当前位置: X-MOL 学术Near Surf. Geophys. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A numerical dispersion‐suppressed method for shallow seismic migration
Near Surface Geophysics ( IF 1.6 ) Pub Date : 2020-11-03 , DOI: 10.1002/nsg.12134
Yanli Liu 1 , Zhenchun Li 1 , Jiao Wang 2 , Miaomiao Sun 1 , Qiang Liu 1
Affiliation  

Reverse time migration can accurately image underground earth structures. However, for shallow seismic exploration, the seismic wave velocity is often lower than the velocity in the middle‐deep layers, which causes numerical dispersion for finite‐difference schemes and leads to poor seismic imaging quality. Suppressing numerical dispersion by grid encryption or increasing the finite‐difference order seriously reduces computational efficiency, which is not the optimal solution. To improve the imaging quality without sacrificing computational efficiency, a regularization factor is added to the acoustic wave equation to correct the phase velocity of high wave‐number components. An appropriate regularization factor can eliminate numerical dispersion that results from large grid interval schemes and can reduce the size of computational grids and improve computational efficiency. For simulations in the frequency domain, reducing the grid size also means reducing computer memory requirements. Numerical experiments indicate that the regularization factor should match the degree of numerical dispersion. Larger regularization factors can suppress serious numerical dispersion. However, excessively large regularization factors may destroy the effective wave field. Different numerical models verify the effectiveness of the improved acoustic equation for suppressing numerical dispersion and maintaining amplitudes and provide a novel means to improve the shallow seismic image quality.

中文翻译:

浅层地震迁移的数值弥散抑制方法

反向时间偏移可以准确地成像地下地球结构。但是,对于浅层地震勘探,地震波速度通常低于中深层的速度,这会导致有限差分方案的数值分散,并导致较差的地震成像质量。通过网格加密抑制数值离散或增加有限差分阶数会严重降低计算效率,这不是最佳解决方案。为了在不牺牲计算效率的情况下提高成像质量,将正则化因子添加到声波方程中以校正高波数分量的相速度。适当的正则化因子可以消除由于大网格间隔方案导致的数值离散,并可以减小计算网格的大小并提高计算效率。对于频域中的仿真,减小网格大小也意味着减少计算机内存需求。数值实验表明,正则化因子应与数值离散度相匹配。较大的正则化因子可以抑制严重的数值离散。但是,过大的正则化因子可能会破坏有效波场。不同的数值模型验证了改进的声学方程在抑制数值离散和保持振幅方面的有效性,并为提高浅层地震图像质量提供了一种新颖的手段。对于频域中的仿真,减小网格大小也意味着减少计算机内存需求。数值实验表明,正则化因子应与数值离散度相匹配。较大的正则化因子可以抑制严重的数值离散。但是,过大的正则化因子可能会破坏有效波场。不同的数值模型验证了改进的声学方程在抑制数值离散和保持振幅方面的有效性,并为提高浅层地震图像质量提供了一种新颖的手段。对于频域中的仿真,减小网格大小也意味着减少计算机内存需求。数值实验表明,正则化因子应与数值离散度相匹配。较大的正则化因子可以抑制严重的数值离散。但是,过大的正则化因子可能会破坏有效波场。不同的数值模型验证了改进的声学方程在抑制数值离散和保持振幅方面的有效性,并为提高浅层地震图像质量提供了一种新颖的手段。较大的正则化因子可以抑制严重的数值离散。但是,过大的正则化因子可能会破坏有效波场。不同的数值模型验证了改进的声学方程在抑制数值离散和保持振幅方面的有效性,并为提高浅层地震图像质量提供了一种新颖的手段。较大的正则化因子可以抑制严重的数值离散。但是,过大的正则化因子可能会破坏有效波场。不同的数值模型验证了改进的声学方程在抑制数值离散和保持振幅方面的有效性,并为提高浅层地震图像质量提供了一种新颖的手段。
更新日期:2020-11-03
down
wechat
bug