Gradient trace gather correlation weighting preprocessing method for least square reverse time migration

Abstract

The invention relates to a gradient trace gather correlation weighting preprocessing method for least square reverse time migration, which belongs to the field of seismic migration imaging, and comprises the following steps of: firstly, obtaining a gradient profile of each shot in the least square reverse time migration, respectively extracting gradient trace gathers at the same position, arranging the gradient trace gathers the according to the distances between traces and shots from small to large to obtain a gradient trace gather at each position, preliminarily cutting off a noise part of a distant trace in each gradient trace gather, correlating all gradient traces in the gradient trace gather with the reference trace one by one, selecting N traces with relatively large correlation values, and superposing to form a new gradient trace, and placing each new gradient track at a corresponding position to obtain a final gradient profile. The method can improve the precision of gradient preprocessing, reduce the interference of low-frequency noise in a gradient profile, improve the continuity of stratums on two sides of the model and the imaging precision of a deep stratum structure, reduce the number of iterations, and solve the problem of large calculated amount of least square reverse time migration.

Description

technical field [0001] The invention belongs to the field of seismic velocity modeling, and in particular relates to a gradient gather correlation weighted preprocessing method of least square reverse time migration. Background technique [0002] Least squares reverse time migration is a high-precision seismic velocity modeling algorithm. In each iteration process, firstly, based on the initial reflection coefficient model and background velocity, the two-way wave equation finite difference numerical simulation technology is used to perform linearization normalization. The simulated seismic record is derived, and then the difference between the simulated seismic record and the measured seismic record is put into L 2 The target functional is established in the norm, and the iterative inversion is performed based on the gradient method to minimize the target functional, so as to obtain the model modification amount and complete one iteration. The entire least squares reverse ...

AI Technical Summary

Problems solved by technology

[0004] When the conventional least squares reverse time migration algorithm calculates the gradient profile, the total gradient profile is directly superimposed by the gradient profile of each shot. In fact, when the distance between the gradient trace and the shot point is very far, the current shot will The gradient of the channel has almost no contribution, on the contrary, it will bring irregular interference, affect the quality of the gradient profile, and then increase the number of iterations and calculation time

At present, there is no good method to select effective gradient shots to improve the convergence efficiency of inversion

Images