Method for converting casting body slice pore radius dimension and reservoir pore structure characterization method

Abstract

This invention relates to a method for converting the pore radius dimension of cast thin sections and a method for characterizing reservoir pore structure, comprising the following steps: conducting pore characteristic experiments on cast thin sections to obtain area frequency data corresponding to different pore radius intervals; performing trend fitting on the discrete distribution data of pore radius area to obtain a continuous distribution curve; calculating the apparent volume frequency data corresponding to different pore radius intervals; correcting the radius from the mathematical expectation value of cutting pores to the true value, and correcting the boundary values ​​of the pore radius intervals corresponding to the apparent volume frequency data to obtain the final pore radius volume frequency distribution of the cast thin section. This application converts the conventional planar pore distribution characteristics of cast thin sections to three-dimensional spatial characteristics, which can eliminate the dimensional difference between the pore distribution obtained from the cast thin section experiment and the pore distribution in the actual three-dimensional space of the formation, enhance its matching with the logging curves that reflect the three-dimensional distribution of pore space, and effectively improve the accuracy of reservoir pore structure characterization.

Description

Technical Field

[0001] This invention relates to the field of methods for converting the pore radius dimension of cast thin sheets, specifically to methods for converting the pore radius dimension of cast thin sheets and methods for characterizing reservoir pore structure. Background Technology

[0002] For tight sandstone, oil and natural gas are hosted in the pores of the rock. During the formation of oil and gas reservoirs, oil and natural gas generated in source rocks often migrate multiple times before entering the pore spaces of the rock with good preservation conditions, thus becoming enriched and forming reservoirs. During oil and gas extraction, oil and natural gas utilize the pressure difference between the formation and the wellbore to enter the wellbore from the rock pores and be extracted to the surface. Therefore, whether it is the formation of oil and gas reservoirs or the extraction of oil and gas, the rock pore structure is an important factor. It determines whether oil and natural gas can migrate and form reservoirs effectively, and also determines whether oil and natural gas are easy to extract and whether oil and gas wells can produce high yields during the extraction process. Pore ​​characteristic experiments on cast thin sections are an effective method for characterizing reservoir pore structure. By injecting dyed resin or liquid glue into the thin section and completing the die casting, the area of ​​the dyed part is analyzed and statistically analyzed to obtain the main range, morphological characteristics, and frequency distribution of pore radius. When characterizing the pore structure of the entire reservoir, the commonly used technique is to calibrate the pore structure of the cast thin section with logging curves (such as nuclear magnetic resonance logging data that can reflect the pore structure from the side), giving the logging curves a quantitative meaning of pore radius. This allows for the characterization of the pore structure characteristics of the entire reservoir, and thus the assessment of the reservoir's storage performance and permeability.

[0003] However, conventional cast thin sections actually observe a two-dimensional plane that cuts through the near-spherical pore space, and the resulting pore radius distribution is an area distribution, with the area being the square of the radius. In reality, pores in a reservoir are distributed in three dimensions, and the nuclear magnetic resonance (NMR) logging relaxation spectrum also reflects the volume distribution of pores in three-dimensional space, with the volume being the cube of the radius. Therefore, there is a significant dimensional difference between the pore distribution of cast thin sections and the pore radius distribution characterized by NMR and other logging curves. This can lead to the calculated reservoir pore size distribution deviating from reality, resulting in low accuracy and causing inconvenience for subsequent exploration and judgment. Summary of the Invention

[0004] The purpose of this invention is to address the significant dimensional difference between existing methods for calculating pore distribution in cast thin films and the actual pore distribution in three-dimensional formation space. This invention provides a method for converting the pore radius dimension of cast thin films and a method for characterizing reservoir pore structure. The pore radius dimension conversion method provided by this invention transforms the conventional planar pore distribution characteristics of cast thin films into three-dimensional spatial characteristics. This eliminates the dimensional difference between the pore radius distribution obtained from cast thin film experiments and the pore radius distribution in actual three-dimensional formation space, enhancing its matching with well logging curves that reflect the three-dimensional distribution of pore space. This brings greater convenience to reservoir characteristic analysis and oil and gas reservoir description.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for converting the pore radius dimension of a cast thin sheet includes the following steps:

[0007] Step 1: Conduct a casting thin section pore characteristic experiment on the core thin section sample to obtain the corresponding area frequency data for different pore radius ranges;

[0008] Step 2: Perform trend fitting on the discrete distribution data of pore radius and area to obtain a continuous distribution curve;

[0009] Step 3: Calculate the apparent volumetric frequency distribution data corresponding to different pore radius ranges.

[0010] Step 31: In the continuous distribution curve obtained in step 2, a certain pore radius interval is divided into several micro-elements at equal intervals, and the area frequency of each micro-element is calculated using the continuous distribution curve function.

[0011] Step 32: Based on the divided micro-element sequence and the area frequency of each micro-element, calculate the average pore radius of the corresponding pore radius interval, and then use the average pore radius and the area frequency of the corresponding pore interval to calculate the apparent volume frequency distribution of the corresponding pore radius interval.

[0012] Step 33: Using the same calculation method as Step 31-32, obtain the apparent volume frequency data for other pore radius ranges;

[0013] Step 4: Correct the radius from the mathematical expectation value of the cut hole to the true value, and correct the boundary value of the pore radius interval corresponding to the apparent volume frequency data to obtain the final pore radius volume frequency distribution of the cast sheet.

[0014] First, casting thin section pore characteristic experiments were conducted on core thin sections to obtain area frequency data corresponding to different pore radius intervals. Then, a fitted continuous distribution curve was obtained with area frequency as the ordinate and pore radius as the abscissa. Next, the area frequency data was converted into apparent volumetric frequency data using the continuous distribution intervals. Finally, the boundary values ​​of the pore radius intervals corresponding to different volumetric frequency data were corrected to obtain the final pore radius volumetric frequency distribution of the casting thin section. The provided method for calculating the pore distribution of casting thin sections transforms the conventional planar pore distribution characteristics of casting thin sections into three-dimensional spatial characteristics. This eliminates the dimensional difference between the pore distribution obtained from casting thin section experiments and the pore distribution in the actual three-dimensional space of the formation, enhancing its matching with well logging curves that reflect the three-dimensional distribution of pore space. This brings greater convenience to reservoir characteristic analysis and oil and gas reservoir description.

[0015] Furthermore, in step 1, the core thin section sample is taken from the end face of the plunger core sample.

[0016] Furthermore, in step 1, during the experiment on the porosity characteristics of the cast thin section, the color of the resin injected into the cast thin section satisfies the following: H channel value ≥ 150, S channel value ≥ 0.6, and V channel value ≥ 0.6 in the three channels of the HSV color space. The purpose of injecting resin is mainly to facilitate the observation of the morphology, size, and other characteristics of the pore space. However, the color of the resin must not be confused with other components of the rock besides the pores, otherwise it will seriously affect the identification of the pores. Since the mineral components such as quartz framework, feldspar, and rock fragments in dense sandstone can exhibit various colors, through statistical analysis of interfering colors, it is believed that interfering colors are usually located in the low H, low S, and low V value range. Therefore, taking relatively high H, S, and V values ​​can avoid the influence of interfering colors. Thus, the color of the resin satisfies H channel value ≥ 150, S channel value ≥ 0.6, and V channel value ≥ 0.6.

[0017] Furthermore, in step 1, during the experiment on the porosity characteristics of the cast thin sections, the sampling interval was no more than 10 μm. Statistical analysis of the dense sandstone cast thin sections revealed that over 65% of the samples had peak area frequencies in the range of 0–10 μm. To ensure the accuracy of fitting and characterization, the sampling interval for experimental analysis was set to no more than 10 μm.

[0018] Furthermore, in step 2, Dirichlet boundary conditions are set for the discrete distribution data of the area frequency of the pore radius to control the value of the boundary points. Based on this, given the frequency distribution characteristics of the area frequency data, which show that "the area frequency of small pores is high, and the area frequency of medium and large pores gradually decreases", in order to obtain a better fitting effect, a skewed distribution function is used to perform trend fitting on the two-dimensional area frequency data of the pore radius to obtain a continuous distribution curve.

[0019] Furthermore, the Dirichlet boundary conditions are set as follows: when the pore radius approaches 0, its area frequency approaches 0 infinitely; when the pore radius approaches infinity, its area frequency approaches 0.

[0020]

[0021] In the formula, r is the capillary radius obtained from the pore structure analysis of the cast sheet, and P is the area frequency corresponding to a certain capillary radius.

[0022] Furthermore, the skewed distribution function is:

[0023] In the formula, F(r) is the area frequency density of the skewed distribution of capillary radius, r is the capillary radius (μm), a is the skewness, μ is the mean of the skewed distribution, and σ is the standard deviation of the skewed distribution.

[0024] Furthermore, in step 32, the average pore radius is calculated using Equation 1;

[0025] Formula 1: In the formula, The average pore radius, in μm, corresponds to the pore radius range. This represents the area frequency within the corresponding pore radius range; It is a vector representing the area frequency of each micro-element within the corresponding pore radius range; It is a vector representing the sequence of infinitesimal pore radii within the corresponding pore interval.

[0026] Furthermore, in step 33, the average pore radius and the area frequency of the corresponding pore interval are substituted into Equation 2 to calculate the apparent volume frequency of the corresponding pore radius interval.

[0027] Formula 2:

[0028] In the formula, This represents the apparent volume frequency within the corresponding pore radius range; This represents the area frequency corresponding to the pore radius range; A represents the average pore radius within the corresponding pore radius range, in μm; n Let N be a vector, corresponding to the area frequency of each infinitesimal element within the pore radius range; n It is a vector representing the sequence of infinitesimal pore radii within the corresponding pore interval.

[0029] Furthermore, in step 4, correcting the radius from the mathematical expectation value of the cut hole to the true value involves dividing the boundary value of the pore radius range by a conversion factor.

[0030] The conversion factor is calculated using Equation 3:

[0031] Formula 3 is: In the formula, Ratio is the conversion factor, and the calculated result is a constant of 0.866; R is the actual pore radius of the near-spherical pores, in μm.

[0032] Another objective of this invention is to provide a method for characterizing reservoir pore structure.

[0033] A method for characterizing reservoir pore structure includes the following steps:

[0034] Step 1: Calculate the porosity distribution of the cast thin sheet;

[0035] Step 2: Test the nuclear magnetic resonance T2 distribution of the same core sample from Step 1;

[0036] Step 3: Establish the correspondence between the porosity distribution of the cast thin sheet obtained in Step 1 and the nuclear magnetic resonance T2 distribution obtained in Step 2;

[0037] Step 4: Using the correspondence obtained in Step 3, convert the entire reservoir nuclear magnetic resonance logging data into a depth-varying pore radius distribution, thus obtaining the characterization results of the reservoir pore structure.

[0038] This invention provides a method for converting the dimensionality of reservoir pore radius, transforming the pore distribution characteristics of conventional planar cast thin sections into three-dimensional spatial characteristics. This eliminates the dimensional difference between the pore distribution obtained from cast thin section experiments and the pore distribution in the actual three-dimensional space of the formation, enhances its matching with the nuclear magnetic resonance T2 distribution, which reflects the three-dimensional distribution of pore space, and effectively improves the accuracy of reservoir pore radius distribution calculation results, bringing more convenience to reservoir feature analysis, oil and gas reservoir description, and other work.

[0039] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0040] 1. The method for converting the pore radius dimension of cast thin sections disclosed in this invention first obtains area frequency data corresponding to different pore radius intervals by conducting pore characteristic experiments on core thin sections. Then, a fitted continuous distribution curve is obtained with area frequency as the ordinate and pore radius as the abscissa. The area frequency data is then converted into apparent volume frequency data using the continuous distribution intervals. Finally, the boundary values ​​of the pore radius intervals corresponding to different volume frequency data are corrected to obtain the final pore radius volume frequency distribution of the cast thin section. This method converts the conventional planar pore distribution characteristics of cast thin sections into three-dimensional spatial characteristics, eliminating the dimensional difference between the pore distribution obtained from the cast thin section experiment and the pore distribution in the actual three-dimensional space of the formation.

[0041] 2. This invention provides a method for converting the pore radius dimension of reservoirs, transforming the conventional planar cast thin-section pore distribution characteristics into three-dimensional spatial characteristics, enhancing its matching with the nuclear magnetic resonance T2 distribution that reflects the three-dimensional distribution of pore space, effectively improving the accuracy of reservoir pore radius distribution calculation results, and bringing more convenience to reservoir feature analysis, oil and gas reservoir description and other work. Attached Figure Description

[0042] Figure 1 A flowchart illustrating a method for converting the pore radius dimension of a cast sheet, as provided in an embodiment of the present invention.

[0043] Figure 2 This is a fitting curve obtained by continuously fitting the skewed distribution function to the pore radius area frequency distribution of sample 1 in an embodiment of the present invention.

[0044] Figure 3 This diagram illustrates the physical meaning of each parameter when calculating the apparent volumetric frequency distribution of pore radius in an embodiment of the present invention.

[0045] Figure 4 This is a histogram comparing the measured pore radius area frequency distribution and the apparent volume frequency distribution of pore radius after dimensional transformation for sample 1 in this embodiment of the invention.

[0046] Figure 5 This is a schematic diagram illustrating the principle of correcting the pore radius from the apparent volume frequency distribution to the final volume frequency distribution in an embodiment of the present invention.

[0047] Figure 6 This is a graph showing the correspondence between the area frequency distribution of pore radius and the nuclear magnetic resonance relaxation time spectrum (T2) obtained in this embodiment of the invention without considering the dimensional limitations of the pore radius distribution of the cast sheet.

[0048] Figure 7 This is a graph showing the correspondence between the pore radius volume frequency distribution obtained after dimensional transformation and correction in this embodiment of the invention and the nuclear magnetic resonance relaxation time spectrum (T2).

[0049] Figure 8 The pore radius distribution map is obtained by establishing a scale relationship between the nuclear magnetic resonance logging T2 data used to calculate the pore radius distribution and the pore distribution data of the cast thin sheet without dimension transformation and the pore distribution data of the cast thin sheet after dimension transformation, respectively, and the nuclear magnetic resonance relaxation time T2.

[0050] Figure 9 This is a comparison chart of the average pore radius values ​​before and after dimensional transformation with the average pore radius values ​​of constant-rate mercury intrusion. Detailed Implementation

[0051] The present invention will now be described in detail with reference to the accompanying drawings.

[0052] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0053] Example 1

[0054] Figure 1 A flowchart illustrating a method for converting the pore radius dimension of a cast thin sheet, as provided in an embodiment of the present invention. Figure 1 As shown in the figure, this embodiment provides a method for converting the pore radius dimension of a cast thin sheet. The specific steps of this method are as follows:

[0055] Step 1: Conduct a casting thin section pore characteristic experiment on the core thin section sample to obtain the corresponding area frequency data for different pore radius ranges;

[0056] Specifically, in step 1, the core thin section sample is taken from the end face of the plunger core sample: the purpose of performing the dimensional transformation of the pore radius distribution of the cast thin section is mainly to match it with data including but not limited to nuclear magnetic resonance relaxation spectrum data. Therefore, it is required that the cast thin section sample be as homologous as possible to the samples from other experiments. Therefore, the end face of the standard plunger core sample is taken as the cast thin section sample.

[0057] Specifically, in step 1, during the experiment on the porosity characteristics of the cast thin section, the color of the resin injected into the cast thin section needs to meet certain conditions: H channel value ≥ 150, S channel value ≥ 0.6, and V channel value ≥ 0.6 in the three channels of the HSV color space. The purpose of injecting resin is mainly to make the shape, size, and other characteristics of the pore space easier to observe. However, the color of the resin should not be confused with other components of the rock besides the pores, otherwise it will seriously affect the identification of the pores. Since the mineral components such as quartz framework, feldspar, and rock fragments in dense sandstone can exhibit various colors, Table 1 shows the statistical results of various interfering colors and their characteristics in the HSV (hue, saturation, brightness) color space. Through the statistics of interfering colors, it is believed that interfering colors are usually located in the low H value, low S value, and low V value range. Therefore, taking relatively high H, S, and V values ​​can avoid the influence of interfering colors. Thus, the resin color meets the requirements of H channel value ≥ 150, S channel value ≥ 0.6, and V channel value ≥ 0.6.

[0058] Table 1. Sources and HSV characteristics of each interfering color component.

[0059]

[0060] Specifically, in step 1, during the experiment on the porosity characteristics of the cast thin sections, the sampling interval was no more than 10 μm. Statistical analysis of the dense sandstone cast thin sections revealed that over 65% of the samples had peak area frequencies in the range of 0–10 μm. To ensure the accuracy of fitting and characterization, the sampling interval for experimental analysis was set to no more than 10 μm.

[0061] Step 2: Perform trend fitting on the discrete distribution data of pore radius and area to obtain a continuous distribution curve;

[0062] Specifically, in step 2, Dirichlet boundary conditions are set for the discrete distribution data of the area frequency of the pore radius to control the value of the boundary points. Based on this, given the frequency distribution characteristics of the area frequency data, which show that "the area frequency of small pores is high, and the area frequency of medium and large pores gradually decreases", in order to obtain a better fitting effect, a skewed distribution function is used to perform trend fitting on the two-dimensional area frequency data of the pore radius to obtain a continuous distribution curve.

[0063] Specifically, the Dirichlet boundary conditions are set as follows: when the pore radius approaches 0, its area frequency approaches 0 infinitely; when the pore radius approaches infinity, its area frequency approaches 0.

[0064]

[0065] In the formula, r is the capillary radius obtained from the pore structure analysis of the cast sheet, and P is the area frequency corresponding to a certain capillary radius.

[0066] For sample 1 in this embodiment, the frequency distribution of its pore radius area distribution exhibits the characteristics shown in Table 2, namely, a high frequency in the low value region of capillary radius, and a gradual decrease in area frequency to 0 as the capillary radius increases. This sample is generally representative of the pore structure distribution of dense sandstone.

[0067] Table 2. Frequency distribution of capillary radius and area of ​​representative samples in this embodiment.

[0068]

[0069] It should be noted that the area frequency of the pore radius exhibits a high frequency characteristic in the low value region and a slow frequency decrease characteristic in the high value region, showing obvious skewed distribution characteristics. Based on the specific Dirichlet boundary conditions mentioned above, it is believed that fitting with a skewed distribution function can satisfy its distribution law.

[0070] Specifically, the skewed distribution function is:

[0071] In the formula, F(r) is the area frequency density of the skewed distribution of capillary radius, r is the capillary radius (μm), a is the skewness, μ is the mean of the skewed distribution, and σ is the standard deviation of the skewed distribution.

[0072] See Figure 2 The figure shows the capillary radius frequency distribution obtained by fitting the skewed distribution function to sample 1. The skewness α value is 49.5. Its distribution characteristics conform to the Dirichlet boundary conditions and have a high degree of agreement with the measured porosity distribution of the cast thin sheet. That is, it has a high area frequency in the range of 0-10μm and 10-20μm, and the area frequency gradually decreases with the increase of capillary radius, which conforms to the typical skewed distribution characteristics.

[0073] Step 3: Calculate the apparent volumetric frequency data corresponding to different pore radius ranges.

[0074] Step 31: In the continuous distribution curve obtained in step 2, a certain pore radius interval is divided into several micro-elements at equal intervals, and the area frequency of each micro-element is calculated using the continuous distribution curve function.

[0075] For near-spherical pores, the difference between volume and area is due to a product factor. Therefore, a crucial step in dimensionality transformation is to obtain the average pore radius within each experimental sampling interval, which serves as the product factor for dimensionality transformation. Given that the continuous area frequency distribution function has already been fitted in the preceding work of this embodiment, this provides an important method for implementing this step: integrating the area and volume within each experimental analysis sampling interval to obtain the average pore radius. Therefore, firstly, the experimental analysis sampling interval is differentiated at equal intervals in step 31. In this embodiment, since the experimental sampling interval is 10 μm, setting the length of each micro-element to Δr = 1 μm already achieves a good differentiation effect.

[0076] In this embodiment, the experimental sampling intervals are sorted from smallest to largest using 0 as the subscript, and R represents the maximum pore radius value of each experimental analysis sampling interval. The maximum pore radius of each sampling interval can be expressed as:

[0077] R0, R1, R2, R3, ..., R i ,……,R n

[0078] Here, we set R0 to 0. The first sampling interval is 0–10 μm, so the maximum pore radius R1 in this interval is 10. The second sampling interval is 10–20 μm, so the maximum pore radius R2 in this interval is 20, and so on. n The value is 80. For the i-th experimental analysis sampling interval, the frequency of the pore radius area frequency distribution observed in the pore structure analysis of the cast thin sheet can be expressed as:

[0079]

[0080] In the formula, To analyze the pore structure of a cast thin sheet within a certain sampling interval, the area frequency of the pore radius is calculated, such as when R² = 20, S... 20 This is equal to 32.27% of the area frequency of pores in the 10–20 μm range in Table 2; r is the pore radius, μm; Δr is the infinitesimal pore radius element, μm; F(r) is the area frequency density of pores with radius r; α r The frequency of the area occupied by a certain pore radius element.

[0081] Figure 3 This diagram illustrates the physical meaning of each parameter when calculating the apparent volume frequency distribution of pore radius in this embodiment. Figure 3 'a' is a schematic diagram illustrating the interrelationships of various physical quantities within the second sampling interval (10μm~20μm). In this sampling interval, each α... r The value is equal to F(r)·Δr, that is, the area frequency of each microelement is equal to the area frequency density at that position multiplied by the width of the microelement; Equal to all α within this interval r The sum of S, for the interval in the graph. 20 =α 11 +α 12 +α 13 +…+α 20 .

[0082] Step 32: Based on the divided micro-element sequence and the area frequency of each micro-element, calculate the average pore radius of the corresponding pore radius interval, and then use the average pore radius and the area frequency of the corresponding pore interval to calculate the apparent volume frequency distribution of the corresponding pore radius interval.

[0083] Furthermore, in step 32, the average pore radius is calculated using the following formula:

[0084]

[0085] In the formula, The average capillary radius, in μm, is the calculated average capillary radius of the i-th pore radius interval obtained by fitting based on a skewed distribution. Indicates that the radius of the capillary tube is R. i-1 ~R i The area frequency of the interval; Let be a vector representing the area frequency of each infinitesimal pore radius element within a certain sampling interval:

[0086]

[0087] for Figure 3 The sampling interval in a, A 20 ={α 11 ,α 12 ,…...,a 20}

[0088] In particular, A n Indicates from R0 to R n Area frequencies of all infinitesimal elements within the range:

[0089]

[0090] In the formula, The area frequency of the infinitesimal element with the largest pore radius;

[0091] Let be a vector representing a infinitesimal pore radius element within a certain sampling interval:

[0092]

[0093] for Figure 3 The sampling interval in a, N 20 ={11,12,……,20}

[0094] In particular, N n Indicates from R0 to R n All pore radius micro-elements within the range.

[0095] N n ={Δr,2Δr,……,r max}

[0096] In the formula, r max The value represents the maximum pore radius, in μm.

[0097] Use the above formula to Figure 3 Average pore radius of the sampling interval Calculations are performed to obtain The apparent volume of this sampling interval is

[0098] Based on the average pore radius within each experimental sampling interval and the experimentally obtained area frequency, the apparent volume frequency distribution of the pore radius is calculated:

[0099]

[0100] In the formula, Represents the apparent volume frequency of the i-th pore radius interval; Indicates that the radius of the capillary tube is R. i-1 ~R i The area frequency of the interval; A represents the average capillary radius (μm) of the i-th pore radius sampling interval calculated in the previous step; n Let N be a vector representing the area frequencies corresponding to a total of n capillary radius infinitesimals; n Let be a vector, representing the set of n infinitesimal capillary radius elements. For Figure 3 The interval in a is considered to have a volume frequency of .

[0101] It should be noted that the volume frequency distribution calculated at this time is called the "apparent volume frequency distribution". This is mainly because the pore radius value in the pore radius range at this time is the mathematical expectation of the pore radius that cuts the near-spherical pores, rather than the actual pore radius. It is necessary to correct it in the subsequent step 4 to obtain the actual pore radius range. Only then will the volume frequency distribution be the final volume frequency distribution. Therefore, the volume frequency distribution here is called the apparent volume frequency distribution.

[0102] In the actual situation of this embodiment, specifically, since Δr = 1μm, A n and N n The possible values ​​are:

[0103]

[0104] N n ={1,2,……,r max}

[0105] Step 33: Using the same calculation method as Step 31-32, obtain the apparent volume frequency data for other different pore radius ranges;

[0106] Figure 3 b is a schematic diagram showing the relationships between various physical quantities, including multiple sampling intervals. For Figure 3 b. In this embodiment, the experimental sampling interval is 10 μm, and the area frequency of each sampling interval is represented by S from left to right. 10 S 20 S 30 ...Description. The area frequency of each interval is calculated separately. and average pore radius The apparent volumetric frequency distribution of the pore radius can then be calculated.

[0107] Table 3 and Figure 4The table shows a comparison of the area frequency and apparent volume frequency of the pore distribution in Sample 1 of this embodiment. As can be seen from the table, due to the addition of the product factor—radius R—the area frequency and apparent volume frequency are significantly different. For the low pore radius range, due to the smaller product factor, the apparent volume frequency is lower than the area frequency, while the opposite is true for the high pore radius range. It should be noted that in this embodiment, because a skewed distribution was used to fit the pore radius area distribution, although the area distribution in the 70–80 μm range is 0, the fitted value inevitably has some error compared to the true value. This is why the apparent volume frequency in this range is slightly greater than 0.

[0108] Table 3 Comparison of capillary radius area versus apparent volume frequency of representative samples in this embodiment.

[0109]

[0110] Step 4: Correct the radius from the mathematical expectation value of the cut hole to the true value, and correct the boundary value of the pore radius interval corresponding to the apparent volume frequency data to obtain the final pore radius volume frequency distribution of the cast sheet.

[0111] In step 1 of this embodiment, since the experiment on the area frequency distribution of pore radius in the cast thin sheet involves cutting the core and observing the thin sheet in a plane, it cannot be guaranteed that the radius of the nearly circular pore plane obtained by cutting is exactly equal to the radius of the actual nearly spherical pore space. Specifically, since the cutting of the nearly spherical pore space is random during the casting process, the probability of any infinitesimal element being cut is equal from the sphere radius of 0 to the maximum sphere radius. Therefore, step 4 is further performed, and its specific operation method is as follows: based on the uniformly distributed probability density function, the expected value of the capillary radius obtained by cutting the spherical pore in a plane in three-dimensional space is obtained, the ratio of the expected value of the capillary radius to the true value is calculated, and the pore radius is corrected according to the volume frequency distribution to obtain the final volume frequency distribution result of the pore radius of the cast thin sheet. Figure 5 The basic principle of this step is demonstrated.

[0112] Specifically, in step 4, correcting the radius from the expected value of the cut pore to the true value involves dividing the boundary value (i.e., the expected value) of the pore radius range by a conversion factor to obtain the true pore radius value, i.e., R = r E / Ratio, the conversion factor, is calculated using the following formula:

[0113]

[0114] In the formula, Ratio is the conversion factor, and the calculated result is a constant of 0.866; R is the actual pore radius of the near-spherical pores, in μm.

[0115] Table 4 shows the frequency distribution of the pore radius of the cast sheet after step 4. As can be seen from the first row of the table, since the Ratio obtained by cutting is less than 1, this step actually widens the radius range of each sampling interval, while the frequency value of each interval does not change.

[0116] Table 4. Volume frequency distribution of true three-dimensional radius range of representative samples in this embodiment.

[0117]

[0118] Example 2

[0119] A method for calculating reservoir porosity distribution, characterized by comprising the following steps:

[0120] Step 1: Obtain the porosity distribution of the cast sheet obtained in Example 1, i.e., the data in Table 4;

[0121] Step 2: Test the nuclear magnetic resonance T2 distribution of the same core sample from Step 1;

[0122] Step 3: Establish the correspondence between the porosity distribution of the cast thin sheet obtained in Step 1 and the nuclear magnetic resonance T2 distribution obtained in Step 2;

[0123] Step 4: Using the correspondence obtained in Step 3, convert the entire reservoir nuclear magnetic resonance logging data into a depth-varying pore radius distribution, thus obtaining the reservoir pore distribution.

[0124] Since core and thin-section data are discrete relative to the entire reservoir, using discrete pore structure information to characterize and evaluate the reservoir cannot reflect its overall picture. While well logging curves can continuously reflect the reservoir, they lack the ability to directly characterize pore radius, only porosity as a macroscopic parameter, thus significantly reducing the accuracy of reservoir characterization. Therefore, an ideal method for refined reservoir evaluation involves using discrete pore structure information derived from cast thin sections to calibrate the well logging curves. This allows the well logging curves to indirectly calculate pore radius. Then, using this calibration model, the pore radius continuously distributed with depth can be calculated from the well logging data, thereby achieving a refined characterization of reservoir features.

[0125] Among all well logging methods, nuclear magnetic resonance (NMR) logging is the only one capable of characterizing pore structure. Although the inverted parameter is a relaxation time spectrum, not the pore radius, the relaxation time spectrum generally has a good positive correlation with the pore radius distribution. Therefore, the relaxation time can be used to calibrate the pore radius. Since the NMR relaxation time spectrum reflects the volumetric frequency distribution of the pore radius, the pore radius of the cast sheet used as the calibration target must also be a volumetric frequency distribution. The following comparisons of the pore radius distributions of the cast sheet without dimensionality transformation and correction, and those after the aforementioned operations, with the relaxation time T2 spectrum demonstrate the beneficial effects of this invention.

[0126] Figure 6 The correspondence between the pore radius distribution and the nuclear magnetic resonance relaxation time spectrum (T2) of two core samples used in this embodiment, obtained using existing technology—that is, without considering the dimensional limitations of the pore radius distribution of cast thin sections. Figure 6 a and Figure 6 (b represents the correspondence diagram of two different core samples). In the oil and gas exploration and development and well logging industries, T2 is generally logarithmically calibrated before being used to calibrate with pore radii. The diagram shows that the correspondence between the pore radius and T2 of the cast thin sheet without dimensional conversion is poor: the former exhibits a clear skewed distribution, with a large proportion of small pores, gradually decreasing towards larger pores; while the latter shows a significantly weaker skewed distribution, exhibiting a low proportion of small and large pores and a higher proportion of medium-sized pores. Therefore, forcibly calibrating and establishing a conversion relationship between the two without dimensional conversion and correction will lead to a significant distortion of the conversion relationship.

[0127] Figure 7 The correspondence between the pore radius distribution obtained from two core samples of all core samples used in this embodiment, obtained by dimensional transformation and correction using the method of this invention, and the nuclear magnetic resonance relaxation time spectrum (T2). Figure 7 a and Figure 7 (b represents the correspondence diagram of two different core samples). In the oil and gas exploration and development and well logging industries, T2 is generally logarithmically calibrated before being used for calibrating with the pore radius. The diagram shows a good correspondence between the pore radius and T2 of the cast thin sheet; both exhibit a low proportion of small and large pores and a high proportion of medium pores, with a high degree of overlap. Compared with existing methods, this significantly improves the matching accuracy of the conversion model.

[0128] exist Figure 7 Based on the correspondence, for all core samples used in this embodiment, the matching relationship between relaxation time T2 and pore radius is found, and the T2 distribution of nuclear magnetic resonance logging can be converted into the pore radius distribution. Figure 8The figures show NMR logging T2 data used to calculate pore radius distribution (a), and pore radius distribution maps (b and c) obtained by establishing a scaling relationship with NMR relaxation time T2 using untransformed and transformed cast thin-section pore distribution data, respectively. The figures show that in the untransformed group, the pore radius in the 4714m-4721m well section is mainly between 2 and 25 μm. The pore radius in the 4724m well section is mainly between 0 and 25 μm, and the pore radius in the 4724m-4728m well section is mainly between 0 and 15 μm. After dimensional transformation, the pore radius in the 4714m-4721m well section is mainly between 10 and 40 μm, the pore radius in the 4721m-4724m well section is mainly between 3 and 40 μm, and the pore radius in the 4724m-4728m well section is mainly between 3 and 25 μm. The pore radius calculated after dimensional transformation is significantly larger than that without transformation.

[0129] To determine which group's calculated pore radius is more accurate, the average pore radius obtained from each group's statistics is compared with the average pore radius obtained from the constant-rate mercury intrusion porosimetry (CRIP) experiment. The group whose average pore radius is closer to the CRIP average pore radius is considered more accurate. CRIP is an experimental method for measuring the volumetric distribution of pore radius with high precision. However, due to its high cost, long experimental cycle, and environmental risks, unlike the cast thin-section experiment which allows for batch preparation and analysis, CRIP typically involves sporadic sampling and analysis within a single reservoir segment. Since the pore radius distribution measured by CRIP is a volumetric frequency distribution, it can be used to verify the technical effectiveness of this invention. Figure 9 In order to be in Figure 8 The average pore radius values ​​obtained by weighted summation of the untransformed and dimensionally transformed groups were compared with the average pore radius value obtained by constant-rate mercury intrusion porosimetry. Because the same calibration method was used, the average pore radius values ​​calculated from the two sets of data have similar shapes, but due to differences in whether or not dimensional transformation was performed, there are significant numerical differences. As can be seen from the figure, the average pore radius of the four constant-rate mercury intrusion porosimetry blocks in this reservoir segment is closer to the average pore radius of the dimensionally transformed group, thus confirming that the pore radius obtained by dimensional transformation is closer to the actual reservoir conditions.

[0130] Therefore, in summary Figure 6 , Figure 7 The matching of the pore radius and relaxation time T2 of the cast thin sheet, and Figure 9 The comparison between the average pore radius calculated in the experiment and the average pore radius of constant-rate mercury intrusion confirms the superiority of the technical method of the present invention over existing methods.

[0131] This invention provides a method for calculating reservoir porosity distribution, which transforms the conventional planar cast thin-section porosity distribution characteristics into three-dimensional spatial characteristics, enhancing its matching with the nuclear magnetic resonance T2 distribution that reflects the three-dimensional spatial distribution of pores. This effectively improves the accuracy of reservoir porosity distribution calculation results and brings more convenience to reservoir feature analysis, oil and gas reservoir description, and other work.

[0132] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for converting the pore radius dimension of a cast thin sheet, characterized in that, Includes the following steps: Step 1: Conduct a casting thin section pore characteristic experiment on the core thin section sample to obtain the area frequency discrete distribution data corresponding to different pore radius ranges; Step 2: Perform trend fitting on the discrete distribution data of pore radius and area to obtain a continuous distribution curve; Step 3: Calculate the apparent volume frequency distribution data corresponding to different pore radius ranges; Step 31: In the continuous distribution curve obtained in step 2, a certain pore radius interval is divided into several micro-elements at equal intervals, and the area frequency of each micro-element is calculated using the continuous distribution curve function. Step 32: Based on the divided micro-element sequence and the area frequency of each micro-element, calculate the average pore radius of the corresponding pore radius interval, and then use the average pore radius and the area frequency of the corresponding pore interval to calculate the apparent volume frequency distribution of the corresponding pore radius interval. The average pore radius is calculated using Equation 1; Formula 1: In the formula, The average pore radius, in μm, corresponds to the pore radius range. This represents the area frequency within the corresponding pore radius range; It is a vector representing the area frequency of each micro-element within the corresponding pore radius range; It is a vector representing a sequence of infinitesimal pore radii within the corresponding pore range; Substituting the average pore radius and the area frequency of the corresponding pore range into Equation 2, the apparent volume frequency of the corresponding pore radius range is calculated. Formula 2: In the formula, This represents the apparent volume frequency within the corresponding pore radius range; This represents the area frequency corresponding to the pore radius range; The average pore radius, in μm, corresponds to the pore radius range. It is a vector, corresponding to the area frequency of each micro-element within the pore radius range; It is a vector representing a sequence of infinitesimal pore radii within the corresponding pore range; Step 33: Using the same calculation method as in Steps 31-32, obtain the apparent volume frequency distribution data for other pore radius ranges; Step 4: Correct the radius from the mathematical expectation value of the cut hole to the true value, and correct the boundary value of the pore radius interval corresponding to the apparent volume frequency data to obtain the final pore radius volume frequency distribution of the cast sheet.

2. The method for converting the pore radius dimension of a cast sheet according to claim 1, characterized in that, In step 1, the core thin section sample is taken from the end face of the plunger core sample; in step 1, during the porosity characteristic experiment of the cast thin section, the color of the gel injected into the cast thin section meets the following requirements: H channel value ≥ 150, S channel value ≥ 0.6, and V channel value ≥ 0.6 in the three channels of the HSV color space; in step 1, during the porosity characteristic experiment of the cast thin section, the analysis sampling interval does not exceed 10 μm.

3. The method for converting the pore radius dimension of a cast sheet according to claim 1, characterized in that, In step 2, Dirichlet boundary conditions are set for the discrete distribution data of pore radius area frequency, and the skewed distribution function is used to perform trend fitting on the pore radius area frequency data to obtain a continuous distribution curve.

4. The method for converting the pore radius dimension of a cast sheet according to claim 3, characterized in that, The Dirichlet boundary condition is as follows: when the pore radius approaches 0, its area frequency approaches 0 infinitely; when the pore radius approaches infinity, its area frequency approaches 0. In the formula, r is the radius obtained from the pore structure analysis of the cast sheet, and P is the area frequency corresponding to a certain pore radius.

5. The method for converting the pore radius dimension of a cast sheet according to claim 3, characterized in that, The frequency density function of the skewed distribution is: In the formula, F(r) is the area frequency density of the pores with radius r, r is the capillary radius (μm); a is the skewness; μ is the mean of the skewed distribution; and σ is the standard deviation of the skewed distribution.

6. The method for converting the pore radius dimension of a cast sheet according to claim 1, characterized in that, In step 4, Correcting the radius from the mathematical expectation value of the cut pore to the true value involves dividing the boundary value of the pore radius range by a conversion factor. The conversion factor is calculated using Equation 3: Formula 3 is: In the formula, The conversion factor is calculated to be a constant of 0.

866. R is the true pore radius of the near-spherical pores, in μm.

7. A method for characterizing reservoir pore structure, characterized in that, Includes the following steps: Step 1: Calculate the pore radius distribution of the cast sheet obtained by the pore radius dimension transformation method as described in any one of claims 1-6; Step 2: Test the nuclear magnetic resonance T2 distribution of the same core sample from Step 1; Step 3: Establish the correspondence between the pore radius distribution of the cast thin sheet obtained in Step 1 and the nuclear magnetic resonance T2 distribution obtained in Step 2; Step 4: Using the correspondence obtained in Step 3, convert the entire reservoir nuclear magnetic resonance logging data into a depth-varying pore radius distribution, thus obtaining the characterization results of the reservoir pore structure.

Images