基于反演思想的最小二乘类偏移方法,具有较常规方法更好的成像保幅性和振幅均衡性,逐渐成为研究的热点.本文将单平方根分步傅里叶反偏移算子和偏移算子引入最小二乘框架,实现了基于单平方根算子的叠前最小二乘分步傅里叶偏移(SSR-LSSSF)算法.同时,为了抑制离散傅里叶变换引起的周期性伪波场,在偏移算法中引入了余弦衰减边界.在实现优化算法的基础上,通过对复杂盐丘模型的成像试处理及与其他方法成像对比可知:(1)本文发展的SSR-LSSSF偏移方法具有较好的迭代收敛特性,成像具有较好的保幅性,成像结果逐渐逼近真实反射系数.(2)通过与RTM、LSRTM及传统SSF偏移方法成像结果对比,表明本文方法在成像精度和偏移噪声压制方面具有一定的优势.
The inversion-based least-squares migration method has become a hot research area in the geophysics field. This is due largely to its better performance on preserving amplitude and balancing amplitude,comparing to conventional migration. In the paper,we introduce split-step Fourier demigration operator and migration operator into the framework of least-squares, and implement prestack least-squares split-step Fourier migration based on single square root propagator( SSR-LSSSF). Besides,in order to suppress periodic artificial-wavefield caused by the discrete Fourier transform( DFT),we adopt cosine damping boundary in our algorithm. Finally, we conduct imaging test on complex SALT model,and have some comparisons with other imaging methods.Thus,we draw conclusions as following:( 1) the SSR-LSSSF algorithm presented in the paper shows a better convergence during iteration,a higher amplitude preservation and gradually closes to real reflection coefficient in imaging results.( 2) By contrast with the RTM,LSRTM and traditional SSF migration,the SSR-LSSSF algorithm has some advantages in imaging precision and migration noise suppression.