工业过程稳态优化中一种改进的迭代学习控制  

An Improved Method for Iterative Learning Control in Industrial Process Steady-state Optimization

作　　者：何善君[1] 程菲[1] 罗键[1] 

机构地区：[1]厦门大学自动化系,福建厦门361005

出　　处：《厦门大学学报（自然科学版）》2007年第3期326-330,共5页Journal of Xiamen University：Natural Science

摘　　要：为了改善工业过程稳态控制的动态品质,对在重复性控制较为实用的PD型迭代学习控制算法进行改进,引入积分环节,以适应工业过程稳态控制;针对带有噪声的系统输出信号进行滤波,改进局部对称双积分型迭代学习控制算法,提出了基于递归最小二乘法(Recursive Least Squares,RLS)滤波的PID型迭代学习控制算法,以期过滤噪声的高频和低频分量,获得更为理想的系统输出.最后给出了相关的仿真试验结果.In order to improve procedure of industrial process steady-state optimization,integral unit was added to PD-type iterative learning controller, which is practical in repeatable control system. Furthermore, an improved PID-type iterative learning control based on RLS adaptive filter was introduced. To do so, high frequency and low frequency of noise were filtered and system output would be more perfect. The numerical simulations showed that all improved algorithms in this paper were necessary and more effective.

关 键 词：迭代学习控制 稳态优化 PID型 递归最小二乘法 滤波 MATLAB 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]