文献类型：期刊文献

中文题名：PTA氧化反应器进料溴离子浓度的软测量

英文题名：The Soft-sensing Measurement of Consistency of Br Ion that Exists in Oxidizing Reactor of PTA Units

作者：朱斌[1];王永富[2];何衍庆[1]

机构：[1]华东理工大学自动化系;[2]扬子石化公司化工厂;

年份：2001

期号：5

起止页码：27

中文期刊名：石油化工自动化

外文期刊名：Automation in Petro-chemical Industry

语种：中文

中文关键词：软测量;神经网络;回归分析;主元回归;偏最小二乘法;PTA;氧化反应器;溴离子;浓度;精对苯二甲酸

外文关键词：soft sensing;neural network;regressive analysis;PCR;PLS;PTA

摘要：以某厂PTA生产车间氧化反应器进料中溴离子浓度的测量为例,探讨了软测量技术在石油化工生产过程中的应用,建立了溴离子浓度的基于BP网络、RBF网络、主元回归法和偏最小二乘法的四种软测量模型. 通过MATL AB工具对这四种模型进行了仿真研究,比较各自的优缺点,并对各个模型进行数据预测,获得了较好的效果.
It is mainly discussed the soft sensing techniques applying to the petrochemical process.Aiming at the online estimate of consistency of Br ion that is in oxidizing reactor in PTA unit,four soft sensing models are built based on BPNN,RBFNN, PCR and PLS.After studying the four model simulations by MATLAB tools,their advantages and shortcoming comes.The four models are forecast by other data and we get good validity.