文献类型：期刊文献

英文题名：Refinery planning optimization based on smart predict-then-optimize method under exogenous price uncertainty

作者：Wang, Cong[1];Peng, Xin[1,2];Zhao, Liang[1];Zhong, Weimin[1]

机构：[1]East China Univ Sci & Technol, Key Lab Smart Mfg Energy Chem Proc, Minist Educ, Shanghai 200237, Peoples R China;[2]East China Univ Sci & Technol, Engn Res Ctr Proc Syst Engn, Shanghai 200237, Peoples R China

年份：2024

卷号：188

外文期刊名：COMPUTERS & CHEMICAL ENGINEERING

收录：;EI(收录号：20242516283693);WOS:【SCI-EXPANDED(收录号:WOS:001259025800001)】；

基金：This work was supported by National Natural Science Fund for Distinguished Young Scholars (61925305) , National Natural Science Foundation of China (62373154) , National Natural Science Foundation of China (62173145) , Fundamental Research Funds for the Central Universities Project of Shanghai Gas Turbine Union Innovation Center and Shanghai AI Lab.r Universities Project of Shanghai Gas Turbine Union Innovation Center and Shanghai AI Lab.

语种：英文

外文关键词：Smart predict-then-optimize; Refinery planning; Temporal graph convolution network; Price uncertainty

摘要：Refinery planning plays a crucial role in optimizing refined resource allocation. However, the exogenous uncertainty represented in product prices often hinders refineries from making timely adjustments to their planning decisions and affecting overall profitability. To address this challenge, a smart predict -then -optimize method is designed for refinery planning under product price uncertainty. The temporal graph convolution network is employed to capture spatio-temporal features in product prices, resulting in enhanced prediction performance. Furthermore, the proposed method effectively integrates the stages of price prediction and refinery planning optimization, with the combined loss contributing to improved decision -making performance. The enhancements in predicting price uncertainty parameters and optimizing profits is demonstrated by multi -period and single -period refinery planning case.