论文部分内容阅读
过去,在海上处理中,惯于在动校正后,用加权叠加增强一次波与多次波之比(P/M)。最近,为了同样的目的,即增大P/M比而在动校正以前进行叠加。这一目的部分是在野外用扩大震源组合的办法,部分是在处理中通过相邻道加权求和的办法达到的。本文试图估计这些方法各自的优点和提供某些最佳化参数。把多次波相对一次波延迟一个随偏移距而增大的时间(△MNO)。这些ΔMNO确定一个与地震脉冲褶积的尖脉冲序列,然后把各尖脉冲加权。处理的目的则是通过改变权系数值使其产生的输出为最小。使P/M最佳化是对一个频率进行的,并忽略了拉伸效应。然而,对一给定的脉冲波形,最佳权系数值可以同包括拉伸效应的脉冲褶积,并观察总的效果。动校正前形成组合也已用来压制多次波。该方法的基础是长组合具有一定方向性,并压制从其它方向到达的信号。因为多次波以不同于一次波的入射角到达组合(或从震源组合射出),所以鉴别一次波和多次波是可能的。最佳角以外的全部其它噪声同样受到压制。对用这种方法进行鉴别作了研究,并与加权共深度点叠加所得到的分辩力作了比较。也分析了它们的联合效果。变化的倾角使形成组合的作用复杂化,并且可以破坏所期望的效果,致使多次波以相对组合基线为最佳的方向到达,而一次波反而不在这个方向,于是减小了P/M比。对一给定的几何形态模拟了这种效果。在动校正前以及在共深度点叠加前形成组合都可以用来压制多次波。为了从动校正前组合获得效益。加权叠加是必不可少的。权系数值的选择应仔细考虑。在某些倾斜层的情况下,叠加前组合对P/M比将起负作用,但这种负作用也许并不严重,而且可以用适当的记录方法加以避免。动校正前适当的组合和加权共深度点叠加的配合使用将产生比单独用其中任一种方法都好的P/M比,并改善资料的总信噪比。
In the past, in maritime processing, it was used to enhance the primary / multiplicative ratio (P / M) with weighted superposition after dynamic correction. Recently, for the same purpose, namely to increase the P / M ratio, the superposition is performed before the dynamic correction. This objective is partly achieved by expanding the combination of hypocenters in the field and partly by weighted summation of adjacent roads during processing. This article attempts to assess the merits of each of these methods and provide some optimization parameters. Delay multiples relative to primary by a time that increases with offset (Δ MNO). These ΔMNO determine a spike sequence that is convoluted with the seismic pulse and then weight each spike. The purpose of processing is to minimize the output produced by changing the weight coefficient value. Optimizing P / M is done on one frequency and ignoring the stretching effect. However, for a given pulse waveform, the optimal weight coefficient may be convoluted with pulses that include the stretching effect and observe the overall effect. The formation of a combination before dynamic correction has also been used to suppress multiple waves. The basis of this approach is that long combinations have a certain directivity and suppress signals arriving from other directions. Because multiple waves arrive at the combination (or are emitted from a combination of sources) at angles of incidence that differ from the primary waves, it is possible to identify primary and multiple waves. All other noises outside of the best corners are also suppressed. This method of identification was studied and compared with the weighted total depth of points obtained by the superposition of the separation made a comparison. Also analyzed their joint effect. The varying dip complicates the effect of forming a combination and can disrupt the desired effect so that the multiple arrives in the direction that is optimal relative to the combined baseline, whereas the primary is not in this direction, reducing the P / M ratio . This effect is modeled for a given geometry. It is used to suppress multiple waves before moving and before forming a total depth. In order to benefit from the pre-correction combination. Weighted superposition is essential. The choice of weight coefficient should be carefully considered. In the case of some inclined layers, the pre-stack combination will have a negative effect on the P / M ratio, but this negative effect may not be severe and may be avoided by suitable recording methods. The combination of a suitable combination of pre-calibration and weighted total depth stacking will produce a P / M ratio better than either method alone and improve the overall signal-to-noise ratio of the data.