采用Adaline神经网络实现增益自适应内模控制  

Gain Adaptive Internal Model Control Based on Adaline Neural Network

作　　者：刘红军[1] 赵鑫[1] 薛浩洋[1] 

机构地区：[1]华北电力大学自动化系,河北保定071003

出　　处：《系统仿真学报》2005年第8期1955-1958,共4页Journal of System Simulation

摘　　要：引入1/1Pade逼近处理纯滞后环节,利用二步法设计内模控制器,运用单层Adaline神经网络参数辨识算法得到被控对象的稳态增益,并用于内模控制算法中,以此来自动调整被控对象模型增益,实现增益自适应控制。大量仿真实验表明此控制方法能克服系统增益变化对控制品质的影响,具有鲁棒性好,抗干扰能力强,控制精度高等特点。After approximating the time delay unit using Pade Series, the intemal model controller was got using the two-step design approach.The Adaline neural network was employed to identify the gain of system, and apply the gain to the IMC to achieve the objective of gain adaptive control. The mass simulation results show that the gain adaptive IMC can overcome the influence resulted from the gain variation of system, and the designed system has good robustness, better ability of anti-disturbance and superior precision.

关 键 词：内模控制(IMC) Adaline神经网络 增益自适应 参数辨识 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]