Study On Seismic Fluid Discrimination Technology Of Tight Gas Reservoir

Tight sandstone gas reservoir is one of the important fields of natural gas exploration and development at present.However,because of strong compaction and late diagenesis,the tight sandstone reservoir have complex pore structure and low permeability,which is different from the conventional reservoir.The variation of seismic response characteristics with the change of porosity and gas saturation is small,which is difficult to be identified by seismic methods,so the identification of favorable reservoirs is always the difficulty of seismic prediction.The problems currently encountered are mainly manifested in two aspects.First,the commonly used petrophysical models are not suitable for tight sandstone reservoirs.Second,the conventional prestack elastic impedance inversion method has low resolution to tight sandstone reservoirs and it is difficult to achieve the application effect.This paper takes the tight sandstone reservoir of the Sulige gas field in Ordos Basin as the object.First,the construction of rock physical model for reservoir is carried out.The bulk modulus model of fluid is optimized by improving the Brie calculation method.The higher precision shear modulus of Krief model,Nur model or Pride-Lee model is combined with the relationship of Kd.y and udry by Han,and the fixed empirical parameters are expressed as the change value with the samples to improve the dry rock model.Then a fluid substitution model is established.On this basis,the shear wave velocity of the lack of logging in the study area is predicted,which provides data basis for seismic data interpretation and inversion.Secondly,the two layer medium models of three types of reservoir and cover combination are established by using the logging response characteristics of the target layer in the log data.The influences of water saturation,porosity and shale content on various elastic parameters and reflection coefficient are analyzed through two layer medium model,and the role of each physical parameter is understood.The accuracies of conventional elastic impedance(El)equation,generalized elastic impedance(GEI)equation and ray elastic impedance(REI)equation are compared by A VO analysis.The reliability and stability of each method are verified by two layer model inversion and well model inversion.Thirdly,after the analysis the cross diagrams and the coefficient of fluid factor sensitivity indicator of multiple parameters,the sensitive parameters for identifying the lithology and fluid are found.Finally,the seismic data of the study area are inverted by ray elastic impedance equation expressed by lithology sensitive parameter and fluid sensitive parameter to obtain the sensitive parameter profiles reflecting lithology and fluid.It has been proved by drilling data that this study has achieved good results in the fluid detection of tight sandstone reservoirs.The reaearch shows:(1)In addition to the volume modulus,water saturation,and pressure of each phase fluid,the bulk modulus of the fluid is also related to the connectivity of the pores.In the case of low effective pressure,the change rule of fluid bulk modulus with the change of fluid saturation is mainly determined by connectivity.Wood model is applicable to calculate fluid bulk modulus of conventional connected reservoir,while a combination of Wood and Domenico model is suitable for tight reservoir;(2)By improving the fluid bulk modulus model and the dry rock model,the average error of predicting S-wave velocity of the gas bearing section is 5.59%.The reflection coefficient calculated by different water saturation models with different errors proves that the predicted S-wave velocity can reach the purpose of detecting the fluid.At the same time lithology index m in Krief model,critical porosity(?)c in Nur model and consolidation parameters c in Pride-Lee model which are corresponding to each sample can be obtained.The values of these parameters can reflect lithology difference,pore structure,compaction degree and other characteristics,which indicate the geological characteristics of the reservoir;(3)Water saturation has little influence on P-and S-wave impedance.It is necessary to find a new fluid identification factor to better detect the fluid.It is found that Poisson's ratio is the most sensitive parameter for lithological identification,and f and f/vs are the sensitive parameters for fluid identification;(4)Due to the impact of tight sandstone reservoir by surrounding rock is large,so we first use Poisson's ratio which is the optimal parameters of lithology identification to identify effective sandstone range,and then use the sandstone range to delineate fluid identification factor for f/vs,and finally get the fluid identification factor data;(5)The precision of GEI formula is higher than El formula in tight sandstone reservoir of Sulige gas field in Ordos Basin,and the amplitude calculated by REI is more obvious than that calculated by GEI,so REI is more favorable for the prediction of the tight sandstone reservoir;(6)The ray elastic impedance inversion method is applied to the fluid identification and prediction of the Sulige tight sandstone reservoir,and the results are in good agreement with the drilling data.