Pore Structure Modeling And Transport Mechanism Of Low Permeability Sandstone Reservoirs

Low-permeability sandstone reservoirs are one of the key points of current oil and gas exploration.The in-depth study of pore structure is a key issue in the characterization of low-permeability sandstone reservoirs.Due to the nature of low permeability,low-permeability sandstones(or natural porous media)generally have rich pore structure types,more complex microscopic structure and rougher grain surface compared with the conventional porous media.The fluid located at pore surface tends to be bound to form irreducible state.Generally,this natural phenomenon is often ignored in the conventional permeability models or other recently presented fractal models.Thus,the permeability model of low-permeable porous media needs to be specially developed.Pore structure of porous media is the crucial attribute in controlling fluid flow.However,direct experimental measurements or numerical reconstructions are commonly expensive and not environmentally friendly,with great uncertainty caused by the complex nature of pore structure.It is very significant to evaluate the three-dimensional pore structure more economically.Archie's law has been widely used for electrical conduction modeling,as well as water saturation estimation.However,its empirical nature has also been extensively explored.Thus,in this work,based on many characteristics of low-permeability sandstone reservoirs including high-content bound water,multi-scale pore structure and multi-scale grain structure,fractal theory and multi-scale characterization techniques are used to construct high-precision pore network model,and to analytically develop the intrinsic transport mechanism of low permeability porous media.Based on the multiscale grain structure,a new model of pore structure typing and a generalized model of grain specific surface area were developed.In the proposed models,a new pore-structure-type indicator was defined to characterize the discrepancies of pore structures and divide pore structure type based on fractal geometry theory and a modified Kozeny-Carman equation.We compared the proposed model and the conventional model with comprehensive experimental tests in 130 low permeability sandstone cores from the Dongfang gas reservoir in the South China Sea.The grain fractal dimension D_g was first accurately calculated through thin section analysis.Result shows that grains of low-permeability sandstone indeed have the bifractal distribution with constant radius boundaries of approximately 30?m,potentially accounting for the multiscale of pore spaces.Compared with the conventional model,considering the complex size distributions of pore and grain as well as irreducible water saturation,the developed model as a generalized expression performs better in pore structure typing with higher correlation coefficients in porosity-permeability and structural coefficients-pore radius relationships due to the newly identified indicator with a wider range.Finally,a multiscale workflow was proposed to analyze the petrological and structural features of pore structures by two-dimensional core images,thin sections,scanning electron microscope,mercury intrusion tests and three-dimensional micro-computed tomography.This work enables us to explain the genetic types and differences of pore structure and may help to characterize the internal connection of grain structure-pore structure-flow mechanism in hydrocarbon exploration and development.Five samples with different pore structure types(porosity is approximately equal,however permeability varies widely)were chosen to conduct micro-computerized tomography test.Then,fractal and multifractal methods were used to characterize complex pore space of chosen samples based on the number-area(N-A)model,radius-volume(R-V)model and number-radius(N-R)model respectively,and the effect of image scale on the fractal dimension was investigated.From N-A model,each pore structure type has similar box dimension and generalized fractal dimension,but has significant difference in multifractal spectrum.Specially,the different values of singularity exponent in the multifractal spectrum have a good relationship with the experimentally measured permeability.From R-V model,there are clear inflections points on the fractal curves.Based on the value of maximum pore throat radius,only one segment(located in the middle)can be used to calculate the fractal dimension.From N-R model,fractal dimension obtained by pore network model does not satisfy the traditional relationship.Whether R-V or N-R model cannot distinguish rock petrophysical types and there is no obvious correlation between fractal dimension and the petrophysical properties.An improved model was theoretically deduced based on the fractal geometry theory with the irreducible water saturation be considered.133 total sandstone samples from a gas reservoir were used to verify the validity of the developed model.The mercury intrusion experiment and the conventional petrophysical property test were carried out,and the thin section of each sample was crafted.Pores were segmented by the proposed color extracted algorithm for calculating pore fractal dimension.Results show that the available fractal model overestimates permeability values.A new form of the classical Kozeny-Carman equation was also developed to accurate permeability estimation.In the low permeable porous media,the empirical Kozeny-Carman constant needs to take 3.5,instead of 5.However,no longer is typical superiority found at very low permeability(<0.5 mD)in the improved fractal model and corresponding reasons were analyzed and discussed.One special"bridge function"was developed as a function of the apparent length and tortuosity fractal dimension,which can characterize the relationship of pore structures between two dimensions(2D)and three dimensions(3D).Bridge function can serve as a conversion bridge of the radius to determine the capillary pressure curve(CPC).Estimations by the proposed method with experimental results obtained by mercury intrusion porosimetry were compared in six typical natural sandstones with varying porosities and permeabilities.Result shows that cross sections of the global pore structure,such as thin section,electronic probe,and micro computed tomography slices,give a reliable estimation of the CPC using the bridge function in porous media with a medium porosity.However,in unconventional porous media with a relatively low porosity(~10%)or extra high porosity(~30%),due to the empirical nature of the equation widely used to calculate the tortuosity fractal dimension,the necessary modification is necessary to obtain the CPC when applying the bridge function in such porous media.This insight can significantly simplify the procedure for obtaining the petrophysical properties of a porous medium,which may shed light on the inherent differences and correlations between the2D and 3D pore structures of porous media.Based on a parallel fractal tortuous capillary bundle model,an analytical electrical conduction model was derived for estimating the electrical conductivity or formation factor of a saturated porous media.A total of 46 sandstone samples from different basins and strata with varying porosity and permeability were selected to verify the validity of the developed model.The proposed model,the existing models and classical Archie's model were compared through rock electrical experiments.The experimental results confirm that each sample should have a specific cementation factor(m)value instead of a group of samples assigned a fitted m.The m is analytically interpreted as the average tortuosity degree of electrical paths.The proposed model without any empirical constants shows obvious superiority over the other famous models in calculating the formation factor at low porosity(<12%)with an error factor of±10.Additionally,the difference between hydraulic conduction and electrical conduction is analyzed through experiments and numerical simulations.Result shows that the electrical conduction appears to have no threshold saturation and pushes forward evenly along the potential drop.However,hydraulic conduction has significant threshold pressure and pore-size dependence for both non-wetting and wetting phases.Finally,a fractal saturation model for estimating water saturation is also developed for unsaturated porous media.The conduction model and saturation model gain insights into the electrical conductivity characteristics of porous media and the intrinsic nature of Archie's law.The physics-based models developed in this work considering more realistic geological factors and complex pore-grain structures help to deepen understanding of intrinsic connection between the microstructures and transport mechanisms of hydraulic and electrical conduction in low-permeability sandstone reservoirs or low-permeability porous media.