2016年8月 SE
大庆石油地质与开发
Petroleum Geology and Oilfield Development in Daqing
DOI:10.3969/J.ISSN.1000-3754.2016.04.025
高陡复杂构造逆时偏移去噪方法及其应用
范飞'李振春'高赞2
Aug，2016 Vel. 35 Na. 4
（1.中国石油大学（华东）地球科学与技术学院，山东青岛266580；2.中国科学院渗流流体力学研究所，河北率坊065007）摘要：高陡复杂构造区域是油气勘探的重点和难点区域，逆时编移成像技术是实现高陡复杂构造成像精度最高的成像技术之一。有效去除偏移剖面中的成像噪声，保持地震振幅信息是实现复杂构造精细成像的关键。在分析逆时偏移噪声产生原理的基础上，从波数域与空间域的关系分析了噪声的特征，提出了高阶微分去噪方法。台阶模型和高陵复杂若丘模型应用结果表明：该方法能有效去除成像噪声，恢复高装复杂构造形态，保持地震振
幅和相位信息，可以应用于构造解释和储层预测中。关键词：高陡复杂构造；逆时偏移；成像噪声
中图分类号：P631
文献标识码：A
文章编号：1000-3754（2016）04-0126-06
DENOISINGMETHODOFTHEREVERSE-TIMEMIGRATIONFOR
HIHGANDSTEEPSTRUCTUREANDITSAPPLICATION
FAN Fei', LI Zhenchun', GAO Zan?
(1. School of the Geosciences, China University of Petroleum, Qingdao 266580, China ;
2.Institute of Porous Flow and Fluid Mechanics, Chinese Academy of Sciences, Langfang 065007, China) oaeeosoaaooaasee verse-time migration imaging technique is one of the imaging technologies with the highest imaging precision for the high and steep complex structure, Effectively removing the imaging noise in the migration sections and keeping the seismic amplitude information are the critical points to realize the imaging of the complex structure. Based on the analyses of reverse-time migration noise generating principles, and from the viewpoint of the relationships between the wave number domain and space domain, the characteristics of the noise are analyzed, and the high-order differ-ential denoising method is put forward. The bench model and high-steep complex rock model application results show that the method can effectively remove the imaging noise, remove the high-steep complicated structural forms, keep the amplitude and phase informations and finally can be adopted in the structural interpretation and reservoir prediction.
Key word: high steep complicated structure; reverse-time migration; imaging noise
复杂高陡构造是世界含油气盆地主要的勘探领域，也是中国实现陆上油气突破的重点领域之一。
收稿日期：2016-03-16
改回日期：2016-04-25
近年来，经过复杂山地地震勘探攻关，高陡构造地震勘探技术取得了较大进步。但是由于复杂的山前
作者简介：范飞，男，1991年生，在读硕士，从事地震波传播与偏移成像方法研究。
E-mail ;zhangyang_antonoil@ 163, com
万方数据