摘要
研究如何在白天观测地脉动存在人为振动干扰情况下准确计算卓越周期.选择一个场地进行连续24 h地脉动观测.应用快速傅里叶变换计算地脉动数据的功率谱,应用Hilbert-Huang变换对地脉动数据进行经验模态分解并计算边际谱.通过实验发现,夜间和白天观测地脉动功率谱具有明显差异.在夜间观测的地脉动主峰明显,容易识别卓越周期,在白天观测的地脉动功率谱存在多个峰值,难以识别卓越周期.边际谱与功率谱基本一致,存在多个峰值,难以由边际谱判别卓越周期.通过实验还发现,可以由某一IMF分量的频谱判别卓越周期,但具体为哪个IMF分量需结合场地条件判断.对原始数据减去IMF分量后差值的频谱分析发现主峰明显,频率和夜间观测的非常接近,可以判别卓越周期.实验结果表明,对于在白天观测存在人为振动干扰的场地,难以通过功率谱和边际谱判别卓越周期,可以应用Hilbert-Huang变换进行经验模态分解,由原始数据减去IMF分量后差值的频谱峰值判别卓越周期.
This paper studies how to calculate the predominant period of microtremor observed during the day under the human vibrational disturbances.Microtremor observation were carried out for 24 hours at the same site,and the microtremor waveforms observed at night and during the day were compared.The power spectrum of microtremor was calculated by fast Fourier transform(FFT),and the marginal spectrum of microtremor was calculated by Hilbert-Huang transform(HHT).The microtremor waveform were decomposed into multiple IMF components by the Empirical Mode Decomposition(EMD).The experimental results show that the power spectrum of microtremor observed at night and during the day are significantly different.The predominant period of microtremor observed at night is easy to distinguish because of the obvious main peak.However,the predominant period of microtremor observed during the day is difficult to distinguish because of its multiple peaks.The marginal spectrum of microtremor observed during the day is basically the same as the power spectrum,both of which have multiple peaks.Therefore,it is difficult to distinguish the predominant period from the marginal spectrum for the microtremor under the human vibrational disturbances during the day.The predominant period can be distinguished by the spectrum of a certain IMF component,but which IMF component should be determined according to the specific site conditions.The frequency spectrum analysis of the remaining data after original data minus intrinsic mode function shows that the main peak is obvious,and the frequency of the main peak is very close to the power spectrum of the microtremor observed at night,so the predominant period can be distinguished.These results indicate that when there are human vibrational disturbances during the day,the Hilbert-Huang transform can be used to determine the predominant period from the peak spectral value of the remaining data from the intrinsic mode function subtracted from the original data.
作者
李传金
渐翔
王强
赵宇尘
姚建雄
林智勇
LI ChuanJin;JIAN Xiang;WANG Qiang;ZHAO YuChen;YAO JianXiong;LIN ZhiYong(School of Civil Engineering,Fujian University of Technology,Fuzhou 350118,China;Institute of Geotechnical Engineering,Fujian University of Technology,Fuzhou 350118,China)
出处
《地球物理学进展》
CSCD
北大核心
2021年第4期1688-1693,共6页
Progress in Geophysics
基金
福建省自然科学基金高校联合资助面上项目(2018J01620,2018J01626)
福建省大学生创新创业训练计划项目(20201388026)联合资助。
