Seismic data reconstruction method based on matrix reduced rank

Abstract

The invention provides a seismic data reconstruction method based on a matrix reduced rank, and the method comprises the steps: obtaining a quadruple Toeplitz matrix built through the seismic frequency slicing data, and storing a part of rows and columns of elements of the quadruple Toeplitz matrix, so as to express the quadruple Toeplitz matrix, wherein the part of rows and columns of elements of the quadruple Toeplitz matrix comprises all different elements of the quadruple Toeplitz matrix; carrying out the reduced-rank processing of the quadruple Toeplitz matrix expressed by the part of rows and columns of elements of the quadruple Toeplitz matrix through employing a random QR reduced-rank decomposition algorithm, so as to carry out the reduced-rank processing of the quadruple Toeplitz matrix; solving the mean value of the diagonal elements of the quadruple Toeplitz matrix through a no-expansion averaging algorithm after reduced-rank processing, and obtaining the reconstruction data of the seismic frequency slicing data; carrying out the Fourier inversion of the reconstruction data of the seismic frequency slicing data, and obtaining the time domain seismic reconstruction data which is used for oil gas exploration. The method can reduce the data storage size and calculation burden, and does not reduce the reconstruction quality.

Description

Technical field [0001] The invention relates to the technical field of oil and gas exploration and development, in particular to a method for reconstruction of seismic data based on matrix reduction. Background technique [0002] With the continuous improvement of my country's oil and gas exploration and development, the objects of seismic exploration are gradually shifting from structural oil and gas reservoirs to lithological oil and gas reservoirs, and seismic exploration targets have become increasingly complex, which places higher requirements on the quality of seismic data processing. However, in the process of seismic data acquisition and processing, irregular missing track seismic data is often encountered. Its existence will adversely affect the correct operation of the subsequent multi-channel processing technology of seismic data and reduce the processing quality of seismic data. [0003] The causes of irregular seismic data are mainly derived from two aspects: First, i...

AI Technical Summary

Problems solved by technology

However, for high-dimensional seismic data reconstruction, especially for pre-stack 5D seismic data reconstruction, it is necessary to construct a large quadruple-block Hankel matrix or quadruple-block Toeplitz matrix. Decreasing rank decomposition of large quadruple-block Hankel matrix or quadruple-block Toeplitz matrix

Gao et al. (Afast reduced-rank interpolation method for prestack seismic volumes that depend on four spatial dimensions, Geophysics, 2013, 78(1):V21–V30) proposed to use Lanczos bidiagonal decomposition instead of random SVD decomposition to realize the quadruple block Hankel Or the rank reduction of the quadruple block Toeplitz matrix, although the Lanczos bidiagonal decomposition method can improve the rank reduction calculation efficiency, but it is still necessary to generate the large low The rank quadruplex Toeplitz matrix leads to a large amount of data storage, and the reconstruction efficiency is still very low

[0007] Therefore, 5D seismic data reconstruction is not only faced with the problem of large amount of calculation, but also faces the challenge of large amount of data storage

Images