唐欢欢^1,2,,
毛伟建^1,2,,,
詹毅³
1. 中国科学院精密测量科学与技术创新研究院计算与勘探地球物理研究中心, 武汉 430077
2. 大地测量与地球动力学国家重点实验室, 武汉 430077
3. 中国石油集团公司东方地球物理公司物探技术研究中心, 涿州 072751

基金项目: 国家重点研发计划（2016YFC0601105），国家自然科学基金（U1562216）和中国石油集团公司"深层与非常规物探新方法新技术"项目（2019A-33）联合资助


详细信息

作者简介: 唐欢欢, 女, 1987年生, 中国石油大学(北京)地球探测与信息技术专业硕士毕业, 主要从事地震数据重建及地震数据处理相关工作.E-mail:tanghuan@whigg.ac.cn

通讯作者: 毛伟建, 男, 研究员, 博士生导师, 主要从事地震数据处理、成像和反演研究.E-mail:wjmao@whigg.ac.cn

中图分类号: P631

收稿日期:2019-02-13
修回日期:2020-05-19
上线日期:2020-09-05

Reconstruction of 3D irregular seismic data with amplitude preserved by high-order parabolic Radon transform

TANG HuanHuan^1,2,,
MAO WeiJian^1,2,,,
ZHAN Yi³
1. Center for Computational and Exploration Geophysics, Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences, Wuhan 430077, China
2. State Key Laboratory of Geodesy and Earth's Dynamics, Wuhan 430077, China
3. Research and Development Center, BGP, CNPC, Zhuozhou 072751, China


More Information

Corresponding author: MAO WeiJian,E-mail:wjmao@whigg.ac.cn

MSC: P631

--> Received Date: 13 February 2019
Revised Date: 19 May 2020
Available Online: 05 September 2020


摘要
摘要:3D地震数据不规则采样缺失重建是地震勘探数据处理流程中的重要问题.本文提出了一种基于具有保幅特性的非均匀高阶抛物Radon变换（NHOPRT）地震数据重建方法.在最小二乘反演方程中引入Delaunay三角网格剖分来计算空间不规则加权系数，从而获得最接近完整规则数据的高阶抛物Radon变换域系数.在用SVD求解反演方程过程中，利用高阶抛物Radon变换算子在频率域为指数函数，具有线性可分解特性，将二维空间的高阶抛物Radon变换算子分解为两个独立的一维空间变换算子，减小了变换算子的矩阵大小，从而很大程度地提高了计算效率.理论模型和实际地震数据重建测试证明了本文方法的有效性以及实用性.
关键词: 不规则地震数据/
3D高阶Radon变换/
指数函数分解/
保幅重建

Abstract:The reconstruction of irregular 3D seismic data is an important step in seismic data processing. This paper proposes a reconstruction method based on Non-uniform High Order Parabolic Radon Transform (NHOPRT) with amplitude preserved. The Delaunay triangulation is used to calculate weighting coefficients related to the irregular grid, which are then introduced into the least square inversion equation to acquire the NHOPRT domain coefficients which are mostly closest to the complete regular data. As the 2D space NHOPRT operator is large and involves matrix multiplication and inversion, it consumes considerable time. By using the property of linear decomposition of the operator in frequency, we decompose the 2D space NHOPRT operator into two independent 1D operators, thus reducing the size of the 2D space NHOPRT operator and greatly improve the computational efficiency. Tests of the synthetic model and field data demonstrate the effectiveness and usefulness of the method.
Key words:Irregular seismic data/
3D high-order Radon transform/
Exponential decomposition/
Amplitude preserved reconstruction



PDF全文下载地址:

http://www.geophy.cn/data/article/export-pdf?id=dqwlxb_15584