利用小波包变换时频谱识别宁夏及邻区的地震和爆破*

(宁夏回族自治区地震局,银川 750001)

小波包变换; 时频谱; 地震; 爆破; 识别; 宁夏及邻区

Discrimination between Earthquakes and Explosions in Ningxia and Its Neighboring Region Using Time-frequency Spectrum Wavelet Packet Transform
ZENG Xian-wei,ZHAO Wei-ming,LI Hong-ting,SHI Hai-kuo,YAO Lin

(Earthquake Administration of Ningxia Hui Autonomous Region,Yinchuan 750001,Ningxia,China)

wavelet packet transform; time-frequency spectrum; earthquakes; explosions; identification; Ningxia and the neighboring region

备注

采用dmey小波基函数分别对地震和爆破事件的垂直向记录信号进行小波包变换,计算各事件信号的归一化时频谱值以及P波和S波时频谱值达到最大时的频率fmp和fms,比较地震信号和爆破信号P波段(0~6.25 Hz)和S波段(0~6.25 Hz)在各相同分解频带内的瞬时谱最大值差异,寻找合适的单项定量识别指标,并综合各单项识别指标形成综合识别判据。运用综合识别判据对银川台记录到的宁夏及邻区14个地震事件和19个爆破事件进行判别,结果表明,各单项定量识别指标的识别率均在80%以上,综合判别结果均与事件的真实类型一致。

Using dmey wavelet-base function,we conduct a wavelet-packet transform of the signals of vertical component recordings of the earthquakes and the explosions to calculate the values of the normalized time-frequency spectrum of the signal of each earthquake and explosion,and to calculate the frequency fmp on condition that the value of time-frequency spectrum of P wave is maximum.We also calculate the frequency fms on condition that the value of time-frequency spectrum of S wave is maximum.Then we compare the maximum value of instantaneous spectrum of the seismic signals with that of the blasting signals in the same decomposition frequency band of P wave within frequency 0-6.25 Hz.And we compare the maximum value of instantaneous spectrum of the seismic signals with that of the blasting signals in the same decomposition frequency band of S wave within frequency 0-6.25 Hz.Thus we get some quantitative indexes to identify the earthquake and the explosion. Combining all these indexes we get a synthetic criterion.Then we apply the single index and the synthetic criterion to the identification test of 14 quakes and 19 explosions recorded by Yinchuan Seismic Station.We find that using the single index,80% of the quakes and the explosions can be identified,and 100% of the quakes and the explosions can be identified using the synthetic criterion.