基于波形相似特征的微地震事件初至拾取及全局校正

Arrival picking and global refinement for microseismic events based on waveform similarity

微地震事件初至拾取是井下微地震监测数据处理的关键步骤之一.初至误差的存在会使微地震震源定位结果产生较大偏差,进而影响后续的压裂裂缝解释.通常初至拾取过程对所有的微地震事件选择相同的特征函数并采用一致的拾取参数进行统一处理,然而当事件的能量、震源机制、传播路径以及背景噪声等存在明显差异时,所得初至拾取结果差别显著.为了提高微地震事件初至拾取标准一致性,本文提出基于波形相似特征的初至拾取及全局校正方法.该方法首先利用互相关函数对每个事件内的各道记录进行时差校正,得到初始初至信息并形成叠加道,再对所有事件的叠加道进行全局互相关得到事件间初至相对校正量,最终初至结果可以通过各个事件的初始初至信息与其相对校正量相加得到.方法将所有微地震事件初至结果作为一个整体处理,从而能够克服常规方法初至拾取标准一致性差的缺陷.实际资料处理结果表明,相比于常规方法,该方法可以有效提高事件初至拾取和定位结果的一致性.

Arrival picking of microseismic events is a fundamental step in processing of downhole microseismic monitoring data.The errors in the arrival times may lead to significant bias in the estimate of hypocenter locations and thus the interpretation of hydraulic fractures.Due to the difference of energy,focal mechanisms,propagation paths and the presence of ambient noise,the arrival results show significant distinction with the same method in conventional picking workflow.We have developed a new method based on waveform similarity in arrival picking and refinement to increase the consistency of arrival picking criteria.This method performs moveout correction for the microseismic records based on cross-correlation functions,the records after moveout correction are superposed and the arrival times are obtained by pick the arrivals for the stack traces.Global refinement is applied to the stack waveforms of different events to calculate the interevent arrival time difference.Finally,the arrival results of each microseismic events can be obtained by adding the arrival information and relative time differences.The proposed method considers the arrival data as a whole and compares recordings from different events,which can overcome the shortcoming of the conventional algorithms in terms of the identic picking criteria.The processing results of microseismic events recorded by borehole array indicate the method can improve the accuracy of the arrival and event location results compared to the traditional processing method.