2012年印尼8.6级地震应变地震波的Hilbert-Huang时频分析

地震 ›› 2014, Vol. 34 ›› Issue (2): 45-54.

2012年印尼8.6级地震应变地震波的Hilbert-Huang时频分析

李文军¹, 张晶¹, 刘琦¹, 池顺良^2,3, 梁建宏⁴, 王武星¹, 曹建玲¹

1.中国地震局地震预测研究所(地震预测重点实验室), 北京 100036;
2.中国科学院计算地球动力学重点实验室, 北京 100049;
3.鹤壁市科技局, 河南鹤壁 458000;
4.中国地震台网中心, 北京 100045

发布日期:2020-07-14
作者简介:李文军(1969-), 男, 湖南怀化人, 副研究员, 博士, 主要从事地震学和地震中长期预测研究。
基金资助:
地震行业科研专项(201108009-3)资助

Time-frequency Analysis of the 2012 Indonesia M8.6 Earthquake Strain Seismic Waves by Means of Hilbert-Huang Transform

LI Wen-jun¹, ZHANG Jing¹, LIU Qi¹, CHI Shun-liang^2,3, LIANG Jian-hong⁴, WANG Wu-xing¹, CAO Jian-ling¹

1. Key Laboratory of Earthquake Prediction, Institute of Earthquake Science, CEA, Beijing 100036, China;
2. Key Laboratory of Computational Geodynamics, Chinese Academy of Sciences, Beijng 100049, China;
3. Hebi Science and Technology Bureau, Henan 458000, China;
4. China Earthquake Networks Center, Beijing 100045,China

Published:2020-07-14

摘要/Abstract

摘要： 本文通过端点效应压制的Hilbert-Huang变换, 对大同及沁源台布置的四分量钻孔应变仪记录的印尼8.6级地震激发的应变地震波形进行时频分析, 结果显示印尼8.6级地震的主震和8.2级余震的应变地震波序列各个震相具有不同的时频特征: ① 地震波到达之前的所谓“环境噪声”部分, 瞬时频率低, 瞬时振幅小; ② P波初至时, 高频成分突然增加, 振幅也随即增强; ③ S波到达时, 频率有所降低而振幅剧烈上升; ④ 面波到达时, 振幅进一步剧烈上升达到整个序列的极大值; ⑤ 尾波部分振幅逐渐降低, 但与噪声部分相比频率依然偏高, 振幅依然偏大。本文也将应变地震波与地震仪记录的地震波进行对比, 虽然应变地震波与地震波波形和Fourier谱具有极高的相关系数, 但从Hilbert-Huang变换得到的边际谱上看, 应变地震波与地震波有显著的区别, 应变地震波比地震波记录的低频成分相对更多。通过Hilbert谱, 有助于更好地了解非平稳信号的局部特征, 对于突变信号的地震波, Hilbert-Huang变换是一个较好的时频分析工具。

关键词: 应变地震波, Hilbert-Huang变换, 经验模分解, 边际谱, 2012年印尼8.6级地震

Abstract: We use the endpoint suppression Hilbert-Huang transform to analyse the Indonesia M8.6 earthquake strain seismic waves recorded by four-component borehole strain meters located at Datong and Qingyuan in time and frequency domain. The result shows that different phases of the strain seismic wave of both the M8.6 main event and M8.2 aftershock eventshows different significant time-frequency features.Before the waveforms arrive, instantaneous frequency is low and instantaneous amplitude is small. At P wave first arrival, high frequency components appear of a sudden increase, at S wave arrival, lower frequency and amplitude rise sharply. When surface wave arrives, the amplitude further rises and reachesthe greatest value of the entire sequence. Coda amplitude decreases gradually, but the frequency is still high compared to the noise portion. The amplitude is still much larger. Although strain seismic wave and seismic wave have a correlation coefficient of up to 0.93, through the Hilbert-Huang transformation, the marginal spectrum shows that strain seismic wave andseismic wave are remarkably different. Strain seismic records have relatively more low frequency component. Hilbert spectrum is helpful in understanding the detailed characterof non-stationary signal. For transient signals of strain seismic wave, Hilbert-Huang transform is an effective time-frequency analysis tool.

Key words: Strain seismic wave, Hilbert-Huang transform, Empirical mode decomposition, Marginal spectrum

中图分类号:

P315.7

李文军, 张晶, 刘琦, 池顺良, 梁建宏, 王武星, 曹建玲. 2012年印尼8.6级地震应变地震波的Hilbert-Huang时频分析[J]. 地震, 2014, 34(2): 45-54.

LI Wen-jun, ZHANG Jing, LIU Qi, CHI Shun-liang, LIANG Jian-hong, WANG Wu-xing, CAO Jian-ling. Time-frequency Analysis of the 2012 Indonesia M8.6 Earthquake Strain Seismic Waves by Means of Hilbert-Huang Transform[J]. EARTHQUAKE, 2014, 34(2): 45-54.

参考文献

[1] 池顺良. 深井宽频钻孔应变地震仪与高频地震学地震预测观测技术的发展方向, 实现地震预报的希望[J]. 地球物理学进展, 2007, 22(4): 1164-1170.
[2] 邱泽华, 等. 钻孔差应变仪观测的苏门答腊大地震激发的地球环型自由振荡[J]. 地球物理学报, 2007, 50(3): 797-805.
[3] 张晶, 刘琦. 钻孔应变差应变耦合系数反演与有关机理的探讨[J]. 大地测量与地球动力学, 2012, 32(2): 1-5.
[4] 高伟民, 蔡惟鑫. 关于应变地震仪的探讨[J]. 地壳形变与地震, 1991, 11(3): 47-54.
[5] Mike E J, Kyle R B. EarthScope Plate Boundary Observatory GPS and Strainmeter Site Permitting: A Perspective Two Years into the Construction Phase[J]. Seismol Res Lett, 2005, 76: 671-672.
[6] 宋茉, 等. 钻孔应变台站记震能力研究[J]. 大地测量与地球动力学, 2010, 30 supp(2): 31-35.
[7] 杨选辉, 等. 测震数采仪记录的钻孔应变[J]. 中国地震, 2011, 27(1): 65-71.
[8] 郭燕平, 等. 钻孔体应变高采样率观测实验[J]. 大地测量与地球动力学, 2012, 32(1): 51-55.
[9] Ding J J. Time frequency analysis and wavelet transform class notes[M]. The Department of Electrical Engineering, National Taiwan University (NTU), Taipei, Taiwan, 2011.
[10] Alan V O, et al. Discrete-Time signal processing[M]. Prentice Hall, 2005.
[11] Qian S, Chen D. Joint time-frequency analysis: Methods and Applications[M]. Prentice Hall, NJ, 1996, Chap 5.
[12] Gabor D. Theory of communication[C]. IEEETransactions On Signal Processing, 1946, 93: 42-457.
[13] Paul S A. The illustrated wavelet transform handbook[M]. Institute of Physics, 2002.
[14] Swami A. Higher-order spectral analysis toolbox[M]. The Mathworks Inc, Second Printing, 1995.
[15] Wood J C, Barry D T. Linear signal synthesis using the Radon-Wigner transform[C]. IEEE Transactions On Signal Processing, 1994, 42: 2105-2111.
[16] Barshan B, Ayrulu B. Fractional Fourier transform preprocessing for neural networks and its application to recognition[J]. Neural Networks, 2002, 15(1): 131-140.
[17] Stockwell R G, et al. Localization of the complex spectrum: the S transform[C]. IEEE Transactions On Signal Processing, 1996, 44(4): 998-1001.
[18] Heisenberg W. Quantum theory and measurement Princeton(in Germany) [J]. Journal of Physics, 1927, 43: 172-198.
[19] Huang N E, et a1. The empirical mode decomposition and Hilbert spectrum for nonlinear and non-stationary time series analysis[M]. Proc.Roy.Soc., London A, 1998, 454: 903-995.
[20] 武安绪, 等. Hilbert-Huang变换与地震信号的时频分析[J]. 中国地震, 2005, 21(2): 207-215.