局部角度域高斯束偏移参数优选研究

Parameters optimization in local angle-domain Gaussian beam migration

高斯束偏移是一种介于Kirchhoff偏移和波动方程偏移之间的替代方法。基于高斯束叠加积分表征的格林函数,在成像条件中引入偏移张角信息,推导了局部角度域高斯束叠前深度偏移的理论公式,并阐述了其原理和方法。在此基础上,定性、定量地分析了初始束宽、束中心间隔、初始射线参数间隔以及偏移张角等参数对高斯束偏移成像质量和计算效率的影响,根据各个参数的作用及特点,给出参数优化的选取准则。加拿大山麓模型、Marmousi模型、盐丘模型以及BP模型的数值应用试验表明:参数优化选取不仅可以改善偏移成像质量(包括可以压制折射波等非反射波互相关成像引起的低频噪声和对棱柱波等有效能量进行成像),而且能够极大地提高计算效率。

Gaussian beam migration is an alternative to Kirchhoff migration or wave-equation migration. Based on Green＇s function by the Gaussian beam summation, we have derived the expression of local angle-domain Gaussian beam migration by incorporating migrated opening angle information into the imaging condition and then illustrated its principle and method. On this basis, we have researched the impact of parameters, such as initial beam width, beam center interval, initial ray parameter spacing and migrated opening angle, on the image quality and computational efficiency by means of qualitative and quantitative analysis. According to the characteristic of each parameter, we have developed an optimal selection criterion of parameters. At last, the numerical experiments on Canadian Foothills, Marmousi, Salt and BP models indicate that optimal selection of pa rameters not only improves image quality including the suppression of low-frequency artifacts caused by the cross-correlation imaging of refracted waves and the imaging of available energy such as prismatic wave, but also greatly enhances the compu- tational efficiency.