A hydrate dissociation mesoscopic gas-liquid migration identification test system

Abstract

The application relates to a hydrate decomposition microcosmic gas-liquid migration identification test system, which comprises a lamellar hydrate preparation module, a collection module and a quantitative analysis module. The lamellar hydrate preparation module comprises a transparent experimental soil tank, a lamellar hydrate sample arranged on the transparent experimental soil tank and a transparent soil tank cover plate covered on the lamellar hydrate sample. The lamellar hydrate sample is in the shape of a round cake, and an annular water permeable stone is arranged on the outer side of the lamellar hydrate sample. The collection module collects images and parameters of the lamellar hydrate sample in the process of gas-liquid migration in real time, inputs the images and parameters into the quantitative analysis module for processing, and obtains effective seepage porosity of the lamellar hydrate sample. Compared with the prior art, the application can solve the problems of real-time observation difficulty, low key information extraction efficiency and quantitative analysis difficulty of microcosmic gas-liquid migration in the hydrate decomposition process.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of natural gas hydrate development, and in particular to a test system for identifying microcosmic gas-liquid migration in hydrate decomposition. BACKGROUND

[0002] Natural gas hydrate is a clathrate compound formed by natural gas molecules (mainly methane) and water molecules under low temperature and high pressure, and is widely distributed in the seabed and permafrost regions. Natural gas hydrate is a strategic replacement energy that is highly valued by countries around the world. On November 3, 2017, China officially listed natural gas hydrate as a new mineral, accelerating the development process of natural gas hydrate resources, with the goal of leading the world in natural gas hydrate mining technology by 2035. The microcosmic gas-liquid migration mechanism in the decomposition process of natural gas hydrate is a key scientific issue for efficient exploitation of natural gas hydrate.

[0003] Gas-liquid migration in hydrate-bearing sediments during decomposition is a complex dynamic process. During the decomposition and secondary generation of hydrates, the skeleton structure of the sediments undergoes mechanical deformation under the coupling of temperature and pressure; at the same time, the solid phase materials in the pores undergo chemical phase change (hydrates) or motion migration (fine soil particles), causing the pore connectivity state of the sediments to change constantly, thereby affecting the macroscopic seepage characteristics of the hydrate reservoir. Therefore, it is urgent to study the microcosmic starting mechanism, transport route and retention characteristics of gas-liquid under the disturbance of hydrate exploitation.

[0004] The existing test devices and technologies mainly conduct experimental research on the seepage characteristics of hydrate-bearing sediment field samples, and infer the change law and control factors of permeability characteristics according to the gas-liquid flow rate at the inlet and outlet. Due to the lack of direct observation of the actual gas-liquid migration microcosmic mode, it is difficult to reveal the complex physical nature of the dynamic change of seepage characteristics in the hydrate decomposition process.

[0005] Some scholars have attempted to use micro-imaging methods such as electrical resistivity tomography, nuclear magnetic resonance, and three-dimensional CT scanning to study the change and motion law of three-phase substances in hydrate-bearing sediment pores. For example, an electrical resistivity tomography instrument is used to observe the imaging of the natural gas hydrate formation process, a nuclear magnetic resonance imaging technique is used to observe the hydrate formation and decomposition in the sandy soil skeleton and the multiphase seepage process, a CT three-dimensional scanning is used to extract the pore structure of the hydrate-bearing sediment sample under different hydrate saturations and the spatial distribution of different phase pore materials, and an X-ray tomography is used to form a hydrate. However, these experimental systems still have considerable difficulties in real-time dynamic observation of gas-liquid migration in hydrate-bearing sediments, and lack effective technical means to accurately extract and quantify the changes in gas-liquid dynamic transport and effective permeability in the decomposition process of natural gas hydrate.

[0006] In summary, the natural gas hydrate decomposition process microscopic visualization experimental device of the prior art has the following three problems:

[0007] (1) It is difficult to observe the dynamic change of the microscopic seepage characteristics in the natural gas hydrate decomposition process;

[0008] (2) There are still great difficulties in real-time dynamic observation of gas-liquid migration in hydrate-containing sediments;

[0009] (3) The key information extraction efficiency of microscopic gas-liquid migration is low and quantitative analysis is difficult. SUMMARY

[0010] The purpose of the present application is to overcome the defects of the prior art and provide a hydrate decomposition microscopic gas-liquid migration identification test system, which can solve the problems of real-time observation difficulty, low key information extraction efficiency and quantitative analysis difficulty in the hydrate decomposition process.

[0011] The purpose of the present application can be achieved by the following technical solutions:

[0012] The present application provides a hydrate decomposition microscopic gas-liquid migration identification test system, which comprises a lamellar hydrate preparation module, a collection module and a quantitative analysis module. The lamellar hydrate preparation module comprises a transparent experimental soil tank, a lamellar hydrate sample arranged on the transparent experimental soil tank and a transparent soil tank cover plate arranged on the lamellar hydrate sample. The lamellar hydrate sample is in the shape of a round cake, and an annular water-permeable stone is arranged on the outside of the lamellar hydrate sample.

[0013] The collection module collects images and parameters of the lamellar hydrate sample in the gas-liquid migration process in real time, and inputs the images and parameters into the quantitative analysis module for processing to obtain the effective seepage porosity of the lamellar hydrate sample.

[0014] Preferably, the annular water-permeable stone is connected with an annular methane gas injection device for forming the lamellar hydrate sample. The annular methane gas injection device continuously and uniformly injects methane gas to synthesize hydrate into the transparent experimental soil tank along the radial direction of the periphery of the annular water-permeable stone, thereby preparing the lamellar hydrate sample.

[0015] Preferably, the annular methane gas injection device comprises a methane gas cylinder, a methane gas pipeline and a temperature control coil connected in sequence, and the end of the temperature control coil is connected with the annular water-permeable stone.

[0016] Preferably, the annular water-permeable stone is arranged in the middle of the lamellar hydrate sample.

[0017] Preferably, the center of the lamellar hydrate sample is provided with a micro depressurization production well penetrating through the transparent soil tank cover plate.

[0018] Preferably, the micro-decompression mining well is a cylindrical permeable stone.

[0019] Preferably, the collecting module comprises a loading steel frame arranged on the top of the transparent soil tank cover plate and a supporting platform arranged on the bottom of the transparent experimental soil tank, both of which are in a cylindrical structure; the inner wall of the bottom of the supporting platform is provided with a remote control industrial camera, the lens of the remote control industrial camera is aimed at the transparent experimental soil tank, and the inner wall of the top of the loading steel frame is provided with an adjustable light source, and the irradiation end of the adjustable light source is aimed at the transparent soil tank cover plate.

[0020] The adjustable light source irradiates on the lamellar hydrate sample, and the remote control industrial camera is used to take real-time images of the lamellar hydrate sample in the gas-liquid migration process.

[0021] Preferably, airtight and light-tight spaces are formed between the loading steel frame and the transparent soil tank cover plate and between the transparent experimental soil tank and the supporting platform.

[0022] Preferably, the processing process of the quantitative analysis module is specifically as follows:

[0023] S1: pre-process all the images of the lamellar hydrate sample in the gas-liquid migration process, obtain an image data set, input the image data set into a first neural network model established and trained in advance, and obtain instance segmentation results of solid-liquid-gas three-phase components in the lamellar hydrate sample image at different time points;

[0024] S2: based on the instance segmentation results of the images at different time points, calculate the pixel proportion of the liquid and gas in the pore network of the lamellar hydrate sample at different time points, i.e. the proportion of the gas-liquid components, and further quantify the effective percolation porosity of the lamellar hydrate sample;

[0025] S3: obtain macroscopic characteristic data of the lamellar hydrate sample at different time points, combine the proportion of the gas-liquid components and the effective percolation porosity of the lamellar hydrate sample at different time points obtained in S2, establish and train a second neural network model, and obtain the quantitative correlation between the macroscopic characteristic data and the proportion of the gas-liquid components and the effective percolation porosity.

[0026] Preferably, the macroscopic characteristic data comprises external temperature, stress and strain, pore pressure, volume change and decomposed methane gas amount of the lamellar hydrate sample.

[0027] Compared with the prior art, the present application has the following beneficial effects:

[0028] 1. This invention provides a system for identifying and testing the microscopic gas-liquid transport during hydrate decomposition. It employs a disc-shaped sheet hydrate sample, ensuring sufficient light transmittance and high-definition image observation capabilities, while eliminating the boundary effects of gas-liquid transport in small samples. An annular permeable stone surrounding the sheet hydrate sample allows methane gas and water produced during hydrate decomposition to seep out, facilitating the observation of microscopic seepage changes during the natural gas hydrate decomposition process.

[0029] 2. This invention provides a microscopic gas-liquid transport identification and testing system for hydrate decomposition. It creates sealed spaces at both the upper and lower parts of the lamellar hydrate sample, and installs an adjustable light source at the top of the inner wall of a loading frame. A remote-controlled industrial camera is installed on the bottom inner wall of the support platform to capture images of the lamellar hydrate sample during gas-liquid transport in real time. This solves the problem of significant difficulties in real-time dynamic observation of gas-liquid transport in hydrate-containing sediments in existing technologies.

[0030] 3. The present invention provides a microscopic gas-liquid transport identification and testing system for hydrate decomposition. Based on image deep learning, it performs dynamic solid-liquid-gas identification on images, quantifies the dynamic changes of effective gas-liquid transport routes in the entire sample pore network, and can more accurately calculate the effective permeable porosity. At the same time, it establishes a quantitative correlation between macroscopic characteristic data, gas-liquid component ratio and effective permeable porosity. Based on the quantitative correlation, the effective permeable porosity can be directly obtained from macroscopic characteristic data, improving the extraction efficiency of key information on microscopic gas-liquid transport. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the structure of a microscopic gas-liquid transport identification and testing system for hydrate decomposition provided by the present invention.

[0032] Figure 2 for Figure 1 A schematic diagram of the cyclic methane injection device in the illustrated embodiment.

[0033] Figure 3 Images of the decomposition process of a lamellar hydrate sample observed using a remote-controlled industrial camera.

[0034] The following are marked in the diagram: 10. Loading steel frame; 11. Adjustable light source; 12. First temperature-controlled ventilation inlet; 13. First temperature-controlled ventilation outlet; 20. Transparent experimental soil tank; 21. Layered hydrate sample; 22. Transparent soil tank cover; 210. Micro pressure-reducing well; 220. Ring-shaped permeable stone; 230. Methane cylinder; 231. First valve; 232. Second valve; 233. Third valve; 234. Vacuum pump; 235. Temperature-controlled coil; 30. Support platform; 31. Remote-controlled industrial camera; 32. Second temperature-controlled ventilation inlet; 33. Second temperature-controlled ventilation outlet. DETAILED DESCRIPTION

[0035] The present application will be described in detail below with reference to the drawings and specific embodiments. The embodiments are implemented on the premise of the technical solutions of the present application, and give detailed implementation modes and specific operation processes, but the protection scope of the present application is not limited to the following embodiments.

[0036] Referring to Figure 1 As shown in the figure, the embodiment provides a hydrate decomposition microscopic gas-liquid migration identification test system, which comprises a lamellar hydrate preparation module, a collection module and a quantitative analysis module. The lamellar hydrate preparation module comprises a transparent experimental soil tank 20, a lamellar hydrate sample 21 arranged on the transparent experimental soil tank 20, and a transparent soil tank cover plate 22 covering the lamellar hydrate sample 21. The lamellar hydrate sample 21 is in the shape of a round cake, and an annular water-permeable stone 220 is arranged outside the lamellar hydrate sample 21. The cross-sectional shape of the transparent experimental soil tank 20 is circular, and a cylindrical cavity accommodating the lamellar hydrate sample 21 is formed between the transparent experimental soil tank 20 and the transparent soil tank cover plate 22. The annular water-permeable stone 220 is located in the transparent experimental soil tank 20.

[0037] The collection module collects images and parameters of the lamellar hydrate sample 21 in the process of gas-liquid migration in real time, and inputs the images and parameters into the quantitative analysis module for processing to obtain the effective seepage porosity of the lamellar hydrate sample 21. The images of the lamellar hydrate sample 21 in the process of gas-liquid migration refer to Figure 3 As shown in the figure.

[0038] The lamellar hydrate sample 21 prepared in the embodiment is in the shape of a round cake, so that the lamellar hydrate sample 21 has sufficient light transmission performance and high-definition image observation performance, and the gas-liquid migration boundary effect of the small sample can be eliminated. The annular water-permeable stone 220 arranged outside the lamellar hydrate sample 21 can seep out the methane gas and water generated by the decomposition of hydrate, thereby facilitating the observation of the change of microscopic seepage.

[0039] As an optional implementation, the annular water-permeable stone 220 is connected with an annular methane gas injection device. The annular methane gas injection device continuously and uniformly injects methane gas to synthesize hydrate along the periphery of the annular water-permeable stone 220 radially to the transparent experimental soil tank 20, thereby preparing the lamellar hydrate sample 21.

[0040] Referring to Figure 2 As an optional implementation, as shown in the figure, the annular methane gas injection device comprises a methane gas cylinder 230, a methane gas pipeline and a temperature control coil 235 connected in sequence, and the end of the temperature control coil 235 is connected with the annular water-permeable stone 220.

[0041] In the embodiment, the methane gas pipeline is sequentially provided with a first valve 231, a second valve 232 and a third valve 233, wherein the output end of the second valve 232 is connected with a pressure gauge, and the input end of the third valve 233 is connected with a vacuum pump 234.

[0042] Preferably, the annular water-permeable stone 220 is arranged at the middle of the sheet hydrate sample 21, and provides an axisymmetric boundary condition for the circular annular methane injection inlet.

[0043] As an optional embodiment, the center of the sheet hydrate sample 21 is provided with a micro-depleted well 210 penetrating through the transparent soil tank cover plate 22.

[0044] Preferably, the micro-depleted well 210 is a high water-permeable and gas-permeable cylindrical water-permeable stone, so that the gas-liquid transportation in the hydrate decompression process is axisymmetric.

[0045] The collecting module includes a loading steel frame 10 arranged at the top of the transparent soil tank cover plate 22 and a supporting platform 30 arranged at the bottom of the transparent experimental soil tank 20, and the loading steel frame 10 and the supporting platform 30 are both cylindrical structures; a closed and light-tight space is formed between the loading steel frame 10 and the transparent soil tank cover plate 22 and between the transparent experimental soil tank 20 and the supporting platform 30; a remote industrial camera 31 is arranged on the inner wall of the bottom of the supporting platform 30, and the lens of the remote industrial camera 31 is aligned with the transparent experimental soil tank 20; and an adjustable light source 11 is arranged on the inner wall of the top of the loading steel frame 10, and the irradiation end of the adjustable light source 11 is aligned with the transparent soil tank cover plate 22.

[0046] Preferably, the adjustable light source 11 is a high-intensity light source.

[0047] Before starting image and video collection, the collection parameters of the remote industrial camera 31 are set, and the collection profile and the exposure time are set; after the parameters are set, the images and videos will be stored and processed according to the parameters. After starting collection, the software can save the collected images and video data in a high-speed storage device in real time, and the storage format adopts the original data format with the highest efficiency.

[0048] As an optional embodiment, the sidewall of the loading steel frame 10 is provided with a first temperature control air inlet 12 and a first temperature control air outlet 13 for controlling the temperature of the closed space formed between the loading steel frame 10 and the transparent soil tank cover plate 22, and the sidewall of the supporting platform 30 is provided with a second temperature control air inlet 32 and a second temperature control air outlet 33 for controlling the temperature of the closed space formed between the transparent experimental soil tank 20 and the supporting platform 30.

[0049] As an optional embodiment, the outer walls of the loading steel frame 10 and the supporting platform 30 are both provided with a thermal insulation layer.

[0050] Specifically, the servo loading system in the GCTS high-pressure low-temperature triaxial test system is used to pressurize the top of the loading frame 10, so as to realize accurate loading control of the outer load of the transparent experimental soil tank 20.

[0051] The collection module and the lamellar hydrate preparation module in the embodiment are based on the GCTS high-pressure low-temperature triaxial test system and are modified. The collection module also includes other data collection modules of the original system. During the test process, the external temperature, stress and strain, pore pressure, volume change, and decomposed methane gas amount of the sample are automatically recorded. After the collection module collects the images and parameters of the lamellar hydrate sample 21 in the gas-liquid migration process in real time, the images and parameters are sent to the quantitative analysis module for quantitative analysis.

[0052] The quantitative analysis module receives the images and parameters of the lamellar hydrate sample 21 in the gas-liquid migration process sent by the collection module and performs quantitative analysis.

[0053] Specifically, the quantitative analysis method of the lamellar hydrate sample 21 includes the following steps:

[0054] S1: Preprocess all the images of the lamellar hydrate sample 21 in the gas-liquid migration process, obtain an image data set, input the image data set into a first neural network model established and trained in advance, and obtain instance segmentation results of solid-liquid-gas three-phase components in the lamellar hydrate sample 21 image at different time points.

[0055] As an optional implementation, the preprocessing process of the images of the lamellar hydrate sample 21 in the gas-liquid migration process is as follows: based on the image data labeling software LabelMe, the images are labeled to obtain an image data set in the Microsoft COCO data format.

[0056] As an optional implementation, in the embodiment, the first neural network model adopts Mask RCNN.

[0057] S2: Based on the instance segmentation results of the images at different time points, the pixel proportion of the gas and liquid in the pore network of the lamellar hydrate sample at different time points is calculated, that is, the gas-liquid component proportion is quantified, and then the effective seepage porosity of the lamellar hydrate sample is quantified.

[0058] S3: Obtain the macroscopic characteristic data of the lamellar hydrate sample at different time points, including: external temperature, stress and strain, pore pressure, volume change, and decomposed methane gas amount, combine the gas-liquid component proportion and the effective seepage porosity of the lamellar hydrate sample at different time points obtained in S2, establish and train a second neural network model, and obtain the quantitative correlation between the macroscopic characteristic data and the gas-liquid component proportion and the effective seepage porosity.

[0059] After obtaining the quantitative correlation between the macroscopic characteristic data and the gas-liquid component proportion and the effective seepage porosity, only the macroscopic characteristic data needs to be collected to obtain the gas-liquid component proportion and the effective seepage porosity in the next experiment.

[0060] The working steps and working principles of the identification test system for hydrate decomposition mesoscopic gas-liquid migration are as follows:

[0061] (1) Connect the methane gas pipeline and remove the impurity gas in the gas pipeline. Open the methane gas cylinder 230 and three valves, fill the methane gas into the gas pipeline, and close the switch of the methane gas cylinder 230 after a certain amount of methane gas is put in;

[0062] (2) Connect the lamellar hydrate sample 21 (which needs to be loaded to the designed stress) with the ring-shaped water-permeable stone 220, and connect with the ring-shaped methane gas injection device, open the vacuum pump 234, and vacuumize the lamellar hydrate sample 21;

[0063] (3) The methane gas required for hydrate formation enters the sample along the periphery of the ring-shaped water-permeable stone 220 in a radial direction, and the gas-liquid migration is axisymmetric.

[0064] (4) The methane gas pressure of the gas pipeline and the experimental temperature are kept stable, and when the hydrate synthesis is completed, all the valves are closed.

[0065] (5) After the preparation of the hydrate sample is completed, the sample is isostatically consolidated to the initial temperature and pressure conditions through the triaxial test system and the loading steel frame, and then the sample is subjected to decompression decomposition test;

[0066] (6) In the hydrate decompression decomposition test, a stepwise temperature and pressure control method is used to simulate the quasi-static decomposition process of the sample, and after the internal and external temperature and pressure of the sample are balanced and stabilized, the next test step is carried out, until the sample is completely decomposed, and the temperature, stress and strain, pore pressure, volume change and decomposed methane gas quantity of the sample are automatically recorded during the test;

[0067] (7) The mesoscopic gas-liquid migration behavior image of the lamellar hydrate sample 21 during the test is shot by using the remote control industrial camera 31;

[0068] (8) The parameters obtained in (6) and the images obtained in (7) are input into the quantitative analysis module for processing, to obtain the quantitative correlation between the macroscopic characteristic data and the gas-liquid component proportion and the effective seepage porosity.

[0069] In summary, the identification test system for hydrate decomposition mesoscopic gas-liquid migration provided by the present application adopts a lamellar circular cake-shaped sample to realize that the sample has sufficient light transmission performance and high-definition observation image.

[0070] In order to eliminate the gas-liquid migration boundary effect of a small sample, the shape of the slice sample is designed as a circle. A high water-permeable and air-permeable water-permeable stone is arranged at the middle height of the sample, and an axisymmetric boundary condition of the circular methane gas inlet is provided.

[0071] The miniature decompression production well adopts a high water-permeable and air-permeable cylindrical water-permeable stone arranged at the center position of the circular slice sample. The gas-liquid transportation in the hydrate decompression process is also axisymmetric. The methane gas and water generated by the hydrate decomposition seep out from the water-permeable stone into the collection pipeline, facilitating the observation of the change of the micro seepage.

[0072] Based on the light transmission characteristics of the slice sample in the hydrate decompression process, the higher the proportion of the hydrate component in the pore of the soil particle skeleton, the stronger the light transmission performance of the sample. After the decomposition of the hydrate, the soil body becomes more loose, and the light is more easily transmitted. Due to the large environmental stress of the hydrate sample, the gas and liquid coexist in the sample after the decomposition of the hydrate, and the gas often gathers to form bubbles that are easily observed and migrate to the decompression production well under the action of high gradient pressure. Therefore, through the irradiation of the light source, the pore channel and the micro seepage migration process in the hydrate sediment sample can be directly observed.

[0073] In order to realize real-time observation of micro visualization, the present application adjusts the thickness of the soil sample layer and the intensity of the visualization light source to achieve the best image contrast effect of the phase change process, the gas-liquid migration route and the development of the transport network, and realizes automatic identification and effective processing of a large amount of image data.

[0074] By arranging airtight and light-tight spaces above and below the slice circular pie-shaped sample, the light source in the upper sealed chamber is a high-intensity light source, and the remote control camera industrial camera in the lower sealed chamber can shoot the micro high-definition images of the dynamic process of the gas-liquid migration of the sample in real time. According to the image definition, the brightness of the light source and the parameter setting of the camera industrial camera can be adjusted to improve the definition and visualization analysis effect of the image.

[0075] The test observation data in the present application are high-definition image data, from which the visualization images of the dynamic evolution of the gas-liquid migration route in the thin-layer sample at any time can be obtained. Based on the image depth learning method, the dynamic identification of solid-liquid-gas can be carried out on the image, the dynamic change of the effective gas-liquid migration route in the whole sample pore network can be quantified, the proportion of the gas-liquid route in the whole pore can be obtained, and the effective seepage porosity can be calculated.

[0076] The quantified micro migration mode information obtained by calculation and the macroscopic permeability data measured by the test can be used for data statistics and correlation analysis, the correlation between the micro gas-liquid migration behavior and the macro seepage characteristics can be established, and the micro parameters controlling the dynamic evolution of the macro seepage characteristics can be identified through the correlation.

[0077] The preferred embodiments of the present application have been described above in detail. It should be understood that modifications and variations to the present application can be affected by those skilled in the art without departing from the scope of the application. Accordingly, it is intended that all of the subject matter of the above description and the claims be interpreted to encompass all such modifications and changes.

Claims

1. A hydrate dissociation mesoscopic gas-liquid migration identification test system, comprising a slice hydrate preparation module, a collection module and a quantitative analysis module, characterized in that, The layered hydrate preparation module includes a transparent experimental soil tank (20), a layered hydrate sample (21) placed on the transparent experimental soil tank (20), and a transparent soil tank cover plate (22) covering the layered hydrate sample (21). The layered hydrate sample (21) is in the shape of a disc and is surrounded by annular permeable stones (220). The acquisition module acquires images and parameters of the sheet hydrate sample (21) in real time during the gas-liquid migration process, and inputs the images and parameters into the quantitative analysis module for processing to obtain the effective permeation porosity of the sheet hydrate sample (21). The annular permeable stone (220) is connected to an annular methane injection device for forming a lamellar hydrate sample (21). The annular methane injection device continuously and uniformly injects methane gas radially into the transparent experimental soil tank (20) along the periphery of the annular permeable stone (220) to synthesize hydrate and prepare the lamellar hydrate sample (21).

2. A hydrate dissociation mesoscopic gas-liquid migration identification test system according to claim 1, characterized in that, The annular methane injection device includes a methane cylinder (230), a methane gas pipeline and a temperature regulating coil (235) connected in sequence, with the end of the temperature regulating coil (235) connected to the annular permeable stone (220).

3. The identification test system for hydrate dissociation mesoscopic gas-liquid migration according to claim 1, characterized in that, The annular permeable stone (220) is located in the middle of the lamellar hydrate sample (21).

4. The hydrate dissociation mesoscopic gas-liquid migration identification test system according to claim 1, wherein, The center of the lamellar hydrate sample (21) has a micro-pressure reduction well (210) that passes through the transparent soil trough cover plate (22).

5. A hydrate dissociation mesoscopic gas-liquid migration identification test system according to claim 4, characterized in that, The micro-pressure reduction well (210) is a cylindrical permeable rock.

6. The identification test system for hydrate dissociation mesoscopic gas-liquid migration according to claim 1, characterized in that, The acquisition module includes a loading steel frame (10) on top of the transparent soil trough cover (22) and a support platform (30) at the bottom of the transparent experimental soil trough (20). Both the loading steel frame (10) and the support platform (30) are cylindrical structures. The bottom inner wall of the support platform (30) is equipped with a remote-controlled industrial camera (31), the lens of which is aimed at the transparent experimental soil trough (20). The top inner wall of the loading steel frame (10) is equipped with an adjustable light source (11), the illumination end of which is aimed at the transparent soil trough cover (22). The adjustable light source (11) illuminates the sheet hydrate sample (21) and captures images of the sheet hydrate sample (21) in real time during the gas-liquid transport process through the remote-controlled industrial camera (31).

7. The identification and testing system for microscopic gas-liquid transport in hydrate decomposition according to claim 6, characterized in that, A closed, opaque space is formed between the loading steel frame (10) and the transparent soil trough cover plate (22), as well as between the transparent experimental soil trough (20) and the support platform (30).

8. The identification and testing system for microscopic gas-liquid transport in hydrate decomposition according to claim 1, characterized in that, The specific processing procedure of the quantitative analysis module is as follows: S1: Preprocess all collected images of the sheet hydrate samples (21) during the gas-liquid transport process to obtain an image dataset. Input the image dataset into the first neural network model that has been pre-established and trained to obtain the instance segmentation results of the solid-liquid-gas three-phase components in the image of the sheet hydrate sample (21) at different times. S2: Based on the instance segmentation results of the images at different times, calculate the pixel proportion of liquid and gas in the pore network of the lamellar hydrate sample at different times, that is, quantify the proportion of gas and liquid components, and then quantify the effective permeation porosity of the lamellar hydrate sample. S3: Obtain macroscopic characteristic data of the lamellar hydrate sample at different times. Combine the gas-liquid component ratio and effective permeation porosity of the lamellar hydrate sample obtained in S2 at different times to establish and train a second neural network model to obtain the quantitative correlation between macroscopic characteristic data and gas-liquid component ratio and effective permeation porosity.

9. The identification and testing system for microscopic gas-liquid transport in hydrate decomposition according to claim 8, characterized in that, The macroscopic characteristic data include the external temperature, stress and strain, pore pressure, volume change and amount of methane gas decomposed from the lamellar hydrate sample (21).

Images