Study Of Seismic Data Interpretation Method Based On Wavelet Transform

Seismic wave energy attenuate when it spreading in underground, attenuationis particularly prominent when the seismic wave spreading in pore medium of highfrequency of seismic signal. So we can use seismic frequency attenuation gradientattribute for hydrocarbon detection. With the development of oil and gas exploration,the precision of traditional frequency attenuation gradient method is more and moreun-meeting the high resolution and high precision demand of exploration. This paperis to seize this problem, focus on the advantages of time and frequency wavelettransform, present frequency attenuation gradient analysis methods based on time-frequency wavelet and texture analysis method based on wavelet model.This work mainly has the breakthrough in the following aspects:①propose anew time-frequency analysis method-time-frequency three parameter wavelettransformation this method might through the adjustment wavelet three parameterchange window box number narrow width, achieve at the same time to the changesignal analysis temporal frequency resolution, on the other hand became thetraditional continual wavelet transformation by the time scaling transform time whenfrequency diagram, the high-frequency unit aliasing energy is pulled down seriously,this created the false appearance to the high frequency weaken explanation.②inorder to describe how does the seismic wave spread in the unground, we take thewave extrapolation method based on wave equation for attenuation of seismic energysimulation shows. The method choose Futterman-model to calculate, using Q tocharacterize the attenuation of formation and fluid, which can be consistent with thephysical test record.③this paper proposes a new analysis technology of frequencyattenuation based on time-frequency three-parameter wavelet transform. Firstly, thismethod uses time-frequency continuous three-parameter wavelet transform tocalculate the amplitude spectrum of target signal. Secondly, calculate the attenuationcoefficient of high frequency amplitude spectrum by non-linear algorithm. Resultsof theoretical calculation proved that the new method improve the resolution in timeand frequency domain of STFT, and overcomes the illusion of high frequency signal energy attenuation due to the energy overlap in continuous wavelet transform. Thepractical results show that, the new method in this paper has higher accuracyand test match rate than the conventional ones, it has important practical value.④thispaper proposes a texture analysis method based on wavelet model. The textureanalysis method presented in this paper is use a wavelet as model to extract therelevant information on wavelet frequency by calculating the similarity coefficientwith actual seismic data. The application shows that the method can improve theresolution of the image, while maintaining the texture can protrude laterallydiscontinuous singularity.