冷连轧机厚度和张力系统的智能解耦控制  

Intelligent decoupling control for thickness and tension system of tandem cold rolling mill

作　　者：李伯群[1] 李娇[1] 李济瀚 

机构地区：[1]辽宁科技大学电子与信息工程学院,辽宁鞍山114051

出　　处：《辽宁科技大学学报》2014年第1期16-21,共6页Journal of University of Science and Technology Liaoning

摘　　要：现代冷连轧机组中的厚度和张力系统具有多变量、强耦合、不确定等特点。以某厂1700 mm冷轧机组为背景,以现场数据为核心,建立了厚度和张力耦合系统数学模型。针对系统模型的参数扰动以及外部扰动等不确定性,提出了基于准对角递归神经网络多变量PID控制策略,该策略具有较强的表达和处理瞬态信息的能力,实现了厚度和张力系统的解耦控制。仿真结果验证了本算法的有效性,其解耦响应速度和抗干扰能力明显优于传统解耦控制效果,适合解决非线性系统的动态问题。The thickness and tension system is multi-variable, strong coupling and uncertain in the modem cold strip mill. The coupling mathematic model of thickness and tension system is established based on field data from 1700 mm cold rolling mill. By considering the uncertainties such as model parameter perturbation and external disturbance, the quasi-diagonal recurrent neural network PID strategy is proposed to realize the thickness and tension decoupling control. It has a strong ability to express and process the transient information. The simulation results show that the effectiveness of the algorithm and its response speed and anti-interference ability are obviously superior to conventional decoupling control,which is suitable to solve the dynamic problems in a nonlinear system.

关 键 词：冷连轧机 厚度和张力系统耦合模型 解耦控制 准对角递归神经网络 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]