论文部分内容阅读
一般情况下,当速度滤波器用于叠前地震数据时,它们能够压制长周期多次波,但它们也损害一次波振幅。由于在数据道集的近炮检距处一次波和多次波的视速度不同可以忽略,速度滤波器在这些区域无效,因为它易于去除和畸变一次波信号。这个问题的常规解决方法是切除或去掉近炮检距数据,但这种方法对低覆盖资料不利。本文中,我们提出了一种解决这个问题的滤波方法,它响应于一次波和多次波之间的平均视速度差而不是瞬时视速度差。这种滤波器称为局部相干滤波器,本质上是一个步长等于空间预测距离的有限差分算子。在局部相干滤波器的小应用时窗内,经过NMO校正的多次波可被预测,而过校正的一次波不可被预测。在道集中速度滤波器失效的区域,此滤波器的差分算子的步长值使得它能够识别一次波和长周期多次波。本文提出了局部相干滤波器,并且在模型和实际资料应用中将它与f-k滤波器进行了对比。结果表明局部相干滤波器没有畸变一次波反射,并且更有效地压制了长周期多次波。
In general, when velocity filters are used for prestack seismic data, they are able to suppress long-period multiples, but they also impair the amplitude of the first wave. Since the apparent velocities of primary and multiple waves are negligible at near offset of the data gathers, the speed filter is not effective in these areas because it is easy to remove and distort the primary signals. A common solution to this problem is to remove or remove near-offset data, but this approach is detrimental to low coverage data. In this paper, we propose a filtering method to solve this problem, which responds to the average apparent velocity difference between primary and multiple waves rather than the instantaneous apparent velocity difference. This filter, called the local coherence filter, is essentially a finite difference operator with a step size equal to the spatial prediction distance. In the small application window of the local coherence filter, NMO-corrected multiples can be predicted, while overcorrected primary waves can not be predicted. In the area where the central speed filter fails, the step size of this filter’s differential operator enables it to identify both primary and long period multiples. In this paper, a local coherent filter is proposed and compared with the f-k filter in the model and actual data applications. The results show that the local coherence filter does not distort the primary reflection and suppress the long period multiples more effectively.