欢迎访问过程工程学报, 今天是

过程工程学报 ›› 2025, Vol. 25 ›› Issue (10): 1030-1038.DOI: 10.12034/j.issn.1009-606X.225005

• 研究论文 • 上一篇    下一篇

基于PAM-GRU的三苯收率预测方法研究及应用

韩永明1， 孙亚帅1， 倪庆旭1,2， 潘峰2， 孙庆峰3， 谭蕾4， 胡渲1*， 耿志强1*   

  1. 1. 北京化工大学信息科学与技术学院，北京 100029 2. 中国石化销售股份有限公司华北分公司，天津 300384 3. 山东港源管道物流有限公司，山东 烟台 264006 4. 昆仑数智科技有限责任公司，北京 100007
  • 收稿日期:2025-01-06 修回日期:2025-03-18 出版日期:2025-10-28 发布日期:2025-10-28
  • 通讯作者: 耿志强 gengzhiqiang@mail.buct.edu.cn
  • 基金资助:
    国家自然科学基金项目;国家自然科学基金项目

BTX yield prediction method study based on pyramid attention mechanism integrating gated recurrent unit and its application

Yongming HAN1,  Yashuai SUN1,  Qingxu NI1,2,  Feng PAN2,  Qingfeng SUN3,  Lei TAN4,  #br# Xuan HU1*,  Zhiqiang GENG1*   

  1. 1. College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China 2. Sinopec Sales Co., Ltd. North China Branch, Tianjin 300384, China 3. Shandong Gangyuan Pipeline Logistics Co., Ltd., Yantai, Shandong 264006, China 4. Kunlun Digital Technology Co., Ltd., Beijing 100007, China
  • Received:2025-01-06 Revised:2025-03-18 Online:2025-10-28 Published:2025-10-28
  • Contact: Zhi-Qiang GENG gengzhiqiang@mail.buct.edu.cn

RichHTML

PDF

摘要： 针对传统三苯收率预测中存在的模型精度低的问题，本工作提出了一种基于金字塔注意力机制的门控循环单元网络(PAM-GRU)的三苯收率预测方法。PAM-GRU通过引入金字塔注意力机制，利用卷积对多特征序列数据建立层次化的注意力结构，解决传统自注意力机制性能增益受限问题，以提取空间特征。同时引入门控循环单元，利用其循环结构捕捉时间序列中的动态信息，挖掘时序数据的时间变化规律。通过融合数据的时空特征，实现对三苯收率的准确预测。所提方法应用在实际连续重整化工生产流程中，在平均绝对百分比误差、平均方根误差、平均绝对误差、决定系数四个指标上与循环神经网络、长短期记忆网络、门控循环单元、基于注意力机制的长短期记忆网络、基于注意力机制的门控循环单元模型对比。结果表明，所提出的PAM-GRU模型能够有效整合连续重整生产过程中的空间和时间特征，实现对三苯收率的高效、准确预测。此外，为应对复杂生产环境，对模型增加了鲁棒性测试，结果表明，所提模型具有较强的鲁棒性，能够有效抑制突发噪声的干扰。

关键词: 金字塔注意力机制, 门控循环单元, 化工生产, 三苯收率预测

Abstract: Aiming to address the challenge of low model accuracy in traditional benzene-toluene-xylene (BTX) yield prediction, this work proposes a novel prediction approach based on a pyramid attention mechanism (PAM) combined with a gated recurrent unit (GRU), referred to as the PAM-GRU model. The PAM of the proposed method can enable the construction of a hierarchical attention structure for multi-feature sequential data, allowing for the extraction of spatial features. In parallel, the GRU is utilized to capture dynamic temporal information within the time-series data through its recurrent structure, enabling the proposed prediction method to uncover underlying temporal variation patterns. By seamlessly integrating both spatial and temporal features, the PAM-GRU method achieves a more accurate and reliable prediction. Finally, the proposed method is applied to a real-world continuous reforming chemical production process, and the proposed model performance is evaluated using various metrics. The proposed PAM-GRU model is compared with models such as the recurrent neural networks (RNN), the long short-term memory networks (LSTM), the gated recurrent units (GRU), the LSTM based on attention mechanisms (Attention-LSTM), and the GRU based on attention mechanisms (Attention-GRU) in terms of four indicators: mean absolute percentage error (MAPE), root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2). The results show that the proposed PAM-GRU model can effectively integrate the spatial and temporal features in the continuous reforming production process, achieving efficient and accurate prediction of the BTX yield. In addition, to cope with the complex production environment, a robustness test is added in the proposed model. The results show that the proposed model has strong robustness and can effectively suppress the interference of sudden noise.

Key words: pyramid attention mechanism, gated recurrent unit, chemical production, BTX yield prediction