摘要
基于双程波方程的叠前逆时偏移法是成像精度较高的偏移方法，但因受存储量大、计算量大等影响，工业化进程缓慢，其中低频噪声对构造解释精度影响很大。因此，采用高阶有限差分算法计算逆时偏移，着重研究了拉普拉斯算子滤波压制低频噪声方法。Marmousi模型的逆时偏移去噪测试结果表明，拉普拉斯算子去噪法在压制低频噪声方面比常规的高通滤波法更有效，既能保持有效信号的振幅、相位特性，又能明显改善剖面的成像效果。
Although prestack reverse time migration （ RTM ） based on two-way wave equation is one kind of migra- ting method with much higher imaging pecision, the large amount of storage, large calculation and so on result in slowly industrialized process, and furthermore the low frequency noise also has much serious influences over the in- terpretation precision of the structures. Therefore the high-order finite difference algorithm is adopted to calculate the reverse time migration in the paper, and low-frequency noise suppressing approach by Laplace operator noise-e- liminating filter is mainly focused. Comparing with the conventional high-pass filter approach, the RTM denoising test on Marmousi model shows that Laplace operator denoising approach can effectively suppress the low-frequency noise： not only preserve the properties of amplitudes and phases of those effective signals, but also obviously im- prove the imaging effects of the profiles.
出处
《大庆石油地质与开发》
CAS
CSCD
北大核心
2012年第5期154-157,共4页
Petroleum Geology & Oilfield Development in Daqing
基金
中国石油科技创新基金项目“碳酸盐岩储层地震叠前逆时成像方法研究”（2011D-5006-0304）、国家自然科学青年基金项目“基于波动方程三维表面多次波压制方法研究”（41104088）.
关键词
逆时偏移
拉普拉斯算子
去噪
有限差分
reverse time migration
Laplace operator
denoising
finite difference
