摘要
针对Himont工艺的大型聚丙烯工业装置质量指标实时控制的复杂性,本文提出了一种通过表观能量平衡反推物料平衡并结合经验数据回归的质量指标实时推断方法。利用基于闭环辨识的鲁棒内模控制方法实现了的大型聚丙烯工业装置的先进控制。所提先进控制解决方案已成功应用于某炼油厂的聚丙烯工业装置,长周期的运行结果证实了所提解决方案可以有效地减少产品质量波动,提高产率并减轻操作人员的工作强度,取得了显著的经济效益。
The advanced process control (APC) solution for Himont Spheripol licensed polypropylene process is proposed. Based on the first principle model, the performance indexes indicating the product quality such as melt flow rate (MFR) and Isotactic Index (Ⅱ) were calculated. Based on the performance calculation and robust internal model control with closed-loop identification method, a strategy for direct quality control was proposed. The benefit of the proposed advanced process control solution to a real industrial polypropylene process was audited as standard control error of reaction temperature was decreased as about 10% as that before APC, product rate was improved 1 ton per hour, on-grade product rate was improved 9% and grade switching time was decreased 2 hours.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2007年第4期507-510,共4页
Computers and Applied Chemistry
基金
北京市科技新星计划(2005B15A)
北京市教委(xk100100435)
北京化工大学(QN0416
A040507)
关键词
聚丙烯
先进控制
推断计算
闭环辨识
内模控制
polypropylene, advanced process control, performance calculation, closed-loop identification, internal model control