches is an open-access, peer-reviewed research paper by AKPAN Nseabasi M. and EGEMBA Kingsley C., published in Volume 15, Issue 8 of the International Journal of Advanced Research in Science and Technology (IJARST), a UGC-approved journal (Print ISSN 2319-1783, Online ISSN 2320-1126).

Author

AKPAN Nseabasi M. and EGEMBA Kingsley C.

Abstract

The vacuum distillation unit (VDU) of a West African refinery was modeled using conventional Feedforward Artificial Neural Network (FF-ANN) and Attention-Based Artificial Neural Network (Attention-ANN) approaches, and the effectiveness of the models in predicting critical product variables were evaluated. Using preprocessed plant data, the models were trained with 14 input process variables and six process outputs. The performance of the models was evaluated using the coefficient of determination (R²), root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and Pearson's correlation coefficient (R), while paired samples t-test was performed over 20 repeated runs. Both models achieved high predictive accuracy, with the Attention-ANN demonstrating  superior overall performance (R² = 0.9970, RMSE = 6.3080, MAE = 4.5341, and MAPE = 4.27%) compared with the FF-ANN (R² = 0.9853, RMSE = 14.0252, MAE = 5.2037, and MAPE = 7.92%), while also achieving a slightly better performance in predicting five of the six individual output variables. The observed differences between the predictions of the models were not statistically significant, though the attention mechanism enhanced model interpretability by identifying the column bottom pressure as the most influential process variable. The predictive performance achieved compared favorably with reported refinery modelling studies for VDU modeling. These findings indicate that attention-based neural networks represent an effective and interpretable framework for multi-output VDU modelling.

Keywords: Vacuum Distillation Unit, Neural Network Modeling, Attention-Based Mechanism, Vacuum Distillation Product Prediction, Model Interpretability

DOI: https://doi.org/10.62226/ijarst20262765

REFERENCES

1. Chandler, B. (2022). Addressing Africa’s Energy Deficit: Gas, Renewables. Mo Ibrahim Foundation. https://mo.ibrahim.foundation/sites/default/files/2022-09/energy-transition.pdf. (Retreieved 9th August 2015)

2. Ozan, S. (2024).” Africa’s Escalating Energy Demand: Fostering Opportunities for Sustainable Growth”. ReportLinker, https://www.reportlinker.com/article/8460#:~:text=The%20Surge%20in%20Energy%20Needs,towards%20innovative%20solutions%20and%20collaborations (retrieved 9th August 2025)
3. Davies, D., A. Cobbinah, C. Sithole, and S. Williams. (2024) “60% of Africa’s Energy Will Be Fossil Fuel-Based by 2040”. Africa Briefing, October 2024. https://africabriefing.com/60-of-africas-energy-will-be-fossil-fuel-based-by-2040/. (Retrieved 9th August 2025).
4. Hawilti. (2024). Refineries Watch: Sub-Saharan Africa. https://hawilti.com/wp-content/uploads/2024/07/Refineries-Watch-2024-Report.pdf. (Retrieved 9th August 2025)
5. Holechek, J. L., H. M. Geli,M. N. Sawalhah, and R. Valdez. (2022). “A Global Assessment: Can Renewable Energy Replace Fossil Fuels by 2050?” Sustainability 14(8): 4792.
6. H’ng, S., L. Ng, D. Ng, and V. Andiappan. (2024). “Optimisation of Vacuum Distillation Units in Oil     Refineries Using Surrogate Models”. Process Integration and Optimization for Sustainability. 8: 1–23.
7. Elayane, J., R. Bchitou, and A. Bouhaouss. (2017)” Study of the Thermal Cracking during the Vacuum Distillation of Atmospheric Residue of Crude Oil”. Scientific Study and Research: Chemistry and Chemical Engineering, Biotechnology, Food Industry. 18: 61–71.
8. Jin, Q., Z. Li, Z. Yan, B. Wang, and Z. Wang. (2022). “Optimization Study on Enhancing Deep-Cut Effect of the Vacuum Distillation Unit (VDU)”. Processes. 10(2): 359.
9. Atta, M., Khan, H., Ali, M., Tariq, R., Yasir, A., Iqbal, M., Din, S., and Krzywanski, J. (2024). “Simulation of Vacuum Distillation Unit in Oil Refinery: Operational Strategies for Optimal Yield Efficiency”. Energies. 17, 3806.
10. Okwonna, O., and A. Obuebite. (2021). “Artificial Intelligence Approach in Crude Distillation Unit Operation”. Global Journal of Engineering and Technology Advances 09(02): 075–082.
11.  Egemba, K., and Igbokwe P. (2024). Artificial Neural Network Model for the Prediction of the Products of a Crude    Distillation Column. Journal of Engineering Research and Reports. 26(8): 1–9.
12. Oster, K., Güttel, S., Chen, L., Shapiro, J. L., and Jobson, M. (2021). “Machine learning-based soft sensors for vacuum distillation unit”. arXiv. https://arxiv.org/abs/2111.11251
13. Buitinck, L., Louppe, G., Blondel, M., Pedregosa, F., Mueller, A., Grisel, O., Niculae, V., Prettenhofer, P.,Gramfort, A., Grobler, J., Layton, R., Vanderplas, J., Joly, A., Holt, B., and Varoquaux, G. (2013). “API Design for Machine Learning Software: Experiences from the Scikit-learn Project”. (arXiv:1309.0238). arXiv. https://arxiv.org/abs/1309.0238.
14. Goodfellow, I., Y. Bengio, and A. Courville. (2016). Deep Learning. MIT Press.
15. Bergstra, J., and Bengio, Y. (2012). “Random search for hyper-parameter optimization”. Journal of Machine Learning Research, 13(2): 281-305.
16. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I. (2017). “Attention is all you need”. In I. Guyon, U. von Luxburg, S. Bengio, H. Wallach, R. Fergus, S. V. N.Vishwanathan, and R. Garnett (Editors), Advances in Neural Information Processing Systems. Volume 30. 7102p : 5998–6008
17. Masters, D. and Luschi, C., (2018). “Revisiting small batch training for deep neural networks”. arXiv. arXiv:1804.07612
18. Prechelt, L. (1998). “Early Stopping - But When?”. In G. B. Orr & K.-R. Müller (Editors.), Neural Networks: Tricks of the Trade. Springer. 55-69
19. Feurer, M. and Hutter, F. (2019). “Hyperparameter Optimization”. In: Hutter, F., Kotthoff, L., Vanschoren, J. (Editors). Automated Machine Learning. The Springer Series on Challenges in Machine Learning. Springer, Cham. https://doi.org/10.1007/978-3-030-05318-5_1
20. Chicco, D., Tötsch, N., and Jurman, G. (2021). “The coefficient of determination R-squared is more   informative than SMAPE, MAE, MAPE, MSE, and RMSE in regression analysis evaluation”. PeerJ Computer Science, 7, Article e623. https://doi.org/10.7717/peerj-cs.623
21. Willard, J., Jia, X., Xu, S., Steinbach, M., and Kumar, V. (2023). Integrating scientific knowledge with machine learning for engineering and environmental systems. ACM Computing Surveys, 55(4): 1 – 37
22. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. and Salakhutdinov, R., (2014). “Dropout: a simple way to prevent neural networks from overfitting”. The Journal of Machine Learning Research, 15(1): 1929–1958
23. Kleinberg, B., Li, Y. and Yuan, Y., (2018). “An alternative view: When does SGD Escape Local Minima?”.International Conference on Machine Learning (ICML). 2698–2707
24. Kister, H. Z. Distillation Design. McGraw-Hill, 1992
25. Halsey, L. G. (2025). “The reign of the p-value is over: What alternative analyses could we employ to fill the power vacuum?” Nature Human Behaviour. 9(2): 141-152

DOI

10.62226/ijarst20262765

PAGES : 2516-2525 | 7 VIEWS | 2 DOWNLOADS

How do you cite this paper?

AKPAN Nseabasi M. and EGEMBA Kingsley C. — “Modeling of the Vacuum Distillation Unit of a Refinery Using Attention-Based and Conventional Neural Network Approaches.” International Journal of Advanced Research in Science and Technology (IJARST), Volume 15, Issue 8. DOI: 10.62226/ijarst20262765.

Read / Download Full Article