化工学报 ›› 2016, Vol. 67 ›› Issue (6): 2488-2494.DOI: 10.11949/j.issn.0438-1157.20151611

• 过程系统工程 • 上一篇    下一篇

针对多变量耦合时滞系统的无模型控制改进算法

牛培峰, 李梦宁, 孙丽朋, 马云鹏, 刘魏岩, 李刚   

  1. 燕山大学电气工程学院, 河北 秦皇岛 066004
  • 收稿日期:2015-10-26 修回日期:2016-02-21 出版日期:2016-06-05 发布日期:2016-06-05
  • 通讯作者: 牛培峰
  • 基金资助:

    国家自然科学基金项目(61573306)。

Improved model-free control algorithm for multivariate coupling time-delay system

NIU Peifeng, LI Mengning, SUN Lipeng, MA Yunpeng, LIU Weiyan, LI Gang   

  1. School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, Hebei, China
  • Received:2015-10-26 Revised:2016-02-21 Online:2016-06-05 Published:2016-06-05
  • Supported by:

    supported by the National Natural Science Foundation of China (61573306).

摘要:

工业生产是国家经济发展的命脉,而在工业生产中过程控制起到非常重要的作用。实际的工业过程控制系统难以避免时滞,再者大部分的工业化生产过程都不只有一个被控量且往往互相耦合,种种原因使得整个控制过程难以建立准确的数学模型,因此一般控制方法难以实现令人满意的控制。无模型自适应控制(model free adaptive control,MFAC)方法无须建立被控对象准确模型,可直接进行控制器的设计,考虑针对被控系统的大时滞特性,将基本MFAC算法中的控制律进行变周期差值分析,增大控制输入差值周期,对原有控制算法进行改进,并对MIMO系统实现直接解耦控制。以循环流化床锅炉燃烧过程模型为例,MATLAB计算机模拟仿真实验表明该改进算法具有良好的控制效果。

关键词: 多变量耦合, 大时滞, 无模型自适应, 过程控制, 改进算法, 计算机模拟, 循环流化床

Abstract:

The industrial production is the lifeline of the country's economic development, and the process control plays an important role in industrial production. Inevitably, the actual industrial process control systems are time-delay system. Besides, there are more than one controlled variable in most of the industrial production processes, and these controlled variables are coupled. All of these factors result in a huge difficulty to build an exact mathematical model for the control system, indicating that the tradition control methods cannot meet the control requirements nicely. With model-free adaptive control (MFAC) method, controller can be directly designed without exact model to reach the control objectives. Given the large time-delay of CFBB, the control law in MFAC is analyzed by changing the period of D-value,and control algorithm should be improved by increasing the period of D-value. Then, the decoupling control strategy of MIMO can be achieved. In the end, taking the circulating fluidized bed boiler combustion process as example, the MATLAB computer simulation results show that the algorithm has better control effect.

Key words: multivariate coupling, large time-delay, model-free adaptive, process control, improved algorithm, computer simulation, circulating fluidized bed

中图分类号: