Method for suppressing high-power-frequency noise in parallel seismic source seismic prospecting data based on principal component analysis

Abstract

The invention discloses a method for suppressing high-power-frequency noise in parallel seismic source seismic prospecting data based on principal component analysis. Parallel seismic source seismic prospecting refers that a plurality of seismic sources are arranged at different shot points in a prospecting area, data is acquired by adopting a nearly simultaneous excitation technology, and the seismic prospecting production efficiency and the seismic data quality can be improved. Original seismic record acquired by parallel seismic source excitation in a human activity area are usually mixed with high-power-frequency noise, a Hankel matrix is built according to a phase space theory for an interference channel, and principal component decomposition and useful signal recovery are performed on the Hankel matrix. It is proved that the method can realize effective high-power-frequency noise suppression for the seismic prospecting data obtained by parallel seismic source acquisition. Compared with a traditional high-power-frequency noise suppression method, the method disclosed by the invention is fast in data processing, does not need to accurately estimate the frequency of the high-power-frequency noise, is free of phase deviation and good in signal protection, can effectively improve the seismic data quality of parallel seismic source prospecting and reduces the parallel seismic source seismic prospecting data processing cost.

Description

Technical field: [0001] The invention relates to a method for suppressing strong power frequency noise of parallel seismic source seismic exploration data based on principal component analysis. The method is used for suppressing strong power frequency noise during parallel seismic source seismic exploration. Background technique: [0002] Parallel source seismic exploration is often affected by strong power frequency noise during the acquisition process. The acquired parallel seismic data is often mixed with strong power frequency noise. Therefore, the quality of the obtained seismic data is not high, which affects the later seismic data interpretation and migration imaging. quality. However, there are few literatures on the suppression method of strong power frequency noise in parallel source seismic exploration data. At present, the methods for suppressing strong power frequency noise in other fields mainly include frequency domain method and time domain method. Frequency ...

AI Technical Summary

Problems solved by technology

[0002] Parallel source seismic exploration is often affected by strong power frequency noise during the acquisition process. The acquired parallel seismic data is often mixed with strong power frequency noise. Therefore, the quality of the obtained seismic data is not high, which affects the later seismic data interpretation and migration imaging. quality

However, there are few literatures on the suppression method of strong power frequency noise in parallel source seismic exploration data. At present, the methods for suppressing strong power frequency noise in other fields mainly include frequency domain method and time domain method. Frequency domain methods such as notch method, The wavelet filtering method and so on all perform power frequency suppression in the frequency domain. Although this type of method is simple and convenient, when the frequency of the useful signal and the frequency of the power frequency overlap, this method will cause a certain degree of damage to the useful signal, and at the same time There will also be a certain phase shift

Time-domain methods such as sine-cosine approximation method, power frequency regression subtraction method, adaptive filtering method, etc. express the power frequency noise as a function with amplitude, frequency, and phase as variables, and then estimate the strong power frequency by the sine-cosine function approximation method. noise, but this type of method needs to accurately estimate the frequency of strong power frequency noise. At the same time, this method needs to be processed multiple times when multiple power frequency harmonics exist, so the algorithm efficiency is low

It can be seen that the above two types of methods are not suitable for the suppression of strong power frequency noise in parallel source seismic exploration data.

Images