摘要
本文提出一种辨识连续Hammerstein模型参数的方法,该方法利用不同幅值的M周期序列作为系统的输入,通过DFT变换,离析出线性子系统的连续模型,并由频域辨识法及时域辨识法直接估计出连续模型的参数和非线性特性的系数。数字仿真结果表明本算法是可行且有效的。
In this paper, a method for estimating the parameters of continuous Hammerstein model is proposed. The Msequences of different magnitude are used as input signals herein. By means of DFT transforms, an equivalent continuous model of the linear subsystem is analysed and separated from the original. The parameters of the continuous model are directly estimated in both frequency and time domain. The simulation results show that the method is feasible and effective.
出处
《控制理论与应用》
EI
CAS
CSCD
北大核心
1990年第1期75-79,共5页
Control Theory & Applications
关键词
连续模型
参数估计
非线性系统
nonlinear system
continuous model
system identification
DFT transform
