Complex Wavelet Spectrum Analysis To 3D Seismic Data

After decades of exploration and exploitation of land in eastern China, oil exploration anddevelopment degree grows gradually. Hydrocarbon exploration has entered the complexreservoir area, large oil fields which have been put into the development have almost entered thehigh water content period. Since exploration of the central and western complex surface area isin a relatively low level, reservoir has gradually become the main exploration targets. Newsituation in the exploration and development presented a serious challenge to the traditionalgeophysical technology.China's land-based oil and gas complex reservoir includes thin interbedded sandstonereservoir, carbonate lithology reservoir and a variety of special lithology reservoir. Complexreservoir is now important for the exploration area. Thin interbedded sandstone reservoir ischaracterized by large area distribution, low abundance and worse permeability. Such reservoirsare the main sides of Songliao Basin Placanticline, Songnan area lithology reservoir, OrdosMesozoic lithology reservoir and Upper Paleozoic lithology gas reservoir, northwest Sichuanshallow gas reservoir. High technology is needed for the identification of the main belt andfracture zones, identifying stratigraphic trap, looking for confections in low-grade reserior.Carbonate reservoirs are mainly distributed in Sichuan, the Ordos Basin and Tarim Basin. Suchreservoirs are characterized that reservoir space for carbonate fracture-hole and porosity and thepowerful inhomogeneity. The key to exploration is to determine carbonate reservoir insider caveand the weathering crust hole, cave, slit the spatial distribution. Complex reservoir predictiontechnology has broad application prospects.Spectrum analysis technology in theory is mainly based on the thin-layer reflection tuningprinciple. According to the theory, for the thin layer whose thickness is less than one fourth ofthe wavelength, in the time domain, with growth of the thickness of the thin, seismic reflectionamplitude gradually increases. When the thickness increases to a quarter wavelength tunablethickness, the amplitude of reflection come to the maximum value. With the increase in thethickness of thin, reflection amplitude decreased gradually. Time domain reflection amplitude ofthe greatest value, which corresponds to the frequency domain amplitude of the largest energyvalue. Thin tuning caused by the amplitude spectrum interference depends on the characteristicsof the acoustic characteristics of thin and its thickness.Spectrum analysis is an interpretation of reservoir technology which based on the frequency. It demonstrated to us a new seismic interpretation methods. Discrete Fourier transform orcomplex wavelet transform seismic data from the time domain into the frequency domain.Conversion calculated amplitude spectrum. The staff can see thin interference characteristics notonly from explained profile, but also from the plane. Such special treatment technology, wedefined as three-dimensional seismic data spectrum analysis techniques. Papers published in thesecond half of 2006 still mostly use short-time Fourier transform method for spectrum analysis.However, if the main frequency of the original seismic record is low, identifying stratigraphicthickness of the time is an equivalent thickness, which is the effects of overall number of thins.Moreover, with this method, the accurate positioning is only the time of window,but it can notestablish the accurate positioning in the window. The author presented a method of complexwavelet spectrum analysis, the result of the method of 3D seismic data in different frequency,full 3D explanation (including horizontal slice explanation)become possible. It can identifylithology and stratigraphic trap, and detect small incontinuity. The methods that have discretefrequency energy body (spectral decomposition techniques) characteristics, have timepositioning accuracy (traditional spectral decomposition technology can only get the timepositioning of the window). Complex wavelet transform highlightings local characteristics ofwavelet transform, which can avoid facing the process of selecting the window function and thesizes of time window when using WFT. Complex wavelet transform overcome the calculationoverload problem caused by sliding window in WFT. The results showed the complex wavelettransform spectrum analysis techniques have higher resolution than WFT. The spectrum offavorable reservoir characteristics is better reflected.This paper begins with complex seismic data transformation based on Hilbert transform toget seismic instantaneous features, then introduces the principle and algorithm of both the WFTand complex wavelet transform method for spectrum analysis, especially Contrasts WFT andcomplex wavelet transform spectrum analysis techniques.(1) It introduced a complex seismic analysis, the principle of Hiibert transform. Complexseismic analysis Using Hilbert transform to compute seismic instantaneous features.(2) It introduced the Fourier transform and window Fourier transform theory. It introduced3D seismic data spectrum analysis technology based on the window Fourier transform method.Details of algorithm and implementation process of this method was given. Through analyzingthe results, try to look for ways to improve this method.(3) It introduced the wavelet transform and complex wavelet transform theory. Definition ofcomplex wavelet is given. It introduced 3D seismic data spectrum analysis technology based onthe complex wavelet transform. Details of algorithm and implementation process of this methodwas given. This method was compared with the window Fourier transform spectrum analysis toshow the difference.During researching the principle and algorithm of Fourier transform spectrum analysis,we realized the limitations of the method, because the important feature of the estimate ofseismic amplitude spectrum is the length of selected time window function. If the selected timewindow is too short, the amplitude spectrum will proceed deconvolution with window function, which made this method lose local frequency characteristics. The other hand, too short a timewindow will make wavelet Sidelobe show for the illusion of a single reflection. Increasing thelength of time window will improve the frequency resolution, but if the time window is too long,flute marks for the features will show on the several reflection in the time window on theamplitude spectrum, so it is difficult to distinguish individual reflection amplitude spectrumcharacteristics. The time window problem of Fourier transform algorithm will make theamplitude spectrum deviate from the estimates. In practical application, it is difficult todetermine the length of the window, and it is impossible to get the quantitative analysis of thedeviation caused by the window length. Wavelet-based spectral analysis technology isbecoming the important tools of the seismic interpretation and reservoir prediction, In manypractical applications which play a greater role.In this paper, we used complex wavelet method for seismic signal spectrum analysis, morletwavelet transforms. We transform the scale domain into the frequency domain and get thespectrum characteristics of different frequency. And the theoretical model to illustrate thecomplex wavelet analysis of spectrum has higher accuracy than the Fourier transform. Finally,the actual 3-D seismic data in the Xinjiang Oilfield and Jianghan Basin Xiefengqiao area wereproceeded with complex wavelet transform spectrum analysis. The results showed complexwavelet this method can reflect reservoir distribution, location and morphology.