基于源层位的层间多次波压制技术及其应用

Technology and application of internal multiple suppression for seismic data based on source layer

低速煤层发育会形成能量较强的层间多次波,使得地震数据中的一次有效反射波信号被淹没,难以识别,导致地震成像精度较低,严重影响了后续的地震解释.针对上述的层间多次波压制问题,本文根据实际工区地震数据的能量、频率在横向上存在较大的差异特征,研发形成了基于源层位的地震资料层间多次波压制技术,并在准中地区进行了应用研究.模型合成数据及实际数据应用结果表明:本文提出的方法技术可有效压制模型数据与实际数据中的层间多次波,同时较好地保护一次反射波,验证了本文方法的正确性.而且,实际数据层间多次波压制处理后的地震数据剖面结构特征更加符合沉积规律,多次波能量、频率在横向上差异减小,为后续的烃源岩预测提供可靠的数据.

The development of low-speed coal seam will form internal multiples with strong energy,which makes primaries in seismic data submerged and difficult to identify,resulting in the low accuracy of seismic imaging,which seriously affects the subsequent seismic interpretation.In view of the above problem of internal multiple suppression,this paper develops and forms an internal multiple suppression technology based on source layer according to the characteristics of the energy and frequency of field data in the work area,and carries out application research in Huaizhong area.In response to the multiple characteristics in the research area,the Radon transform was first used in the early processing stage to suppress the multiples in the pre-stack data.However,due to significant differences in the energy and frequency of multiples in the lateral direction,it is difficult to completely suppress these internal multiples in pre-stack data.Therefore,there are still strong multiples on the post-stack profile,and appropriate multiple suppression methods need to be selected before suppression.The results of model synthesis data and filed data application show that the method and technology proposed in this paper.In field data processing,first of all,we analyze the source layer generated by internal multiples.Then,we extract the virtual events of the primaries generated by this source layer.Next,we construct the relevant virtual events and internal multiples with the primaries generated below the source layer.Fininally,we use adaptive matching subtraction for multiples suppression.The proposed approach can effectively suppress internal multiples between model data and field data,while also effectively protect primaries.And the correctness of the proposed method is verified.Moreover,the structural characteristics of the field data profile after internal multiples suppression are more in line with sedimentary laws,and the lateral differences in multiples energy and frequency are reduced,providing reliable data for subsequent hydrocarbon source rock prediction.