基于快速匹配追踪算法的地震道集剩余时差校正

为提取正确的AVO信息和获得高信噪比、高分辨率的叠加剖面,需要消除CMP道集上的剩余时差,校平同相轴。为此,给出了一种基于匹配追踪(Matching Pursuit,MP)算法的地震道集剩余时差校正方法。利用匹配追踪算法在地震信号分解与重构方面的优势,将CMP道集信号分解为不同的子波,对各个子波分别进行剩余时差校正,再用校正后的子波重构出消除剩余时差后的地震道。同时,为提高匹配追踪算法的计算效率,利用相位展开法得到稳定的瞬时频率,以瞬时频率为中心进行扫描,快速得到最优匹配频率。理论模型和实际资料的测试结果表明,这种方法不仅可以提高匹配追踪算法的计算效率,而且还能很好地校平CMP道集同相轴。

In order to extract the correct AVO information and obtain the stacking section with high SNR and high resolution,we need to eliminate residual moveout of CMP gathers by leveling reflection events.This paper presents a seismic gather residual moveout correction method based on matching pursuit algorithm.Using the advantage of the seismic signal decomposition and reconstruction of the matching pursuit algorithm,the CMP gathers signals are decomposed into different wavelets.Residual moveout correction is respectively conducted onto each wavelet,and then the corrected wavelets are used for seismic wavelet reconstruction.Moreover,in order to improve computation efficiency of the matching pursuit algorithm,we use the phase unwrapping method to get the stable instantaneous frequency and take the instantaneous frequency at envelope peak as the center to scan,and the optimal matching frequency is obtained rapidly.The theoretical model and actual data tests show that,this method not only improve the calculation efficiency,but also is good at leveling reflection events on CMP gather.

国家重点基础研究发展计划(973计划)项目(2013CB228604)、国家自然科学基金项目(41404088)和中国石油大学(华东)研究生创新工程项目(YCX2015006)联合资助。