【摘要】 水力压裂微地震事件的自动识别是微地震数据处理流程中的一项关键步骤。目前常用于微地震事件识别的STA/LTA方法仅利用单道的振幅(或能量)特征差异进行事件识别,因此,容易遗漏低信噪比事件。针对这一问题,提出了一种基于多道相似系数的微地震事件自动识别新方法。该方法的基本思路是在一个滑动时窗内对分段记录进行时差校正后计算其多道相似系数,并利用该相似系数作为检测微地震事件存在与否的依据。合成数据试算结果表明,该方法能够成功识别出信噪比仅为-2.5dB的有效事件。将此方法应用于实际资料的处理,微地震事件识别的准确率达到90%以上。通过与STA/LTA方法处理结果进行对比分析,证明了该方法是一种更为有效的微地震事件识别方法。更多还原

【Abstract】 Automatic microseismic event detection for hydraulic fracturing is a crucial step in microseismic data processing.Conventional event detection methods,such as the STA/LTA method,detect the events only by using the amplitude(or energy)difference of single channel,which usually fail to identify certain events when the signal-to-noise ratio is low.In order to solve this problem,we develop a new method based on multi-channel semblance coefficient to detect low SNR microseismic events.By using our method,we can calculate a multi-channel semblance coefficient for the record segment within a sliding time window after moveout correction,and the coefficient is applied as a detector for the existence of a microseismic event.To examine the effectiveness of our method,we apply it on synthetic and field data.The result of the synthetic data demonstrates that our method can successfully identify the event with SNR for only-2.5dB;and for the field data example,our method can detect the microseismic events with the accuracy over 90%.The above results are superior to those obtained by using the STA/LTA method,which proves our method a more effective tool for microseismic event detection.更多还原