高级检索

    王杰, 陈博, 刘松, 欧阳福生, 戴宁锴, 赵明洋. 石油化工过程的静态与时序数据组合建模[J]. 华东理工大学学报(自然科学版), 2023, 49(4): 489-497. DOI: 10.14135/j.cnki.1006-3080.20220304001
    引用本文: 王杰, 陈博, 刘松, 欧阳福生, 戴宁锴, 赵明洋. 石油化工过程的静态与时序数据组合建模[J]. 华东理工大学学报(自然科学版), 2023, 49(4): 489-497. DOI: 10.14135/j.cnki.1006-3080.20220304001
    WANG Jie, CHEN Bo, LIU Song, OUYANG Fusheng, DAI Ningkai, ZHAO Mingyang. Combined Modeling of Static and Sequential Data for Petrochemical Process[J]. Journal of East China University of Science and Technology, 2023, 49(4): 489-497. DOI: 10.14135/j.cnki.1006-3080.20220304001
    Citation: WANG Jie, CHEN Bo, LIU Song, OUYANG Fusheng, DAI Ningkai, ZHAO Mingyang. Combined Modeling of Static and Sequential Data for Petrochemical Process[J]. Journal of East China University of Science and Technology, 2023, 49(4): 489-497. DOI: 10.14135/j.cnki.1006-3080.20220304001

    石油化工过程的静态与时序数据组合建模

    Combined Modeling of Static and Sequential Data for Petrochemical Process

    • 摘要: 传统的石油化工过程建模中仅使用静态数据,而未能充分考虑连续生产过程中时序信息对建模指标的影响。本文提出了一种静态与时序数据组合网络(CNSS)模型,使用前馈神经网络提取静态数据的信息,使用Bi-LSTM(Bidirectional-Long Short Term Memory)和自注意力机制提取操作变量时序数据中的信息,其中Bi-LSTM提取操作变量在时序逻辑上的信息,自注意力机制提取操作变量之间的交叉信息,通过静态和时序数据信息的组合以获得更好的模型预测性能;并使用CNSS模型分别对S Zorb装置精制汽油辛烷值(RON)、催化裂化烟气脱硝系统氮氧化物(NOx)的出口质量浓度进行预测,结果表明:CNSS模型的预测精度明显高于仅使用静态数据的机器学习模型,其对精制汽油RON预测的平均绝对误差和平均绝对百分比误差分别为0.1091、0.12%,对NOx出口质量浓度预测的平均绝对误差和平均绝对百分比误差分别为2.4430 mg/m3、5.60%。对于因工艺参数波动较大而需要考虑时序信息的石油化工过程,CNSS模型可以为其建立机器学习模型提供重要参考。

       

      Abstract: The traditional modeling of petrochemical processes only uses static data, but fails to fully consider the impact of sequential information on modeling indicators. In this study, a combination network of static and sequential data (CNSS) for petrochemical process was proposed. The feedforward neural network was used to extract the information of the static data. The bidirectional long short-term memory (Bi-LSTM) and the self-attention mechanism were used to extract the chronological logic and cross information of the operating variables, respectively. Through the combination of the information from static and sequential data, a better prediction performance of CNSS can be obtained. The model was verified through the research octane number (RON) prediction of refined gasoline from a S Zorb unit and the concentration prediction of outlet NOx from flue gas denitration system in fluid catalytic cracking (FCC) unit, respectively. The results show that CNSS has a much higher prediction accuracy than the machine learning model with static data. Its average absolute error and average absolute percentage error for RON prediction of refined gasoline are 0.1091 units and 0.12%, and those for NOx outlet mass concentration prediction are 2.4430 mg/m3and 5.60%, respectively. The CNSS model can be an important reference to establish a machine learning model for petrochemical process, in which timing information should be considered due to the large fluctuations in the values of process parameters.

       

    /

    返回文章
    返回