首页 | 本学科首页   官方微博 | 高级检索  
     

基于混叠震源和频率组编码的频率域自适应全波形反演(英文)
引用本文:韩淼,韩立国,刘春成,陈宝书. 基于混叠震源和频率组编码的频率域自适应全波形反演(英文)[J]. 应用地球物理, 2013, 10(1): 41-52. DOI: 10.1007/s11770-013-0367-3
作者姓名:韩淼  韩立国  刘春成  陈宝书
作者单位:吉林大学地球探测科学与技术学院;中海油研究总院
基金项目:financially supported by the National Natural Science Foundation of China(No.41074075/D0409);the National Science and Technology Major Project(No.2011ZX05025-001-04)
摘    要:
全波形反演是一种高精度的地震成像方法,可以对地下介质物性参数模型进行准确的重构。然而在实际应用中,尤其是在三维复杂介质反演中,计算成本太大是该方法的一个重要缺陷。将混叠震源技术引入到频率域全波形反演中可以大幅度地降低计算成本,提高反演效率。但是使用震源编码技术也带来了两个问题:一方面,参与编码的各个震源之间会产生"串扰噪声",导致反演结果中出现假象;另一方面,基于震源编码的频率域全波形反演方法周围噪声较为敏感,使该方法对含噪数据反演质量较差。本文引入一种频率组编码方法来压制"串扰噪声",并基于震源编码技术提出一种频率域自适应全波形反演方法,通过一个与频率相关的自适应选择机制,将常规频率域全波形反演方法和基于震源编码的全波形反演方法联合起来,在保证反演质量的同时也最大程度地提高了反演效率。

关 键 词:全波形反演  混叠震源  频率组编码  自适应

Frequency-domain auto-adapting full waveform inversion with blended source and frequency-group encoding
, Han Miao , Han Li-Guo , Liu Chun-Cheng , Chen Bao-Shu. Frequency-domain auto-adapting full waveform inversion with blended source and frequency-group encoding[J]. Applied Geophysics, 2013, 10(1): 41-52. DOI: 10.1007/s11770-013-0367-3
Authors:, Han Miao , Han Li-Guo , Liu Chun-Cheng , Chen Bao-Shu
Affiliation:1. College of Geo-Exploration Science and Technology, Jilin University, Changchun 130026, China
2. CNOOC Research Institute, Beijing 100027, China
Abstract:
As a high quality seismic imaging method, full waveform inversion (FWI) can accurately reconstruct the physical parameter model for the subsurface medium. However, application of the FWI in seismic data processing is computationally expensive, especially for the three-dimension complex medium inversion. Introducing blended source technology into the frequency-domain FWI can greatly reduce the computational burden and improve the efficiency of the inversion. However, this method has two issues: first, crosstalk noise is caused by interference between the sources involved in the encoding, resulting in an inversion result with some artifacts; second, it is more sensitive to ambient noise compared to conventional FWI, therefore noisy data results in a poor inversion. This paper introduces a frequency-group encoding method to suppress crosstalk noise, and presents a frequency-domain auto-adapting FWI based on source-encoding technology. The conventional FWI method and source-encoding based FWI method are combined using an auto-adapting mechanism. This improvement can both guarantee the quality of the inversion result and maximize the inversion efficiency.
Keywords:Full waveform inversion  Frequency-domain  Blended source  Frequency-group encoding  Auto-adapting
本文献已被 CNKI 万方数据 SpringerLink 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号