基于经验模态分解的地震波特征提取的研究
|
摘要
本文原始数据为35个天然地震和27个人工爆破事件的离震中最近的5个台站的垂直分量波形数据.从原始波形数据分别提取最大振幅对应的周期、倒谱的方差、自相关函数的最大值3个特征.依据希尔伯特-黄变换原理,用经验模态分解方法把原始波形信号分解为10个左右的本征模态函数分量后,再从每个分量中分别提取这3个特征.接着对所获取的特征样本集合采用随机划分法分为学习样本集与检验样本集,然后再通过支持向量机进行分类识别,如此反复进行多次样本划分和分类识别.结果表明经验模态分解后的分量信号提取的这3个特征具有更高的识别率,说明了经验模态分解有利于识别天然地震和人工爆破事件,值得进一步深入研究.
The research on seismic signal processing,analysis,and further discrimination of earthquakes and explosions plays a fundamental role in the development of seismology and other geosciences.And explosion event recognition,especially nuclear explosion recognition,is indispensable for public welfare and world peace.This paper introduces some important results of our research project:"Classification and Decision Support System for Recognizing Natural Earthquake Events and Explosion Events".The selected raw data are the up-down components of 5 observatory stations which are the nearest ones to the epicenters of 35 earthquake events occurred during the years of 2003~2007 and 27 explosion events occurred during the years of 2002~2008.From the raw waveform data,3 features:ratio of the maximal amplitude of S-Wave and P-Wave,cepstral variance,maximum of autocorrelation,are extracted.And based upon Hilbert-Huang Transform(HHT) principle,by means of empirical mode decomposition(EMD) method,original waves are decomposed into about 10 intrinsic mode functions(IMFs).These 3 features are also extracted from each of the IMFs for every selected wave signals.The acquired about features samples are randomly partitioned into training sample set and testing sample set.Finally,Support vector machine(SVM) classifier for discriminating earthquake and explosion events are built and trained by training sample set and tested by testing sample set.Such a process of samples-partition,training,testing are reiterated 100 times.The results show that these 3 features extracted from IMFs after EMD is better than those extracted from raw signal with higher correct recognition rate.This implies EMD processing of seismic wave is beneficial for recognition of earthquake and explosion events and deserves more intensive research.
The research on seismic signal processing,analysis,and further discrimination of earthquakes and explosions plays a fundamental role in the development of seismology and other geosciences.And explosion event recognition,especially nuclear explosion recognition,is indispensable for public welfare and world peace.This paper introduces some important results of our research project:"Classification and Decision Support System for Recognizing Natural Earthquake Events and Explosion Events".The selected raw data are the up-down components of 5 observatory stations which are the nearest ones to the epicenters of 35 earthquake events occurred during the years of 2003~2007 and 27 explosion events occurred during the years of 2002~2008.From the raw waveform data,3 features:ratio of the maximal amplitude of S-Wave and P-Wave,cepstral variance,maximum of autocorrelation,are extracted.And based upon Hilbert-Huang Transform(HHT) principle,by means of empirical mode decomposition(EMD) method,original waves are decomposed into about 10 intrinsic mode functions(IMFs).These 3 features are also extracted from each of the IMFs for every selected wave signals.The acquired about features samples are randomly partitioned into training sample set and testing sample set.Finally,Support vector machine(SVM) classifier for discriminating earthquake and explosion events are built and trained by training sample set and tested by testing sample set.Such a process of samples-partition,training,testing are reiterated 100 times.The results show that these 3 features extracted from IMFs after EMD is better than those extracted from raw signal with higher correct recognition rate.This implies EMD processing of seismic wave is beneficial for recognition of earthquake and explosion events and deserves more intensive research.
引文
[1]安镇文,郭祥云,边银菊,等.核爆炸与地震识别研究进展.国际地震动态,2008,(8):22-31.An Z W,Guo X Y,BianY J,et al.Recent advances indiscrimination of nuclear explosion and earthquake.RecentDevelopments in World Seismology(in Chinese),2008,(8):22-31.
[2]和雪松,李世愚,沈萍,等.用小波包识别地震和矿震.中国地震,2006,22(4):425-434.He X S,Li S Y,Shen P,et al.A wavelet packet approach towave classification of earthquakes and mining shocks.Earthquake Research in China(in Chinese),2006,22(4):425-434.
[3]刘庆敏,杨午阳,李书平,等.希尔伯特-黄变换在地震资料处理中的应用.西北地震学报,2009,31(3):211-216.Liu Q M,Yang W Y,Li S P,et al.Application of the hilbert-huang transform to seismic data processing.NorthwesternSeismological Journal(in Chinese),2009,31(3):211-216.
[4]Huang N E,Shen Z,Long S R,et al.The empirical modedecomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis.Proceedings of the RoyalSociety of London,1998,454(1971):903-995.
[5]Hassan H H.Empirical Mode Decomposition(EMD)ofpotential field data:airborne gravity data as an example.SEGExpanded Abstracts,2005:704-706.
[6]M agrin-Chagnolleau I,Baraniuk R G.Em pirical m odedecomposition based time-frequency attributes.SEG Meeting,1999.
[7]Huang N E,Shen Z,Long S R.A new view of nonlinear waterwaves:the Hilbert Spectrum.Annual Review of FluidMechanics,1999,31:417-457.
[8]Huang N E,Wu M L C,Long S R,et al.A confidence limit forthe empirical mode decomposition and Hilbert spectralanalysis.Proceedings of the Royal Society of London,2003,459(2037):2317-2345.
[9]李夕兵,张义平,刘志祥,等.爆破震动信号的小波分析与HHT变换.爆炸与冲击,2005,25(6):528-535.Li X B,Zhang Y P,Liu Z X,et al.Wavelet analysis andHilbert-Huang Transform of blasting vibration signal.Explosion and Shock Waves(in Chinese),2005,25(6):528-535.
[10]孙涛,刘晶璟,孔凡,等.小波变换和希尔伯特-黄变换在时频分析中的应用.中国水运(理论版),2006,4(11):17-19.Sun T,Liu J J,Kong F,et al.In time frequency analysisapplication.China Water Transport(in Chinese),2006,4(11):17-19.
[11]黄汉明,边银菊,卢世军,等.天然地震与人工爆破的波形小波特征研究.地震学报,2010,32(3):270-276.Huang H M,Bian Y J,Lu S J,et al.A wavelet featureresearch on seismic waveforms of earthquakes andexplosions.Acta Seismologica Sinica(in Chinese),2010,32(3):270-276.
[12]包翠玲.天然地震波形与工业化爆破波形振幅比对比的研究.内陆地震,2007,21(3):267-274.Bao C L.Contrast research on amplitude ratio betweenseismic wave and explosion wave.Inland Earthquake(inChinese),2007,21(3):267-274.
[13]魏富胜,黎明.震源性质的倒谱分析.地震学报,2003,25(1):47-54.Wei F S,Li M.Cepstrum analysis of source character.ActaSeismologica Sinica(in Chinese),2003,25(1):47-54.
[14]崔凤林,勾永峰,蔡朝晖.用于火山岩预测的地震属性提取及有效性分析方法研究.石油物探,2005,44(6):598-600.Cui F L,Gou Y F,Cai Z H.The abstraction of seismicattributes and its effectiveness in volcanic prediction.Geophysical Prospecting for Petroleum(in Chinese),2005,44(6):598-600.
[15]Huang H M.Pattern classifiers for analysis of mineralelements related to human health.Hong Kong:Department ofPhysics and Materials Science,City University of HongKong,2001.
[16]陈林,宋海斌.基于经验模态分解的地震瞬时属性提取.地球物理学进展,2008,23(4):1179-1185.Chen L,Song H B.Seismic instantaneous attribute extractionbased on empirical mode decomposition.Progress inGeophysics(in Chinese),2008,23(4):1179-1185.
[17]杨培杰,印兴耀,张广智.希尔伯特-黄变换地震信号时频分析与属性提取.地球物理学进展,2007,22(5):1585-1590.Yang P J,Yin X Y,Zhang G Z.Seismic signal time-frequencyanalysis and attributes extraction based on HHT.Progress inGeophysics(in Chinese),2007,22(5):1585-1590.
[18]刘喜武,刘洪,李幼铭,等.局域波分解及其在地震信号时频分析中的应用.地球物理学进展,2007,22(2):365-375.Liu X W,Liu H,Li Y M,et al.Local waves decompositionand its applications to seismic signal time-frequency analysis.Progress in Geophysics,2007,22(2):365-375.
[19]周挚,山秀明,张立,等.基于HHT提取昆明、下关重力固体潮的地震前兆信息.地球物理学报,2008,51(3):836-844.Zhou Z,Shan X M,Zhang L,et al.The gravity tide ofKunming&Xiaguan based on the HHT.Chinese J.Geophys.(in Chinese),2008,51(3):836-844.
[20]龚志强,邹明玮,高新全,等.基于非线性时间序列分析经验模态分解和小波分解异同性的研究.物理学报,2005,54(8):3947-3957.Gong Z Q,Zou M W,Gao X Q,et al.On the differencebetween empirical mode decomposition and waveletdecomposition in the nonlinear time series.Acta PhysicaSinica(in Chinese),2005,54(8):3947-3957.
[21]于彩霞,魏文博,景建恩,等.希尔伯特-黄变换在海底大地电磁测深数据处理中的应用.地球物理学进展,2010,25(3):1046-1056.Yu C X,Wei W B,Jing J E,et al.Application of Hilbert-Huang transformation to marine magnetotelluric soundingdata processing.Progress in Geophysics(in Chinese),2010,25(3):1046-1056.
[22]胥德平,邓兴,郭科,等.基于特征矩阵联合近似对角化和经验模态分解的储层识别.地球物理学进展,2011,26(2):572-578.Xu D P,Deng X,Guo K,et al.Reservoir detection based onJADE and EMD.Progress in Geophysics(in Chinese),2011,26(2):572-578.
[23]边肇祺,张学工.模式识别.北京:清华大学出版社,2000:284-304.Bian Z Q,Zhang X G.Pattern Recognition(in Chinese).Beijing:Tsinghua University Press,2000:284-304.
[24]郭丽娟,孙世宇,段修生.支持向量机及核函数研究.科学技术与工程,2008,8(2):487-490.Guo L J,Sun S Y,Duan X S.Research for support vectormachine and kernd function.Science Technology andEngineering(in Chinese),2008,8(2):487-490.
[25]张军,李洪奇,郭海峰,等.基于支持向量机的测井资料幅度异常验收方法研究.地球物理学进展,2009,24(6):2181-2185.Zhang J,Li H Q,Guo H F,et al.Abnormal amplitudechecking method of log data based on SVM.Progress inGeophysics(in Chinese),2009,24(6):2181-2185.
[26]王碧泉,陈祖荫.模式识别:理论、方法和应用.北京:地震出版社,1989:192-201.Wang B Q,Chen Z Y.Pattern Recognition:Theory Methodand Application(in Chinese).Beijing:Earthquake Press,1989:192-201.
[27]毕明霞,黄汉明,边银菊,等.天然地震与人工爆破波形信号HHT特征提取和SVM识别研究.地球物理学进展,2011,26(4):1157-1164.Bi M X,Huang H M,Bian Y J,et al.A study on seismicsignal HHT features extraction and SVM recognition ofearthquake and explosion.Progress in Geophysics(inChinese),2011,26(4):1157-1164.
[2]和雪松,李世愚,沈萍,等.用小波包识别地震和矿震.中国地震,2006,22(4):425-434.He X S,Li S Y,Shen P,et al.A wavelet packet approach towave classification of earthquakes and mining shocks.Earthquake Research in China(in Chinese),2006,22(4):425-434.
[3]刘庆敏,杨午阳,李书平,等.希尔伯特-黄变换在地震资料处理中的应用.西北地震学报,2009,31(3):211-216.Liu Q M,Yang W Y,Li S P,et al.Application of the hilbert-huang transform to seismic data processing.NorthwesternSeismological Journal(in Chinese),2009,31(3):211-216.
[4]Huang N E,Shen Z,Long S R,et al.The empirical modedecomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis.Proceedings of the RoyalSociety of London,1998,454(1971):903-995.
[5]Hassan H H.Empirical Mode Decomposition(EMD)ofpotential field data:airborne gravity data as an example.SEGExpanded Abstracts,2005:704-706.
[6]M agrin-Chagnolleau I,Baraniuk R G.Em pirical m odedecomposition based time-frequency attributes.SEG Meeting,1999.
[7]Huang N E,Shen Z,Long S R.A new view of nonlinear waterwaves:the Hilbert Spectrum.Annual Review of FluidMechanics,1999,31:417-457.
[8]Huang N E,Wu M L C,Long S R,et al.A confidence limit forthe empirical mode decomposition and Hilbert spectralanalysis.Proceedings of the Royal Society of London,2003,459(2037):2317-2345.
[9]李夕兵,张义平,刘志祥,等.爆破震动信号的小波分析与HHT变换.爆炸与冲击,2005,25(6):528-535.Li X B,Zhang Y P,Liu Z X,et al.Wavelet analysis andHilbert-Huang Transform of blasting vibration signal.Explosion and Shock Waves(in Chinese),2005,25(6):528-535.
[10]孙涛,刘晶璟,孔凡,等.小波变换和希尔伯特-黄变换在时频分析中的应用.中国水运(理论版),2006,4(11):17-19.Sun T,Liu J J,Kong F,et al.In time frequency analysisapplication.China Water Transport(in Chinese),2006,4(11):17-19.
[11]黄汉明,边银菊,卢世军,等.天然地震与人工爆破的波形小波特征研究.地震学报,2010,32(3):270-276.Huang H M,Bian Y J,Lu S J,et al.A wavelet featureresearch on seismic waveforms of earthquakes andexplosions.Acta Seismologica Sinica(in Chinese),2010,32(3):270-276.
[12]包翠玲.天然地震波形与工业化爆破波形振幅比对比的研究.内陆地震,2007,21(3):267-274.Bao C L.Contrast research on amplitude ratio betweenseismic wave and explosion wave.Inland Earthquake(inChinese),2007,21(3):267-274.
[13]魏富胜,黎明.震源性质的倒谱分析.地震学报,2003,25(1):47-54.Wei F S,Li M.Cepstrum analysis of source character.ActaSeismologica Sinica(in Chinese),2003,25(1):47-54.
[14]崔凤林,勾永峰,蔡朝晖.用于火山岩预测的地震属性提取及有效性分析方法研究.石油物探,2005,44(6):598-600.Cui F L,Gou Y F,Cai Z H.The abstraction of seismicattributes and its effectiveness in volcanic prediction.Geophysical Prospecting for Petroleum(in Chinese),2005,44(6):598-600.
[15]Huang H M.Pattern classifiers for analysis of mineralelements related to human health.Hong Kong:Department ofPhysics and Materials Science,City University of HongKong,2001.
[16]陈林,宋海斌.基于经验模态分解的地震瞬时属性提取.地球物理学进展,2008,23(4):1179-1185.Chen L,Song H B.Seismic instantaneous attribute extractionbased on empirical mode decomposition.Progress inGeophysics(in Chinese),2008,23(4):1179-1185.
[17]杨培杰,印兴耀,张广智.希尔伯特-黄变换地震信号时频分析与属性提取.地球物理学进展,2007,22(5):1585-1590.Yang P J,Yin X Y,Zhang G Z.Seismic signal time-frequencyanalysis and attributes extraction based on HHT.Progress inGeophysics(in Chinese),2007,22(5):1585-1590.
[18]刘喜武,刘洪,李幼铭,等.局域波分解及其在地震信号时频分析中的应用.地球物理学进展,2007,22(2):365-375.Liu X W,Liu H,Li Y M,et al.Local waves decompositionand its applications to seismic signal time-frequency analysis.Progress in Geophysics,2007,22(2):365-375.
[19]周挚,山秀明,张立,等.基于HHT提取昆明、下关重力固体潮的地震前兆信息.地球物理学报,2008,51(3):836-844.Zhou Z,Shan X M,Zhang L,et al.The gravity tide ofKunming&Xiaguan based on the HHT.Chinese J.Geophys.(in Chinese),2008,51(3):836-844.
[20]龚志强,邹明玮,高新全,等.基于非线性时间序列分析经验模态分解和小波分解异同性的研究.物理学报,2005,54(8):3947-3957.Gong Z Q,Zou M W,Gao X Q,et al.On the differencebetween empirical mode decomposition and waveletdecomposition in the nonlinear time series.Acta PhysicaSinica(in Chinese),2005,54(8):3947-3957.
[21]于彩霞,魏文博,景建恩,等.希尔伯特-黄变换在海底大地电磁测深数据处理中的应用.地球物理学进展,2010,25(3):1046-1056.Yu C X,Wei W B,Jing J E,et al.Application of Hilbert-Huang transformation to marine magnetotelluric soundingdata processing.Progress in Geophysics(in Chinese),2010,25(3):1046-1056.
[22]胥德平,邓兴,郭科,等.基于特征矩阵联合近似对角化和经验模态分解的储层识别.地球物理学进展,2011,26(2):572-578.Xu D P,Deng X,Guo K,et al.Reservoir detection based onJADE and EMD.Progress in Geophysics(in Chinese),2011,26(2):572-578.
[23]边肇祺,张学工.模式识别.北京:清华大学出版社,2000:284-304.Bian Z Q,Zhang X G.Pattern Recognition(in Chinese).Beijing:Tsinghua University Press,2000:284-304.
[24]郭丽娟,孙世宇,段修生.支持向量机及核函数研究.科学技术与工程,2008,8(2):487-490.Guo L J,Sun S Y,Duan X S.Research for support vectormachine and kernd function.Science Technology andEngineering(in Chinese),2008,8(2):487-490.
[25]张军,李洪奇,郭海峰,等.基于支持向量机的测井资料幅度异常验收方法研究.地球物理学进展,2009,24(6):2181-2185.Zhang J,Li H Q,Guo H F,et al.Abnormal amplitudechecking method of log data based on SVM.Progress inGeophysics(in Chinese),2009,24(6):2181-2185.
[26]王碧泉,陈祖荫.模式识别:理论、方法和应用.北京:地震出版社,1989:192-201.Wang B Q,Chen Z Y.Pattern Recognition:Theory Methodand Application(in Chinese).Beijing:Earthquake Press,1989:192-201.
[27]毕明霞,黄汉明,边银菊,等.天然地震与人工爆破波形信号HHT特征提取和SVM识别研究.地球物理学进展,2011,26(4):1157-1164.Bi M X,Huang H M,Bian Y J,et al.A study on seismicsignal HHT features extraction and SVM recognition ofearthquake and explosion.Progress in Geophysics(inChinese),2011,26(4):1157-1164.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心