基于干涉成像条件的弹性波震源逆时定位研究

Study on elastic wave source localization by using time-reversal methods with interferometric imaging condition

震源逆时成像是震源定位中的一种有效方法,但由于逆时定位是基于波场在逆时延拓过程中的相干叠加,获得准确的逆时定位成像结果需要密集的检波器分布与精确的速度模型.实际应用中,空间采样稀疏且无法获得精确速度模型,加之地震数据信噪比低等因素使得震源逆时定位精度降低.为改善震源逆时定位效果,本文将弹性波逆时定位算法与干涉成像条件相结合,探讨多种模型下的弹性波震源逆时定位效果.根据简单模型与复杂模型的试算表明:该方法在稀疏采样、速度模型不精确及信噪比低的情况下,能获得较为准确的震源定位,且能有效的压制传统震源逆时定位成像的背景噪声.

Time-reversal methods have significant potential for solving the source-location problem. The principle of time-reversal methods,which used for source localization,is the coherence stack of the wavefields,which come from the process of the time-reversal extrapolation. It is conditional to obtain an exact source localization result. An accurate result of source-location can be obtained,only when the geophones are distributed around the source densely and the velocity model is accurate. In the actual field construction,the interval between each geophone is large,and in most cases the geophones are only arranged on the ground. Meanwhile,the structure of the geologic body underground is complicated and the exact velocity of the geologic body cannot be obtained. Furthermore,the seismic data,which received by geophones,have a low signal-tonoise ratio. All of these factors lead to the decrease of the accuracy of source localization by using time-reversal methods. In order to improve the results of source localization,we make some improvements to the conventional algorithm of time-reversal methods. In this paper,the algorithm of elastic wave source localization by using time-reversal methods is combined with interferometric imaging condition. With these algorithms the results of elastic wave source localization are discussed under various models. According to the results of the layered isotropic medium model with velocity disturbance and the complicated model with velocity disturbance,the source location is more accurate by using time-reversal methods with interferometric imaging condition when the geophones are arranged sparsely and the velocity of the model cannot be obtained and the seismic data have a low signal-to-noise ratio.Meanwhile,the background noise appeared in the conventional source localization image using time-reversal methods can effectively be removed by using time-reversal methods with interferometric imaging condition.