基于神经网络逆系统的循环流化床内模控制  

Internal Model Control of Circulating Fluidized Bed Based on Neural Network Inverse System

作　　者：王杰[1] 陈春宇[1] 

机构地区：[1]郑州大学电气工程学院,河南郑州450001

出　　处：《郑州大学学报（工学版）》2013年第2期71-75,共5页Journal of Zhengzhou University（Engineering Science）

基　　金：国家自然科学基金资助项目(60905039)

摘　　要：在循环流化床锅炉(Circulating Fluidized Bed Boiler,CFBB)燃烧控制系统中,维持正常的床温和主蒸汽压力是循环流化床锅炉稳定、经济运行的关键.笔者针对这两个变量的强耦合特性,引入神经网络逆系统方法,实现系统的线性化解耦,并针对线性化解耦的非理想性以及参数、干扰等不确定因素对系统的影响,采用具有较好稳定性和抗干扰能力的内模控制策略对系统进行闭环控制.仿真结果表明,基于神经网络逆系统的内模控制方法不仅能够实现系统解耦,获得优良的静、动态特性,且具有良好的鲁棒稳定性和抑制扰动的能力.In the circulating fluidized bed boiler（CFBB） combustion control system, to maintain normal bed temperature and main stream pressure is key to the stable and economic operation of CFBB. In view of the strong coupling relationship between this the two variables, this paper used neural network inverse system method to solve decoupling problem of the system. Furthermore, in view of the imperfection of the decoupling and parameter perturbation, interference and other factors which affect the system, a closed-loop internal mod- el control strategy which has good stability and anti-jamming ability control is carried out. The simulation re- sult shows that internal model control based on neural network inverse system method can not only realize de- coupling of the system, get excellent static and dynamic characteristics, but also have good robust stability and disturbance rejection capability.

关 键 词：循环流化床锅炉 神经网络逆系统 解耦 内模控制 

分 类 号：TK229.6[动力工程及工程热物理—动力机械及工程]