一种微震震相到时自动拾取方法

An Automatic Picking Method for Onset Time of Microseismic Phase

震相初至到时的自动拾取是微震监测中一个关键问题,是实现震源定位及震源机制解释的前提和基础。针对微震信号固有的低信噪比、非线性、非平稳等特征,提出了一种震相初至自动拾取方法。首先,应用小波阈值降噪法对采集到的含噪微震信号进行降噪处理,降低人工采集时机械类振动噪声对微震震相初至拾取的干扰;然后使用STA/LTA对降噪后的微震信号进行预拾取,获得震相初至的大致时刻;以该时刻为基准,分别向前向后各截取500个采样点构成一个时间窗口,在该时间窗口内应用AIC准则法精确求取震相初至。为验证本文方法的有效性,以人工拾取的到时为基准,误差在±10ms之内视为准确拾取。实验结果显示,本文方法对低信噪比微震信号拾取的准确率为95.79%,较单纯使用STA/LTA及AIC方法准确率分别提高了9.78%及10.08%。

As a key problem in microseismic monitoring,the automatic picking of the onset time of seismic phase is the premise and foundation for the focal location and focal mechanism interpretation.This paper proposed a picking method for the onset time of seismic phase in view of the low signal-to-noise ratio,nonlinear and non-stationary characteristics inherent in seismic signals.Firstly,the wavelet threshold de-noising method was used to de-noise the original micro-seismic signals.The purpose was to reduce the disturbance of mechanical vibration noise in manual acquisition to the picking of the onset time of seismic phase.Then,STA/LTA was used to pre-pick up the microseismic signals so as to get the rough onset time.With this moment as the center,500 sampling points were intercepted forward and 500 sampling points were intercepted backward to form a time window.Finally,the AIC criterion method was used to obtain the accurate onset time of seismic phase within the time window.To verify the effective-ness of the method,with the manually picked onset time as the benchmark,the picking with the internal error of ±10 ms was considered as accurate picking.The experimental results show that the picking accuracy of low SNR signals with the proposed method is 95.79%,improving 9.78% and 10.08% repectively compared with that of the STA/LTA method and AIC method.