化工学报 ›› 2020, Vol. 71 ›› Issue (12): 5696-5705.DOI: 10.11949/0438-1157.20200401

• 过程系统工程 • 上一篇    下一篇

基于最近邻与神经网络融合模型的软测量建模方法

杨逸俊1(),王振雷1(),王昕2   

  1. 1.华东理工大学化工过程先进控制和优化技术教育部重点实验室,上海 200237
    2.上海交通大学电工与电子技术中心,上海 200240
  • 收稿日期:2020-04-16 修回日期:2020-07-11 出版日期:2020-12-05 发布日期:2020-12-05
  • 通讯作者: 王振雷
  • 作者简介:杨逸俊(1996—),男,硕士研究生,2508567995@qq.com
  • 基金资助:
    国家重点研发计划项目(2018YFB1701103)

Soft sensor modeling method based on hybrid model of nearest neighbor and neural network

YANG Yijun1(),WANG Zhenlei1(),WANG Xin2   

  1. 1.Key Laboratory of Advanced Control and Optimization for Chemical Processes, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China
    2.Center of Electrical & Electronic Technology,Shanghai Jiao Tong University, Shanghai 200240, China
  • Received:2020-04-16 Revised:2020-07-11 Online:2020-12-05 Published:2020-12-05
  • Contact: WANG Zhenlei

摘要:

软测量建模能够有效地解决生产过程中在线分析仪表测量滞后大、价格昂贵、维护保养复杂等问题。目前,基于数据驱动的神经网络是软测量建模的主要工具之一。而在建模数据的采集过程中,主导变量的采集相对辅助变量要困难得多,由此产生了大量缺失标签的数据。但传统的软测量建模方法却忽视了这些无标签数据,只利用少量的有标签数据建模,从而影响了模型的预测精度。为了解决标签缺失的问题,采用最近邻算法对无标签数据进行伪标记,同时设计了由卷积操作与门限循环单元神经网络(GRU)结合的网络结构来进一步利用无标签数据,提取不同时刻数据中的动态特征,提高神经网络的预测精度。最后将该方法应用于丙烯精馏塔塔顶丙烷浓度的预测,实验结果表明该模型能有效处理非线性动态系统的标签缺失问题,具有更高的预测精度。

关键词: 软测量, 动态建模, 过程系统, 最近邻算法, 门限循环单元神经网络

Abstract:

Soft-sensing modeling can effectively solve the problems of large measurement lag, high price, and complex maintenance of online analytical instruments in the production process. At present, neural network based on data-driven is one of the main tools of soft sensor. In the process of modeling data collection, the collection of dominant variables is much more difficult than that of auxiliary variables, resulting in a large amount of unlabeled data. However, traditional soft sensor modeling methods ignore these unlabeled data and only use a small amount of labeled data for modeling, which has negative effect on the prediction accuracy of the model. To solve the problem of label missing, the nearest neighbor algorithm is used to pseudo label the unlabeled data. At the same time, a network structure is designed by combining convolution operation and gated recurrent unit neural network (GRU) to further utilize the unlabeled data, extract the dynamic feature from data at different time, and improve the prediction accuracy of the neural network. Finally, the method is applied to the prediction of propane concentration on the top of propylene distillation column. The results show that the model can solve the problem of label missing in the nonlinear dynamic system and has higher prediction accuracy.

Key words: soft sensor, dynamic modeling, process systems, nearest neighbor algorithm, gated recurrent unit neural network

中图分类号: