摘要
在简要介绍时程信号的小波分析和Hilbert Huang变换 (HHT)理论的基础上 ,通过地震波和其它时程信号实例 ,对比分析了小波变换和HHT变换结果 .比较显示 :HHT变换和小波变换均能用于对非平稳的信号进行分析 ,并能捕捉到信号变化的主要特征 ;与受所选母波影响较大的小波分析不同 ,HHT变换得到的固有模态函数是直接从原始时程数据中分离出来的 ,它更能反映原始数据的固有特性 ;小波分析得到的谱的能量在频率范围内分布较广 ,而HHT变换的Hilbert谱的大部分能量都集中在一定的时间和频率范围内 ,能清晰地刻画信号能量随时间、频率的分布 .因此 ,Hilbert Huang变换不仅是对非平稳信号进行分析的有效方法 。
After introducing the brief theories of wavelet analysis and Hibert Huang Transform (HHT), several signal data were analyzed by using HHT and wavelet analysis methods. From the comparison of the two kinds results, we can get conclusions as following: Both HHT and wavelet transform can be used to analyze non stationary signal and grasp main characteristics of signal. In HHT the intrinsic mode function (IMF) is directly decomposed from original data, while wavelet components are decomposed according to mother wavelet, which are influenced strongly by the selected mother wavelet. So that IMF can reflect the intrinsic physical characteristics of original data more efficiently. Hilbert spectrum can clearly express the energy distribution with time and frequency. Most energy in Hilbert spectrum from HHT is much more concentrated in the definitive range of time and frequency than that in Morlet spectrum from wavelet analysis. Therefore we consider that HHT is not only one efficient method for analyzing non stationary and non linear data, but also is a useful tool for examining detail characters of time history signal.
出处
《地震学报》
CSCD
北大核心
2003年第4期398-405,共8页
Acta Seismologica Sinica
基金
国家自然科学基金 ( 5 0 1780 5 5 )资助项目
