论文部分内容阅读
地震前兆观测数据中干扰信号频繁出现,大大增加了地震前兆异常信息提取难度。本文选用基于傅里叶变换的功率谱、傅里叶谱以及基于希尔伯特-黄变换的边际谱方法,针对黑龙江地区数字化水位和竖直摆倾斜测项中的典型干扰信号进行分析,提取了水位高频干扰、周期干扰以及竖直摆倾斜的风扰、同震应变等常见干扰信号的频谱特征。通过对比3种频谱分析方法的差异性,发现功率谱法和傅里叶谱对提取干扰和噪声等平稳信号更加有效,而边际谱法对提取同震应变波等非平稳信号的效果更加显著。
The frequent occurrence of interference signals in the earthquake precursory observation data greatly increases the difficulty of extracting the anomalous information of earthquake precursors. In this paper, the power spectrum based on Fourier transform, Fourier spectrum and edge spectrum based on Hilbert-Huang transform are selected to analyze the typical interference signals in the digitized water level and vertical tilt in Heilongjiang region. The spectrum characteristics of common interference signals such as high-frequency interference at water level, periodic interference and wind disturbance with vertical pendulum tilt and coseismic strain. By comparing the differences of the three kinds of spectrum analysis methods, it is found that the power spectrum method and the Fourier spectrum are more effective in extracting stationary signals such as noise and interference, while the marginal spectrum method is more effective in extracting non-stationary signals such as coseismic strain waves.