Comprehensive characterization method for heterogeneity of low permeability tight sandstone reservoirs

Abstract

The present invention discloses a method for comprehensively characterizing heterogeneity of low-permeability tight sandstone reservoirs, which is characterized by comprising the following steps: S1, using a numerical integration method to establish a comprehensive fractal dimension D of a rock sample; c and a new model for calculating the contribution rate of pores of different scales to the comprehensive fractal dimension; S2, collecting mercury injection data, preprocessing the mercury injection data by combining interpolation and derivation methods, and establishing a mercury injection data relationship curve; S3, calculating the sample comprehensive fractal dimension D based on the mercury injection data relationship curve and the new model c and the contribution of pores of different scales to the comprehensive fractal dimension; S4, the comprehensive fractal dimension D of the sample based on the calculation c The present invention combines interpolation, derivation and numerical integration techniques to more easily and accurately calculate the fractal dimension of the dynamic changes of pores of different scales and analyze the contribution of pores of different scales to the overall heterogeneity of the sample.

Description

Technical Field

[0001] The invention belongs to the field of low-permeability tight sandstone development, and in particular relates to a method for comprehensive characterization of heterogeneity of low-permeability tight sandstone reservoirs. Background Art

[0002] Low-permeability tight sandstone reservoirs have always faced many challenges in the development of oil and gas resources due to their complex pore structure, low permeability, and strong heterogeneity. The characterization of the heterogeneity of such complex reservoirs is of great significance for correctly understanding reservoir characteristics and formulating reasonable development policies. Since low-permeability tight sandstone reservoirs have multi-scale pore structures, complex mineral compositions, and highly variable permeability, traditional reservoir characterization methods have limitations in dealing with these complexities and cannot correctly understand the heterogeneity characteristics of such reservoirs. In recent years, fractal theory, as an effective mathematical tool, has gradually been applied to the fields of geology and oil and gas engineering to characterize complex reservoir structures and fluid behaviors. Fractal theory was first proposed by French mathematician Mandelbrot to describe the structural characteristics of complex objects in nature (Mandelbrot, 1975). Fractal theory of porous media is a theoretical method for describing and analyzing the complex internal structure of porous media. The application of fractal theory in porous media can be used to describe the characteristics of pore size distribution, surface area, volume, and connectivity of pore structure. Fractal geometry is a discipline that studies geometric structures that are self-similar, irregular, and scale-free. The pore structure of porous media usually exhibits complex and irregular characteristics, which makes it difficult to accurately describe traditional Euclidean geometry. Fractal theory quantitatively describes this complexity and irregularity through fractal dimension. Fractal dimension is the core concept of fractal geometry and is used to describe the complexity of fractal structures. Low-permeability tight sandstone reservoirs usually have complex pore structures, which not only show obvious heterogeneity in spatial distribution, but also show self-similar characteristics at different scales. Fractal theory can describe the heterogeneity and multi-scale characteristics of low-permeability tight sandstone reservoirs through the heterogeneous characterization parameter-fractal dimension.

[0003] There are many methods to obtain fractal dimension, such as image analysis, gas adsorption, and mercury injection. Since the mercury injection method can measure the volume change of mercury entering the rock pores at different pressures, and capture the pore characteristics of different scales in the reservoir in a relatively detailed manner. Therefore, it has become the most widely used method for quantifying pore structure characteristics. Through the mercury injection experiment, a relationship curve between mercury saturation and capillary pressure can be drawn. This curve usually shows a linear relationship in the double logarithmic coordinate system, and its slope can be used to calculate the fractal dimension and quantify the complexity and heterogeneity of the pore structure. Compared with other methods, the mercury injection method can more accurately characterize the distribution and morphology of micron- to nanometer-scale pores, and has the advantages of large characterization scale, high accuracy, and rich data, so that the fractal dimension calculated using mercury injection data covers more comprehensive information on pore structure characteristics. Whether it is conventional rock samples or unconventional tight sandstone samples, the mercury injection method can be applied, and the test results have good repeatability. Through mercury injection tests on multiple samples, the heterogeneity and multi-scale characteristics of the reservoir can be systematically studied, providing a scientific basis for reservoir classification and evaluation. Therefore, the most widely used method is to use regression to calculate fractal dimension using mercury injection test data. However, when using regression methods to process mercury injection data, due to the complex characteristics of mercury injection experimental data of different types of rock samples, linear regression methods are usually difficult to obtain good fitting results. Some scholars have proposed the use of piecewise linear regression to artificially segment the mercury injection curve according to the general shape, but this method is highly subjective, the boundary between large pores and small pores is not clear, and the geological significance of the segmented fractal dimension is unclear. In addition, due to the significant differences in the heterogeneity of rock samples with different pore structure characteristics, it is difficult to perform comparable scale division based on the fractal dimension obtained by piecewise regression. This not only makes it difficult for the fractal dimension results to reflect the overall fractal (or heterogeneity) characteristics of the rock sample, but also makes it difficult to clearly understand the heterogeneity of the rock sample through reliable fractal dimension quantitative calculation results, and it is also impossible to perform comparable analysis of the fractal dimension characteristics of different samples in different scale ranges. The above problems make it difficult for the fractal dimension calculated by the conventional fractal dimension calculation method and its improved method based on piecewise fitting to effectively guide engineering practice. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the method for comprehensive characterization of heterogeneity of low-permeability tight sandstone reservoirs provided by the present invention solves the problem that the prior methods cannot accurately reflect the overall fractal characteristics and multi-scale heterogeneity of the reservoir.

[0005] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a comprehensive characterization method for heterogeneity of low-permeability tight sandstone reservoirs, the technical route is as follows Figure 1 As shown, the specific steps include:

[0006] S1. Use numerical integration method to establish the comprehensive fractal dimension D of rock samples c A new model for calculating the contribution of pores of different scales to the comprehensive fractal dimension;

[0007] S2, collecting mercury injection data, pre-processing the mercury injection data by combining interpolation and derivation methods, and establishing a mercury injection data relationship curve;

[0008] S3. Calculate the comprehensive fractal dimension D of the sample based on the mercury injection data relationship curve and the new model c and the contribution of pores of different scales to the comprehensive fractal dimension;

[0009] S4. Calculated sample comprehensive fractal dimension D c And the contribution rate of pores of different scales to the comprehensive fractal dimension are used to complete the comprehensive characterization of the heterogeneity of low-permeability tight sandstone reservoirs.

[0010] Further: In the step S1, a new model for calculating the comprehensive fractal dimension Dc of rock samples and the contribution rate of pores of different scales to the comprehensive fractal dimension is established by using a numerical integration method, including the following steps:

[0011] Step 1: Using the minimum capillary pressure data P c,min Calculate the maximum pore radius r max .

[0012]

[0013] Where: P c,min is the minimum capillary pressure, usually the displacement pressure P e ; r max is the maximum pore radius.

[0014] Step 2: Using the maximum capillary pressure data P c,max Calculate the minimum pore radius r min .

[0015]

[0016] Where: P c,max is the maximum capillary pressure; r min is the minimum pore radius.

[0017] Step 3: From the minimum pore radius r min To the maximum pore radius r max Between, take n linearly uniformly distributed values.

[0018] Step 4: Use the Young-Laplace equation to calculate the linearly uniformly distributed pore radius r i The corresponding capillary pressure data P c,i .

[0019]

[0020] Where: P c,i is the radius of the ith pore r i The corresponding capillary pressure; r i is the i-th pore radius.

[0021] Step 5: Subdivide the P c Data were log-transformed.

[0022]

[0023] Where: lgP c,i is the radius of the ith pore r i The corresponding capillary pressure is the logarithm to base 10.

[0024] Step 6: Convert the original interpolation data (d(lgS Hg ) / d(lgP c ), lgP c ) at the new segmentation data point lgP c,i A quadratic interpolation calculation is performed on the i Corresponding to d(lgS Hg ) i / d(lgP c ) i This can more accurately describe d(lgS Hg ) i / d(lgP c ) i The variation of derivative values ​​at different pore size scales improves the accuracy and reliability of the calculation of the fractal dimension contribution rate of porous media.

[0025] Step 7: Plot the relationship between the pore radius and the slope value (r i ~d(lgS Hg ) i / d(lgP c ) i ). i is the independent variable, d(lgS Hg ) i / d(lgP c ) i are the dependent variables, plotted on the horizontal and vertical axes respectively.

[0026] Step 8: Use numerical integration method to establish comprehensive fractal dimension D c Compute the new model.

[0027] Assume that d(lgS Hg ) i / d(lgP c ) i With r i The interpolation data has n discrete data points. The local element fractal dimension D i (D i =k i +3 = d(lgS Hg ) i / d(lgP c ) i +3, i = 1, 2, ..., n) The area enclosed by the data point and the horizontal axis of the pore radius r is A r The integral formula using the trapezoidal method can be written as:

[0028]

[0029] Where: D i is the local element fractal dimension, i = 1, 2, ..., n, indicating the number of elements; A r is the local microelement segment fractal dimension D i The area enclosed by the data point and the horizontal axis of the pore radius r; Δr is the pore throat radius range, Δr = r max -r min ; n represents the number of discrete data points, which is equal to the minimum pore radius r min To the maximum pore radius r max The number of linear uniform distributions obtained between; f(D c ) is the comprehensive fractal dimension function; g(k) is the double logarithm (lgS Hg ,lgP c ) Scatter linear regression slope or derivative function.

[0030] Step 9: Using the same method, establish a new model for calculating the contribution of pores of different scales to the comprehensive fractal dimension.

[0031] Local element fractal dimension D i The area enclosed by the data point and the horizontal axis of the pore radius r is A r , the area A can be r The area is divided into different pore size ranges.

[0032] For micropores (r < 0.1 μm), the contribution of micropore-scale pores to the overall heterogeneity of the sample can be calculated using the following formula:

[0033]

[0034] For mesopores (0.1 μm to 1 μm), the contribution of mesopore-scale pores to the overall heterogeneity of the sample can be calculated using the following formula:

[0035]

[0036] For macropores (1 μm to 10 μm), the contribution of macropore-scale pores to the overall heterogeneity of the sample can be calculated using the following formula:

[0037]

[0038] For super-large pores (>10μm), the following formula can be used to calculate the comprehensive fractal dimension of super-large pore scale:

[0039]

[0040] Where: A r,M , A r,I , A r,L , A r,U The local microelement fractal dimension D in the pore size range of micropores (r≤0.1μm), mesopores (0.1μm~1μm), macropores (1μm~10μm) and ultra-macropores (>10μm) i (D i =k i +3 = d(lgS Hg ) i / d(lgP c ) i +3, i = 1, 2, ..., m1) The area enclosed by the data point and the horizontal axis of the pore radius r; ζ M , I , L , U are the contribution rates of micropores (r≤0.1μm), mesopores (0.1μm~1μm), macropores (1μm~10μm) and ultra-macropores (>10μm) to the comprehensive fractal dimension; Δr1 is the pore size range of micropores (r≤0.1μm), Δr1=r(0.1)-r min , r(0.1)=0.1μm; Δr2 is the pore size range of mesopores (0.1μm~1μm), Δr2=r(1)-r(0.1), r(1)=1μm, r(0.1)=0.1μm; Δr3 is the pore size range of macropores (1μm~10μm), Δr3=r(10)-r(1), r(10)=10μm, r(1)=1μm; Δr4 is the pore size range of ultra-large pores (>10μm), Δr4=r max-r(10), r(10) = 10μm; m1 is the number of data points with linear uniform distribution of pore throat radius in the range from the minimum value to 0.1μm; m2 is the number of data points with linear uniform distribution of pore throat radius in the range from 0.1μm to 1μm; m3 is the number of data points with linear uniform distribution of pore throat radius in the range from 1μm to 10μm; m4 is the number of data points with linear uniform distribution of pore throat radius in the range from 10μm to the maximum pore throat radius.

[0041] Establishing a clear pore size range can obtain good comparable results when comparing and analyzing the fractal characteristics of different porous media and different pore size scales, which is convenient for comprehensive evaluation and comparative study. Figure 5 The relationship curve between mercury saturation and capillary pressure measured by mercury injection experiment is shown, and the process of mercury injection into different pore sizes is described. The fractal characteristics of porous media have multi-scale differentiation characteristics. When conducting mercury injection experiments, mercury, as a non-wetting phase, will gradually invade the porous media in the order of pore size. The process of mercury preferentially invading porous media is subject to pore development and scale effects. Schematic diagram of the physical meaning of fractal dimensions of different pore scales, such as Figure 5 Specifically:

[0042] Ultra-large pores (>10μm): Ultra-large pores have large volumes and good connectivity, and are usually filled in the initial stages of mercury intrusion. Mercury intrusion process: Since these pores have larger diameters, the required capillary pressure is lower. Therefore, mercury will enter these pores first.

[0043] Macropores (1μm to 10μm): Macropores are relatively large, can accommodate more mercury, and have better connectivity. As capillary pressure increases, mercury will gradually enter these smaller pores. Macropores are filled in the early to mid-stage of mercury intrusion.

[0044] Mesopores (0.1 μm to 1 μm): Mesopores are of medium size and are evenly distributed. They may be connected to micropores. After the capillary pressure increases further, mercury begins to enter these medium-sized pores. These pores are filled in the middle and late stages of mercury intrusion.

[0045] Micropores (<0.1 μm): Micropores are very small, usually connected to other pores, and small in size. Mercury can only enter these tiny pores under very high capillary pressure. Micropores are the last area to be invaded by mercury.

[0046] In summary, in the mercury injection experiment, the intrusion process of mercury causes pores of different pore sizes to be filled in sequence, forming a mercury injection curve. Since there are multiple pore sizes in porous media, these pore sizes have different complexities and heterogeneities in spatial distribution and structure. Fractal theory provides an effective method to quantify the heterogeneity and complexity of pores of different scales by calculating the fractal dimension contribution rate. The present invention can capture this multi-scale pore characteristic. According to the new method proposed in the present invention, the combined method of interpolation, derivation, and integration, the calculated comprehensive fractal dimension reflects the complexity and heterogeneity of the pore structure, and can reveal the characteristics of pores at different scales, thereby comprehensively characterizing the multi-scale heterogeneity of the reservoir.

[0047] Further: in said S2, the mercury injection data includes mercury saturation and capillary pressure, collecting the mercury injection data, pre-processing the mercury injection data by combining interpolation and derivation methods, and establishing a mercury injection data relationship curve, including the following sub-steps:

[0048] S21, performing logarithmic transformation on the mercury saturation and the capillary pressure, and establishing a scatter plot of the relationship between the mercury saturation and the capillary pressure in a double logarithmic coordinate system;

[0049] Among them, in the double logarithmic coordinate system, the horizontal axis is the logarithmic value of mercury saturation, and the vertical axis is the logarithmic value of capillary pressure;

[0050] S22, obtaining continuous interpolation data from a scatter plot of the relationship between mercury saturation and capillary pressure in a double logarithmic coordinate system by an interpolation method, performing a derivative operation on the interpolation data, and obtaining a mercury injection data relationship curve;

[0051] The mercury injection data relationship curves include: a relationship curve between the slope value of the interpolated data after derivation and the capillary pressure, a relationship curve between the slope value of the interpolated data after derivation and the mercury saturation, and a relationship curve between the slope value of the interpolated data after derivation and the pore radius.

[0052] Further: In S3, the comprehensive fractal dimension D of the sample is calculated based on the mercury injection data relationship curve and the new model. c and the contribution rate of pores of different scales to the comprehensive fractal dimension, including the following steps:

[0053] S31, calculating the capillary pressure corresponding to the linearly uniformly distributed pore radius by the Young-Laplace equation, and taking it as the capillary pressure corresponding to the subdivided pore radius;

[0054] S32, performing logarithmic conversion on the capillary pressure corresponding to the subdivided pore radius, and performing interpolation calculation on the data in the relationship curve between the slope value of the interpolated data after differentiation and the capillary pressure according to the result of the logarithmic conversion, so as to obtain the relationship between the pore radius and the slope value of the interpolated data after differentiation;

[0055] S33, based on the relationship between the pore radius and the slope value of the interpolated data after derivation, draw a relationship graph between the pore radius and the slope value;

[0056] Wherein, the horizontal axis of the relationship graph between the pore radius and the slope value is the pore radius, and the vertical axis is the slope value of the interpolated data after derivation;

[0057] S34, based on the established rock sample comprehensive fractal dimension D c A new model is developed to calculate the contribution rate of pores of different scales to the comprehensive fractal dimension, and then the comprehensive fractal dimension of the sample and the contribution rate of pores of different scales to the comprehensive fractal dimension are calculated based on the relationship diagram between pore radius and slope value.

[0058] The beneficial effects of the present invention are:

[0059] (1) The present invention provides a comprehensive characterization method for the heterogeneity of low-permeability tight sandstone reservoirs. By combining interpolation, derivation, and numerical integration techniques, the method can more accurately calculate the fractal dimension of dynamic changes in different pore scales, thereby improving the calculation accuracy of the fractal dimension. Combining interpolation, derivation, and numerical integration techniques can effectively control and reduce errors in the calculation process, improve the reliability of the results, and lay the foundation for more accurately calculating the comprehensive fractal dimension of the sample, analyzing the contribution of pores of different pore scales to the overall heterogeneity of the sample, and comprehensively evaluating the heterogeneity of the reservoir.

[0060] (2) The present invention realizes automated processing and efficient calculation, avoids the subjectivity and uncertainty of manual data segmentation analysis, and makes data processing more automated and objective. The new method of the present invention significantly improves the calculation efficiency while improving the calculation accuracy.

[0061] (3) The present invention has strong adaptability and practicality, is suitable for porous media of different types and structures, can handle fractal characteristic analysis of different pore size scales, and is widely used in the calculation of fractal dimension of porous media in practical engineering and scientific research. It has strong versatility and adaptability, and provides an effective method for the characterization and analysis of porous media.

[0062] (4) The present invention realizes the multi-scale fractal property refinement analysis. By generating a more detailed pore size range, the method can capture the subtle changes in the pore structure at different scales and provide more detailed analysis results. When conducting comparative analysis of the fractal properties of different porous media and different pore size scales, the method can provide results with good comparability, which is convenient for comprehensive evaluation and comparative research.

[0063] (5) The present invention analyzes the water lock characteristics of gas reservoirs through fractal dimension, which can quantify the complexity and heterogeneity of reservoir pore structure, reveal the fluid flow pattern, and provide theoretical guidance for reducing the water lock effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same meaning. In the accompanying drawings:

[0065] Figure 1 Technical roadmap of the method of the present invention

[0066] Figure 2 lgS Hg ~lgP c Schematic diagram of various forms of curves

[0067] Figure 3 Numerical integration to calculate lgS Hg ~lgP c Schematic diagram of the geometric meaning of the average slope of a curve

[0068] Figure 4 Schematic diagram of the steps of calculating fractal dimension using the new method of the present invention

[0069] Figure 5 Mathematical and geometrical significance of numerical method for calculating comprehensive fractal dimension

[0070] Figure 6 Schematic diagram of the physical significance of the contribution of pores of different pore sizes to the overall heterogeneity of the sample

[0071] Figure 7 Porosity and permeability relationship diagram

[0072] Figure 8 Schematic diagram of the diversity of mercury injection curves

[0073] Fig. 9 Relationship between fractal dimension Df of pore structure and various parameters calculated by numerical method

[0074] Fig.10 Relationship between fractal dimension Dτ of pore tortuosity and various parameters calculated by numerical method

[0075] Fig.11 The relationship between the comprehensive fractal dimension Dc and various parameters calculated by numerical method

[0076] Fig.12 Relationship between the contribution of pore structure and pore tortuosity to overall reservoir heterogeneity and porosity and permeability

[0077] Fig.13 Relationship between the contribution of pores of different scales to the overall heterogeneity of rock samples and porosity

[0078] Fig.14Relationship between the contribution rate of pores of different scales to the overall heterogeneity of rock samples and PG

[0079] Fig.15 Relationship between the contribution of pores of different scales to the overall heterogeneity of rock samples and permeability

[0080] Fig.16 Relationship between the contribution rate of pores of different scales to the overall heterogeneity of rock samples and RQI

[0081] Fig.17 Relationship between water-locking damage and fractal dimension of gas phase relative permeability under irreducible water saturation DETAILED DESCRIPTION

[0082] The embodiment of the present invention provides a method for comprehensive characterization of heterogeneity of low permeability tight sandstone reservoirs. The implementation process of the method is as follows: Figure 1 shown.

[0083] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0084] S1. Using numerical integration method, a new model for calculating the comprehensive fractal dimension Dc of rock samples and the contribution rate of pores of different scales to the comprehensive fractal dimension is established. The specific steps include:

[0085] Part of the data in the example analysis of the present invention is applied to some existing basic models of porous medium fractal dimensions for calculation. Therefore, before conducting the example analysis, it is necessary to explain the basic theory and basic model of porous medium fractal.

[0086] (1) Basic theoretical analysis and basic model of porous media fractals

[0087] The fractal theory model was originally derived from the basic theory of probability theory. Generally speaking, the cumulative distribution function is a probability function that describes the random variable X is less than or equal to a certain value x. Mathematically, the cumulative distribution function F(x) is defined as: F(x) = P(X≤x) where P represents probability. The cumulative distribution function provides an overall description of the distribution of random variables, reflecting the probability accumulation of variables within different value ranges. In the fractal theory of porous media, the cumulative distribution function is used to describe the geometric properties of pores and capillaries, such as the number of pores, length, etc. The cumulative probability distribution provides a comprehensive description of the pore structure, which can reflect the overall characteristics of pores in different size ranges, rather than just the local characteristics of a specific size.

[0088] Assuming that rock pores are composed of capillary bundles, according to fractal theory, the relationship between mercury saturation and capillary force in the data obtained from mercury injection experiments can be expressed by formula (a-1):

[0089] lg(S Hg )∝(D L +D f -3)lg(P c )(a-1)

[0090] Where: lg represents the logarithm with base 10; D f is the fractal dimension of the pore space; D L is the fractal dimension of the pore length.

[0091] From formula (a-1), we can see that for a homogeneous reservoir, the double logarithmic diagram (lgS Hg ~lgP c ), theoretically, mercury saturation and capillary pressure are linearly related. Since porous media can be regarded as capillaries of different pore sizes, for capillary clusters of different sizes, the relationship between mercury saturation and capillary pressure is not a specific linear relationship, but changes continuously according to the pore size range.

[0092] Comprehensive fractal dimension D c The fractal dimension D of the pore space f Fractal dimension D of pore length L The composition is expressed as

[0093] D c =D f +D L (a-2)

[0094] In the pore structure analysis, two fractal dimensions D are considered comprehensively. f and D L , which can more fully describe the fluid flow and storage characteristics of the reservoir.

[0095] Combining equation (a-1) and equation (a-2), we can get

[0096] lg(S Hg )∝(D c -3)lg(P c )(a-3)

[0097] In the double logarithmic plot (lgS Hg ~lgP c ) has a slope of D f +D L -3 or D c -3. From this slope, we can determine the fractal dimension D f and DL The slope can be expressed as k, and the linear fitting relationship in the double logarithmic graph can be expressed as follows:

[0098] y=kx+b(a-4)

[0099] Where: y = lg (S Hg ), x = lg(P c ), k=D c -3 = D f +D L -3, due to D f The range is usually between 2 and 3. L The range is usually between 1 and 3, so D c The range of is between 3 and 6; the range of k is between 0 and 3.

[0100] Furthermore, D f It can also be calculated by the method provided by Yu et al., and the expression is as follows:

[0101]

[0102] Where: D min and D max is the minimum and maximum value of the aperture, this formula needs to satisfy D min / D max <0.01 condition.

[0103] Furthermore, D L It can also be calculated by the method provided by Yu et al., and the expression is as follows:

[0104]

[0105] Where: τ av is the average tortuosity, dimensionless; D av is the average pore diameter. Formula (a-6) describes how to calculate the fractal dimension of pore tortuosity by knowing the average tortuosity and the average pore diameter.

[0106] D L Another simple calculation can also be used:

[0107]

[0108] From the perspective of parameter dependence, this formula does simplify the calculation because it only depends on one parameter, porosity. This simplifies the parameter acquisition and processing process in practical applications. This makes it more advantageous in practical calculations and applications, especially when data is limited or it is difficult to obtain additional parameters.

[0109] The above is the basic derivation of fractal theory, which is the theoretical basis for obtaining fractal dimension. It is also the most widely used method for obtaining fractal dimension using mercury injection experimental data.

[0110] On this basis, many researchers found that when using mercury injection experimental data to calculate the fractal dimension, in the double logarithmic graph (logS Hg ~lgP c ) presents segmented characteristics. The saturation weighted method and capillary pressure weighted method are proposed. The following shows the comprehensive fractal dimension D c Two different weighting calculation methods.

[0111] Mercury saturation weighted method:

[0112] D c =D1S Hg,in +D2(S Hg,max -S Hg,in )(a-8)

[0113] Capillary pressure weighted method:

[0114]

[0115] Where: D c is the comprehensive fractal dimension; S Hg,in For the lgS Hg ~lgP c The inflection point in the double logarithm mainly corresponds to the mercury saturation; S Hg,in is the maximum mercury saturation; D1 is lgS Hg ~lgP c The first segment of the double logarithm is the comprehensive fractal dimension; D2 is lgS Hg ~lgP c The second comprehensive fractal dimension in the double logarithm; ΔP c,1 lgS Hg ~lgP c The first segment of the double logarithm corresponds to the absolute value of the capillary pressure change of the capillary pressure curve segment; ΔP c,2 IgS Hg ~lgP c The second segment of the double logarithm corresponds to the absolute value of the capillary pressure change of the capillary pressure curve segment; ΔP c,1+2 It represents the total change of capillary pressure in porous media after mercury injection.

[0116] In the process of actually applying the above methods to calculate the comprehensive fractal dimension, it was found that the above-mentioned conventional method is more applicable to relatively homogeneous porous media, that is, a general linear regression method can be used to obtain the comprehensive fractal dimension for relatively homogeneous porous media, while for porous media with a certain degree of heterogeneity or pore development of different scales, the segmented weighted average method (mercury saturation weighted method, capillary pressure weighted method) can be used to calculate the comprehensive fractal dimension. Previous studies usually divided the double logarithmic calculation data into two segments, representing two different scales of pore structure and fluid flow characteristics. However, for actual porous media with more complex pore structures, it is not entirely possible to present two-segment segmented characteristics with a strong linear relationship. There may be 2 or more double logarithmic curve characteristics, and there may also be characteristics that do not have an obvious linear relationship (such as Figure 3 ). This complexity makes the above method not generally applicable to the calculation of fractal dimensions of various types of porous media. In addition, when using regression method and segmented regression method to process mercury injection experimental data, the mercury injection curve and lgS Hg ~lgP c The general shape of the double logarithmic curve relies on artificial segmentation, the inflection point of the curve is highly subjective, the boundary between large pores and small pores is not clear, and the geological significance of the segmented fractal dimension is not clear enough.

[0117] (2) Embodiments of the present invention

[0118] In the present invention, the comprehensive fractal dimension D c The process of establishing a new model for calculating the contribution rate of pores of different scales to the comprehensive fractal dimension includes the formula used in the preprocessing of mercury injection data and the process of establishing a new model using the numerical integration method. The relevant calculation formulas used in the preprocessing of mercury injection data in steps 1-1 to 1-7 are as follows:

[0119] Step 1-1: Using the minimum capillary pressure data P c,min Calculate the maximum pore radius r max .

[0120]

[0121] Where: P c,min is the minimum capillary pressure, usually the displacement pressure P e ; r max is the maximum pore radius.

[0122] Step 1-2: Using the maximum capillary pressure data P c,max Calculate the minimum pore radius r min .

[0123]

[0124] Where: Pc,max is the maximum capillary pressure; r min is the minimum pore radius.

[0125] Step 1-3: From the minimum pore radius r min To the maximum pore radius r max Between, take n linearly uniformly distributed values.

[0126] Step 1-4: Calculate the linearly uniformly distributed pore radius r using the Young-Laplace equation i The corresponding capillary pressure data P c,i .

[0127]

[0128] Where: P c,i is the radius of the ith pore r i The corresponding capillary pressure; r i is the i-th pore radius.

[0129] Step 1-5: Subdivide the P c Data were log-transformed.

[0130]

[0131] Where: lgP c,i is the radius of the ith pore r i The corresponding capillary pressure is the logarithm to base 10.

[0132] Step 1-6: Convert the original interpolation data (d(lgS Hg ) / d(lgP c ), lgP c ) at the new segmentation data point lgP c,i A quadratic interpolation calculation is performed on the i Corresponding to d(lgS Hg ) i / d(lgP c ) i This can more accurately describe d(lgS Hg ) i / d(lgP c ) i The variation of derivative values ​​at different pore size scales improves the accuracy and reliability of the calculation of the fractal dimension contribution rate of porous media.

[0133] Step 1-7: Plot the relationship between the pore radius and the slope value (r i ~d(lgS Hg ) i / d(lgP c )i ). i is the independent variable, d(lgS Hg ) i / d(lgP c ) i are the dependent variables, plotted on the horizontal and vertical axes respectively.

[0134] Further: Based on steps 1-1 to 1-7, a numerical integration method is used to calculate the average value of the fractal dimension within the entire pore radius, that is, the comprehensive fractal dimension D c .

[0135] Assume that d(lgS Hg ) i / d(lgP c ) i With r i The interpolation data has n discrete data points. The local element fractal dimension D i (D i =k i +3 = d(lgS Hg ) i / d(lgP c ) i +3, i = 1, 2, ..., n) The area enclosed by the data point and the horizontal axis of the pore radius r is A r The integral formula using the trapezoidal method can be written as:

[0136]

[0137] Where: D i is the local element fractal dimension, i = 1, 2, ..., n, indicating the number of elements; A r is the local microelement segment fractal dimension D i The area enclosed by the data point and the horizontal axis of the pore radius r; Δr is the pore throat radius range, Δr = r max -r min ; n represents the number of discrete data points, which is equal to the minimum pore radius r min To the maximum pore radius r max The number of linear uniform distributions obtained between; f(D c ) is the comprehensive fractal dimension function; g(k) is the double logarithm (lgS Hg ,lgP c ) Scatter linear regression slope or derivative function.

[0138] Step 1-9: Similarly, we can also calculate the contribution rate of the fractal dimension of samples with different pore throat radius scales.

[0139] Local element fractal dimension D iThe area enclosed by the data point and the horizontal axis of the pore radius r is A r , the area A r The area is divided into different pore size ranges.

[0140] For micropores (r < 0.1 μm), the contribution of micropore-scale pores to the overall heterogeneity of the sample can be calculated using the following formula:

[0141]

[0142] For mesopores (0.1 μm to 1 μm), the contribution of mesopore-scale pores to the overall heterogeneity of the sample can be calculated using the following formula:

[0143]

[0144] For macropores (1 μm to 10 μm), the contribution of macropore-scale pores to the overall heterogeneity of the sample can be calculated using the following formula:

[0145]

[0146] For super-large pores (>10μm), the following formula can be used to calculate the comprehensive fractal dimension of super-large pore scale:

[0147]

[0148] S2. Collect mercury injection data, pre-process the mercury injection data by combining interpolation and derivation methods, and establish a mercury injection data relationship curve

[0149] Step 2-1: Core data collection and organization

[0150] The porosity and permeability data of 53 groups of cores were collected and sorted out, as shown in Table 1. The porosity and permeability relationship is as follows: Figure 2 As shown, there are 18 groups of samples with K≥10mD, 24 groups of samples with 1mD≤K<10mD, and 12 groups of samples with K<1mD. The number of samples is sufficient. There is a certain positive correlation between porosity and permeability, but the correlation is poor, indicating that the reservoir has strong heterogeneity.

[0151] Table 1 Basic data of samples

[0152]

[0153]

[0154]

[0155] Step 2-2: Basic mercury injection curve analysis

[0156] The characteristics of core mercury injection curves of different permeability levels are as follows: Figure 3 As shown in the figure, the reservoir properties corresponding to the type I mercury injection curve are the best, and the reservoir properties corresponding to the type IV are the worst. The overall morphological characteristics of the type I and type II mercury injection curves are consistent, but the corresponding maximum mercury saturations are different, and the mercury saturation range corresponding to the flat section of the type I mercury injection curve is significantly larger and distributed within the range of lower capillary force, that is, the overall pore size distribution of the type I reservoir is more concentrated and the pore size is larger. The type III mercury injection curve shows a transition from type II to type IV, and there are both concave curves and convex curves in the curve. When the type III mercury injection curve transitions from a concave curve to a convex curve, the concentration of the corresponding reservoir pore size distribution gradually improves, but it does not mean that the reservoir properties are better, but that the overall pore size becomes smaller, and the distribution is relatively concentrated in a smaller pore size range, so the reservoir properties further deteriorate. The maximum mercury saturation corresponding to the type IV mercury injection curve begins to gradually decrease, and the corresponding capillary force is large and the range of variation is small. On the one hand, it means that the overall pore size is very small, and on the other hand, it means that the porosity is significantly reduced and the reservoir properties are very poor.

[0157] From the perspective of the morphology of different types of mercury injection curves, the pore size distribution concentration points corresponding to different types of mercury injection curves are significantly different, and the reservoir shows obvious strong heterogeneity, but the change in the overall distribution range of pore size corresponding to different types of mercury injection curves is not obvious. Therefore, the range of reservoir heterogeneity caused by pore structure for different types of reservoirs is small. From the perspective of the same type of mercury injection curve, the mercury injection curve has only one flat section, indicating that the pore size distribution is in the form of a single peak. Although the overall distribution range of pore size is large, it has obvious characteristics of clustered distribution.

[0158] S3. Calculate the comprehensive fractal dimension D of the sample based on the mercury injection data relationship curve and the new model c and the contribution of pores of different scales to the comprehensive fractal dimension

[0159] Step 3-1: Numerical integration method to obtain the comprehensive fractal dimension D c

[0160] Firstly, the numerical integration method is used to calculate the pore structure fractal dimension and the comprehensive fractal dimension based on formulas (a-5) and (6), respectively. Then, the pore tortuosity fractal dimension is obtained using formula (a-2). The calculation results are shown in Table 2.

[0161] Table 2 Comprehensive fractal dimension of samples

[0162]

[0163]

[0164]

[0165] Step 3-2: Contribution of pores of different scales to the overall heterogeneity of rock samples

[0166] First, based on the numerical integration method and using formulas (8), (10), (12) and (14), the contribution of pores of different pore sizes to the overall heterogeneity of the rock sample is calculated. As shown in Table 3, the comprehensive fractal dimension is the sum of the fractal dimensions of pores of different pore sizes. The comprehensive fractal dimension obtained by this method is inconsistent with the comprehensive fractal dimension calculated by formula (6). The two methods consider different angles and have different weighting forms. Therefore, the calculation results are not comparable, but both have corresponding rationality.

[0167] Table 3 Contribution of pores of different scales to the overall heterogeneity of rock samples

[0168]

[0169]

[0170] S4. Calculated sample comprehensive fractal dimension D c The contribution rate of pores of different scales to the comprehensive fractal dimension is used to complete the comprehensive characterization of the heterogeneity of low-permeability tight sandstone reservoirs.

[0171] Step 4-1: Heterogeneity analysis of low permeability tight sandstone reservoir

[0172] The pore type and pore size distribution of tight sandstone lead to differences in fractal dimensions and pore structure parameters. Figures 9 to 12 It can be seen that the pore structure fractal dimension calculated by the numerical method is mainly distributed between 1 and 1.5; the pore tortuosity fractal dimension calculated by the numerical method is mainly distributed between 1.5 and 3. A small number of tortuosity fractal dimensions greater than 3 may be caused by the large interval of mercury injection data of the corresponding samples, but the impact on the entire data body is small; the comprehensive fractal dimension calculated by the numerical method is mainly distributed between 3 and 6, and the difference in pore tortuosity has a more significant impact on reservoir heterogeneity. The specific data of pore structure, pore tortuosity and comprehensive fractal dimension are listed in Table 2.

[0173] Fig. 9 The fractal dimension of pore structure is related to porosity, permeability, rock quality index (RQI), porosity group (PG), average pore throat radius R and pore throat radius R corresponding to mercury saturation of 15%. 15There is a positive correlation between them, and there is a strong correlation with porosity and porosity group (PG). The overall results show that the better the reservoir properties, the larger the fractal dimension of the pore structure, and the stronger the reservoir heterogeneity caused by the difference in pore structure. The better the reservoir properties, the larger the pore size distribution range, and micropores to super-large pores are developed, so the pore structure is relatively more complex, and the larger the fractal dimension of the pore structure, the greater the impact on reservoir heterogeneity. The worse the reservoir properties, the smaller and more concentrated the pore size distribution range, so the pore structure distribution is relatively simpler, the smaller the fractal dimension, and the smaller the impact on reservoir heterogeneity.

[0174] Fig.10 The fractal dimension of pore tortuosity is related to porosity, permeability, RQI, PG, average R and R 15 The worse the reservoir properties, the more small-diameter micropores and mesopores are developed, the smaller the pore size of the fluid seepage channel, the longer the actual flow path of the seepage process, the greater the pore tortuosity, the stronger the reservoir heterogeneity caused by the pore tortuosity, and the larger the fractal dimension.

[0175] Fig.11 In the study, the fractal dimension was combined with porosity, permeability, RQI, PG, average R and R 15 There is a negative correlation between the two, indicating that the better the reservoir properties, the weaker the overall heterogeneity of the reservoir. The variation characteristics of the comprehensive fractal dimension are consistent with the fractal dimension of pore tortuosity, so it is believed that the reservoir heterogeneity caused by pore tortuosity plays a controlling role in the overall heterogeneity.

[0176] Fig.12 In the figure, the contribution rate of pore structure and pore tortuosity to the overall heterogeneity of the reservoir is described under different porosity and permeability. It can be seen that the contribution rate of pore tortuosity to the overall heterogeneity of the reservoir is higher than that of pore structure, and when the porosity is less than 16.5% or the permeability is lower than 1mD, the difference between the two contribution rates increases with the deterioration of physical properties. When the porosity is greater than 16.5% or the permeability is higher than 1mD, the ratio of the contribution rate of pore structure and pore tortuosity to the overall heterogeneity of the reservoir is about 4:6.

[0177] For pores of different scales, the difference in pore fractal dimensions will also make their contribution rates to the overall heterogeneity of rock samples different. Figures 13-16It can be seen that the contribution rate of micropores to the overall heterogeneity of rock samples is mostly less than 10%, and the data of a small number of samples is greater than 10%; the number of samples with a contribution rate of medium pores to the overall heterogeneity of rock samples higher than 10% has increased significantly; the contribution rate of large pores to the overall heterogeneity of rock samples is mainly distributed between 10% and 90%; the contribution rate of super-large pores to the overall heterogeneity of rock samples is mainly distributed between 40% and 90%, but the number of samples with super-large pores has decreased significantly. The contribution rate of pores of different scales to the overall heterogeneity of rock samples is different, and there is also overlap, which further shows that reservoir heterogeneity is not only affected by factors such as pore size, pore type and pore size distribution, but also by pore tortuosity.

[0178] Fig.13 The contribution rate of micropores with R≤0.1μm to the overall heterogeneity of rock samples is weakly negatively correlated with porosity, while the fractal dimension corresponding to pores with R>0.1μm has no obvious correlation with porosity. The contribution rate of micropores to the overall heterogeneity of rock samples is concentrated below 10%, and the influence is relatively small, but the porosity shows obvious sensitivity to the change of the contribution rate of micropores to the overall heterogeneity of rock samples. Medium pores mainly contribute to rock samples with porosity greater than 15%, and the contribution rate to the overall heterogeneity is less than 10%, and a small amount of contribution rate is higher than 10%. Both large pores and super large pores mainly contribute to reservoirs with porosity greater than 15%, and the former's contribution rate to the overall heterogeneity of rock samples is mainly distributed between 10% and 90%. The latter's contribution rate to the overall heterogeneity of rock samples is mainly distributed between 40% and 90%. It also shows that super-large pores are mainly developed in samples with porosity greater than 15%, while super-large pores are almost non-existent in samples with porosity less than 15%. Porosity does not change significantly with the contribution of mesopores, macropores and super-large pores to the overall heterogeneity of rock samples.

[0179] The relationship between fractal dimension and PG in different aperture ranges is shown in Fig.14 In the figure, the various relationships shown are consistent with the porosity. In addition, it also shows that the higher the contribution rate of micropores and mesopores to the overall heterogeneity of rock samples, that is, the stronger the heterogeneity of micropores and mesopores, the weaker the storage capacity of the reservoir; while the change in the contribution rate of macropores and ultra-macropores to the overall heterogeneity of rock samples has little effect on the storage capacity of the reservoir.

[0180] Fig.15The contribution rate of micropores and mesopores with pore size R≤1μm to the overall heterogeneity of rock samples has no obvious effect on the change of core permeability. From mesopores to super-large pores, the overall distribution of the contribution rate of pores to the overall heterogeneity of rock samples tends to increase, and the permeability range also increases. In addition, when the contribution rate of super-large pores to the overall heterogeneity of rock samples is higher than 40%, as the contribution rate continues to increase, the core permeability tends to increase. Fig.16 In the figure, the relationship between the contribution rate of pores with different pore sizes to the overall heterogeneity of rock samples and RQI shows a characteristic that is consistent with permeability.

[0181] Comprehensive analysis shows that (1) pore size, pore type and pore size distribution greatly affect pore complexity, thereby determining fractal dimension, among which pore tortuosity fractal dimension is mainly affected. (2) The development of large-size pores and fractal characteristics jointly control the seepage capacity of low-permeability tight sandstone reservoirs, while micropores and mesopores mainly affect the storage capacity of low-permeability tight sandstone reservoirs. (3) A specific type of pore has a main distribution range of fractal dimension. The pores represented by the fractal dimension are mainly composed of one or more pore types. The pore types and pore distributions of different fractal dimensions are not exactly the same, which is the reason for the poor correlation between fractal dimensions. (4) There is a strong correlation between the fractal dimension of pore structure and porosity and PG, indicating that porosity has an important influence on the complexity of pore structure. The larger the fractal dimension of pore structure, the higher the reservoir storage capacity. (5) The fractal dimension of pore structure increases with the increase of pore size, and the reservoir heterogeneity caused by pore structure becomes stronger. (6) The fractal dimension of pore tortuosity is higher than that of pore structure, which means that the reservoir heterogeneity caused by pore tortuosity is stronger and is the main reason for reservoir heterogeneity.

[0182] Step 4-2: Apply fractal dimension to analyze water lock characteristics of gas reservoirs

[0183] By studying the differences in the contribution of fractal dimensions and pores of different scales to the overall heterogeneity of rock samples, we can understand the impact of the water lock effect in gas wells. The higher the fractal dimension, the more complex the pore characteristics, and the complex pore characteristics will directly affect the flow path and flow resistance of water and gas in the reservoir.

[0184] Through the differences in the fractal dimension of pore structure, the fractal dimension of pore tortuosity, the comprehensive fractal dimension and the contribution rate of pores of different pore sizes to the overall heterogeneity of rock samples, the main influencing factors of the strength of water locking effect at different pore scales are identified, and the degree of reservoir heterogeneity is quantitatively described from different angles. The higher the heterogeneity, the more significant the water locking effect.

[0185] Regarding the influence of water lock effect, from the perspective of porosity and permeability, it is generally believed that the lower the porosity and permeability, the more serious the influence of water lock effect on gas phase permeability will be. Fig. 9 It can be seen that the reservoir heterogeneity caused by the pore structure increases with the improvement of reservoir physical properties, thereby aggravating the water locking effect. Fig.10 It can be seen that the reservoir heterogeneity caused by pore tortuosity weakens as the reservoir physical properties improve, thereby reducing the water lock effect. It can be seen that the change in reservoir heterogeneity caused by pore structure is opposite to the change in reservoir heterogeneity caused by pore tortuosity, and the comprehensive fractal dimension is composed of the two. The fractal dimension of pore tortuosity is obviously higher than that of pore structure, and the change characteristics of the comprehensive fractal dimension are consistent with those of the fractal dimension of pore tortuosity. Therefore, the reservoir heterogeneity caused by pore tortuosity plays a dominant role in the overall heterogeneity of the reservoir.

[0186] For the study area, Fig.12 It can be seen that when the porosity is less than 16.5% or the permeability is less than 1mD, the contribution of pore tortuosity to the overall rock heterogeneity begins to increase significantly with the deterioration of reservoir properties, while the contribution of pore structure to the overall rock heterogeneity begins to decrease. The change in the fractal dimension of pore tortuosity leads to the extension of the actual path of fluid flow. Although the reservoir heterogeneity caused by the pore structure has been reduced, the overall reservoir heterogeneity is still increasing, and the fluid seepage resistance has increased, so the corresponding impact of water lock will be greater, and the degree of damage to the gas phase relative permeability caused by the water lock effect will also be greater.

[0187] In addition, by Fig.12 It can also be shown that when the porosity is greater than 16.5% or the permeability is greater than 1 mD, the ratio of the contribution of pore tortuosity and pore structure to the overall heterogeneity of the rock is about 6:4. Figures 9 to 12 It can be seen that as the reservoir properties improve, the influence of pore tortuosity on reservoir heterogeneity is more significantly reduced, and the change trend of the comprehensive fractal dimension is consistent with the change trend of the pore tortuosity fractal dimension, and opposite to the change trend of the pore structure fractal dimension. Therefore, although the influence of pore tortuosity is still greater as the reservoir properties improve, the overall rock heterogeneity caused by pore tortuosity is decreasing, while the overall rock heterogeneity caused by pore structure is increasing, but the former has a greater impact, so the overall rock heterogeneity is decreasing. Furthermore, the seepage resistance and flow path length of the fluid seepage process will be reduced, and the influence of the water lock effect will be smaller.

[0188] Through the gas-water relative permeability data under different physical properties, we obtained the corresponding gas phase relative permeability fractal dimension and plotted the relationship between the fractal dimension and the gas phase relative permeability damage value under irreducible water saturation ( Fig.17). The larger the fractal dimension of gas phase relative permeability, the stronger the heterogeneity of the reservoir, and the larger the water lock damage value (1-Krgmax), which is consistent with our analysis. Different cores have different physical properties, and the water lock characteristics will also have certain differences. This is because the physical properties of the core, such as pore structure, permeability and porosity, directly affect the flow behavior of the fluid in the reservoir.

[0189] In the description of the present invention, it is necessary to understand that the orientation or positional relationship indicated by the terms "center", "thickness", "upper", "lower", "horizontal", "top", "bottom", "inner", "outer", "radial", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", and "third" are used only for descriptive purposes, and cannot be understood as indicating or implying the relative importance or the number of implicitly specified technical features. Therefore, the features defined by "first", "second", and "third" may explicitly or implicitly include one or more of the features.

Claims

1. A comprehensive characterization method for heterogeneity of low-permeability tight sandstone reservoirs, characterized in that: The following steps are involved: S1. Use numerical integration method to establish the comprehensive fractal dimension of rock samples D c A new model for calculating the contribution of pores of different scales to the comprehensive fractal dimension; S2, collecting mercury injection data, pre-processing the mercury injection data by combining interpolation and derivation methods, and establishing a mercury injection data relationship curve; S3. Calculate the comprehensive fractal dimension of the sample based on the mercury injection data relationship curve and the new model D c and the contribution of pores of different scales to the comprehensive fractal dimension; S4. Calculation-based sample comprehensive fractal dimension D c and the contribution rate of pores of different scales to the comprehensive fractal dimension, completing the comprehensive characterization of the heterogeneity of low-permeability tight sandstone reservoirs; In S1, the comprehensive fractal dimension of rock samples is established D c The new model is calculated as: in, In the formula, D c is the comprehensive fractal dimension; To interpolate the slope value of the data after differentiation, is the logarithmic value of mercury saturation, is the logarithm of capillary pressure, r max is the maximum pore radius, r min is the minimum pore radius; D i is the local element fractal dimension, i = 1, 2, …, n , represents the number of microelements; A r is the local microelement segment fractal dimension D i Data points and pore radius r The area enclosed by the horizontal axis; Δ r is the pore throat radius range, Δ r = r max - r min ; n Represents the number of discrete data points, equal to the minimum pore radius r min To the maximum pore radius r max The number of linear uniform distributions obtained between; f ( D c ) is the comprehensive fractal dimension function; g( k ) is the double logarithm (lg S Hg , lg P c ) Scattered linear regression slope or derivative function; k is the linear regression slope, k = D c -3 = D f + D L -3.

2. The method for comprehensive characterization of heterogeneity of low-permeability tight sandstone reservoir according to claim 1, characterized in that: In S1, the pore scale is divided into micropores, mesopores, macropores and super-large pores according to the scale of pore radius, wherein the pore radius of micropores is less than 0.1 μm, the pore radius of mesopores is in the range of 0.1 μm to 1 μm, the pore radius of macropores is in the range of 1 μm to 10 μm, and the pore radius of super-large pores is greater than 10 μm. The new models for calculating the contribution rate of pores of different scales to the comprehensive fractal dimension are: Calculate the contribution of micropore-scale pores to the comprehensive fractal dimension The specific expression is: Calculate the contribution of mesopore-scale pores to the overall fractal dimension The specific expression is: Calculate the contribution of macropore-scale pores to the comprehensive fractal dimension The specific expression is: Calculate the contribution of ultra-large pores to the comprehensive fractal dimension The specific expression is: Where: A r,M , A r,I , A r,L , A r,U The local micro-element fractal dimensions within the pore size range of micropores, mesopores, macropores and super-macropores are D i , D i = k i +3 = d(lg S Hg ) i / d(lg P c ) i + 3. i = 1, 2, …, m 1. Data points and pore radius r The area enclosed by the horizontal axis; ζ M , ζ I , ζ L , ζ U are the contribution rates of micropores, mesopores, macropores and super-macropores to the comprehensive fractal dimension; Δ r 1 is the pore size range of micropores, Δ r 1 = r (0.1)- r min , r (0.1) = 0.1 μm; Δ r 2 is the pore size range of the mesopores, Δ r 2 = r (1)- r (0.1), r (1) = 1 μm, r (0.1) = 0.1 μm; Δ r 3 is the pore size range of macropores, Δ r 3 = r (10)- r (1) r (10) = 10 μm, r (1) = 1 μm; Δ r 4 is the pore size range of super large pores, Δ r 4 = r max - r (10) r (10) = 10 μm; m 1 is the number of data points with linear uniform distribution of pore throat radius in the range from the minimum value to 0.1 μm; m 2 is the number of data points with linear uniform distribution of pore throat radius in the range of 0.1 μm to 1 μm; m 3 is the number of data points with linear uniform distribution of pore throat radius in the range of 1 μm to 10 μm; m 4 is the number of data points with linear uniform distribution of pore throat radius in the range of 10 μm to the maximum pore throat radius.

3. The method for comprehensive characterization of heterogeneity of low permeability tight sandstone reservoir according to claim 1, characterized in that: In S2, the mercury injection data includes mercury saturation and capillary pressure. Preprocessing the mercury injection data to establish a mercury injection data relationship curve includes the following steps: S21, performing logarithmic transformation on the mercury saturation and the capillary pressure, and establishing a scatter plot of the relationship between the mercury saturation and the capillary pressure in a double logarithmic coordinate system; Among them, in the double logarithmic coordinate system, the horizontal axis is the logarithmic value of mercury saturation, and the vertical axis is the logarithmic value of capillary pressure; S22, obtaining continuous interpolation data from a scatter plot of the relationship between mercury saturation and capillary pressure in a double logarithmic coordinate system by an interpolation method, performing a derivative operation on the interpolation data, and obtaining a mercury injection data relationship curve; The mercury injection data relationship curves include: a relationship curve between the slope value of the interpolated data after derivation and the capillary pressure, a relationship curve between the slope value of the interpolated data after derivation and the mercury saturation, and a relationship curve between the slope value of the interpolated data after derivation and the pore radius.

4. The method for comprehensive characterization of heterogeneity of low-permeability tight sandstone reservoir according to claim 1, characterized in that: The S3 comprises the following sub-steps: S31, calculating the capillary pressure corresponding to the linearly uniformly distributed pore radius by the Young-Laplace equation, and taking it as the capillary pressure corresponding to the subdivided pore radius; S32, performing logarithmic conversion on the capillary pressure corresponding to the subdivided pore radius, and performing interpolation calculation on the data in the relationship curve between the slope value of the interpolated data after differentiation and the capillary pressure according to the result of the logarithmic conversion, so as to obtain the relationship between the pore radius and the slope value of the interpolated data after differentiation; S33, based on the relationship between the pore radius and the slope value of the interpolated data after derivation, draw a relationship graph between the pore radius and the slope value; Wherein, the horizontal axis of the relationship graph between the pore radius and the slope value is the pore radius, and the vertical axis is the slope value of the interpolated data after derivation; S34, based on the comprehensive fractal dimension of rock samples D c A new model is developed to calculate the contribution rate of pores of different scales to the comprehensive fractal dimension, and then the comprehensive fractal dimension of the sample and the contribution rate of pores of different scales to the comprehensive fractal dimension are calculated based on the relationship diagram between pore radius and slope value.

Images