Coal rock dynamic capillary force determination method

Abstract

This invention relates to the field of unconventional oil and gas field development technology, specifically to a method for determining dynamic capillary force in coal and rock. The method includes: selecting target core samples to prepare coal samples and obtaining the full-scale pore size distribution of the coal samples; constructing a three-dimensional random pore network model based on the full-scale pore size distribution; obtaining a topology-corrected static pore network model; establishing a dynamic model of contact angle variation with pore pressure; establishing dynamic deformation equations for pore and throat radii variation with effective stress; and simulating the gas-driven water process under different pore pressures based on an intrusion permeation algorithm, outputting capillary force-saturation curves under different pore pressure conditions, and forming a dynamic capillary force surface plot. This application has significant application value and promising prospects in the field of unconventional oil and gas field development, significantly improving the scientific nature of reservoir seepage behavior characterization and the rationality of development engineering decisions.

Description

Technical Field

[0001] This invention relates to the field of unconventional oil and gas field development technology, specifically to a method for determining dynamic capillary force in coal and rock. Background Technology

[0002] In coalbed methane development, the core process involves depressurizing and draining water to desorb methane from the coal seam. This process is accompanied by a continuous decrease in reservoir pore pressure, leading to a significant increase in effective stress acting on the coal seam and altering the thermodynamic state of the gas-liquid-solid three-phase interface. Unlike conventional sandstone reservoirs, coal seams are highly stress-sensitive. During depressurization and drainage, the increased effective stress forces the closure of fractures (cleavages) and matrix contraction, resulting in dynamic compression of the pore structure. Simultaneously, changes in pressure and density alter the wettability of the coal seam surface, manifested as a dynamic drift in the contact angle. However, current standard methods for measuring rock capillary forces, such as high-pressure mercury intrusion porosimetry or centrifugation, are based on static assumptions: experiments are conducted under constant pressure, assuming the rock is a rigid body with fixed wettability. These methods completely ignore the two key physical effects of dynamic compression of the pore structure and dynamic evolution of wettability throughout the entire lifecycle of coalbed methane drainage.

[0003] This static characterization is disconnected from the dynamic nature of reservoirs, leading to serious engineering problems: First, it underestimates the flow resistance during the water production period because it fails to consider that the narrowing of the pore throat due to increased stress raises the capillary force threshold; second, it causes distortion in gas-water production prediction because ignoring changes in wettability affects the judgment of gas breakthrough time; finally, for soft materials such as coal, the high-pressure mercury intrusion method misjudges the compressed volume of the coal matrix as the mercury ingress volume in the high-pressure section, thus systematically overestimating the micropore volume and distorting the true pore size distribution.

[0004] Therefore, existing static capillary force characterization methods cannot accurately reflect the actual seepage behavior of coalbed methane reservoirs. In order to achieve accurate prediction of production capacity and optimization of development plans, it is urgent to establish a dynamic capillary force determination method that can couple stress deformation, dynamic wettability and actual pore structure. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a method for determining the dynamic capillary force of coal and rock, which solves the problem that conventional capillary force measurement methods are unable to reflect the evolution law of capillary force caused by the increase of effective stress, compression of pore structure, and changes in wettability parameters with pressure during the drainage and depressurization process of coal and rock.

[0006] To solve the above problems, the technical solution of the present invention is as follows: A method for determining the dynamic capillary force of coal and rock, comprising: S1. Select the target core to prepare coal samples and obtain the full-scale pore size distribution of the coal samples; S2. Construct a three-dimensional random pore network model based on the full-scale pore size distribution; S3. The real pore network topology of coal and rock is obtained by focusing ion beam scanning electron microscopy, and the topology of the three-dimensional random pore network model constructed in S2 is corrected to obtain the topology-corrected static pore network model. S4. Conduct coal and rock wetting angle experiments under varying pressure conditions to establish a dynamic model of the contact angle as a function of pore pressure. S5. Conduct coal and rock stress sensitivity experiments and establish dynamic deformation equations for pore and throat radii as a function of effective stress. S6. Based on the intrusion percolation algorithm, combined with the static pore network model described in S3, the dynamic model described in S4, and the dynamic deformation equation described in S5, the gas-driven water process under different pore pressures is simulated, and the capillary force-saturation curves under different pore pressure conditions are output to form a dynamic capillary force surface plot.

[0007] Furthermore, S1 includes: S11. Perform high-pressure mercury intrusion testing on the coal sample. Correct the high-pressure mercury intrusion test data for coal matrix compression. Based on the corrected mercury intrusion pressure data, use the Washburn equation to convert the mercury intrusion pressure into an equivalent pore throat radius to obtain the high-pressure mercury intrusion pore size distribution. ; S12. Conduct low-temperature liquid nitrogen adsorption tests on coal samples, determine nitrogen adsorption isotherms, and analyze the adsorption isotherms based on the BJH model to obtain the low-temperature liquid nitrogen adsorption pore size distribution. ; S13, will and Data fusion was performed to obtain the full-scale aperture distribution.

[0008] Furthermore, in S11, the coal matrix compression correction for the high-pressure mercury intrusion test data includes: Let the volume compressibility of coal matrix be... The sample volume is The initial pressure of mercury porosimetry is When the mercury porosimetry pressure is hourly coal matrix compression volume , The cumulative mercury injection volume recorded by the mercury porosimeter is denoted as The actual cumulative mercury ingress volume in the pores after removing the influence of coal matrix compression is then calculated. .

[0009] Furthermore, in S11, the expression for the Washburn equation is: , in, The equivalent orifice throat radius, For mercury-gas interfacial tension, The contact angle of mercury on the coal surface is denoted as .

[0010] Furthermore, in S13, data fusion includes: Determine the overlapping pore size range between cryogenic liquid nitrogen adsorption and high-pressure mercury intrusion data. 1, 2]; A weighted average method was used to fuse the apertures within the overlapping aperture range. The resulting full-scale aperture distribution was as follows: ,in, , They are respectively , Weight; Single-source data is used in non-overlapping intervals when < Take at 1 o'clock ,when > Take at 2 o'clock .

[0011] Furthermore, S2 includes: The full-scale pore size distribution is transformed into three types of discrete parameters: pore volume size distribution, throat size distribution, and connectivity. For each connecting pore body and The larynx, larynx length Calculated by the following formula: ,in, , The coordinates of the pore body center are: , The radius of the pore volume; Throat radius Based on the pore-throat ratio Assigning values ​​using the random sampling method: .

[0012] Furthermore, in S3, the topology correction of the three-dimensional random porous network model includes: Calculate the mean coordination number of the stochastic model ; True average coordination number obtained by focused ion beam scanning electron microscopy The objective function is to iteratively adjust the stochastic model by randomly disconnecting or adding throat connections between pores until... We obtain the static porosity network model after topology correction.

[0013] Furthermore, S4 includes: The coal sample, after surface polishing and pre-treatment by cleaning and drying, was fixed in the sealed sample cell of a high-temperature and high-pressure contact angle measuring instrument. Simulated formation water was injected into the sample cell, and the formation temperature was set. Preset time for balancing; Adjust the sample cell pressure to different preset pressures. The contact angle was measured using the bubble method at each pressure point. A dynamic model of contact angle variation with pore pressure was established. .

[0014] Furthermore, S5 includes: A triaxial rock mechanics testing system was used to conduct porosity and permeability tests under varying confining pressure and varying pore pressure conditions, and the effective stress was calculated. ,in, For confining pressure, The effective stress coefficient; Establish the dynamic deformation equations for pore and throat radii as a function of effective stress: ,in, The pore or throat radius under the current effective stress. The initial aperture, The initial effective stress, The porosity is the pore compressibility coefficient. This represents the current effective stress.

[0015] Furthermore, in S6, the simulation of the gas-driven water process under different pore pressures includes: For each pore pressure point Update effective stress The pore and throat radii are updated using the dynamic deformation equation. The contact angle is updated through the contact angle dynamic model. ; Calculate the capillary force threshold for each larynx ,in, For air-water interfacial tension, For pressure The radius of the lower larynx; An intrusive percolation algorithm was used to simulate gas-driven water and record the capillary force-saturation relationship.

[0016] Compared with existing technologies, this invention has the following advantages: By integrating low-temperature liquid nitrogen adsorption and high-pressure mercury intrusion data, and correcting the coal matrix compression effect in the high-pressure section of mercury intrusion, the problem of overestimation of micropore volume caused by the strong stress sensitivity of coal and rock in traditional high-pressure mercury intrusion methods is solved. The matrix compression correction formula is used to eliminate the contribution of coal skeleton deformation to mercury intrusion volume, so that the obtained pore size distribution is closer to the real stratum pore structure, providing an accurate geometric and statistical basis for subsequent pore network modeling, and significantly improving the accuracy of multi-scale pore characterization of coal and rock.

[0017] Based on the generation of random pore networks, a three-dimensional reconstruction technique using focused ion beam scanning electron microscopy is introduced to extract the coordination number statistical characteristics of real coal and rock. The network connectivity is then targeted for correction. Through iterative adjustment, the error between the model's average coordination number and the real structure is made less than the allowable value. This overcomes the defect of systematic deviation between traditional random network topology and real coal and rock, making the pore network model consistent with real coal and rock in terms of pore size distribution, porosity, and coordination number distribution, thus significantly improving the realism and representativeness of the network model.

[0018] By simultaneously coupling the evolution of pore and throat dimensions induced by effective stress with the evolution of contact angle related to pore pressure into the pore network model, a dynamic capillary force solution framework was established. An exponential pore size-effective stress deformation equation was established through stress sensitivity experiments, and a dynamic function of contact angle-pore pressure was established through variable pressure wetting angle experiments. This enabled the synchronous dynamic updating of geometric parameters and wettability parameters in capillary force calculation, breaking through the limitations of traditional static capillary force measurement methods that assume rigid skeleton and constant wettability.

[0019] By simulating permeability quasi-static displacement through intrusion, capillary force-saturation relationships are output at multiple discrete pore pressure points, ultimately yielding a dynamic capillary force surface plot covering the entire coalbed methane drainage process. This plot directly reveals the evolution of capillary force during drainage and pressure reduction, providing crucial constitutive data for accurately predicting gas breakthrough time, assessing flow resistance during water production, and optimizing drainage regimes. It effectively solves the problems of distorted gas and water production predictions and biased engineering decisions caused by neglecting dynamic effects in existing technologies, significantly improving the reliability of dynamic prediction for coalbed methane development. Attached Figure Description

[0020] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the full-scale aperture distribution fusion of the present invention; Figure 3 This is a schematic diagram showing the comparison of FIB-SEM topology correction before and after step S3 of the present invention. Figure 4 This is a dynamic capillary force surface diagram of the present invention. Detailed Implementation

[0021] The method for determining dynamic capillary force in coal and rock as described in this invention addresses the technical shortcomings of existing static capillary force measurement methods, which cannot match the dynamic evolution characteristics of coal and rock reservoirs during coalbed methane drainage. Through a progressive approach involving precise full-scale pore characterization, pore network topology correction, dynamic parameter coupling modeling, and accurate permeability simulation, a dynamic capillary force determination system adapted to the entire lifecycle of coalbed methane drainage is constructed. Figure 1 As shown, the steps are as follows: S1. Select the target core to prepare coal samples and obtain the full-scale pore size distribution of the coal samples; S2. Construct a three-dimensional random pore network model based on the full-scale pore size distribution; S3. The real pore network topology of coal and rock is obtained by focusing ion beam scanning electron microscopy, and the topology of the three-dimensional random pore network model constructed in S2 is corrected to obtain the topology-corrected static pore network model. S4. Conduct coal and rock wetting angle experiments under varying pressure conditions to establish a dynamic model of the contact angle as a function of pore pressure. S5. Conduct coal and rock stress sensitivity experiments and establish dynamic deformation equations for pore and throat radii as a function of effective stress. S6. Based on the intrusion percolation algorithm, combined with the static pore network model described in S3, the dynamic model described in S4, and the dynamic deformation equation described in S5, the gas-driven water process under different pore pressures is simulated, and the capillary force-saturation curves under different pore pressure conditions are output to form a dynamic capillary force surface plot.

[0022] The specific steps are detailed below: S1: Select the target core to prepare coal samples and obtain the full-scale pore size distribution of the coal samples. This step aims to solve the problem of accurate characterization of the multi-scale pore structure of coal and rock, especially to overcome the distortion of pore size distribution caused by the compression effect of the coal matrix in the high-pressure mercury intrusion porosimetry. It is the foundation for all subsequent modeling and simulation work. The output full-scale pore size distribution directly determines the geometric and statistical characteristics of the pore network model in S2, as follows: S11: High-pressure mercury intrusion testing and matrix compression correction Select representative core samples from the target coal seam, drill plunger samples with a diameter of 25 mm and a length of 50 mm, and some fragment samples. Place the samples in a vacuum drying oven and perform vacuum drying under reservoir temperature conditions. Take out the samples and weigh them every 1 hour until the mass change of the coal sample between two adjacent weighings does not exceed 0.01 g, to ensure that the samples reach a constant weight state.

[0023] The processed lumpy coal sample was loaded into a mercury injection sample tube and sealed. After vacuum degassing, mercury was injected to establish an initial volume baseline. Then, the mercury injection test was carried out by gradually increasing the pressure according to the preset procedure, from low pressure to above 200 MPa, in order to cover the tiny pores and throats in the coal and rock. After maintaining each pressure level until the mercury injection volume stabilized, the cumulative mercury injection volume and the corresponding pressure were recorded to obtain the original pressure-mercury injection volume data.

[0024] Because coal and rock are highly stress-sensitive, the coal matrix will undergo compression deformation under high pressure during high-pressure mercury intrusion. This results in the mercury intrusion volume recorded by the mercury intrusion instrument containing two parts: the actual pore mercury intrusion volume and the false volume caused by matrix compression. If the matrix compression effect is not corrected, the micropore volume will be systematically overestimated.

[0025] Let the volume compressibility of coal matrix be... The sample volume is The initial pressure of mercury porosimetry is When the mercury porosimetry pressure is At that time, the compressed volume of coal matrix Calculate using the following formula: The cumulative mercury injection volume recorded by the mercury porosimeter is denoted as The actual cumulative mercury ingress volume in the pores after removing the influence of coal matrix compression is then calculated. for: , Based on the corrected mercury inlet pressure data, the Washburn equation was used to convert the mercury inlet pressure into an equivalent pore throat radius, thus obtaining the high-pressure mercury injection pore size distribution. The expression for the Washburn equation is: , in, The equivalent orifice throat radius, The interfacial tension between mercury and gas (usually taken as 485 mN / m). This is the contact angle of mercury on the coal surface (usually taken as 130°-150°). The applied pressure for mercury injection.

[0026] S12: Low-Temperature Liquid Nitrogen Adsorption Test After stabilizing the quality of 60-80 mesh coal powder samples under low-temperature vacuum degassing, they were placed in a 77K liquid nitrogen environment to measure the nitrogen adsorption isotherm. The adsorption amount under different relative pressures was recorded, and the specific surface area and pore volume parameters were calculated. Furthermore, the adsorption isotherm was analyzed based on the BJH model to obtain the pore size distribution of low-temperature liquid nitrogen adsorption. This distribution mainly covers the microporous to mesoporous segment (2-50 nm).

[0027] S13: Data fusion to obtain full-scale aperture distribution High-pressure mercury intrusion data mainly cover the mesoporous to macroporous range (>10 nm), while low-temperature liquid nitrogen adsorption data mainly cover the microporous to mesoporous range (<50 nm). There is overlap between the two in the 10 nm to 100 nm range. To obtain a continuous full-scale pore size distribution, it is necessary to... and Perform data fusion; Determine the overlapping pore size range of cryogenic liquid nitrogen adsorption and high-pressure mercury intrusion data. ,in, , A weighted average method was used to fuse the apertures within the overlapping regions, resulting in the full-scale aperture distribution. Defined as: , in, , They are respectively , The weights, and ; To ensure a smooth and continuous transition within the overlapping intervals, the weighting function adopts a piecewise linear form on the logarithmic aperture scale: , Single-source data is used in non-overlapping intervals: when Time to take ,when Time to take ; Through the above fusion, a pore size distribution covering the entire scale, from micropores to macropores, is obtained. This distribution will serve as the core input parameter for constructing the three-dimensional random pore network model in S2. A schematic diagram of the full-scale pore size distribution fusion is shown below. Figure 2 As shown in Table 1 below, the overlapping interval weights and fused PSDs illustrate the distribution of cryogenic liquid nitrogen adsorption within the overlapping interval [10nm, 100nm]. High pressure mercury intrusion distribution The fusion process, at d=10nm, , The data is entirely from LTNA; at d=100nm, , The data was obtained entirely from MIP data; equal-weight fusion was used in overlapping regions to ensure a smooth transition in aperture distribution, resulting in the final fused data. This will be used as input for the full-scale aperture distribution in subsequent steps.

[0028] Table 1. Overlapping Interval Weights and Fuded PSD S2: Based on the full-scale pore size distribution, a three-dimensional random pore network model is constructed. A three-dimensional network structure consistent with the statistical characteristics of real coal and rock is generated through a random sampling algorithm. The output three-dimensional random pore network model in this step is the object of topology correction in S3. Its geometric parameters (pore volume radius, throat radius, throat length) will be dynamically updated in S5 according to stress changes, as follows: Using the full-scale pore size distribution characteristics obtained from S1, the pore network model space is defined. ( and target porosity With the target average coordination number The full-scale pore size distribution is transformed into three types of discrete parameters: pore volume size distribution, throat size distribution, and connectivity. Represented as the aperture probability density function, randomly selected Equivalent radius of each pore and with the equivalent volume of a porous sphere Calculate the total pore volume By adjusting the number of pores to increase porosity satisfy ( (for allowable error) The coordinates of the pore body center are obtained by using three-dimensional random point placement. For each pore body Search its The nearest neighbor pores are identified and candidate connections are established, or a candidate edge set is generated using 3D Delaunay partitioning, with the target average coordination number as the objective. To constrain the network, the average coordination number is adjusted by randomly deleting or adding edges. ( (for the number of larynxes) ; For each connecting pore body and The larynx, larynx length Calculated by the following formula: , in, , The coordinates of the pore body center are: , The radius of the pore volume; Throat radius Based on the pore-throat ratio Assigning values ​​using the random sampling method: , in, Following a logarithmic distribution, the above steps yield a preliminary three-dimensional random pore network model, including the locations of the pores. Pore ​​radius larynx larynx radius With larynx length .

[0029] S3: The true pore network topology of coal and rock is obtained using focused ion beam scanning electron microscopy, and the topology of the three-dimensional random pore network model constructed in S2 is corrected to obtain a topology-corrected static pore network model. This step aims to solve the problem that the random network generated in S2 may deviate from the real coal and rock in terms of topology (connectivity). The true topology parameters are obtained through FIB-SEM three-dimensional reconstruction, and the random network is corrected to make the model consistent with the real coal and rock in a statistical sense. The output topology-corrected static pore network model is the geometric framework for dynamic capillary force simulation in S6. Its coordination number distribution will directly affect the fluid flow path in the percolation simulation, as detailed below: Representative structural regions were selected from coal samples at the same strata as S1 and S2, and cut into small blocks suitable for the microscopic stage, avoiding visible cracks, fracture zones and obvious mineral interlayers. After drying, the samples were fixed on the sample stage and their surfaces were ground and polished to obtain a flat cross-section.

[0030] After determining the region of interest (ROI), an ion beam is used to etch layer by layer at a set step thickness (nanometer level). After each etching is completed, an image of the cross section is acquired using SEM to obtain a set of continuous sequence slice images. The sequence images are then subjected to drift correction and registration, and preprocessing such as grayscale normalization, noise reduction and stripe removal, and charging artifact suppression to obtain voxelized three-dimensional grayscale volume data.

[0031] The 3D grayscale data is binarized into porous and solid phases. The maximum sphere algorithm is then used to represent the porous binary data in a network: Euclidean distance transformation is performed on the porous phase, and the distance field from each pore voxel to the nearest solid boundary is calculated; the maximum inscribed sphere is searched within the distance field, and its center and radius are used as the scale skeleton representation of the pore structure; connectivity is established between the centers of the maximum spheres, and skeletonization and graph structuring are performed to identify pores and throats, and the coordination number of each pore is extracted. Defined as with porous bodies The number of connected larynxes.

[0032] Based on this, the average coordination number of real coal and rock can be calculated. and coordination number distribution frequency ; Calculate the average coordination number of the random model generated in S2. The true average coordination number obtained by FIB-SEM As the objective function, the stochastic model is iteratively adjusted by randomly disconnecting or adding throat connections between pores until the following condition is met: , in To allow for an error margin, typically 0.1 is used to obtain the static porosity network model after topology correction. A comparison diagram of FIB-SEM topology correction before and after is shown below. Figure 3 As shown.

[0033] S4: Conduct coal-rock wetting angle experiments under varying pressure conditions to establish a dynamic model of the contact angle as a function of pore pressure. This step aims to establish the dynamic relationship between wettability parameters and reservoir pressure, overcoming the limitation of traditional methods that assume constant wettability. The dynamic model established in this step... The contact angle will be updated in real time in S6 based on the current pore pressure, which will affect the calculation of the capillary force threshold, as detailed below: After surface polishing, cleaning, and drying, the coal sheet was fixed in the sealed sample cell of a high-temperature, high-pressure contact angle measuring instrument. Simulated formation water was injected into the sample cell, ensuring full contact between the coal sheet surface and the formation water. The formation temperature was then set. (e.g., 323.15K) The equilibrium preset time is 12h to allow the coal and rock surface to fully interact with water to reach an interface equilibrium state.

[0034] Adjust the sample cell pressure to the first preset pressure point. For example, at a pressure of 10.0 MPa, after the temperature and pressure stabilize, a tiny bubble is introduced at the coal-formation water interface using the bubble method. The bubble volume is controllable. High-speed camera systems are used to continuously acquire bubble contour images. After the morphology stabilizes, a steady-state contour is selected, and the contact angle at that pressure point is obtained by fitting the contour to the Young's-Laplace equation using image analysis software. ; Following the same procedure, the pressure was sequentially adjusted to different environmental pressures, such as 7.0 MPa, 5.0 MPa, 3.0 MPa, 2.0 MPa, 1.0 MPa, and 0.5 MPa, to simulate the pressure drop process during drainage. Multiple measurements were taken at each pressure point, and the average value was calculated. and standard deviation .

[0035] Based on the contact angle measurement data at each pressure point, a dynamic model of the contact angle changing with pore pressure is established. The model can take the form of a linear model, an exponential model, or a power-law model. The model parameters are determined through nonlinear regression to achieve the highest goodness of fit.

[0036] S5: Conduct coal and rock stress sensitivity experiments to establish dynamic deformation equations for pore and throat radii as a function of effective stress. This step aims to establish the dynamic relationship between pore structure geometric parameters and effective stress, overcoming the limitations of traditional methods that assume a rigid skeleton. The dynamic deformation equations established in this step will be updated in real time in S6 based on the current effective stress to determine the pore and throat radii, thereby affecting the capillary force threshold and fluid flow resistance, as detailed below: Porosity and permeability tests were conducted using a triaxial rock mechanics testing system under varying confining pressure and pore pressure conditions. Coal and rock plunger samples from the same formation were selected, loaded into the triaxial testing chamber, and connected to upstream and downstream fluid pipelines. The temperature was controlled at the target formation temperature. Apply a preset constant confining pressure (e.g., 20MPa) to the target value and stabilize.

[0037] Regulate pore pressure using a pressure control system From the initial pore pressure (e.g., 10.0 MPa) Start by gradually reducing the pressure to each preset step level. For example, pressures of 7.0 MPa, 5.0 MPa, 3.0 MPa, 2.0 MPa, 1.0 MPa, and 0.5 MPa were applied. At each pressure point, a preset time of 2 hours was maintained to allow the sample deformation and seepage response to stabilize. After stabilization, the porosity at each pressure point was recorded. With penetration rate And calculate the corresponding effective stress: , in, This is the effective stress coefficient, also known as the Biot coefficient, which is typically taken as 0.8-1.0 for coal and rock. The porosity compressibility of coal and rock was fitted based on experimental data. Establish the dynamic deformation equations for the pore and throat radii as a function of effective stress: , in, The pore or throat radius under the current effective stress. The initial aperture corresponds to the initial effective stress. or initial pore pressure , is the pore compressibility coefficient.

[0038] S6: Based on the intrusion percolation algorithm, combined with the static pore network model described in S3, the dynamic model described in S4, and the dynamic deformation equation described in S5, the gas-driven water process under different pore pressures is simulated, and the capillary force-saturation curves under different pore pressure conditions are output, forming a dynamic capillary force surface plot. This step integrates and outputs all the previous steps. Using the topology-corrected static pore network model established in S3 as the geometric framework, combined with the contact angle dynamic model established in S4 and the dynamic deformation equation established in S5, the gas-driven water process is simulated through the intrusion percolation algorithm, and finally, a dynamic capillary force surface plot for the entire coalbed methane drainage process is output, as detailed below: Set pressure path: Based on the same pressure setpoint sequence in S5 For example, 10.0MPa, 7.0MPa, 5.0MPa, 3.0MPa, 2.0MPa, 1.0MPa, and 0.5MPa represent the pressure drop process during drainage, for each pressure point. Perform the following cyclic simulation: Update effective stress: Calculate the current effective stress. ; Update the geometric topology: Using the dynamic deformation equations established by S5, update the radii of all pores and throats in the network model. ; Update wettability: Update the current contact angle using the contact angle dynamic model established in S4. ; Calculate the capillary force threshold: Calculate the intrusion pressure (capillary force threshold) for each throat in the network: , in, For air-water interfacial tension, For pressure The radius of the lower larynx; Perform intrusion percolation simulation: Inject a non-wetting phase (gas) from the inlet end, and use a quasi-static intrusion percolation algorithm to simulate the current capillary force. Below, identify all intrusive trachea, satisfying Furthermore, it is adjacent to areas that have already been invaded, and thus has priority in intrusion. The smallest larynx, recording the saturation of the non-wetting phase. or wetting phase saturation With capillary force The process of change; Integrate simulation data from all pressure points, with each pressure point corresponding to one... Plot these curves in three-dimensional space, with the horizontal axis representing the wetting phase saturation. The vertical axis represents capillary force. The third axis represents pore pressure. Generate a three-dimensional dynamic capillary force map ( (curved surface), such as Figure 4 As shown, this dynamic capillary force surface plot can directly serve the numerical simulation of two-phase flow in coalbed methane, calibration of phase permeability parameters, prediction of gas breakthrough time, assessment of flow resistance during water production, and optimization design of drainage and production systems.

[0039] By setting the same initial formation parameters, the calculation logic and results of the prior art and the method of the present invention in seven key stages of simulating coalbed methane drainage were compared, thus intuitively demonstrating the dynamism and superiority of the present invention. The comparison between the prior art and the method of the present invention is shown in Table 1 below.

[0040] Table 1 Comparison of Prior Art and This Application This invention addresses the technical challenge of accurately determining the dynamic capillary force of coal and rock in coalbed methane development. It overcomes the limitations of traditional static characterization methods by fusing multi-scale pore characterization and data correction, constructing a three-dimensional random pore network and accurately correcting its topology, and measuring dynamic wettability and stress-sensitive deformation parameters. Combined with an intrusion permeation algorithm, it achieves multi-parameter coupled gas-driven water simulation during drainage, ultimately generating a dynamic capillary force surface map covering the entire lifecycle of coal and rock drainage. This method accurately reconstructs the evolution of capillary force under different pore pressures, effectively solving engineering problems such as underestimating water production resistance, distorted gas-water production predictions, and errors in micropore structure characterization inherent in traditional methods. It provides accurate and reliable parameter support for numerical simulation of two-phase flow in coalbed methane, prediction of gas breakthrough time, and optimization of development schemes. Furthermore, its technical approach can be extended to capillary force measurement in other unconventional oil and gas reservoirs such as shale gas and tight gas, demonstrating significant application value and promising prospects in the development of unconventional oil and gas fields. It significantly improves the scientific rigor of reservoir seepage behavior characterization and the rationality of development engineering decisions in unconventional oil and gas reservoirs.

[0041] The above specific embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to examples, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for determining dynamic capillary force in coal and rock, characterized in that, include: S1. Select the target core to prepare coal samples and obtain the full-scale pore size distribution of the coal samples; S2. Construct a three-dimensional random pore network model based on the full-scale pore size distribution; S3. The real pore network topology of coal and rock is obtained by focusing ion beam scanning electron microscopy, and the topology of the three-dimensional random pore network model constructed in S2 is corrected to obtain the topology-corrected static pore network model. S4. Conduct coal and rock wetting angle experiments under varying pressure conditions to establish a dynamic model of the contact angle as a function of pore pressure. S5. Conduct coal and rock stress sensitivity experiments and establish dynamic deformation equations for pore and throat radii as a function of effective stress. S6. Based on the intrusion percolation algorithm, combined with the static pore network model described in S3, the dynamic model described in S4, and the dynamic deformation equation described in S5, the gas-driven water process under different pore pressures is simulated, and the capillary force-saturation curves under different pore pressure conditions are output to form a dynamic capillary force surface plot.

2. The method for determining dynamic capillary force in coal and rock according to claim 1, characterized in that, S1 includes: S11. Perform high-pressure mercury intrusion testing on the coal sample. Correct the high-pressure mercury intrusion test data for coal matrix compression. Based on the corrected mercury intrusion pressure data, use the Washburn equation to convert the mercury intrusion pressure into an equivalent pore throat radius to obtain the high-pressure mercury intrusion pore size distribution. ; S12. Conduct low-temperature liquid nitrogen adsorption tests on coal samples, determine nitrogen adsorption isotherms, and analyze the adsorption isotherms based on the BJH model to obtain the low-temperature liquid nitrogen adsorption pore size distribution. ; S13, will and Data fusion was performed to obtain the full-scale aperture distribution.

3. The method for determining dynamic capillary force in coal and rock according to claim 2, characterized in that, In S11, the coal matrix compression correction for the high-pressure mercury intrusion test data includes: Let the volume compressibility of coal matrix be... The sample volume is The initial pressure of mercury porosimetry is When the mercury porosimetry pressure is hourly coal matrix compression volume , The cumulative mercury injection volume recorded by the mercury porosimeter is denoted as The actual cumulative mercury ingress volume in the pores after removing the influence of coal matrix compression is then calculated. .

4. The method for determining dynamic capillary force in coal and rock according to claim 3, characterized in that: In S11, the expression for the Washburn equation is: , in, The equivalent orifice throat radius, For mercury-gas interfacial tension, This represents the contact angle of mercury on the coal surface.

5. The method for determining dynamic capillary force in coal and rock according to claim 4, characterized in that: In S13, data fusion includes: Determine the overlapping pore size range between cryogenic liquid nitrogen adsorption and high-pressure mercury intrusion data. 1, 2]; A weighted average method was used to fuse the apertures within the overlapping aperture range. The resulting full-scale aperture distribution was as follows: ,in, , They are respectively , Weight; Single-source data is used in non-overlapping intervals when < Take at 1 o'clock ,when > Take at 2 o'clock .

6. The method for determining dynamic capillary force in coal and rock according to claim 5, characterized in that, S2 include: The full-scale pore size distribution is transformed into three types of discrete parameters: pore volume size distribution, throat size distribution, and connectivity. For each connecting pore body and The larynx, larynx length Calculated by the following formula: ,in, , The coordinates of the pore body center are: , The radius of the pore volume; Throat radius Based on the pore-throat ratio Assigning values ​​using the random sampling method: .

7. The method for determining dynamic capillary force in coal and rock according to claim 6, characterized in that, In S3, topology correction of the three-dimensional random porous network model includes: Calculate the mean coordination number of the stochastic model ; True average coordination number obtained by focused ion beam scanning electron microscopy The objective function is to iteratively adjust the stochastic model by randomly disconnecting or adding throat connections between pores until... We obtain the static porosity network model after topology correction.

8. The method for determining dynamic capillary force in coal and rock according to claim 7, characterized in that, S4 include: The coal sample, after surface polishing and pre-treatment by cleaning and drying, was fixed in the sealed sample cell of a high-temperature and high-pressure contact angle measuring instrument. Simulated formation water was injected into the sample cell, and the formation temperature was set. Preset time for balancing; Adjust the sample cell pressure to different preset pressures. The contact angle was measured using the bubble method at each pressure point. A dynamic model of contact angle variation with pore pressure was established. .

9. The method for determining dynamic capillary force in coal and rock according to claim 8, characterized in that, S5 include: A triaxial rock mechanics testing system was used to conduct porosity and permeability tests under varying confining pressure and varying pore pressure conditions, and the effective stress was calculated. ,in, For confining pressure, The effective stress coefficient; Establish the dynamic deformation equations for pore and throat radii as a function of effective stress: ,in, The pore or throat radius under the current effective stress. The initial aperture, The initial effective stress, The porosity is the pore compressibility coefficient. This represents the current effective stress.

10. The method for determining dynamic capillary force in coal and rock according to claim 9, characterized in that, In S6, the simulation of gas-driven water processes under different pore pressures includes: For each pore pressure point Update effective stress The pore and throat radii are updated using the dynamic deformation equation. The contact angle is updated through the contact angle dynamic model. ; Calculate the capillary force threshold for each larynx ,in, For air-water interfacial tension, For pressure The radius of the lower larynx; An intrusive percolation algorithm was used to simulate gas-driven water and record the capillary force-saturation relationship.

Images