Research On Fractional-order Synchronous Extraction S Transformation Algorithm And Its Application In Reservoir Identification

Fractional domain time-frequency analysis has obvious advantages over traditional methods,and is currently a popular method for analyzing and researching time-frequency domain.The traditional time-frequency wideband product limits the minimum time-frequency resolution.The use of a fractional Fourier transform can break through this limitation and make the time-frequency energy of the obtained signal better.Fractional Generalized S Transform(FRGST)combines the Fractional Fourier Transform with the Generalized S Transform method,which can have both advantages.This method can use the frequency to adjust the narrow time window width.The fractional frequency axis can be adjusted by matching the order parameters.Rotate to a proper position to obtain better time-frequency resolution,but the time-spectrum accuracy obtained by this method is not high.Based on the FRSGT algorithm,this paper draws on the idea of the high-precision time-frequency analysis method-Synchronous Extraction Transformation(SET),and proposes a fractional-order synchronous extraction S-Transform(FRTE)algorithm.This algorithm is processed by the FRGST method to obtain the optimal time-frequency spectrum.The SET method is used to extract the time-frequency coefficients on the time-frequency ridges in the timefrequency spectrum,thereby obtaining a new time-frequency spectrum.The accuracy of frequency analysis has been greatly improved.On the basis of theoretical simulation analysis,the new method is applied to the research of seismic signal reservoir identification,and good results have been obtained.The main research contents are:(1)Literature reading and method research.Read and research the literature on signal processing,especially the traditional time-frequency analysis methods: STFT,FRST,GST,and SET,etc.,understand its principles and formulas,and study the timefrequency characteristics of signal processing.Detailed introduction,analysis of their advantages and disadvantages,and corresponding simulation analysis.In order to make the new method better in principle and formula,and thus get better results.(2)Fractional-order synchronous extraction S transform(FRTE)is proposed.By studying the principles and formulas of the SET and FRGST methods,the signal is processed through FRGST,the optimal rotation angle is determined according to the degree of signal energy accumulation,and the fractional instantaneous frequency estimate is obtained.Based on the instantaneous frequency estimate and the SET algorithm,Fractional SEO.Use fractional order SEO to extract the time spectrum of the signal,retain the energy near the time-frequency ridge in the time-frequency plane,remove the excess divergent energy,obtain the fractional-order synchronous extraction of the generalized S transform value,and then modulate it to finally obtain the FRTE time Spectrum.(3)Analysis and comparison of GST,FRGST,SET and FRTE methods for processing multi-component signals and simulated signals.From the comparison,we can know that the time spectrum obtained by GST and FRGST is relatively rough,SET and FRTE have high accuracy,but FRTE It can better characterize multi-component signals with close frequencies,and the time spectrum obtained by SET will be interfered to obtain better results.The actual seismic data was used to verify the effectiveness of the new method.The results show that FRTE has high-precision time-frequency resolution and strong flexibility.It has a good effect on the high and low frequency characteristics of seismic signals.