论文部分内容阅读
叠前逆时偏移是目前精度较高的成像方法,然而严重的低频噪声降低了逆时偏移的构造成像精度,偏移噪声的压制是逆时偏移必需要考虑的问题。分析了低频噪声的产生机理,并根据声波方程Poynting矢量的方向指示地震波场的传播方向的原理,分离出上、下、左、右行波,该方法计算量和存储量都远远小于常用的二维傅里叶变换分离方法,进而提出归一化的波场分离互相关成像条件,以压制逆时偏移低频噪声,提高成像精度。实现了Marmousi模型的试算,表明在波场延拓过程中利用Poynting矢量能够较好的分离上、下、左、右行波,与常规方法、拉普拉斯滤波、二维傅里叶变换波场分离的成像结果对比表明,成像时使用归一化的波场分离互相关成像条件能更好的压制偏移噪声,得到精度更高的逆时偏移成像结果。
Prestack inverse time migration is the most accurate imaging method at present. However, severe low frequency noise reduces the imaging accuracy of reverse time migration. Suppression of offset noise is an issue that must be considered in reverse time migration. The generation mechanism of low frequency noise is analyzed. According to the principle of Poynting vector of sound wave equation indicating the propagation direction of seismic wavefield, the up, down, left and right traveling waves are separated. The calculation and storage of this method are far less than the commonly used Dimensional Fourier transform separation method, and then put forward normalized wavefield separation and cross-correlation imaging conditions to suppress reverse-time low frequency noise and improve the imaging accuracy. The trial of the Marmousi model is realized, which shows that the Poynting vector can be used to separate the up, down, left and right traveling waves well, and the conventional method, Laplacian filter, two-dimensional Fourier transform The comparison of wavefield imaging results shows that using the normalized wavefield cross-correlation imaging conditions can suppress the offset noise better and obtain more accurate inverse time migration imaging results.