论文部分内容阅读
本文将以前2D叠前弹性波偏移和3D叠前声波偏移结合起来,并加以推广后,提出了一种全3D弹性叠前深度偏移算法。首先用3D弹性有限差分法外推记录数据,然后用3D声波有限差分法计算3D数据体的各个点的成象时间。算法是在三分量矢量波场共炮点数据集上运行的结果得到了在三分量矢量反射系数的分布。用一次反射P波可实现P-S转换反射波的自动成象。由于使用了全波波动方程,该方法没有倾角限制。算法已应用于由三个模型产生的合成数据,即水平反射层,截断楔状倾斜层上覆有小平反射层,以及经典的French双丘及断层模型。
In this paper, we combine the 2D prestack elastic wave migration and the 3D prestack acoustic migration, and then generalize them. Then we propose a full 3D elastic prestack depth migration algorithm. First, the 3D elastic finite difference method is used to extrapolate the recorded data, and then the 3D acoustic wave finite difference method is used to calculate the imaging time of each point of the 3D data volume. The result of the algorithm running on a common shot point data set of a three-component vector wave field results in the distribution of the three-component vector reflection coefficient. The P-S converted reflected wave can be automatically imaged with a single reflection of the P wave. Due to the use of a full-wave wave equation, this method has no dip restrictions. The algorithm has been applied to the synthetic data produced by the three models, namely the horizontal reflection layer, the truncated wedge-shaped inclined layer overlaid with a small flat reflection layer, and the classic French double-hummer and fault model.