A method for characterizing pore size distribution of tight sandstone based on nuclear magnetic resonance and peak coupling technology

Abstract

This invention discloses a method for characterizing the pore size distribution of tight sandstone based on nuclear magnetic resonance (NMR) and peak coupling technology, belonging to the field of tight sandstone pore size characterization technology. The method includes obtaining the fourth derivative curve of the full pore size distribution curve of the tight sandstone sample using Peakfit software, identifying possible sub-peak positions; using the Peakfit software to deconvolve the full pore size distribution curve using a log-normal distribution, and iterating using an iterative peak sharpening algorithm until the chi-square goodness-of-fit standard value no longer changes; and finally, using the correlation coefficient r... 2 The goodness of fit between the fitted model and the original measurement data is evaluated using standard error (SE) and F statistical goodness of fit. This invention is applied to the identification and characterization of pore size distribution in tight sandstone, overcoming various limitations and problems of traditional nuclear magnetic resonance T2 relaxation time testing in characterizing the pore size distribution of tight sandstone. It features fewer limitations and faster, more accurate analytical results.

Description

Technical Field

[0001] This invention belongs to the field of tight sandstone characterization technology, and particularly relates to a method for characterizing the pore size distribution of tight sandstone based on nuclear magnetic resonance and peak coupling technology. Background Technology

[0002] Tight sandstone reservoirs contain abundant tight oil and gas resources, but their complex micro- and nano-pore structures pose significant challenges to the prediction and production of tight sandstone oil reservoirs. Typically, the distribution and accumulation of tight oil and gas largely depend on the pore space of the tight reservoir; therefore, a comprehensive and detailed characterization of the pore size distribution of tight sandstone is crucial for assessing its storage and permeability. Currently, nuclear magnetic resonance (NMR) technology is widely used because it can non-destructively obtain the pore size distribution of the entire pore size range. However, the pore size distribution curve of tight sandstone obtained by converting NMR-T2 curves is often just an envelope formed by the superposition of multiple single peaks, which cannot accurately characterize the pore size distribution characteristics of different tight sandstones, thus hindering accurate prediction and assessment.

[0003] Chinese patent CN12284999A discloses a method for determining the pore size distribution of sandstone. This method obtains the T2 relaxation time distribution curve through nuclear magnetic resonance and the pore throat radius distribution curve through high-pressure mercury intrusion porosimetry. Based on the cumulative distribution curves of pore throat radius and T2 relaxation time, a polynomial fitting equation is established. A longitudinal equal-interval interpolation algorithm is used to obtain the T2 relaxation time and pore throat radius information corresponding to the same cumulative ratio. Then, by introducing the concept of fractal dimension, the micropore throat segment and the cumulative comparison range are determined. The inflection point of the fractal curve is used as the critical inflection point from micropores to macropores, i.e., the right end of the cumulative comparison interval. On this basis, a linear fitting algorithm is used to obtain the conversion coefficient C between T2 relaxation time and pore throat radius under water-saturated conditions. Finally, based on the conversion coefficient C and T2 relaxation time, the full pore size distribution of the sandstone sample is obtained. This method has the advantages of strong versatility and high quantification.

[0004] However, the pore size distribution of tight sandstone reservoirs is often the result of a long and comprehensive geological process, forming a continuum from sub-millimeter to nanometer scale. Larger pore throats can store and transport oil and gas, while smaller throats act as seals. In other words, different pore sizes and their proportions determine the likelihood of containing oil and gas resources. On the other hand, petroleum geologists are accustomed to using porosity and permeability to characterize reservoir rocks. However, for tight reservoirs, using pore size as a scale is more suitable than using porosity-permeability scale to consider oil and gas charging, because capillary pressure is inversely proportional to pore throat size. Therefore, accurately characterizing the full pore size distribution of tight sandstone is particularly important for providing valuable information for evaluation and prediction. However, current pore size distribution curves obtained using nuclear magnetic resonance (NMR) technology cannot highlight the advantages of micropores, mesopores, or macropores, thus failing to reflect the true pore size distribution characteristics. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the technical problem this invention aims to solve is that the current method of obtaining the full pore size distribution of tight sandstone using nuclear magnetic resonance (NMR) technology results in an envelope composed of multiple hidden single peaks, which cannot reflect the true pore size distribution. This invention proposes a method for rapidly and precisely fitting the pore size distribution curve obtained from NMR technology using peak-splitting coupling techniques. This method is of significant practical importance for accurately identifying and characterizing the full pore size distribution of tight sandstone, and for improving the prediction, exploration, and development of tight oil and gas reservoirs.

[0006] To solve the aforementioned technical problem, the technical solution adopted by the present invention is as follows:

[0007] This invention provides a method for characterizing the pore size distribution of tight sandstone based on nuclear magnetic resonance and peak-splitting coupling techniques, comprising:

[0008] The fourth derivative curves of the full pore size distribution curve of the dense sandstone sample were obtained using Peakfit software, and the positions of possible sub-peaks were identified.

[0009] The Peakfit software is used to deconvolve the full aperture distribution curve using a log-normal distribution, and an iterative peak sharpening algorithm is used to iterate until the chi-square goodness standard value no longer changes. After completion, the correlation coefficient r is used to... 2 The standard error SE and F statistical goodness of fit are used to evaluate the goodness of fit between the fitted model and the original measurement data.

[0010] Preferably, the fourth derivative curve of the full pore size distribution curve of the dense sandstone sample is obtained using Peakfit software, and the upward peak in the fourth derivative curve is the position of the possible sub-peak.

[0011] Preferably, the full pore size distribution curve is obtained by the following method:

[0012] Based on the cumulative distribution curve of the pore throat radius and the cumulative distribution curve of the T2 relaxation time, the conversion coefficient C between the T2 relaxation time and the pore throat radius is obtained by using the maximum correlation principle, the least squares method and the interpolation method, and the full pore diameter distribution curve is obtained.

[0013] Preferably, the cumulative distribution curve of the pore throat radius is obtained by testing the dense sandstone sample using a high-pressure mercury intrusion porosimeter; the cumulative distribution curve of the T2 relaxation time is obtained by testing the dense sandstone sample using a low-field nuclear magnetic resonance spectrometer.

[0014] Preferably, the method further includes a pretreatment step before testing the dense sandstone sample. The pretreatment step includes: performing pre-oil washing treatment on the dense sandstone sample using Soxhlet extraction, measuring the porosity and permeability values ​​of the dense sandstone sample, and observing the pore-throat size distribution characteristics of the micropore throat segment of the dense sandstone sample.

[0015] Preferably, the pretreatment step further includes performing low-field nuclear magnetic resonance (NMR) pretreatment and high-pressure mercury intrusion (HMI) pretreatment on the dense sandstone sample after the oil washing pretreatment.

[0016] Preferably, the pretreatment of the dense sandstone sample by Soxhlet extraction includes: placing the dense sandstone sample into an extraction tube and connecting it to a flask containing a mixed solvent of dichloromethane and methanol, introducing cooling water, and immersing the flask in a constant temperature water bath at 60-65°C for at least 72 hours until the solvent in the extraction tube is colorless and transparent.

[0017] Preferably, the pretreatment for the low-field nuclear magnetic resonance experiment includes: drying the dense sandstone sample in an oven at 100°C for 12 hours, and then pressurizing it with saturated liquid at 20 MPa for 24 hours in a vacuum pressurization saturation device.

[0018] Preferably, the pretreatment for the high-pressure mercury intrusion test includes: drying the dense sandstone sample in an oven at 100°C for 12 hours before conducting the high-pressure mercury intrusion test.

[0019] Preferably, the formula for calculating the conversion factor between T2 relaxation time and pore throat radius is as follows:

[0020] T2=C*r

[0021] In the above formula, T2 is the transverse relaxation time measured by nuclear magnetic resonance, in ms; r is the pore throat radius measured by high-pressure mercury intrusion, in μm.

[0022] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0023] This invention provides a method for characterizing the pore size distribution of tight sandstone based on nuclear magnetic resonance (NMR) and peak-splitting coupling technology. Based on the MNR-T2 spectrum, this method employs peak-splitting fitting technology to rapidly and precisely identify and characterize the pore size distribution of tight sandstone. The results of peak-splitting fitting are used to reveal and evaluate the differences in pore size distribution among different tight sandstones. This method is suitable for samples like tight sandstone, where pores span a large scale from sub-millimeter to nanometer. It effectively improves the precision and accuracy of NMR analysis of the full pore size distribution of tight sandstone, and features rapid, simple, and accurate analytical results. Attached Figure Description

[0024] Figure 1This is a scanning electron microscope image of the pore and throat size distribution of the M2001-1 dense sandstone sample from the Quantou Formation of the Cretaceous in the Songliao Basin, provided in an embodiment of the present invention.

[0025] Figure 2 The image shows the pore and throat size distribution of the FN272-1 Cretaceous Quantou Formation tight sandstone sample from the Songliao Basin, provided in an embodiment of the present invention.

[0026] Figure 3 The distribution curve of the saturated T2 spectrum of sample M2001-1 based on nuclear magnetic resonance analysis provided in this embodiment of the invention;

[0027] Figure 4 The distribution curve of the saturated T2 spectrum of sample FN272-1 based on nuclear magnetic resonance analysis provided in this embodiment of the invention;

[0028] Figure 5 The high-pressure mercury intrusion pore size distribution map and the pore size distribution map after T2 spectrum conversion of sample M2001-1 provided in the embodiments of the present invention;

[0029] Figure 6 The high-pressure mercury intrusion pore size distribution map and the pore size distribution map after T2 spectrum conversion of sample FN272-1 provided in the embodiments of the present invention are shown.

[0030] Figure 7 This is a schematic diagram of the nuclear magnetic resonance T2 spectrum and aperture r conversion of sample FN272-1 provided in an embodiment of the present invention;

[0031] Figure 8 The fourth derivative plot of the pore size distribution curve of sample FN272-1 obtained using Peakfit software is provided in an embodiment of the present invention.

[0032] Figure 9 The above is a diagram showing the deconvolution and peak fitting aperture distribution of sample FN272-1 provided in this embodiment of the invention.

[0033] Figure 10 The fourth derivative plot and deconvolution pore size distribution plot of sample M2001-1 provided in the embodiments of the present invention;

[0034] Figure 11 The fitting statistic r provided in this embodiment of the invention for different numbers of deconvolution sub-peaks 2 Graph showing the changes in the values ​​of standard error (SE) and F statistic;

[0035] Figure 12 The pore size distribution diagram of sample CPG-1 based on nuclear magnetic resonance analysis provided in the embodiments of the present invention;

[0036] Figure 13The pore size distribution diagram of sample CPG-2 based on nuclear magnetic resonance analysis provided in the embodiments of the present invention;

[0037] Figure 14 The pore size distribution diagram of the CPG-mix sample provided in the embodiment of the present invention is based on nuclear magnetic resonance analysis.

[0038] Figure 15 The fourth derivative plot of the pore size distribution curve of the CPG-mix sample provided in the embodiment of the present invention;

[0039] Figure 16 The diagram shows the deconvolution and peak fitting aperture distribution of the CPG-mix sample provided in this embodiment of the invention. Detailed Implementation

[0040] The technical solutions in specific embodiments of the present invention will be described in detail and completely below. Obviously, the described embodiments are only some specific implementations of the overall technical solution of the present invention, and not all implementations. Based on the overall concept of the present invention, all other embodiments obtained by those skilled in the art fall within the protection scope of the present invention.

[0041] This invention provides a method for characterizing the pore size distribution of tight sandstone based on nuclear magnetic resonance (NMR) and peak-splitting coupling techniques. The method includes: obtaining the fourth derivative curve of the full pore size distribution curve of a tight sandstone sample using Peakfit software, and identifying possible sub-peak positions; using the Peakfit software to deconvolve the full pore size distribution curve using a log-normal distribution to obtain multiple independent pore size groups, and iterating using an iterative peak sharpening algorithm until the chi-square goodness-of-fit standard no longer changes; and finally, using the correlation coefficient r... 2The standard error (SE) and F-squared goodness-of-fit are used to evaluate the fit between the fitted model and the original measurement data. Specifically, using the "Check Fourth Derivative" option in Peakfit software, the fourth derivative curve of the full aperture distribution curve is obtained to identify the possible sub-peak positions. In this step, the fourth derivative is calculated after two smoothing operations. In most cases, a first-level algorithm is not needed to reveal hidden peaks. In particularly difficult cases, a first-level Fourier domain procedure, such as FFT filtering or Gaussian convolution, may be required. The second-level smoothing algorithm is Savitzky-Golay. Solving the fourth derivative of the full aperture distribution curve can well indicate the minimum number of log-normal distributions. The specific distribution position is represented by the upward peak in the fourth-order plot. Using the "III Deconvolution Automatic Peak Identification" option in Peakfit software, the log-normal distribution function is selected to deconvolve the full aperture distribution curve, and an iterative "Peak Sharpening" algorithm is used to iterate until the chi-square goodness-of-fit standard value no longer changes. After completion, the correlation coefficient r can be used to determine the peak position. 2 The standard error (SE) and F-squared goodness-of-fit are used to evaluate the goodness of fit between the fitted model and the original measurement data. In this step, the generated fitting statistics can be used for trial-and-error deconvolution and peak fitting processes. The test criterion is r. 2 When the F-statistic reaches its maximum value or the standard error reaches its minimum value, increasing or decreasing the number of modeled distributions will lead to a worse fit. This method provides more reliable and realistic results for characterizing and studying the pore size distribution of tight sandstone, which is of great significance for the evaluation and prediction of tight sandstone reservoirs. Generally speaking, peak fitting is a subjective judgment, but the solution provided by this invention is a more objective and explicit method. It combines modal constraints derived from the fourth derivative with iterative deconvolution and peak fitting procedures to most accurately recreate the highest priority quantitative information. It has the characteristics of wide applicability, small limitations, and fast and accurate analysis results, and solves the various limitations and problems of traditional nuclear magnetic resonance T2 relaxation time testing in characterizing the pore size distribution of tight sandstone.

[0042] In a preferred embodiment, the full pore size distribution curve is obtained by the following method: based on the cumulative distribution curve of pore throat radius and the cumulative distribution curve of T2 relaxation time, the conversion coefficient C between T2 relaxation time and pore throat radius is obtained using the maximum correlation principle, least squares method, and interpolation method, thus obtaining the full pore size distribution curve. Specifically, the cumulative distribution curve of pore throat radius is obtained by testing the tight sandstone sample using a high-pressure mercury intrusion porosimeter; the cumulative distribution curve of T2 relaxation time is obtained by testing the tight sandstone sample using a low-field nuclear magnetic resonance spectrometer. Specifically, firstly, the cumulative distribution curve of pore throat radius based on high-pressure mercury intrusion porosimetry and the cumulative distribution curve of T2 relaxation time based on nuclear magnetic resonance are constructed. The conversion coefficient C between T2 relaxation time and pore throat radius is obtained using the maximum correlation principle, least squares method and interpolation method. Finally, the pore size distribution curve is obtained using the conversion coefficient T2 = C*r, where T2 is the transverse relaxation time measured by nuclear magnetic resonance, in ms; and r is the pore throat radius measured by high-pressure mercury intrusion porosimetry, in μm.

[0043] In a preferred embodiment, the method further includes a pretreatment step before testing the dense sandstone sample. The pretreatment step includes: performing a pre-washing treatment on the dense sandstone sample using Soxhlet extraction, measuring the porosity and permeability of the dense sandstone sample, and observing the pore-throat size distribution characteristics of the micropore throat segments of the dense sandstone sample. In this technical solution, the sample is first pre-washed for oil removal, then porosity and permeability tests are conducted. The micropore and throat size and distribution characteristics of the dense sandstone are observed under a field emission scanning electron microscope to verify the correspondence between the macroscopic physical properties and the micropore size distribution of the dense sandstone sample. Specifically, the pre-washing treatment of the dense sandstone sample using Soxhlet extraction includes: placing the dense sandstone sample in an extraction tube and connecting it to a flask containing a mixed solvent of dichloromethane and methanol, introducing cooling water, and immersing the flask in a constant temperature water bath at 60-65°C for at least 72 hours until the solvent in the extraction tube is colorless and transparent. Specifically, after placing the dense sandstone sample into the extraction tube, it is connected to a flask containing a mixed solvent of dichloromethane and methanol (93:7). Cooling water is introduced, and the flask is immersed in a constant temperature water bath at 60-65℃. The mixed solvent is continuously refluxed for extraction for at least 72 hours. During this period, the mixed solvent is replenished in time according to the evaporation situation until the solvent in the extraction tube is colorless and transparent, which means that all the oil in the pores of the dense sandstone has been extracted.

[0044] In a preferred embodiment, the pretreatment step further includes performing low-field nuclear magnetic resonance pretreatment and high-pressure mercury intrusion testing on the dense sandstone sample after the oil washing pretreatment.

[0045] In the above technical solution, the low-field nuclear magnetic resonance (NMR) experimental pretreatment includes: drying the dense sandstone sample in an oven at 100°C for 12 hours, and then pressurizing it with saturated liquid at 20 MPa in a vacuum pressurization saturation device for 24 hours; the high-pressure mercury intrusion (HMI) experimental pretreatment includes: drying the dense sandstone sample in an oven at 100°C for 12 hours, and then conducting HMI testing. Specifically, the sample after oil washing undergoes low-field NMR experimental pretreatment, including drying the dense sandstone sample in an oven at 100°C for 12 hours, then pressurizing it with saturated liquid at 20 MPa in a vacuum pressurization saturation device for 24 hours, and then placing the sample at the center of the NMR spectrometer sample chamber for testing. The CPMG sequence testing parameters are: echo interval of 0.08 ms, 10,000 echoes, waiting time of 10,000 ms, and scan time of 64 scans.

[0046] To more clearly and in detail introduce the method for characterizing the pore size distribution of tight sandstone based on nuclear magnetic resonance and peak coupling technology provided in the embodiments of the present invention, the following description will be made in conjunction with specific embodiments.

[0047] Example 1

[0048] Taking the Cretaceous Quantou Formation tight sandstone in the Songliao Basin as an example, this invention describes in detail the method for rapidly and accurately identifying and characterizing the pore size distribution of tight sandstone.

[0049] S1: Obtain the porosity and permeability of dense sandstone

[0050] The two dense sandstone samples, numbered M2001-1 and FN272-1, were measured to have porosity and permeability values ​​of 11.47% and 12.18% and 0.321 mD and 0.129 mD, respectively, after oil washing.

[0051] S2: Image-based method for determining the distribution characteristics of micropore-throat size

[0052] Micropore-throat characteristics of dense sandstone samples were observed using field emission scanning electron microscopy (FET). The FET was a Zeiss Sigma 500, and a QUORUM Q150 coating system was used for sputtering platinum (Pt) films. Figure 1 and Figure 2 The distribution reveals the two-dimensional pore throat characteristics of dense sandstone samples M2001-1 and FN272-1. Sample M2001-1 exhibits numerous irregular intergranular dissolution pores and intragranular dissolution pores connected by dissolution expansion fractures. In contrast, sample FN272-1 shows a significantly smaller pore diameter and significantly poorer pore throat connectivity.

[0053] S3: Pore size distribution of tight sandstone obtained based on nuclear magnetic resonance and high-pressure mercury intrusion experiments

[0054] Figure 3 and Figure 4 The distribution characteristics of the nuclear magnetic resonance T2 spectra of dense sandstone samples M2001-1 and FN272-1 are shown respectively. The former shows a typical bimodal distribution with the right peak being the dominant peak, indicating that sample M2001-1 is mainly composed of intergranular dissolution pores, followed by abundant intragranular dissolution pores; the latter shows a continuous single-peak distribution. Figure 5 and Figure 6 The results revealed the pore size distribution characteristics based on high-pressure mercury intrusion experiments. Sample M2001-1 showed a weak bimodal distribution, while FN272-1 showed a unimodal distribution.

[0055] S4: Determine the conversion factor C to obtain the full pore size distribution curve of the dense sandstone sample.

[0056] The conversion coefficient C between NMR-T2 relaxation time and pore throat radius was obtained using interpolation and least squares methods, thus acquiring the full pore size distribution curve. Taking sample M2001-1 as an example, the cumulative distribution curve of pore throat radius based on high-pressure mercury intrusion porosimetry and the cumulative distribution curve of T2 relaxation time based on nuclear magnetic resonance were first constructed. Figure 7 Using the cumulative frequency S(i) at any mercury porosimetry pore size r(i) as the standard, T2 is interpolated to obtain T2(i) corresponding to r(i); then, based on the least squares principle, r(i)-T2(i) is linearly fitted to obtain the C value with the smallest error; finally, the pore size distribution curve is obtained using the formula T2=C*r, which calculates the conversion coefficient between T2 relaxation time and pore throat radius. Following the above method, the C values ​​for samples M2001-1 and FN272-1 are 0.013 and 0.02 respectively, and the pore size distribution curves based on nuclear magnetic resonance can be obtained. Figure 5 , Figure 6 ).

[0057] S5: Peak fitting of the pore size distribution curve of tight sandstone was performed using Peakfit software.

[0058] After converting the pore size distribution curve data of the sample to a ".txt" format file, import the data into the Peakfit software interface by clicking the File-Import option and then perform peak fitting. Click Prepare-Section-Logarithmic XAxis, and then input appropriate abscissas (Xi and Xf) to ensure a certain horizontal distance on both sides of the peak to be fitted, and to center the peak as much as possible. Then click Apply New. Next, click Prepare-Inspect FourthDerivative to obtain the corresponding fourth derivative curve. Taking sample FN272-1 as an example, its fourth derivative curve displays five peaks within the effective range. Figure 8The pore size distribution curves of the CPG-mix are labeled D1-D5, meaning there may be five hidden sub-peaks. Then, click the AutoFit-AutoFit PeaksⅢDeconvolution option, select the Statistical-Log Normal Area function for Peak Type, add sub-peaks at the corresponding positions of D1-D5, and click Addl Adjust to iterate until the chi-square goodness-of-fit standard value or the r2 value no longer changes. Figure 9 This represents the highest degree of fit. For sample FN272-1, the correlation coefficient r between the fitted data and the original data is... 2 The value reached 0.9999, with a standard error (SE) of 3.14 and a statistical goodness of fit (F) of 113162, indicating a high degree of fit. The same method was used to process and perform peak fitting on the data for sample M2001, with the following results: Figure 10 As shown.

[0059] S6: Reliability analysis of fitting results

[0060] Using the correlation coefficient r 2 The optimal number of distributions for the peak splitting scheme is determined using three parameters: standard error (SE), statistical goodness of fit (F), and mean square error (r). This verifies the reliability of the peak splitting fit. 2 The larger the F value and the smaller the SE value, the higher the reliability. Figure 11 This demonstrates the correlation coefficient r under peak fitting with different numbers of sub-peaks. 2 The changes in standard error (SE) and statistical goodness of fit (F) were investigated. Examples were used with samples M2001-1 and FN272-1, attempting a 3-9 peak fit. It was found that the two samples achieved their maximum values ​​(r) with 4-peak and 5-peak distributions, respectively. 2 The F-statistic or minimum (SE) indicates that using the fourth derivative to find the number of hidden subpeaks is reliable.

[0061] Example 2

[0062] Taking controllable aperture glass with main apertures of 38nm and 50nm as examples, the reliability of peak fitting results obtained by modal constraints derived from the fourth derivative and iterative deconvolution and peak fitting procedures is verified.

[0063] S1: Sample preparation

[0064] Weigh 1g of controllable pore size glass with main pore sizes of 38nm and 50nm respectively, place them in glass bottles with a diameter of 2.5cm and label them CPG-1 and CPG-2 respectively. Then weigh 0.5g of controllable pore size glass with main pore sizes of 38nm and 50nm respectively and place them in glass bottles of the same size, labeling them CPG-mix. Place the glass bottles containing the samples in a vacuum saturation apparatus and saturate them with water for 12 hours.

[0065] S2: Using nuclear magnetic resonance experiments to test the pore size distribution of samples

[0066] The three prepared samples were placed in an NMRC12-010V nuclear magnetic resonance instrument manufactured by Numai Company for testing. The CPMG sequence test parameters were: echo interval of 0.08 ms, 10,000 echo sequences, waiting time of 10,000 ms, and scan time of 64 scans. Figure 12 , Figure 13 and Figure 14 The pore size distribution characteristics of samples CPG-1, CPG-2, and CPG-mix are shown respectively. Figure 12 and Figure 13 The main peak positions are consistent with those of CPG-1 and CPG-2 samples, with the main peak of CPG-mix located at 43 nm.

[0067] S3: Peak fitting using Peakfit software

[0068] After converting the pore size distribution curve data of CPG-mix to a ".txt" format file, import the data into the Peakfit software interface by clicking the File-Import option and then perform peak fitting. Click the Prepare-Section-Logarithmic XAxis option, and then input appropriate x-coordinates (Xi and Xf) to ensure a certain horizontal distance on both sides of the peak to be fitted, and to center the peak as much as possible. For this sample data, Xi and Xf are 0.01 and 1 respectively. Then click Apply New. Next, click the Prepare-Inspect Fourth Derivative option to obtain the corresponding fourth derivative curve (…). Figure 15The fourth derivative curve shows two peaks within the effective range, D1 and D2, which means that the aperture distribution curve of CPG-mix may have two hidden sub-peaks. Therefore, the x-coordinate positions of peaks D1 and D2 were recorded. The AutoFit-AutoFit PeaksⅢDeconvolution option was clicked, and the Peak Type was set to the Statistical-Log Normal Area function. Two sub-peaks were added at positions D1 and D2, and then the Addl Adjust function was clicked to iterate until the chi-square goodness-of-fit standard value or r was reached. 2 The value no longer changes. Figure 16 Finally, the correlation coefficient r 2 The value was 0.99, the standard error SE was 0.012, and the statistical goodness of fit F was 3763.76, indicating a high degree of fit. Furthermore, the peak positions of D1 and D2 were basically consistent with the main peak positions of samples CPG-1 and CPG-2.

Claims

1. A method for characterizing the pore size distribution of tight sandstone based on nuclear magnetic resonance and peak-splitting coupling techniques, characterized in that, include: The fourth derivative curves of the full pore size distribution curve of the dense sandstone sample were obtained using Peakfit software, and the positions of possible sub-peaks were identified. The Peakfit software is used to deconvolve the full aperture distribution curve using a log-normal distribution, and an iterative peak sharpening algorithm is used to iterate until the chi-square goodness standard value no longer changes. After completion, the correlation coefficient r is used to... 2 The standard error SE and F statistical goodness of fit are used to evaluate the goodness of fit between the fitted model and the original measurement data.

2. The method according to claim 1, characterized in that, The fourth derivative curve of the full pore size distribution curve of the dense sandstone sample was obtained using Peakfit software. The upward peak in the fourth derivative curve is the position of the possible sub-peak.

3. The method according to claim 1, characterized in that, The full aperture distribution curve was obtained through the following method: Based on the cumulative distribution curve of the pore throat radius and the cumulative distribution curve of the T2 relaxation time, the conversion coefficient C between the T2 relaxation time and the pore throat radius is obtained by using the maximum correlation principle, the least squares method and the interpolation method, and the full pore diameter distribution curve is obtained.

4. The method according to claim 3, characterized in that, The cumulative distribution curve of the pore throat radius was obtained by testing the tight sandstone sample using a high-pressure mercury intrusion porosimeter; the cumulative distribution curve of the T2 relaxation time was obtained by testing the tight sandstone sample using a low-field nuclear magnetic resonance spectrometer.

5. The method according to claim 4, characterized in that, It also includes a pretreatment step before testing the dense sandstone sample, the pretreatment step including: pre-washing the dense sandstone sample with oil using Soxhlet extraction, measuring the porosity and permeability of the dense sandstone sample, and observing the pore-throat size distribution characteristics of the micropore throat segment of the dense sandstone sample.

6. The method according to claim 5, characterized in that, The pretreatment step further includes performing low-field nuclear magnetic resonance pretreatment and high-pressure mercury intrusion pretreatment on the dense sandstone sample after the oil washing pretreatment.

7. The method according to claim 5, characterized in that, The pretreatment of the dense sandstone sample by Soxhlet extraction includes: placing the dense sandstone sample into an extraction tube and connecting it to a flask containing a mixed solvent of dichloromethane and methanol, introducing cooling water, and immersing the flask in a constant temperature water bath at 60-65°C for at least 72 hours until the solvent in the extraction tube is colorless and transparent.

8. The method according to claim 6, characterized in that, The pretreatment for the low-field nuclear magnetic resonance experiment includes: drying the dense sandstone sample in an oven at 100°C for 12 hours, and then pressurizing it with saturated liquid at 20 MPa for 24 hours in a vacuum pressurization saturation device.

9. The method according to claim 6, characterized in that, The pretreatment for the high-pressure mercury intrusion test includes: drying the dense sandstone sample in an oven at 100°C for 12 hours before conducting the high-pressure mercury intrusion test.

10. The method according to claim 3, characterized in that, The formula for calculating the conversion factor between T2 relaxation time and orifice throat radius is: T2=C*r In the above formula, T2 is the transverse relaxation time measured by nuclear magnetic resonance, in ms; r is the pore throat radius measured by high-pressure mercury intrusion, in μm.

Images