系统仿真学报 ›› 2015, Vol. 27 ›› Issue (3): 598-602.

• 信息、控制、决策与仿真 • 上一篇    下一篇

基于KPLS的延迟焦化柴油95%点的预测与估计

程剑, 张凌波, 顾幸生   

  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
  • 收稿日期:2014-04-16 修回日期:2014-10-30 出版日期:2015-03-08 发布日期:2020-08-20
  • 作者简介:程剑(1989-),男,浙江杭州人,硕士生,研究方向为系统建模与优化;张凌波(1974-),男,湖南永州人,副教授,研究方向为过程控制。
  • 基金资助:
    中央高校基本科研业务费专项基金:上海市重点学科项目(B504)

Prediction and Estimation of 95% Distillation Point Diesel in Delayed Coking Unit Based on Kernel Partial Least Squares

Cheng Jian, Zhang Lingbo, Gu Xingsheng   

  1. Key Laboratory of Advanced Control and Optimization for Chemical Processes, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China
  • Received:2014-04-16 Revised:2014-10-30 Online:2015-03-08 Published:2020-08-20

摘要: 为了节约人工分析柴油95%点的成本和时间、提高延迟焦化装置调整的效率,在研究延迟焦化装置工艺流程和分析柴油95%点影响因素的基础上,提出了一种基于核偏最小二乘的柴油95%点的预测方法。仿真结果表明该方法能够有效、快速地处理各变量之间的非线性关系,并且预测结果也明显优于偏最小二乘方法等方法。所提出的方法可用于指导实际生产过程,提高企业效益。

关键词: 柴油95%点, 分馏塔, 偏最小二乘法, 核偏最小二乘法

Abstract: In order to save labor cost and time of 95% distillation point of diesel analysis, and improve the efficiency of adjusting the device, an algorithm was proposed to predict 95% distillation point of diesel based on Kernel Partial Least Squares(KPLS), on the basis of studying the process of delayed coking unit and analyzing the main factors related to 95% distillation point of diesel. The experiment results show that this algorithm can deal with the non-linear relationship among the variables effectively and quickly. Furthermore, the results also show that the model based on KPLS is better than that based on partial least squares. So this algorithm can be used to guide production and improve the profit of enterprise in practice.

Key words: 95% distillation point diesel, fractionating tower, partial least squares, kernel partial least squares

中图分类号: