Parallel Marchenko Imaging Method for Low-Frequency Reconstruction

Abstract

The invention relates to a low-frequency reconstruction parallel Marchenko imaging method, and the method enables the technology of passive low-frequency information reconstruction active source low-frequency information based on the frequency advantages to be applied in Marchenko imaging, and can prevent the passive source low-frequency noise from affecting the reconstructed broadband seismic data to maximum extent. The method achieves the recovery of broadband seismic data, and also keeps the high-frequency part of the active source data, so the method accords more with the reality than a method of manually controlling and adjusting the weight of a low-frequency signal in broadband seismic data, saves a lot of time, and remarkably improves the calculation efficiency of Marchenko imaging. Finally, a reflection interface is more focused in the Marchenko imaging of the broadband seismic data. The method employs the stronger high-speed shielding layer penetration capability of low-frequency information and a Marchenko imaging algorithm to eliminate multiple waves between layers, and obtains the high-precision image of a subsurface structure.

Description

technical field [0001] The present invention relates to an underground imaging method for seismic exploration, especially high-precision Marchenko imaging for complex high-speed shielded underground structures by seismic interferometry passive source low-frequency reconstruction seismic beyond interferometry active source data, sub-shot parallel direct wave estimation and data drive . Background technique: [0002] In the process of seismic imaging inversion, low-frequency information is particularly important for structural imaging of subsurface media, and it directly affects the inversion effect of fine structure of subsurface media. Low-frequency signals have a strong ability to penetrate high-speed shielding layers, and low-frequency signals can be used to reduce the scattering and shielding effect of high-speed shielding layers on seismic signals and improve the signal-to-noise ratio of deep-layer signals. In practical applications, due to the inherent low-frequency ad...

AI Technical Summary

Problems solved by technology

Such low-frequency reconstruction methods may produce false information on seismic records in practical applications, and ultimately affect imaging results

The other is to perform seismic interferometry processing on the passive source data, and then perform low-frequency reconstruction on the active source: Zhang matched the active source and passive source data from the perspective of energy (amplitude) in 2015, but even the high-signal Noise Ratio Noise still exists in low-frequency data, and this limitation will also affect the final imaging results

In addition, this passive source low-frequency supplement active source data method is based on artificial gain coefficients, which is highly subjective and cannot be based on data-driven

[0004] The Marchenko imaging algorithm relies on the Green's function of each point in the target imaging area, and a small grid size will cause a huge amount of calculation

The traditional single-shot serial calculation is far from meeting the requirements of calculation efficiency

Images