First arrival chromatography near-surface modeling effective model depth control method and system

Abstract

The invention discloses a first arrival chromatography near-surface modeling effective model depth control method and system. The method comprises the following steps: obtaining ray maximum depth andoptimization speed models of the first iteration and the second iteration; setting a threshold value, and carrying out the Nth iteration: 1, carrying out the preprocessing of the maximum depth of theray of the (N-2) th iteration and the maximum depth of the ray of the (N-1) th iteration; 2, calculating a depth difference; 3, calculating the effective model depth of the Nth iteration; 4, obtainingan optimization speed model; 5, whether N is equal to a threshold value or not is judged, if yes, a near-surface velocity model is obtained, and if not, the step 6 is executed; and 6, optimizing whether the speed model reaches a preset target or not, if so, obtaining a near-surface speed model, and if not, carrying out (N + 1) th iteration. According to the method, the effective model depth of the iteration is determined through the maximum ray depth of the latest secondary iteration in the first arrival chromatography iteration process, and the speed units outside the effective model depth are excluded, so that the forward modeling efficiency is improved.

Description

technical field [0001] The present invention relates to the field of data processing of seismic exploration data, and more specifically relates to a method and system for controlling the effective model depth of first-arrival tomographic near-surface modeling. Background technique [0002] The first-arrival tomographic near-surface modeling technique using the first-arrival time to invert the near-surface velocity model has been widely used in seismic data processing. First-arrival tomographic near-surface modeling technology is an iterative process, and each iteration includes two basic steps: forward modeling and inversion. The forward modeling process uses ray tracing technology to calculate the ray path and time experienced by the seismic waves that are excited at each shot point and propagate through a given initial velocity model and reach each receiving point; the inversion is based on the forward modeling time and the actual The difference in the first arrival time ...

AI Technical Summary

Problems solved by technology

[0003] The forward modeling process is the step that consumes the most CPU computing time in the first-arrival tomographic near-surface modeling technique. Although different ray tracing techniques are available, the CPU computing time consumed by ray tracing is equal to that used to calculate the ray path during the ray tracing process. The range of the velocity model used is related to the range of the velocity model used. The larger the range of the velocity model used, the more CPU computing time will be consumed. However, if the range of the velocity model used is too small, it will lead to wrong ray tracing results.

The existing technology is to fix a large velocity model range. Although this will not lead to wrong ray tracing results for each iteration, it will also consume more CPU computing time, because not every iteration Use that fixed larger velocity model range

Images