Journal of System Simulation
Abstract
Abstract: An intelligent control strategy framework combining offline training and on-site deployment for regenerative thermal oxidizer (RTO) operation control based on RL is proposed to solve the problem that fluctuations in exhaust gas concentration and flow, as well as changes in working conditions, make it difficult for RTO to maintain the system operating parameters within the economically optimal range in practical applications. In the offline phase, RL is employed to enable autonomous policy learning and global optimization. In the online phase, an adaptive policy correction mechanism based on real-time data is introduced. RL control strategy is integrated with the operation monitoring and visualization system to intuitively present operational status and control effects. Simulation results demonstrate that the CO emissions per unit of steam are reduced by approximately 1.4%, and the cumulative economic 2 benefit over a six-month operation period is increased by 0.83%. The steam production is increased by 5.18% in actual industrial conditions.
Recommended Citation
Li, Yuxin; Li, Haijie; Han, Shangbo; Gao, Kai; Zhang, Qingyuan; Xiao, Yanwen; and Liu, Zhengcai
(2026)
"Chemical Plant Control Based on Dynamic Simulation and Reinforcement Learning,"
Journal of System Simulation: Vol. 38:
Iss.
10, Article 3.
DOI: 10.16182/j.issn1004731x.joss.26-0140
Available at:
https://dc-china-simulation.researchcommons.org/journal/vol38/iss10/3
First Page
2802
Last Page
2818
CLC
TP391
Recommended Citation
Li Yuxin, Li Haijie, Han Shangbo, et al. Chemical Plant Control Based on Dynamic Simulation and Reinforcement Learning[J]. Journal of System Simulation, 2026, 38(10): 2802-2818.
DOI
10.16182/j.issn1004731x.joss.26-0140
Included in
Artificial Intelligence and Robotics Commons, Computer Engineering Commons, Numerical Analysis and Scientific Computing Commons, Operations Research, Systems Engineering and Industrial Engineering Commons, Systems Science Commons