基于匹配追踪分解和tau-p域拾取的快速高斯束偏移方法  

Fast Gaussian beam migration method based on matching-pursuit decomposition and events-picking in tau-p domain

作　　者：王旭谦 杨继东[2] 黄建平[2] 李振春[2] 

机构地区：[1]中海油田服务股份有限公司物探事业部物探研究院,天津300451 [2]中国石油大学(华东)地球物理系,青岛266580

出　　处：《地球物理学进展》2017年第2期745-752,共8页Progress in Geophysics

基　　金：油气重大专项(2016ZX05002-005-07HZ;2016ZX05014-001-008HZ);国家重点基础研究发展计划(973)子课题(2014CB239006);国家自然基金(41274124)联合资助

摘　　要：交互式的速度分析是复杂构造探区速度模型建立的一种有效方法,为了使不同的速度模型在短时间内得到检测和标定,本文基于匹配追踪分解算法和tau-p域同相轴拾取技术,发展了一种快速叠前深度域高斯束偏移方法.根据tau-p域地震同相轴的振幅特征,本文采用匹配追踪算法对地震记录进行时间和空间上地稀疏分解,并将分解的时频原子应用于高斯束偏移,以达到压缩地震数据和提高偏移速率的目的.模型试算和实际资料处理结果表明:本文方法不仅能够对地下复杂构造和高陡反射面清晰成像,而且可使叠前地震数据压缩近数十倍,使深度域偏移的计算效率提高一个数量级以上.Interactive velocity analysis is an effective method to build the accurate velocity model for exploration areas with complex structures. In order to test and calibrate the different velocity models in a short time, we developed a fast method of pre-stack depth Gaussian beam migration based on the technique of matching-pursuit and event-picking in tau-p domain in this paper. According to the amplitude features of seismic events in tau-p domain, we decompose the seismic record in time and spatial using matching-pursuit algorithm and apply the decomposed time-frequency atoms to Gaussian beam migration to compress the seismic data volume and accelerate the migration processing. Numerical examples and the processing results of field data demonstrate that our method not only could image the subsurface complex structures and steep-dip reflectors clearly but also has a compression ratio of nearly dozens of times and an acceleration of one order of magnitude at least.

关 键 词：匹配追踪 tau-p域同相轴拾取 高斯束 快速偏移 

分 类 号：P631[天文地球—地质矿产勘探]