A method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs

Abstract

The present invention proposes a method for characterizing the microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs, and belongs to the field of carbon storage measurement technology. A first CT scan is performed on a core sample of a sandstone reservoir saturated with potassium iodide KI formation water to simulate the actual underground formation environment to displace the sample with carbon dioxide. When the sample just produces water, a second CT scan is performed, and carbon dioxide is continuously injected until it is displaced to carbon dioxide saturation. The water output is measured, and the sample is subjected to a third CT scan; the rock particles, water phase, and carbon dioxide in the CT image are ternary-valued to generate a three-dimensional model, and then the image value of carbon dioxide is extracted separately to form a carbon dioxide model, and the geometric parameters of the segmented carbon dioxide interconnect are used to characterize the microscopic characteristics of carbon dioxide in the core experimental sample, and the carbon dioxide storage form and the carbon dioxide storage amount in the sample are determined. This method has simple steps, high precision, and good use effect.

Description

Technical Field

[0001] The present invention belongs to the technical field of carbon sequestration measurement, and in particular relates to a method for characterizing microscopic characteristics of carbon dioxide sequestration in coal-bearing sandstone reservoirs. Background Art

[0002] Carbon dioxide geological storage is an effective negative carbon technology for addressing global climate change. Demonstration projects have been carried out in coal-bearing sandstone reservoirs. The occurrence characteristics of carbon dioxide in coal-bearing sandstone reservoirs, including the occurrence amount, occurrence form and its changing characteristics, are the key scientific basis for evaluating the geological storage potential of carbon dioxide, selecting storage sites, and selecting the formation for injecting carbon dioxide.

[0003] The storage characteristics of carbon dioxide in coal-bearing sandstone reservoirs are studied through physical simulation experiments of carbon dioxide displacement of water phase in core experimental samples. There are currently two main methods for measuring and analyzing carbon dioxide storage characteristics:

[0004] Based on the flow measurement of the carbon dioxide injection process, the characteristics of carbon dioxide in the sandstone reservoir are indirectly analyzed. This method measures the water output at the outlet of the displacement core test sample and uses the displacement water output to approximately replace the carbon dioxide storage in the core test sample. However, this method can only obtain one data, the carbon dioxide storage, which is indirectly calculated from the water output. It cannot characterize the differentiated storage form and change law of carbon dioxide in the core test sample (with heterogeneity).

[0005] Based on the nuclear magnetic resonance monitoring data of core test samples during the carbon dioxide injection process, the characteristics of carbon dioxide in sandstone reservoirs are indirectly analyzed. This method uses the nuclear magnetic resonance signals of the carbon dioxide injection process to analyze the distribution of water phase in pores of different sizes in the core test samples and infer the occurrence characteristics of carbon dioxide in the core test samples. This method cannot directly obtain the characteristic data of injected carbon dioxide in the core test samples, and can only indirectly describe the occurrence characteristics of carbon dioxide through water phase characteristics. It cannot characterize the occurrence amount, occurrence form and change characteristics of carbon dioxide in the core test samples. Summary of the invention

[0006] In view of the shortcomings of the existing technology, a method for characterizing the microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs is provided. Based on the CT scanning monitoring data of stages during the injection of carbon dioxide into core experimental samples, rock particles, water phase and carbon dioxide are ternary segmented, and a three-dimensional rock pore and fluid model is constructed. Then, the carbon dioxide fluid model is picked up to characterize its occurrence, occurrence form and change characteristics in the core experimental samples. That is, it can solve the problems of what form the carbon dioxide stored in the coal-bearing sandstone reservoir exists in and how much carbon dioxide in various forms there is, and realize direct quantitative characterization of the microscopic characteristics of carbon dioxide storage in the coal-bearing sandstone reservoir.

[0007] In order to achieve the above technical objectives, the present invention discloses a method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs, the steps of which are as follows:

[0008] Obtain core test samples of coal-bearing sandstone reservoirs from the reservoir;

[0009] The core test sample is placed in a sealed space, and the core test sample is vacuumed and saturated with formation water containing potassium iodide KI;

[0010] The core experimental sample in a sealed space and saturated with formation water containing potassium iodide KI is placed into the displacement support system, wherein a one-way carbon dioxide injection port and a one-way displacement outlet are provided on the sealed space, and the core experimental sample at this time is subjected to the first CT scan to obtain the status information of the core experimental sample saturated with formation water;

[0011] The displacement experiment was started at a constant pressure of 14MPa and a temperature of 35°C to simulate underground temperature and pressure conditions, and the formation water in the core test sample in the sealed space was displaced through the one-way carbon dioxide injection port; when water was observed to flow out of the displacement outlet, the core test sample was immediately subjected to a second CT scan to obtain the state information of the core test sample at the initial stage of carbon dioxide displacement; carbon dioxide was continuously injected into the core test sample until no water flowed out of the displacement outlet, at which time the core test sample was saturated with carbon dioxide, and the water output from the displacement outlet was measured, which was equal to the volume of carbon dioxide stored in the core test sample; the core test sample in the carbon dioxide saturated state was subjected to a third CT scan to obtain the occurrence characteristics of carbon dioxide in the core test sample in the carbon dioxide saturated state; the positions and angles of the three CT scans of the core test sample were the same;

[0012] Each CT scan acquires 1000 grayscale slice images of the core experimental sample. Each grayscale image consists of three parts: rock particles, water phase and carbon dioxide. All CT scan image data are ternary, and rock particles, water phase and carbon dioxide are represented by gray with different thresholds. The gray thresholds of rock particles, water phase and carbon dioxide are set according to the grayscale displayed by the CT scan data.

[0013] In the computer, the grayscale slice images of the three CT scans are used to construct the three-dimensional rock pore and fluid models corresponding to the three periods of the core experimental samples in the displacement experiment, and the three-valued grayscale segmentation thresholds of rock particles, water phase and carbon dioxide in the three-dimensional rock pore and fluid models are unified until the porosity of the three-dimensional rock pore and fluid model is close to the measured porosity of the core experimental samples, ensuring that the porosity of the three-dimensional rock pore and fluid model is reasonable; the three-dimensional rock pore and fluid model in the original state is established using the first CT scan information, and the three-dimensional rock pore and fluid model generated by the second and third CT scan information is used to obtain the change law of the microscopic occurrence characteristics of carbon dioxide;

[0014] The three-dimensional rock pore and fluid model generated by the third CT scanning information is divided into a large number of carbon dioxide models, and the carbon dioxide interconnected body is extracted from each carbon dioxide model. The geometric parameters of the carbon dioxide interconnected body are used to determine the occurrence type of the carbon dioxide interconnected body. The number of various carbon dioxide interconnected body occurrence types in the carbon dioxide model is counted to determine the occurrence type of the carbon dioxide interconnected body in all carbon dioxide models of the core experimental sample, and then the carbon dioxide occurrence form and carbon dioxide occurrence amount of the core experimental sample are determined according to the geometric parameters of the carbon dioxide interconnected body.

[0015] Furthermore, for the data of the first CT scan, since carbon dioxide has not been injected into the core experimental samples, the grayscale threshold is used to perform binary segmentation of the water phase and rock particles on each CT scan data image, wherein the water phase contains potassium iodide KI solution. The grayscale value of potassium iodide KI solution on the CT image is significantly higher than that of rock particles, and a three-dimensional rock pore and fluid model of the core experimental sample is constructed. The model porosity calculation result of the three-dimensional rock pore and fluid model should be equal to the measured porosity, otherwise the grayscale segmentation threshold is readjusted until the model porosity calculation result is the same as the measured porosity of the core experimental sample. According to actual conditions, the error between the model porosity calculation result and the measured porosity of the core experimental sample is allowed to exist within the preset threshold range.

[0016] Further, during the process of performing three-value segmentation on the data of the third CT scan for rock particles, water phase, and carbon dioxide to construct a three-dimensional rock pore and fluid model, since the rock particle part is exactly the same as the three-dimensional rock pore and fluid model constructed from the data of the first CT scan, it is only necessary to segment carbon dioxide from the water phase part. The volume of the segmented carbon dioxide needs to be the same as the water output measured at the displacement outlet end. An error within a preset threshold range is allowed between the two according to the actual situation. Otherwise, readjust the gray-scale segmentation threshold of carbon dioxide and the water phase until the volume of the segmented carbon dioxide is equal to the water output measured at the displacement outlet end. Denote the gray-scale thresholds after the three-value segmentation of the water phase and carbon dioxide as x, and apply x to the data of the second CT scan. Perform three-value segmentation on the data of the second CT scan and construct a three-dimensional rock pore and fluid model, so that the gray-scale levels of rock particles, water phase, and carbon dioxide in the three-dimensional rock pore and fluid models formed by the three CT scans are consistent.

[0017] Further, based on the three-dimensional rock pore and fluid model generated from the information of the third CT scan, pick out the part with a threshold less than x in the pore range of the three-dimensional rock pore and fluid model to generate a carbon dioxide model. Based on the fluid block function in the Avizo software, divide the carbon dioxide model into a large number of non-touching carbon dioxide connected bodies. Extract the carbon dioxide connected bodies in each carbon dioxide model, and use the geometric parameters of the carbon dioxide connected bodies to characterize the microscopic characteristics of carbon dioxide. The geometric parameters of the carbon dioxide connected bodies include: equivalent radius Q, shape factor G, Euler number E, and sphericity value S. The calculation formulas are as shown in formulas (1) to (4):

[0018] (1)

[0019] (2)

[0020] (3)

[0021] (4)

[0022] In the formulas, V is the volume of carbon dioxide in the carbon dioxide connected body, A is the surface area of carbon dioxide in the carbon dioxide connected body, C is the number of connected pore throats occupied by carbon dioxide in the carbon dioxide connected body, and H is the number of isolated pore throats occupied by carbon dioxide in the carbon dioxide connected body.

[0023] Further, divide each carbon dioxide connected body into four occurrence forms according to the calculation results of the geometric parameters of the carbon dioxide connected body:

[0024] When Q ≤ 20 μm, G < 2, 1 < S < 10, and E ≥ 0 are satisfied, it is determined that the carbon dioxide connected body is in a dispersed spherical state;

[0025] When Q ≤ 20 μm, G > 2, S > 10, and E ≥ 0 are satisfied, it is determined that the carbon dioxide connectivity is banded;

[0026] When 20 < Q < 50 μm, 2 < G < 100, 1 < S < 10, and E ≥ 0 are satisfied, it is determined that the carbon dioxide connectivity is membranous;

[0027] When Q > 50 μm, G > 100, S > 10, and E ≤ -1 are satisfied, it is determined that the carbon dioxide connectivity is in a clustered state;

[0028] After classifying the occurrence forms of the determined carbon dioxide connectivity, the occurrence amounts and volume ratios of various carbon dioxide occurrence forms in the carbon dioxide connectivity can be directly obtained using Avizo software; by comparing the three-dimensional rock pore and fluid models of the initial injection of carbon dioxide in the core experimental sample and the three-dimensional rock pore and fluid models in the carbon dioxide saturation state, the change in the carbon dioxide volume ratio is shown, thus revealing the change process of carbon dioxide from initial injection to saturation.

[0029] Furthermore, carbon sequestration site selection is carried out according to the occurrence forms of the carbon dioxide connectivity in the core experimental sample:

[0030] Core experimental samples are collected from the reservoir. If more than half of the carbon dioxide connectivity in all the carbon dioxide connectivity in the core experimental sample is in the clustered form, a reservoir with better pore connectivity than that of the core experimental sample and with a permeability exceeding that of the core experimental sample needs to be selected as the target sequestration layer; if more than half of the carbon dioxide connectivity in all the carbon dioxide connectivity in the core experimental sample is in the membranous form, a reservoir with a specific surface area exceeding that of the core experimental sample needs to be selected as the target sequestration layer.

[0031] Furthermore, the occurrence forms corresponding to the carbon dioxide connectivity in the core experimental sample are used to judge the safety of carbon sequestration:

[0032] If the core experimental sample shows that the stored carbon dioxide is mainly in the clustered form, it indicates that the carbon sequestration stability of the reservoir type represented by the core experimental sample is good and the safety risk of carbon dioxide leakage is small; if the core experimental sample shows that the stored carbon dioxide is mainly in the spherical form, it indicates that the carbon sequestration stability of the reservoir type represented by the core experimental sample is poor and the safety risk of carbon dioxide leakage is large.

[0033] A computer device includes: a processor, a memory, and a network interface;

[0034] The processor is connected to the memory and the network interface. Among them, the network interface is used to provide network communication functions, the memory is used to store program codes, and the processor is used to call the program codes to execute the method for characterizing the microscopic characteristics of carbon dioxide sequestration in coal measure sandstone reservoirs.

[0035] A computer-readable storage medium stores a computer program, which is suitable for being loaded by a processor and executed by the method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs.

[0036] Beneficial effects: This method can visualize the state of carbon dioxide storage in coal-bearing sandstone reservoirs: based on CT scanning data, a three-dimensional rock pore and fluid model is constructed, and the carbon dioxide injected into the core test sample is intuitively displayed in the three-dimensional space of the core test sample; this method realizes the quantitative characterization of the carbon dioxide occurrence amount and occurrence form in the coal-bearing sandstone reservoir: according to the three-dimensional rock pore and fluid model, the quantitative standard of different occurrence forms is determined based on geometric parameters, and the occurrence amount difference of carbon dioxide in different occurrence forms can be obtained; and the volume proportion change of various forms of carbon dioxide in the core test sample at different injection stages is revealed, The law of carbon dioxide occurrence change in the coal-bearing sandstone reservoir is revealed; based on the CT scanning monitoring data of the core test sample before the injection of carbon dioxide, this method constructs a three-dimensional rock pore and fluid model, and can arbitrarily extract the carbon dioxide occurrence amount at a local position of the core or the entire core space, as well as the occurrence amount of carbon dioxide in different occurrence forms. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flow chart of a method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs in an embodiment of the present invention.

[0038] FIG. 2 ( a ) is an original image of carbon dioxide storage in a coal-bearing sandstone reservoir scanned by CT in an embodiment of the present invention.

[0039] FIG. 2( b ) is a schematic diagram of the ternary-valued carbon dioxide storage map in the coal-bearing sandstone reservoir scanned by CT in an embodiment of the present invention.

[0040] FIG. 3( a ) is a schematic diagram of a connected body of carbon dioxide in a spherical state picked up from a coal-bearing sandstone reservoir in an embodiment of the present invention.

[0041] FIG3( b ) is a schematic diagram of a connected body of carbon dioxide in a strip-like state picked up from a coal-bearing sandstone reservoir in an embodiment of the present invention.

[0042] FIG3( c ) is a schematic diagram of a connected body of carbon dioxide in a film-like state picked up from a coal-bearing sandstone reservoir in an embodiment of the present invention.

[0043] FIG3( d ) is a schematic diagram of a connected body of carbon dioxide in a clustered state picked up from a coal-bearing sandstone reservoir in an embodiment of the present invention.

[0044] Figure 4 It is a histogram of the volume proportions of connected bodies of carbon dioxide in different storage forms in the embodiments of the present invention. DETAILED DESCRIPTION

[0045] The implementation of the present invention will be further described below in conjunction with the accompanying drawings:

[0046] like Figure 1 As shown, the present invention discloses a method for characterizing the microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs. The method characterizes the carbon dioxide storage form and storage amount based on CT scanning monitoring data and modeling during the injection of carbon dioxide into core experimental samples. The specific steps are as follows:

[0047] Preparation of core experimental samples: cores of typical coal-bearing sandstone reservoirs were selected from the Permian Lower Shihezi Formation in Huainan Coalfield, Anhui Province, and made into plug-like core experimental samples with a length of 5 cm and a diameter of 2.5 cm. The grayscale value of the KI-containing water phase on the CT image was significantly high, and the simulated formation water was dyed with KI. The core experimental samples were vacuumed and saturated with KI-containing formation water to facilitate the segmentation of the water phase in the CT information of the core experimental samples.

[0048] Carbon dioxide injection and CT scanning: In order to obtain the characterization data of carbon dioxide storage in sandstone reservoirs, the core test samples need to be scanned by CT three times; the core test samples are placed on the displacement support system, and the core test samples are scanned by CT for the first time before carbon dioxide is injected; then carbon dioxide is injected into the core test samples at a constant pressure of 14MPa and a temperature of 35°C. After water is observed to flow out of the displacement outlet, the core test samples are immediately scanned by CT for the second time; after the scan, carbon dioxide is continued to be injected until no water flows out of the displacement outlet. The core test samples are saturated with carbon dioxide, the water output is measured, and the core test samples are scanned by CT for the third time. The three CT scans are in the same position.

[0049] Tri-valued CT scan data: The data obtained from each CT scan of the core experimental sample is 1000 grayscale images, each of which consists of three parts: rock particles, water phase and carbon dioxide. Different grayscale values ​​are set for rock particles, water phase and carbon dioxide respectively, which can facilitate the segmentation of them by grayscale threshold, that is, the tri-valued CT scan data. If Figure 2 (a) shows the original image generated by CT scanning of the core experimental sample taken from the coal-bearing sandstone reservoir, Figure 2 (b) is the image after the tri-valued original image; for the data of the first CT scan, since carbon dioxide has not been injected into the core experimental sample at this time, the grayscale threshold is used to perform binary segmentation of the water phase and rock particles. The water phase is formation water containing KI solution, and its grayscale value on the CT image is significantly higher than that of the rock particles. After completing the binary segmentation of each image obtained by this CT scan, a three-dimensional rock pore and fluid model corresponding to the second CT scan core experimental sample is constructed. The three-dimensional rock pore and fluid model The porosity calculated by the volume model should be approximately the same as the measured porosity of the core experimental sample. Since the values ​​of the two can only be approximate and not completely the same in actual work, as long as the data error between the two is within the preset threshold range, the values ​​of the two are considered to be the same, otherwise the grayscale segmentation threshold is readjusted until the model porosity is similar to the measured porosity of the core experimental sample; then the data of the third CT scan are ternary segmented into rock particles, water phase and carbon dioxide, and a three-dimensional rock pore and fluid model is constructed: since the rock particle part does not change during the process of using carbon dioxide to displace formation water, there is no need to consider the rock particle part, and it is only necessary to segment carbon dioxide from the water phase part. In the CT image, the grayscale value of carbon dioxide is set between the water phase and the rock particles; the volume of segmented carbon dioxide needs to be approximately the same as the water output measured at the displacement outlet, otherwise the grayscale segmentation threshold of carbon dioxide and water phase is readjusted until the volume of segmented carbon dioxide is similar to the water output measured at the displacement outlet. Assume that the grayscale threshold of water phase and carbon dioxide is x, the grayscale threshold x is greater than x for water phase, and less than x for carbon dioxide; the grayscale threshold x is applied to the data of the second CT scan to construct a three-dimensional rock pore and fluid model, so that the grayscale levels of rock particles, water phase and carbon dioxide in the three-dimensional rock pore and fluid model formed by three CT scans remain consistent.

[0050] Characterization of CO2 microscopic characteristics: The CO2 model was extracted from the three-dimensional rock pore and fluid model constructed by the third CT scan using the grayscale threshold x. The fluid segmentation function in the avizo software was used to segment the CO2 model into multiple non-contacting CO2 connected bodies. The geometric parameters of each CO2 connected body were calculated, including the equivalent radius Q, shape factor G, Euler number E, and sphericity value S. The calculation formulas are as follows: (1) to (4).

[0051] (1)

[0052] (2)

[0053] (3)

[0054] (4)

[0055] Wherein, V is the volume of carbon dioxide in the carbon dioxide connectivity, A is the surface area of carbon dioxide in the carbon dioxide connectivity, C is the number of connected pore throats occupied by carbon dioxide in the carbon dioxide connectivity, and H is the number of isolated pore throats occupied by carbon dioxide in the carbon dioxide connectivity.

[0056] Classify each carbon dioxide connectivity according to the calculation results of the geometric parameters of the carbon dioxide connectivity, and there are four types of occurrence forms in total:

[0057] When Q ≤ 20um, G < 2, 1 < S < 10, and E ≥ 0 are satisfied, it is judged that the carbon dioxide connectivity is in a dispersed spherical state, as shown in Fig. 3(a);

[0058] When Q ≤ 20um, G > 2, S > 10, and E ≥ 0 are satisfied, it is judged that the carbon dioxide connectivity is in a banded state, as shown in Fig. 3(b);

[0059] When 20 < Q < 50um, 2 < G < 100, 1 < S < 10, and E ≥ 0 are satisfied, it is judged that the carbon dioxide connectivity is in a membranous state, as shown in Fig. 3(c);

[0060] When Q > 50um, G > 100, S > 10, and E ≤ -1 are satisfied, it is judged that the carbon dioxide connectivity is in an aggregated cluster state, as shown in Fig. 3(d);

[0061] The above four occurrence forms are specifically shown in Fig. 3(a), Fig. 3(b), Fig. 3(c) and Fig. 3(d). Visual markings are made on the carbon dioxide connectivities with different occurrence forms using different colors for easy distinction:

[0062] Based on the carbon dioxide fluid model, it can not only characterize carbon dioxide with different occurrence forms, but also directly obtain the occurrence amount and its volume ratio of carbon dioxide in various occurrence forms. By comparing the digital models of the initial injection stage and the saturated carbon dioxide stage of carbon dioxide, the change in the volume ratio of carbon dioxide in various occurrence states can show the change characteristics of the occurrence characteristics during the carbon dioxide injection process; for example, from the initial injection stage to the saturated stage, the volume ratio of cluster-shaped carbon dioxide increases from 20% to 95%, indicating that the sequestration of carbon dioxide in the coal-bearing sandstone reservoir is a process of gradually aggregating from a dispersed state into cluster-shaped carbon dioxide.

[0063] Embodiment: Selecting a carbon storage site based on the occurrence morphology of carbon dioxide interconnected bodies in core test samples: Collect core test samples from reservoirs. If more than half of the occurrence morphologies of all carbon dioxide interconnected bodies in the core test samples are in clustered morphology, a reservoir with a permeability higher than that of the core test sample and better pore connectivity needs to be selected as the target layer for storage. Using the sandstone core sample experiment of the Lower Shihezi Formation in the Huainan Coalfield, a pore structure model was constructed based on the first CT scan data, and the model porosity (4.02%) was close to the measured porosity (4.09%). Figure 4 As shown in the figure, in the initial CO2 injection stage, the volume proportions of strip-shaped and film-shaped CO2 are 48.8% and 49.4% respectively; in the CO2 saturation stage, the volume proportion of cluster-shaped CO2 is 81.2%, indicating that the injected CO2 in the core experimental sample changes from strip-shaped and film-shaped to cluster-shaped. The sandstone of the Lower Shihezi Formation in the Huainan Coalfield where the sample is located mainly stores CO2 in the form of clusters. The reservoir with a permeability exceeding that of the core experimental sample is a favorable storage target. If more than half of the CO2 interconnected bodies in the core experimental sample are in the form of membrane, then the reservoir with a specific surface area exceeding that of the core experimental sample should be selected as the storage target layer. Using the core experimental data, the CO2 in the core experimental sample is more than 100% of the CO2 in the core experimental sample. The safety of carbon storage is judged by the occurrence form of the connected carbon dioxide bodies in the test samples: if the core test samples show that the stored carbon dioxide is mainly in clustered form, it means that the reservoir type represented by the core test samples has good carbon sequestration stability, and the safety risk of stored carbon dioxide escaping is small; using the sandstone core sample of the Lower Shihezi Formation of the Huainan Coalfield, in the carbon dioxide saturation stage, the volume proportion of clustered carbon dioxide is large, which is 81.2%, indicating that the carbon sequestration stability of the reservoir is good; if the core test samples show that the stored carbon dioxide is mainly in spherical form, it means that the reservoir type represented by the core test samples has poor carbon sequestration stability, and the safety risk of stored carbon dioxide escaping is high.

Claims

1. A method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs, characterized in that: Here are the steps: Obtain core test samples of coal-bearing sandstone reservoirs from the reservoir; The core test sample is placed in a sealed space, and the core test sample is vacuumed and saturated with formation water containing potassium iodide KI; The core experimental sample in a sealed space and saturated with formation water containing potassium iodide KI is placed into the displacement support system, wherein a one-way carbon dioxide injection port and a one-way displacement outlet are provided on the sealed space, and the core experimental sample at this time is subjected to the first CT scan to obtain the status information of the core experimental sample saturated with formation water; The displacement experiment was started at a constant pressure of 14MPa and a temperature of 35°C to simulate underground temperature and pressure conditions, and the formation water in the core test sample in the sealed space was displaced through the one-way carbon dioxide injection port; when water was observed to flow out of the displacement outlet, the core test sample was immediately subjected to a second CT scan to obtain the state information of the core test sample at the initial stage of carbon dioxide displacement; carbon dioxide was continuously injected into the core test sample until no water flowed out of the displacement outlet, at which time the core test sample was saturated with carbon dioxide, and the water output from the displacement outlet was measured, which was equal to the volume of carbon dioxide stored in the core test sample; the core test sample in the carbon dioxide saturated state was subjected to a third CT scan to obtain the occurrence characteristics of carbon dioxide in the core test sample in the carbon dioxide saturated state; the positions and angles of the three CT scans of the core test sample were the same; Each CT scan acquires 1000 grayscale slice images of the core experimental sample. Each grayscale image consists of three parts: rock particles, water phase and carbon dioxide. All CT scan image data are ternary, and rock particles, water phase and carbon dioxide are represented by gray with different thresholds. The gray thresholds of rock particles, water phase and carbon dioxide are set according to the grayscale displayed by the CT scan data. In the computer, the grayscale slice images of the three CT scans are used to construct the three-dimensional rock pore and fluid models corresponding to the three periods of the core experimental samples in the displacement experiment, and the three-valued grayscale segmentation thresholds of rock particles, water phase and carbon dioxide in the three-dimensional rock pore and fluid models are unified until the porosity of the three-dimensional rock pore and fluid model is close to the measured porosity of the core experimental samples, ensuring that the porosity of the three-dimensional rock pore and fluid model is reasonable; the three-dimensional rock pore and fluid model in the original state is established using the first CT scan information, and the three-dimensional rock pore and fluid model generated by the second and third CT scan information is used to obtain the change law of the microscopic occurrence characteristics of carbon dioxide; The three-dimensional rock pore and fluid model generated by the third CT scanning information is divided into a large number of carbon dioxide models, and the carbon dioxide interconnected body is extracted from each carbon dioxide model. The geometric parameters of the carbon dioxide interconnected body are used to determine the occurrence type of the carbon dioxide interconnected body. The number of various carbon dioxide interconnected body occurrence types in the carbon dioxide model is counted to determine the occurrence type of the carbon dioxide interconnected body in all carbon dioxide models of the core experimental sample, and then the carbon dioxide occurrence form and carbon dioxide occurrence amount of the core experimental sample are determined according to the geometric parameters of the carbon dioxide interconnected body.

2. The method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs according to claim 1, characterized in that: For the data of the first CT scan, since carbon dioxide has not been injected into the core experimental sample yet, each CT scan data image is binarized and segmented into water phase and rock particles using a gray-scale threshold. The water phase contains potassium iodide (KI) solution. By taking advantage of the fact that the gray-scale value of the potassium iodide (KI) solution on the CT image is significantly higher than that of the rock particles, a three-dimensional rock pore and fluid model of the core experimental sample is constructed. The calculated result of the model porosity of the three-dimensional rock pore and fluid model should be equal to the measured porosity. Otherwise, the gray-scale segmentation threshold is readjusted until the calculated result of the model porosity is equal to the measured porosity of the core experimental sample. According to the actual situation, an error within a preset threshold range is allowed between the calculated result of the model porosity and the measured porosity of the core experimental sample.

3. The method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs according to claim 2, characterized in that: During the process of performing ternary segmentation of rock particles, water phase, and carbon dioxide on the data of the third CT scan and constructing a three-dimensional rock pore and fluid model, since the rock particle part is exactly the same as the three-dimensional rock pore and fluid model constructed from the data of the first CT scan, only carbon dioxide needs to be segmented out from the water phase part. The volume of the segmented carbon dioxide should be the same as the measured water output at the displacement outlet end. According to the actual situation, an error within a preset threshold range is allowed between the two. Otherwise, the gray-scale segmentation thresholds of carbon dioxide and the water phase are readjusted until the volume of the segmented carbon dioxide is equal to the measured water output at the displacement outlet end. Denote the gray-scale thresholds after ternary segmentation of the water phase and carbon dioxide as x, and apply x to the data of the second CT scan. Perform ternary segmentation on the data of the second CT scan and construct a three-dimensional rock pore and fluid model, so that the gray-scale color levels of rock particles, water phase, and carbon dioxide in the three-dimensional rock pore and fluid models formed by the three CT scans are consistent.

4. The method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs according to claim 3, characterized in that: Based on the three-dimensional rock pore and fluid model generated from the information of the third CT scan, pick out the part with a threshold less than x in the pore range of the three-dimensional rock pore and fluid model to generate a carbon dioxide model. And based on the fluid block function in the Avizo software, divide the carbon dioxide model into a large number of non-touching carbon dioxide connected bodies. Extract the carbon dioxide connected bodies in each carbon dioxide model, and use the geometric parameters of the carbon dioxide connected bodies to characterize the microscopic characteristics of carbon dioxide. The geometric parameters of the carbon dioxide connected bodies include: equivalent radius Q, shape factor G, Euler number E, and sphericity value S. The calculation formulas are shown in Equations (1) to (4): （1） （2） （3） （4） In the formulas, V represents the volume of carbon dioxide in the carbon dioxide connected body, A represents the surface area of carbon dioxide in the carbon dioxide connected body, C represents the number of connected pore throats occupied by carbon dioxide in the carbon dioxide connected body, and H represents the number of isolated pore throats occupied by carbon dioxide in the carbon dioxide connected body.

5. The method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs according to claim 4, characterized in that: According to the calculation results of the geometric parameters of the carbon dioxide connected bodies, divide each carbon dioxide connected body into four occurrence forms: When Q ≤ 20 μm, G < 2, 1 < S < 10, and E ≥ 0, it is determined that the carbon dioxide connected body is in a dispersed spherical state; When Q ≤ 20 μm, G > 2, S > 10, and E ≥ 0, it is determined that the carbon dioxide connected body is in a banded state; When 20 < Q < 50 μm, 2 < G < 100, 1 < S < 10, and E ≥ 0 are satisfied, it is determined that the carbon dioxide connectivity is membranous; When Q > 50 μm, G > 100, S > 10, and E ≤ -1 are satisfied, it is determined that the carbon dioxide connectivity is in a clustered state; After classifying the occurrence forms of the determined carbon dioxide connectivity, the occurrence amounts and their volume ratios of various carbon dioxide occurrence forms in the carbon dioxide connectivity can be directly obtained using Avizo software; by comparing the three-dimensional rock pore and fluid models of the initial injection of carbon dioxide in the core experimental sample and the three-dimensional rock pore and fluid models under the carbon dioxide saturation state, the change in the carbon dioxide volume ratio is shown, thus revealing the change process of carbon dioxide from initial injection to saturation.

6. The method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs according to claim 5, characterized in that: Conduct carbon sequestration site selection based on the occurrence forms of the carbon dioxide connectivity in the core experimental sample: Collect core experimental samples from the reservoir. If more than half of the carbon dioxide connectivity in all the carbon dioxide connectivity in the core experimental sample is in a clustered form, a reservoir with better pore connectivity than the core experimental sample, with a permeability exceeding that of the core experimental sample, needs to be selected as the target storage layer; if more than half of the carbon dioxide connectivity in all the carbon dioxide connectivity in the core experimental sample is in a membranous form, a reservoir with a specific surface area exceeding that of the core experimental sample needs to be selected as the target storage layer.

7. The method for characterizing microscopic characteristics of carbon dioxide storage in coal-bearing sandstone reservoirs according to claim 6, characterized in that: Use the occurrence forms corresponding to the carbon dioxide connectivity in the core experimental sample to judge the safety of carbon sequestration: If the core experimental sample shows that the stored carbon dioxide is mainly in a clustered form, it indicates that the carbon sequestration stability of the reservoir type represented by the core experimental sample is good, and the safety risk of the stored carbon dioxide escaping is small; if the core experimental sample shows that the stored carbon dioxide is mainly in a spherical form, it indicates that the carbon sequestration stability of the reservoir type represented by the core experimental sample is poor, and the safety risk of the stored carbon dioxide escaping is large.

Images