论文部分内容阅读
在地震资料的处理和解释过程中,确定速度和深度非常重要。我们提出了一种从未叠加的资料中确定速度—深度模型的方法。这种方法可被表示成一个迭代算法,其产生的模型使沿射线追踪旅行时计算的相关值达到最大。在这个模型中各个界面都是用三次样条函数来表示的,并假设每一层的速度为常数。其反演包括确定各层的速度和样条函数结点的位置。反演是用迭代方法一层一层来实现的;在每次迭代过程中都要计算所研究界面的人工合成传播时间曲线。用人工合成旅行时间形成一个刻划波场主要相关特性的函数。并且假定,当人工合成旅行时曲线与实际同相轴的旅行时曲线吻合时这样函数有最大值。该函数的最大值可以用有效的非线性计算程序得到。该反演算法有不少优越性,如不需要在未叠加资料上进行同相轴拾取,也不要用旅行时双曲线逼近曲线拟合。把这种方法用于人工合成资料和野外资料都已取得成功。
In the process of seismic data processing and interpretation, to determine the speed and depth is very important. We propose a way to determine the velocity-depth model from data that has not been superimposed. This method can be expressed as an iterative algorithm that produces a model that maximizes the correlation values calculated along ray-traced trips. In this model, each interface is represented by a cubic spline function, and the speed of each layer is assumed to be constant. The inversion includes determining the velocity of each layer and the position of the spline function node. Inversion is done iteratively layer by layer; the synthetic propagation time curve of the studied interface is calculated during each iteration. Synthetic travel times form a function that characterizes the major pertinent characteristics of the wave field. It is also assumed that such a function has a maximum when the synthetic travel curve coincides with the travel time curve of the actual in-phase axis. The maximum value of this function can be obtained with a valid non-linear calculation program. The inversion algorithm has many advantages, such as not needing to pick up the events on the non-superimposed data and do not use hyperbola to approximate curve fitting when traveling. The use of this method for both synthetic and field data has been successful.