摘要
带钢连续热镀锌镀层厚度控制过程具有多变量耦合、非线性、时变大滞后等特性,难以建模和自动控制。目前国内生产线多采用手动控制,存在镀层厚度偏差大、调节时间长、锌原料浪费等问题。在分析带钢连续热镀锌生产工艺的基础上,提出了一种工艺设定和智能控制相结合的控制方法。针对控制对象多变量耦合、非线性等问题,制定刀距、刀高工艺设定值表,并在镀层厚度机理模型的基础上建立刀压配方数据库,将多变量系统简化为工作点上单变量线性时滞系统。针对镀层厚度检测纯滞后问题,应用迭代学习控制算法去逼近理想刀压控制量,消除镀层厚度偏差。采用Win AC软PLC开发镀层厚度控制系统,实现了镀层厚度自动控制。应用效果表明,控制系统具有良好的控制性能,缩短了镀层厚度调节时间,提高了镀层厚度精度。
The thickness control process of strip continuous hot dip galvanizing has multiple variable coupling,nonlinear and time delay,which is difficult to model and control. At present,the domestic production line adopts manual control,and there are problems such as large thickness deviation,long adjustment time and waste of zinc raw materials.Based on the analysis of continuous hot-dip galvanizing production process,a control method combining process setting and intelligent control was presented. For the multi-variable coupling and non-linearity of the control object,the table of knife-blade and knife-height setting is established. Based on the mechanism model of the coating thickness,a database of knife-pressure formula is established and the multivariable system is simplified to a single-variable linear time-delay system. To deal with the problem of pure lag in coating thickness measurement,an iterative learning control algorithm is applied to approach the ideal knife pressure control to eliminate the coating thickness deviation. The coating thickness control system was developed by using Win AC soft PLC,and the coating thickness was automatically controlled. The application results show that the control system has good control performance,shorten the adjustment time and improve the coating thickness precision.
作者
秦大伟
刘宏民
王军生
张岩
费静
张栋
QIN Da-wei;LIU Hong-min;WANG Jun-sheng;ZHANG Yan;FEI Jing;ZHANG Dong(Education Ministry Engineering Research Center of Rolling Equipment and Complete Technology,Yanshan University,Qinhuangdao 066004,Hebei,China;Ansteel Beijing Research Institute,Ansteel Group,Beijing 102211,China;Ansteel Iron and Steel Research Institute,Ansteel Group,Anshan 114009,Liaoning,China)
出处
《钢铁》
CAS
CSCD
北大核心
2018年第8期62-67,共6页
Iron and Steel
基金
国家自然科学基金资助项目(50604006)
钢铁共性技术协同创新资助项目(2011计划)
关键词
连续热镀锌
镀层厚度
迭代学习控制
配方数据库
hot-dip galvanizing
coating weight
iterative learning control
formula database
