论文部分内容阅读
在2008年5月12日汶川MS8.0地震中,四川数字强震台网共获取了133组三分向加速度记录.本文选取了一些不同断层距的台站所获取的强震动记录进行了处理和分析.在数据处理中,采用基于聚类经验模态分解(EEMD)提取信号时频特性的方法,有效获得了信号能量的时频分布,提取了中心频率、Hilbert能量、最大振幅对应的时频等特性,并与傅里叶变换、小波变换进行了对比研究.研究结果表明,对非线性的强震记录采用聚类经验模态分解(EEMD)能抑制经验模态分解(EMD)中存在的模态混叠问题;与傅里叶变换和小波变换相比发现,HHT边际谱在低频处幅值高于傅里叶谱;与小波变换受到所选取的母波强烈影响不同,HHT直接从强震记录中分离出固有模态函数(IMF),更能反映出原始数据的固有特性,Hilbert谱反映出大部分能量都集中在一定的时间和频率范围内,而小波谱的能量却在频率范围内分布较为广泛.因此,基于EEMD的HHT在客观性和分辨率方面都具有明显的优越性,能提取到更多强震加速度记录的时频特性.
In the Wenchuan MS8.0 earthquake on May 12, 2008, a total of 133 sets of trichothetical acceleration records were obtained from the Sichuan Digital Strong Earthquake Network.This paper selected strong ground motion records obtained from some stations with different fault distances And analysis.In the data processing, the time-frequency distribution of signal energy is obtained by extracting the time-frequency characteristics of signal based on clustering empirical mode decomposition (EEMD), and the center frequency, Hilbert energy and the corresponding maximum amplitude are extracted Frequency and other characteristics, and compared with Fourier transform and wavelet transform.The results show that clustering empirical mode decomposition (EEMD) can suppress the existence of experiential mode decomposition (EMD) in strong earthquake records Compared with Fourier transform and wavelet transform, it is found that the amplitude of HHT marginal spectrum is higher than that of Fourier spectrum at low frequency. Different from the strong influence of the selected mother wave, HHT directly from Strong earthquake records separate the intrinsic mode function (IMF), which can better reflect the inherent characteristics of the original data. Hilbert spectrum reflects that most of the energy is concentrated in a certain time and frequency range, while the energy of the wavelet spectrum is at the frequency Range Therefore, the HHT based on EEMD has obvious superiority in both objectivity and resolution, and can extract more time-frequency characteristics of acceleration records of strong earthquakes.