主成分分析法和局部互相关追踪法在地震电磁信号提取与分析中的应用

发布时间：2018-03-09 12:08

本文选题：地震　切入点：电磁异常　出处：《地震地质》2017年03期 　论文类型：期刊论文

【摘要】：将主成分分析法和局部互相关追踪法应用于地震电磁数据的处理分析,尝试从较强干扰背景中提取相对较弱的电磁异常信息。主成分分析方法能够将不同频段的原始信号由强到弱投影到不同的轴上,使不同的信号分离开来,从而在一定程度上解决在较强干扰背景中识别相对较弱信号的难题;局部互相关追踪法相对于经典互相关方法更适用于非平稳信号的处理,该方法基于不同极低频电磁观测台站对应磁场分量之间的空间互相关性对相关系数异常进行拾取,从而达到弱信号识别的目的。以云南省景谷县2015年11月14日4.6级地震为例,分别运用主成分分析法和局部互相关追踪法对震中附近的极低频观测台观测的电磁资料进行处理分析。首先,将主成分分析方法应用于台站磁场分量数据的处理,得到各主成分及其所占能量比随时间的变化,结果表明震前1周左右与地震相关的第2主成分所占能量比显著增加;其次,利用局部互相关追踪法对该震例进行处理分析并与主成分分析法处理的结果进行对比,结果显示SN及EW向磁场分量的局部互相关结果于地震前1周左右均出现相关系数异常,与主成分分析方法的处理结果基本一致,并探讨了这些异常可能与地震活动性之间存在的关系。虽然目前电磁异常现象的产生与地震之间的确切关系尚缺乏直接的证据与理论支持,但2种方法对电磁异常信号的有效提取将有助于加深对地震电磁现象的认知、理解与进一步研究。
[Abstract]:Principal component analysis method and local cross-correlation tracing method are applied to the processing and analysis of seismic electromagnetic data. The principal component analysis (PCA) method can project the original signals of different frequency bands from strong to weak to different axes, so that different signals can be separated. In order to solve the problem of identifying relatively weak signals in strong interference background to some extent, the local cross-correlation tracing method is more suitable for processing non-stationary signals than classical cross-correlation methods. The correlation coefficient anomaly is picked up based on the spatial correlation between the corresponding magnetic field components of different very low frequency electromagnetic observation stations, so as to achieve the purpose of weak signal recognition. The earthquake of M = 4.6 in Jinggu County, Yunnan Province, on November 14th 2015 is taken as an example. The principal component analysis (PCA) and the local cross-correlation tracing method are used to process the electromagnetic data observed by the very low frequency observation station near the epicenter. Firstly, the principal component analysis method is applied to the processing of the magnetic field component data of the station. The changes of principal components and their energy ratios with time are obtained. The results show that the energy ratio of the second principal component related to earthquakes increases significantly one week before the earthquake. The local cross-correlation method is used to process and analyze the earthquake case and the results of principal component analysis (PCA) are compared. The results show that the correlation coefficient anomalies of SN and EW magnetic field components occur about one week before the earthquake. The results are consistent with that of principal component analysis (PCA). The relationship between these anomalies and seismicity is also discussed. Although the exact relationship between electromagnetic anomalies and earthquakes is still lacking direct evidence and theoretical support. However, the effective extraction of electromagnetic anomaly signals by two methods will help to deepen the cognition, understanding and further study of seismic electromagnetic phenomena.
【作者单位】：中国地震局地质研究所地震动力学国家重点实验室;
【基金】：极低频探地工程地震预测分系统项目(15212Z0000001)资助
【分类号】：P315.7

【相似文献】