A Quantitative Frequency Raising Method

Abstract

The invention provides a quantitative frequency extraction method which belongs to the field of seismic data signal processing. The quantitative frequency extraction method comprises the steps of (1) constructing an analytic function Z1(t) by means of Hilbert transform f(t)<^> between an original signal gather f(t) and an original signal gather f(t); (2) in a frequency extraction range (fmin,fmax) Hz, supposing the increased frequency of a fixed quantity is deltaf, and supposing the other analytic function which is constructed by means of a trigonometric function is Z2(t); (3) obtaining a new analytic function Z(t) from the analytic functions Z1(t) and Z2(t); (4) taking the real part of the new analytic function Z(t) as a signal output after frequency increase; and (5) in the preset frequency range (fmin,fmax)Hz, wherein the deltaf starts from a minimal frequency and iteratively circulates with a preset step, outputting and accumulating the signals with different frequencies, and finishing a quantitative frequency extraction process until reaching a maximal frequency.

Description

technical field [0001] The invention belongs to the field of seismic data signal processing, and in particular relates to a quantitative frequency raising method. Background technique [0002] Spectrum analysis is an important means in modern signal processing and analysis. It expands the signal strength sent by the signal source in order of frequency, making it a function of frequency, and investigates the change law, which is called spectrum analysis. It transforms the time-domain signal In the frequency domain, the purpose is to decompose the complex time history waveform into several single harmonic components through Fourier transform for research, so as to obtain the frequency structure of the signal and the information of each harmonic and phase. If the signal frequency is too low, the resolution of the seismic data will be low, and the signal frequency can be increased by certain digital signal processing means, so that the main frequency of the signal becomes higher...

AI Technical Summary

Problems solved by technology

The spectral whitening method mainly has the following problems: After frequency division, the data of each frequency band can not be satisfied with the time gain. The relative amplitude and waveform relationship before spectral whitening can be maintained in the time direction; when the data gains of each frequency band after frequency division are superimposed, the amplitude of each frequency cannot be determined proportion of components;

However, the accuracy of the inverse Q filter is closely related to the correctness of the quality factor Q value. Due to the complex underground structure, there are many factors that affect the attenuation of seismic waves, and the Q value is often difficult to obtain accurately, resulting in inaccurate inverse Q filter.

Images