用Hartley变换实现连续动力学系统参数模型直接辨识

针对大多数物理系统是由一组微分方程表示的连续时间系统,难以获得信号的各阶微分,连续动力学系统参数模型无法直接辨识的问题,借助Hartley正、逆变换及其微分性质,在频域对信号去噪,获得信号及其微分的估计值,建立参数模型直接估计的最小二乘算法。通过数字仿真分别辨识一个二阶系统和四阶系统,研究频率指数M和时间T与辨识精度的关系,给出辨识时参数选择的基本原则。此外,运用此方法对某冷轧机液压自动厚控系统参数模型进行辨识,结果表明:该方法对噪声具有较强的抑制能力,是一种简单实用的辨识方法。

Direct identification of the parameter model of continuous dynamic system using Hartley transformation

Based on most of the real dynamic systems are continuous in time and can be represented by a set of differential equations,it is difficult to acquire the differentials of input and output signal,the parameter model of continuous time system cannot be identified directly,according to Hartley transformation,inverse transformation and its differential property,the noise was filtered in frequency domain,and the estimators of differential signal were obtained. So,the least-squares algorithm for direct identification of parameter model was formulated.Two models with second order and fourth order were estimated,and the relationship between the frequency index M and the accuracy of identification was investigated,and some tips about design of parameter of Hartley transformation were outlined.Finally, the parameter model of a hydraulic automatic gage control system (HAGC) of cold rolling mill was identified.The result indicates that the method can be used to suppress noise effectively,and it is a simple and practical method.