摘要
由于同时震源数据包含串扰很难直接成像,因此需要对混叠的炮记录进行分离,为了将混叠的波场进行分离,笔者建立了基于稀疏域反演的阈值迭代模型,将混合采集数据的共检波点道集转换到Radon域进行稀疏表示,利用阈值算子进行约束,通过迭代反演得到分离结果。实验算例表明通过较少的迭代即可获得信噪比较高的分离结果,验证了反演方法的有效性,并且该算法对随机噪声有很强的压制效果。
It is difficult to directly image the simultaneous source data that contains crosstalk,so the blended data should be separated into single shot data. In order to deblend the mixed wave field,a threshold iteration model based on sparse domain inversion is set up,which transformed common receiver gather data of blended acquisition into Radon domain for sparse representation,and constrained the model by threshold operator,then used iterative inversion to get separating results. The experimental examples show that the deblended results with high SNR can be obtained with less iterations,which verifies the effectiveness of the inversion method and its strong suppression on random noise.
作者
朱翔宇
毛伟建
杨庆节
唐欢欢
唐博文
ZHU Xiang-yu;MAO Wei-jian;YANG Qing-jie;TANG Huan-huan;TANG Bo-wen(Center for Computational & Exploration Geophysics,Institute of Geodesy and Geophysics,Chinese Academy of Sciences,Wuhan 430017,China;State Key Laboratory of Geodesy and Earth s Dynamic,Wuhan 430017,China;University of Chinese Academy of Sciences,College of Earth and Planetary Sciences, Beijing 100049, China;Research & Development Center,BCP,China National Petroleum Corporation,Zhuozhou 072751,Hebei,China)
出处
《世界地质》
CAS
2018年第3期920-929,共10页
World Geology
基金
中国石油集团公司项目(2016A--33)
国家自然科学基金重点项目(U1562216)联合资助
关键词
同时震源
RADON变换
阈值迭代
整形正则化
波场分离
反演
相干
simultaneous source
Radon transform
threshold iteration
shaping regularization
deblend
inversion
coherence
