基于小波分解与Akaike信息准则的微地震初至拾取方法

Automatic detection method of microseismic event based on wavelet decomposition and Akaike information criteria

(中国石油大学(华东)地球资源与信息学院,山东青岛266555)

Song Weiqi,College of Geo-resource and Information,China University of Petroleum (East China),Qingdao 266555,China

微地震震源的定位要求精确确定初至,人工拾取微地震有效事件需要很大的工作量。首先讨论了Akaike信息准则(AIC)初至拾取方法;然后根据微地震信号在相邻小波尺度上连续的特点,将基于AIC的初至拾取方法与小波多尺度分析方法相结合,对微地震资料进行多尺度分析;最后利用AIC拾取初至,并根据初至的分布特点确定地震记录中是否存在有效的微地震事件。克服了传统AIC法由于噪声影响使初至点模糊而难以准确拾取的缺点。模型与实际资料的应用表明,基于小波分解与AIC相结合的初至拾取方法能够从信噪比低的资料中较准确地识别出有效微地震事件。

Positioning of microseismic source requires accurate first-break picking-up,and manually identifying effective microseismic event costs plenty of time.Firstly,we discussed the first-break picking-up method on the basis of Akaike information criteria (AIC).Then,according to the continuous characteristics of microseismic signal over neighbouring wavelet scale,the first-break picking-up based on AIC was combined with wavelet multi-scale analysis to carry out multi-scale analysis on microseismic data.Finally,AIC was used to pick up first break,and the distribution of first break was analyzed to identify whether existing effective microseismic event or not in seismic record.When using traditional AIC,the first-break points are obscure because of the impact of noise,so the first break cannot be accurately picked up,our method overcomes the shortcoming.Model and actual data application results indicate that the first-break picking-up method based on wavelet decomposition and AIC can relatively accurately pick up effective microseismic event.