.png){kind=icon}
返回
Closed-loop gasoline blending scheduling based on real-time optimized slack feedback
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ces.2025.121426.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Gasoline blending is a complex process that involves mixing various components to produce refined gasoline, directly affecting the economic efficiency and product quality of the refinery. To produce gasoline that meets quality standards with minimal excess in property specifications and at the lowest cost, meticulous development of blending scheduling plans is essential. This ensures reasonable allocation of resources, such as component oils, storage tanks, pumps, and pipelines. Additionally, to overcome fluctuations in oil properties and flow rates, maximize blending efficiency, and ensure product quality standards are met, it is necessary to introduce real-time optimization control technology. However, traditional gasoline blending scheduling optimization methods often overlook the connection with real-time optimization, leading to a disjunction between scheduling plans and actual operations. Therefore, this paper proposes a closed-loop scheduling strategy based on real-time optimization slack feedback. The real-time optimization model incorporates slack parameters into the property constraints. When the real-time optimization process causes the refined gasoline quality to fall short of standards because of fluctuations in component oil properties or flow rates. Slack parameters are used to feedback and adjust the constraints and objective function of the scheduling model, re-solving the subsequent blending plans. Case analysis results show that the proposed closed-loop scheduling strategy can adjust the scheduling model based on slack feedback, correcting non-compliant gasoline properties.
Keyword:
Gasoline blending scheduling
Uncertainties
Real-time optimization
Slack feedback
期刊
IF:
4.3
论文数:
2.3W
被引数:
5.5W
机构
引用论文
.png){kind=icon}
A deep reinforcement learning approach to gasoline blending real-time optimization under uncertainty不确定条件下汽油调合实时优化的深度强化学习方法
PREAI
分享收藏
查看更多内容