论文部分内容阅读
速度和深度估算是地震资料处理和解释中的一个重要步骤。本文介绍了一种由叠前资料估算速度—深度模型的方法。这一方法是由一种产生模型的迭代算法构成,产生的这种模型能使沿着由追踪穿过模型的射线产生的旅行时计算的某种相干性量度最大。在模型中,界面用三次样条函数表示,并且假设各层的速度为常数。反演包括求取所有层位的速度和样条结点的位置。处理的输入数据包括叠前地震资料和一个初始速度—深度模型。这个模型通常以附近的井下信息和叠加剖面的解释结果为依据。通过逐层迭代处理进行反演,在各次迭代过程中,都要计算所研究的界面的合成时距曲线。然后沿合成波至时间形成表征波场主相关特性的泛函。文中假设当合成波至时间曲线与野外道集同相轴的波至时间相一致时泛函达到最大。函数的最大值是用非线性程序设计的有效算法求得的。这种反演算法的优点是不需要在叠前资料上拾取同相轴,并且不是以波至时间的双曲线近似的曲线拟合为依据的。这种方法已成功地用于合成资料和野外资料。
Estimation of velocity and depth is an important step in seismic data processing and interpretation. This article describes a method for estimating velocity-depth models from prestack data. This method consists of an iterative algorithm that produces a model that produces a measure of some coherency that is calculated along the travel by tracking the rays that pass through the model. In the model, the interface is represented by a cubic spline function, and the velocity of each layer is assumed to be constant. Inversion includes finding the velocity of all the horizon and the position of the spline node. The input data processed includes prestack seismic data and an initial velocity-depth model. This model is usually based on the interpretation of nearby downhole information and overlay profiles. By iteratively iteratively processing the inversion, the composite time-distance curve of the studied interface is calculated during each iteration. Then, along the composite wave to the time to form the main characterization of the wave field functional correlation function. It is assumed in the paper that the functional reaches the maximum when the wave from the synthetic wave to the time curve coincides with the time to the phase of the epoch of the field gathers. The maximum value of the function is obtained using a valid algorithm of nonlinear programming. The advantage of this inversion algorithm is that it does not require the picking up of events on the prestack data and is not based on a curve-fitting of the hyperbola approximation of wave-to-time. This method has been successfully used to synthesize data and field data.