一种适于存在极性反转的微震初至到时拾取方法

2020年 59卷 第No. 3期
阅读:78
查看详情
Picking arrival times of microseismic events from surface monitoring data with waveform polarity reversals
1.中国石油化工股份有限公司胜利油田分公司科技处,山东东营257001;2.海洋科学与探测技术教育部重点实验室,中国海洋大学海洋地球科学学院,山东青岛266100;3.中国石油化工股份有限公司胜利油田分公司物探研究院,山东东营257001
1.Science &Technology Department of Shengli Oilfield,SINOPEC,Dongying 257001,China;2.Key Lab of Submarine Geosciences and Prospecting Techniques.MOE.,College of Marine Geoscience,Ocean University of China,Qingdao 266100,China;3.Geophysical Research Institute of Shengli Oilfield,SINOPEC,Dongying 257001,China

地面微震监测资料的信噪比往往很低,从中识别微震事件和拾取微震到时数据仍然是个难题。当不同道的微震存在初至极性反转时,常用的多道微震记录波形相似系数函数会导致微震事件的识别与拾取遗漏。为此,提出一种适于存在极性反转的微震初至到时拾取方法。利用所有道两两之间的互相关函数最大幅值对应的延时数据,求取各道微震事件的相对到时;用各道的相对到时对微震事件初至到时进行时差校正后,利用相邻道零延迟相乘再叠加得到参考道,并用长短时窗能量均值比(STA/LTA)算法确定参考道上微震初至到时;最后,将各道的微震相对到时分别与参考道初至到时相加得到各道微震的绝对到时。合成数据和实际资料的处理结果表明,利用相邻道零延迟相乘能有效克服初至极性反转对微震识别和初至拾取带来的不利影响,为微震事件识别和到时拾取提供信噪比较高的参考道,提高了微震事件识别和初至到时拾取的精度。

Detecting microseismic events and picking arrival times from surface microseismic monitoring data with low signal-to-noise ratio(SNR) is challenging.The waveform semblance function commonly used for microseismic event detection is influenced by the reversed waveform polarities,resulting in the missed detection of some microseismic events.To address this issue,a method for picking arrival times of microseismic events with waveform polarity reversals was proposed.The arrival time of an event in a trace is treated as the sum of the average arrival times for all traces and the relative arrival time in the trace,the latter being defined as the difference between the actual arrival time in a trace and the average one for all traces.The delay of an event between two traces is obtained as the maximum amplitude of their cross-correlation.The equation for the relative arrival times in all traces is constructed using the delays evaluated between each couple of traces.The relative arrival times in each trace are then calculated by evaluating the equation using the least squares method.Subsequently,each trace is corrected in the time domain using its corresponding relative arrival times;then,a reference trace is obtained by stacking the products of each pair of adjacent traces with zero delays.The average arrival time is determined from the reference trace using the STA/LTA ratio algorithm.Finally,the arrival time of an event in each trace is obtained by summing the average arrival time and the corresponding relative arrival time.Tests on synthetic and real data showed that the proposed approach can effectively eliminate the effects of polarity reversals and improve the event detection and arrival time picking by providing the reference traces with high SNR.

微地震; 到时拾取; 极性反转; 互相关函数; 相邻道相乘; 参考道;
microseismic;; arrival-time picking;; polarity reversal;; cross-correlation function;; product of adjacent traces;; reference trace;

国家科技重大专项“渤海湾盆地济阳坳陷致密油开发示范工程”(2017ZX05072)、山东省重点研发计划项目“压裂裂缝的地面微地震成像关键技术研究”(2017GSF16103)共同资助。

10.3969/j.issn.1000-1441.2020.03.003