基于HHT边际移动谱的结构参数时变特性识别

Time-varying identification of structural modal parameters based on HHT marginal shift spectrum

为了识别结构模态参数及其时变特性,以HilbertHuang变换(HHT)方法为基础,结合传递函数的概念,定义了HHT边际谱、HHT边际谱频响函数及HHT边际移动谱。相比使用恒定频率和模态振幅而言,HHT方法可以更好地处理信号的非线性和非平稳性,它由经验模态分解和Hilbert谱分析组成。首先对信号进行经验模态分解,得到若干个固有模态函数(IMF),然后对IMF进行Hilbert变换得到信号的边际谱。相比Hilbert边际谱,边际谱频响函数能够识别高一阶模态参数,而边际移动谱可以识别结构模态参数的时变特性。对一个3层钢筋混凝土框架结构三线型退化模型在强震作用下的各层输出加速度响应进行HHT边际移动谱分析,识别出了结构在强震作用下刚度折减、自振频率随着地震加速度的增加而改变的过程,表明边际移动谱可以识别结构参数的时变特性。

Based on Hilbert-Huang transform （HHT）, the HHT marginal spectrum, marginal spectrum frequency response function and marginal shift spectrum were defined to identify structural modal parameters and their time-varying characteristics. The HHT analysis can better deal with the nonlinear and non-stationary characteristics of signals than traditional modal analyses, which consists of the empirical mode decomposition （EMD） and the Hilbert spectral analysis. The EMD was used to obtain a number of intrinsic mode functions （IMFs） of the signal, with and the signal marginal spectrum then defined by Hilbert transform of the IMF. Compared with the marginal spectrum, the marginal spectrum frequencyresponse function can identify higher-order modal parameters, while the marginal mobile spectrum can identify the time-varying characteristics of structural modal parameters. Acceleration responses of each floor of a three-floor reinforced concrete frame structure with three-linear degradation model under strong earthquake were analyzed using the HHT marginal shift spectrum analysis method, with the structural stiffness and natural frequency changing with increasing seismic acceleration amplitude. The results show that the marginal shift spectrum can identify the time-varying characteristics of structural parameters.