可控受限多变量耦合系统的智能控制  

Intelligent control of limited controllability multivariable coupling control system

作　　者：刘英晖[1] 刘贺平[1] 

机构地区：[1]北京科技大学信息工程学院,北京100083

出　　处：《计算机工程与应用》2009年第21期209-211,221,共4页Computer Engineering and Applications

摘　　要：针对可控受限多变量耦合系统,提出了一种基于对角递归神经网络(DRNN)整定的PID混合解耦控制。采用对角递归神经网络来辨识系统模型,进而对PID控制器参数进行整定,实现多变量解耦控制。通过对多变量耦合控制系统的设计和实时控制,实际控制结果达到了解耦控制的要求,并具有无超调、响应速度快、控制精度高等特点。According to the limited controllability of the muhivariable coupling system,a PID self-tuning mixed decoupling control method based on DRNN is put forward.It can realize decoupling multivariable system with identifying system model with DRNN and tuning the parameters of PID controller.It is used to real-time control of the multivariable furnace temperature control system.The experimental results indicate that mixed decoupling method based on DRNN has the advantage of non-overshoot,high speed and high control precision compared with common PID control.

关 键 词：可控受限多变量耦合系统 对角递归神经网络(DRNN) 混合解耦控制 PID控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]