Application Research Of Synchronous Squeezing Wavelet Transform In High-resolution Processing Of Seismic Signals

Seismic signal is a kind of nonlinear non-stationary signal.We need to understand its frequency composition at a particular time and the time or amplitude distribution at a particular frequency.The time-frequency analysis method is a method suitable for describing the spectral characteristics within the local time range of a seismic signal.However,traditional time-frequency analysis methods are limited by the uncertainty principle and have a low resolution,so the are difficult to accurately describe the details in seismic signal.The empirical mode decomposition is not limited by the uncertainty principle and has a high time-frequency resolution.However,the empirical mode decomposition method is susceptible to mode mixing,leading to the result is inaccurate and the physical meaning is not clear.Synchrosqueezed wavelet transform is a similar method developed on the basis of empirical mode decomposition analysis method.It is a combination of wavelet transform and frequency rearrangement scheme,with strict mathematical foundation and clear physical meaning.In the paper,we starts with the principle of synchrosqueezed wavelet transform and analyzes the application of SWT in high resolution processing of seismic signals.First,we verify the high time-frequency resolution characteristics of the synchrosqueezed wavelet transform by using time-frequency analysis of the simulated signal.In the research of harmonic extraction of signals,it was found that the SWT algorithm can effectively reduce the extraction error caused by mode mixing,and the extraction effect is better than the empirical mode decomposition.At the same time,we studied the effect of the mother wavelet on the results of time-frequency analysis,the results show that the different choices of mother wavelet have limited impact on the synchrosqueezed wavelet transform.In terms of denoising,we compares several traditional denoising methods with denoising methods based on synchrosqueezed wavelet transforms.It is found that the denoising method based on SWT is better than the traditional denoising method,and the proper mother wavelet can further enhance the denoising effect.In the seismic prospecting for oil and gas,some oil-gas target layers are adjacent to the coal seam and are affected by the strong amplitude of the coal seam,which makes it difficult to identify the response of the reservoir in the seismic section and the calculation of geophysical parameters is inaccurate.For this reason,a method for suppressing strong amplitude in coal seam based on synchrosqueezed wavelettransform is designed in this paper..Through the processing of model data and actual seismic data in the Sulige area,the results show that this method not only can effectively suppress the strong amplitude of the coal seam,but also has better precision and accuracy than the method based on empirical mode decomposition.In order to overcome the shortcomings of traditional spectral whitening methods in improving seismic resolution,a spectral whitening method based on SWT was proposed in this paper proposes.The processing of actual seismic signals shows that this method can effectively enhance the local details of seismic signals and improve the time-frequency resolution.Finally,Matlab graphical user interface is used to implement the SWT-based seismic signal processing software.The software has the functions of time-frequency analysis,reconstruction,harmonic extraction,denoising,strong amplitude suppression and weak signal enhancement of seismic signals.