Seismic Impedance Inversion And Its Application In Reservoir Prediction

Seismic impedance inversion is a special process and interpretation techniquewhich uses impedance,the product of rock density and p-wave velocity;seismicimpedance inversion throve from the70s of20century,and change conventionalseismic data into lithology data,which provide important information for oil-gasprospecting.Seismic impedance inversion integrates the resolution of the wells in thevertical and the continuty of the seismic in the horizon.Lithology data can contrast withwells directly,and can be interpreted on the lithology layer.The technology can actuallyreflect underground geological structured and reservoir rock character, and can greatlyraise the percentage of drilling hits.This article describes he basic principles of the several commonly used seismicimpedance inversion methods(model-based inversion,recursive inversion,constrainedsparse spaike inversion,neural network inversion),the applicable conditions,and theadvantages and disadvantages.Recursive trace inversion reflects the seismic data morefaithfully, but its effect subjects to the initial seismic data. It cannot meet the requests ofthin reservoir description. Model based inversion makes full use of seismic, loggingand geology data, greatly improves the resolution of inversion results, better able toreflect lithological characters and physical properties of reservoir. However, it has therelatively large impact of the initial model. Constrained spare spike inversion is able toreflect the lateral change of reservoirs and faith to seismic data. Because of the limit ofassumptions, it cannot reflect slight vertical change of reservoirs.Wavelet affects theeffects of horizon calibration and finally results of the inversion. The author analyzesthe length of wavelet, time windows, take-off time and other factors, continues to carryout horizon calibration and error analysis, estimate the optimal wavelet. And the errorof the final inversion results is the smallest.Then we combine with the real seismicdata,logging data and geological data characteristics of the study area SYB,and respectively use the inversion methods evaluation reef reservoir on the changxingformation-feixianguan formation of the study area through the analysis of vertical andhorizontal resolution and agreement with logging data,then we can obtain the rightinversion method for the area,that is model-based inversion method,then for themethod of inversion results, we should combine with the actual data for geologicalinterpretation.At the same time, this paper also researches the reef reservoir geological andgeophysical basis, and uses conventional porosity inversion method to obtain theporosity data volume coincides with impedance volume. In addition, according to thereef reservoir heterogeneity characteristics, we propose a new prediction method ofporosity based on reef pore structure simulation; the method combines the prestackinversion, also taking into account the effect of pore structure parameters on porosity.The paper statistics the characteristic relation between the rock pore structureparameters and the shear modulus, the volume modulus, poisson’s ratio byPetrophysical test data and well log data, and derives better coherence by comparingpetrophysical statistical fitting the law with logging statistics fitting the law, and alsoanalyzes the influence of the size of the rock pore structure parameter errors on theprediction of porosity; to SYB region, and then using the predicted pore structureparameters to predict the porosity,then we derive that porosity prediction method whichgets into the rock the pore structure parameters information to predict the results withthe measured porosity of the goodness of fit is significantly higher than that ofconventional porosity prediction with the measured porosity of the goodness of fit.Sothis article conducting porosity forecast,based on the simulation of rock pore structureporosity prediction methods.The specific approach is well data yb12well of the SYBregion,fitting the relationship between pore structure parameters and elasticparameters,to obtain the pore structure parameters and poisson’s ratio of the correlationis the best,then use the prestack inversion to obtain the poisson’s ratio of the cube,andthen access the data volume of rock pore structure,and finally to predict the porosity ofthe rock.Judging from the porosity prediction results,we can seen the porosityprediction method based on reef pore structure model simulation in the reef,is moreobvious in the shoal reservoir porosity prediction,and is able to improve the accuracy ofthe porosity prediction.