Fracture time-frequency representation based matching pursuit frequency division method

Abstract

The invention relates to a fracture time-frequency representation based matching pursuit frequency division method, which comprises the steps of S1, calculating a complex seismic trace by Hilbert transform on the basis of a fracture network medium model, S2, calculating the instantaneous envelope, the instantaneous phase and the instantaneous frequency of the complex seismic trace, S3, finding an envelope maximum value and a corresponding time position, and acquiring an estimated value of Morlet wavelet amplitude, S4, calculating the instantaneous phase and the instantaneous frequency at the corresponding time, and acquiring estimated values of a Morlet wavelet phase angle and the main frequency of fracture medium optimization matching, S5, subtracting a real part of the determined Morlet wavelet from the original seismic trace, and S6, repeating the steps S1-S5 until the left seismic trace amplitude is less than a given threshold. The beneficial effect lies in that the fracture time-frequency representation based matching pursuit frequency division method is good in adaptivity, high in noise resistance, and good in time and frequency resolution controllability.

Description

technical field [0001] The invention relates to a matching tracking frequency division method based on time-frequency representation of cracks. Background technique [0002] Seismic exploration has always been the main method to reduce oil and gas exploration risks and production costs. Since the 1980s, research on how to use seismic data to quantitatively detect fractured oil and gas reservoirs has gradually attracted the attention of major oil companies in the world. By the late 1990s, the academic and industrial circles basically reached a consensus that using seismic data (including longitudinal waves and shear waves) can effectively predict the direction and relative distribution of fractures. In terms of fracture detection methods, the effect of shear wave exploration should be better than that of longitudinal waves Good, but more difficult, such as strong destructive source, low signal-to-noise ratio, long record, difficult static correction, high cost, etc. Neverth...

AI Technical Summary

Problems solved by technology

Conventional spectral decomposition technology can no longer meet the frequency division needs of fractured reservoirs

The time-frequency resolution of the short-time Fourier transform (STFT) is determined by the length of the window function, and different window lengths will produce large differences, which will bring difficulties to the application

Wavelet transform (WT) has been widely used due to its multi-resolution characteristics, but it is a time-frequency distribution unfolded on the time-scale plane, and the relationship between scale and frequency is difficult to understand

Wavelet transform has better time resolution at high frequency and better frequency resolution at low frequency, but wavelet transform cannot obtain better frequency resolution in the frequency range from intermediate frequency to high frequency

Images