The Methods Of Weak Signal Seismic Reservoir Prediction In Shanghe Sand Four Period

With the continuous development of oil and gas field exploration and exploitation and theunceasing improvement of exploitation level in each big oilfield, the technician increasinglypaid close attention to the exploration and exploitation of complex reservoirs. Beach bar sandreservoirs as a kind of complex reservoirs, has good prospects in oil and gas exploration andexploitation. The characteristics of Beach bar sand body are nearly source, low maturity,rapid accumulation, longitudinal diversification of deposition thickness and lithic facies, andit is composed of multiphase fan layer stack. All of the characteristics cause the strongheterogeneity. In this paper, we researched Weak Signal Identification, sand body and sandbody boundary identification with Shanghe sand four sections.Based on the careful analysisand study of seismic, geology, logging and drilling data,we detailedly analyzed and studiedthe well logs, seismic reflection characteristics of beach bar sands in objective interval.Based on the sedimentary characteristics, structural characteristics and forming rules ofbeach bar sands in shanghe region, we studied and designed signal extraction and sand bodyboundary identification method in view of the objective interval thin sand shale interbedding.Due to complex formation superposed relationship aspect of the reservoir and small diversityof seismic reflection characteristics caused by vertical and horizontal changes, the sand bodysignal is weak and difficult to identify. Therefore, this paper carried out the method ofcombining third-order statistics and independent variables based on high order statisticsanalysis method. Third-order cumulants can effectively detect non-Gaussian strangeinformation, and then use the independent component de-noising. A combination ofthird-order statistics and independent component can effectively overcome the disadvantagethat traditional denoise method weakened the useful signal, it improved the resolution of theseismic section and the identification of singular boundary information. Verification of thetheoretical model and the actual data, the method has achieved good results.Aim at the difficult to clearly distinguish the boundary of sand body characteristics, theburg spectrum method and the three-dimensional structure attribute based on high orderstatistics estimate has presented in this paper. The resolution of higher-order spectrum is higher than conventional spectral, but the higher-order spectrum has lower signal-to-noiseratio. When generalized gaussian distribution of seismic signal mixed with non gaussian inthe records, or nonlinear low frequency signal, the seismic signal higher-order spectrum hasobvious zoning band. It can well separate the low frequency part and high frequency part ofseismic signal. It has certain effect to portray boundary, thereby increasing the lateralresolution. The three-dimensional structure attribute VCM seismic structural concept isdifferent from the concept of general attribute. VCM seismic structure can detect theabnormal seismic reflection mode by evaluating the different directions of the VCM along theseismic body. The identification of sand bodies fault can be improve by VCM seismicstructure. The two methods presented in this paper has more effect in tiny sand recognitionthan conventional attribute.