Dynamic Process Control based on an AI Model Cluster for Atmospheric and Vacuum Distillation

Jun Liu

doi:10.6919/ICJE.202605_12(5).0019

Authors

Jun Liu

DOI:

https://doi.org/10.6919/ICJE.202605_12(5).0019

Keywords:

Atmospheric and Vacuum Distillation; Process AI Model Cluster; Dynamic Optimization; Intelligent Control System; Expert Knowledge.

Abstract

From November 2024 to August 2025, a data-driven real-time closed-loop optimization project was carried out for a 10 Mt/a atmospheric and vacuum distillation unit. The study was intended to address the strong coupling, frequent operating-mode switching, and limited long-term effectiveness of conventional optimization methods in large crude distillation systems. A process-oriented artificial intelligence (AI) model cluster was therefore developed by using full-process modular artificial neural networks (ANNs) to characterize the coupled states of the unit and reflect changes in operating conditions. On this basis, an intelligent execution system (iES) was built to translate plant-wide optimization results into dynamic control targets acceptable to the field, which were then implemented through multivariable advanced control schemes at the distributed control system (DCS) level. Industrial application shows that, within quality and safety boundaries, this method can transfer optimization solutions to the production unit in a safe and stable manner, demonstrating good industrial applicability.

Downloads

Download data is not yet available.

References

[1] M. Vaccari, R. Bacci Di Capaci, E. Brunazzi, et al.: Optimally Managing Chemical Plant Operations: An Example Oriented by Industry 4.0 Paradigms, Industrial & Engineering Chemistry Research, Vol. 60 (2021) No. 21, p.7853-7867.

[2] Y.D. Su: Improvement and Optimization of Advanced Control for Crude Distillation Units, Petrochemical Automation, Vol. 59 (2023) No. 1.

[3] Y.D. Su: Application of Soft Sensor Technology in Refining and Chemical Units, Petrochemical Automation, Vol. 59 (2023) No. 2.

[4] Y.D. Su: Exploration and Application of Online Real-Time Optimization for Refining and Chemical Units, Petroleum Refinery Engineering, Vol. 53 (2023) No. 5, p.12-15.

[5] C.H. Yang, B. Sun, Y.G. Li, K.K. Huang and W.H. Gui: Cooperative Optimization and Intelligent Control of Complex Production Processes, Acta Automatica Sinica, Vol. 49 (2023) No. 3, p.528-539.

[6] Y. Jiang, S. Yin, J. Dong, et al.: A Review on Soft Sensors for Monitoring, Control, and Optimization of Industrial Processes, IEEE Sensors Journal, Vol. 21 (2021) No. 11, p.12868-12881.

[7] Y. Yu, M. Lee, C. Lee, et al.: Estimating APC Model Parameters for Dynamic Intervals Determined Using Change-Point Detection in Continuous Processes in the Petrochemical Industry, Processes, Vol. 11 (2023) No. 8, p.2229.

[8] Y.R. Li, C.J. Yang, H.W. Zhang and J.F. Li: Discussion on Key Technologies of Digital Twins for Process Industry, Acta Automatica Sinica, Vol. 47 (2021) No. 3, p.501-514.

[9] P. Kumar, J.B. Rawlings and S.J. Wright: Industrial, Large-Scale Model Predictive Control with Structured Neural Networks, Computers & Chemical Engineering, Vol. 150 (2021), p.107291.

[10] Y. Xiong, X. Shi, Y. Ma, et al.: Optimization Design of Crude Oil Distillation Unit Using Bi-Level Surrogate Model, Frontiers in Control Engineering, Vol. 4 (2023), p.1162318.

[11] F. Fiedler, B. Karg, L. Luken, et al.: do-mpc: Towards FAIR Nonlinear and Robust Model Predictive Control, Control Engineering Practice, Vol. 140 (2023), p.105676.

[12] Y.S. Perera, D.A.A.C. Ratnaweera, C.H. Dasanayaka, et al.: The Role of Artificial Intelligence-Driven Soft Sensors in Advanced Sustainable Process Industries: A Critical Review, Engineering Applications of Artificial Intelligence, Vol. 121 (2023), p.105988.