Natural gas hydrate exploitation method and system based on color recognition

Abstract

The invention provides a natural gas hydrate exploitation method and system based on color recognition. The natural gas hydrate exploitation method comprises the steps that natural gas hydrate reservoir parameters are obtained through an actual measurement monitoring device; recording and storing the spatial distribution cloud atlas of the natural gas hydrate reservoir information in the public information at different depths and positions; loading the cloud picture by using image processing software; determining a working area which is effectively identified in the cloud picture, and performing grid division on the working area; the RGB value of each grid in the working area is identified and captured, an identifier is distributed, and an identifier text in matrix distribution is generated; mapping the color features to reservoir parameters, replacing the identifiers with the corresponding reservoir parameters, and generating reservoir texts in matrix distribution; the method comprises the following steps: inputting reservoir text data into hydrate simulation software, carrying out numerical simulation, fitting with an existing pilot production numerical simulation result, adjusting to obtain a heterogeneous geologic model, correcting the heterogeneous geologic model to obtain a mining model, carrying out numerical simulation on the mining model, and selecting a mining scheme for mining.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of natural gas hydrate exploitation, and particularly relates to a natural gas hydrate exploitation method and system based on color identification. BACKGROUND

[0002] Due to the significant difference in hydrate resource distribution, when formulating an exploitation scheme, local test production data need to be combined and analyzed through numerical simulation. For areas without test production, test production data of adjacent areas can be referred to or simulation analysis can be performed.

[0003] The simulation of hydrate reservoirs requires the establishment of corresponding geological reservoir parameters. When establishing a model of geological reservoir parameters, the commonly used methods of exploitation include drilling, profiling, geophysical data analysis, and analysis techniques based on discrete points. These traditional methods have encountered challenges in dealing with actual reservoir parameters, such as in limited reference literature parameters, it is difficult to derive accurate heterogeneous parameters. In the study of natural gas hydrate exploitation, homogeneous models, as a model consistent with statistics, are commonly used to simulate the simulation of hydrate exploitation. This model assumes that the physical parameters in the study area are uniform and consistent, simplifying the complex geological conditions and physical and chemical processes, making simulation relatively easy to perform. Wang et al. (Wang R, Zhang J, Wang T, et al. Numerical Simulation of Improved Gas Production from Oceanic Gas Hydrate Accumulation by Permeability Enhancement Associated with Geomechanical Response[J]. Journal of Marine Science and Engineering, 2023, 11(7): 1468.) used a homogeneous model to simulate the South China Sea hydrate reservoir. This model assumes that the physical parameters in the study area are uniform and consistent, while the geological conditions in nature are often extremely complex, making it difficult to exist a truly homogeneous hydrate reservoir. The formation of hydrates is influenced by the surrounding temperature, pressure, gas source, gas composition, and other factors, and the formation of the region is not fixed. A large amount of field exploration data shows that there are significant differences in initial reservoir parameters of different locations and different components, which makes the homogeneous model have many limitations in practical application. When simulating calculations, the homogeneous model only uses the average value of the parameters to give the entire reservoir, ignoring the influence of spatial heterogeneity on fluid flow, heat transfer, and chemical reactions. The hydrate reservoir in the Shenhu area of the South China Sea has obvious non-uniform characteristics, and the occurrence of hydrates is diverse and varies in thickness. Logging data indicates that thin or dispersed hydrates, thick hydrates, and dispersed hydrates appear near the well, and several types of hydrates appear alternately in depth. Therefore, when using a homogeneous model for calculation, there are inevitably various limitations, which may lead to a large deviation between the simulation results and the actual situation, affecting the output effect and exploitation efficiency of hydrates. In contrast, the heterogeneous model can more realistically restore the physical parameters of the reservoir and be closer to the actual geological conditions.

[0004] To establish a heterogeneous model, Tamaki et al. (Tamaki M, Suzuki K, Fujii, T, etc. Prediction and validation of gas hydrate saturation distribution in the eastern Nankai Trough, Japan: Geostatistical approach integrating well-log and 3D seismic data[J]. Interpretation, 2016, 4(1):SA83-SA94.) used well logging data and 3D seismic data as input data, and the seismic data provided the lateral hydrate saturation distribution information, and the well logging data provided the vertical hydrate saturation distribution information. By integrating well logging data and seismic data through geostatistical methods, the sequential Gaussian and cokriging method was applied to generate a heterogeneous natural gas hydrate saturation map. In the case of good correlation between the two variables, the well logging data obtained from exploration and production tests were used as the main variable, and the inverted seismic impedance data were used as the secondary variable for simulation. Since the heterogeneous hydrate saturation parameters obtained by this method cannot be obtained through public channels, researchers face a large data obstacle when analyzing the hydrates in this area, and have to continue to use the homogeneous model to study the hydrate distribution and exploitation potential in this area. For the heterogeneous hydrate reservoirs whose distribution maps have been published, it is still necessary to spend a lot of effort to re-do the geological modeling using the existing methods, which usually includes multiple links such as data collection, parameter correction and model optimization, and is time-consuming and laborious. SUMMARY

[0005] The present application is a natural gas hydrate exploitation method and system based on color recognition. Since there are significant differences in the distribution of hydrate resources, it is necessary to develop an exploitation plan according to the distribution characteristics. This method can invert heterogeneous reservoir parameters according to geological pictures and other information, realize digital modeling of geological characteristics, and is especially suitable for geological environments with complex hydrate distribution and significant heterogeneous characteristics. It can analyze and study the results of hydrate exploitation, further optimize the exploitation plan, and improve the recovery rate. Compared with the traditional manual data processing method, the use of color recognition for data processing significantly improves the efficiency, reduces human intervention and errors, and makes the data processing more accurate. The generated geological characteristic parameters can correspond to the grids divided by the hydrate simulation software, and can be easily used for two-dimensional and three-dimensional numerical simulation to select the optimal exploitation plan.

[0006] The object of the present application is achieved at least by one of the following technical solutions.

[0007] The natural gas hydrate exploitation method based on color recognition comprises the following steps: S1, obtaining geological parameters through a real-time monitoring device, the geological parameters including gas saturation, liquid saturation, hydrate saturation, porosity, and permeability parameters of the natural gas hydrate reservoir; S2, simulating the exploitation process of the hydrate reservoir by using a heterogeneous geological model established by color recognition to obtain the mean and median values of the gas saturation, liquid saturation, hydrate saturation, porosity, and permeability parameters; S3, comparing the geological parameters obtained by the real-time monitoring device with the mean and median values of the gas saturation, liquid saturation, hydrate saturation, porosity, and permeability parameters obtained by numerical simulation, and correcting the heterogeneous geological model to obtain an exploitation model, using the exploitation model for numerical simulation to select an optimal exploitation scheme for exploitation.

[0008] Further, in step S2, the heterogeneous geological model is established by using color recognition technology, comprising the following steps: S2.1, collecting natural gas hydrate reservoir information and geological picture information, saving the natural gas hydrate reservoir information as a geological reference parameter text, and the natural gas hydrate reservoir information including thermal, chemical, mechanical, and hydraulic geological parameters in the thermal-hydraulic-mechanical-chemical four-field coupling process, wherein the hydraulic geological parameters include saturation parameters of different substances; S2.2, inputting the collected information into the established geological module to obtain corresponding heterogeneous geological parameters using the geological module; S2.3, performing numerical simulation on the obtained heterogeneous geological parameters, and fitting with existing production test numerical simulation results to obtain the corresponding heterogeneous geological model after adjustment.

[0009] Further, in step S2.1, the natural gas hydrate reservoir information specifically includes average hydrate saturation , average water saturation , average gas saturation , average thermal conductivity , average porosity , average permeability , average Young's modulus , and average solubility and other parameters; The geological picture information includes a distribution cloud diagram of reservoir parameter information.

[0010] Further, in step S2.2, the establishment of the geological module comprises the following steps: S2.2.1, recording and saving the spatial distribution cloud diagram of the natural gas hydrate reservoir information at different depths and positions; S2.2.2, load the spatial distribution cloud chart using image processing software; S2.2.3, determine the working area in the spatial distribution cloud chart, and divide the working area into grids; S2.2.4, denoising processing is performed on the noise signal of the working area; S2.2.5, identify and capture the RGB value of each grid in the working area, each RGB value marks an identifier, and generate an identifier text of matrix distribution; S2.2.6, map the color features to the reservoir parameters, convert the identifiers of the identifier text to the corresponding reservoir parameters, and form a reservoir text; S2.2.7, combine the natural gas hydrate reservoir information collected in step S2.1 with the spatial distribution cloud chart of the natural gas hydrate reservoir information saved in step S2.2.1 at different depths and positions, repeat steps S2.2.2 to S2.2.6, obtain various geological parameters of the required heterogeneous hydrate reservoir, and form a corresponding reservoir text.

[0011] Further, in step S2.2.3, the effective recognition working area is determined by selecting four boundary lines in the image processing software, and the area outside the boundary line is not recognized.

[0012] Further, in step S2.2.5, the RGB value of the color of each grid area is input or automatically captured in the image processing software, and each RGB value marks an identifier; a total of RGB values are selected and marked as 0, 1, 2, …, -1; First, the white color scale is selected, and white is used as the background color. Then, the multiple colors corresponding to the image working area are selected for color recognition and marking. If the color scale does not have a corresponding color in the image, the RGB value corresponding to the color scale is directly supplemented in the image.

[0013] Further, in step S2.2.6, the weight of each identifier in the identifier text is counted, and the weight corresponding to the identifier is marked as , , …, …, , is the weight corresponding to the th identifier, then , wherein the first identifier is the background color and is not counted in the statistics; The identifier of the identifier text is converted to the corresponding reservoir parameter as follows: Set the maximum saturation of hydrate to , and there are The first identifier is 0, which is not counted in statistics, and the The first identifier corresponds to the saturation , And Then according to , it is concluded that , wherein is the average value of the hydrate saturation given in the text of the geological reference parameter, is the weight of the th identifier; Each identifier is converted, the first identifier is converted to , and the th identifier is converted to The specific value, so that all the identifiers in the identifier text are converted into reservoir parameters, and the reservoir text is stored separately.

[0014] Further, in step S2.3, the obtained heterogeneous geological parameters are input into the hydrate simulation software, including inputting all the reservoir text data in step S2.2.6 and step S2.2.7 into the hydrate simulation software, inputting all the reservoir parameters in the reservoir text into the corresponding position in the geological module of the hydrate simulation software for numerical simulation, and fitting with the existing production numerical simulation results, and adjusting to obtain the corresponding heterogeneous geological model.

[0015] The system for implementing the natural gas hydrate exploitation method based on color recognition includes a real-time monitoring device for obtaining geological parameters and a geological model. The real-time monitoring device includes exploration and drilling equipment, reservoir core pressure maintaining sampling equipment and sample testing equipment. The geological model includes a heterogeneous geological model, and the obtained geological parameters are used to correct the heterogeneous geological model to obtain an exploitation model, and the exploitation model is used to select an optimal exploitation scheme.

[0016] The computer device of the present application comprises a memory, a processor and a computer program stored in the memory, and when the computer program is executed on the processor, the method is realized Compared with the prior art, the present application has the following advantages: 1、The present application adopts color recognition technology to inverse the heterogeneous reservoir parameters according to geological pictures and other information, converts the color information in the geological image into quantifiable parameters by fully exploiting existing public data resources and combining with part of default data, and then performs parameter statistics and data analysis to generate representative heterogeneous reservoir geological characteristics, and formulates the mining scheme according to the geological parameters and the test mining data, thereby not only enhancing the capturing ability of the complexity and heterogeneity of the geological body, but also reducing the dependence on the field survey data, making the modeling process more intelligent, greatly saving the geological modeling time and reducing the cost.

[0017] 2、The present application effectively simplifies the collection and processing of geological data, realizes digital modeling of geological characteristics, makes the originally time-consuming and labor-consuming geological modeling process efficient and easy to operate, is suitable for large-scale sea area and small-scale laboratory research, and is especially suitable for complex hydrate distribution and significant heterogeneous characteristics of geological environment, and is suitable for expanding and developing research on hydrate reservoirs. 3、The present application modularizes the process of generating a heterogeneous geological model, corrects the heterogeneous geological model according to the geological parameters obtained by the measured monitoring device, thereby obtaining a mining model, adopts the mining model to perform numerical simulation, selects an optimal mining scheme, and performs mining; the mining result of the hydrate can be analyzed and researched, and the mining scheme can be further optimized to improve the recovery rate; the present application uses color recognition technology to process data, is more intelligent and efficient, and reduces human intervention and errors. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is a natural gas hydrate mining method flowchart based on color recognition in the embodiment of the present application.

[0019] Figure 2 is a heterogeneous natural gas hydrate reservoir hydrate saturation original distribution cloud chart and a heterogeneous natural gas hydrate reservoir hydrate saturation distribution cloud chart obtained by color recognition restoration in the embodiment of the present application.

[0020] Figure 3 is an effect diagram after noise reduction processing in the embodiment of the present application.

[0021] Figure 4 is an identifier text schematic diagram of a generated matrix distribution in the embodiment of the present application.

[0022] Figure 5 is a measured monitoring device schematic diagram in the embodiment of the present application. DETAILED DESCRIPTION

[0023] For the person skilled in the art to better understand the present application, the following will be further described in detail in combination with the drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by the person skilled in the art without creative labor are within the protection scope of the present application.

[0024] As shown in the color recognition-based natural gas hydrate exploitation method of the present embodiment, according to the information of geological pictures and the like, the color recognition can be used to invert the parameters of the heterogeneous reservoir, which can be conveniently used for two-dimensional and three-dimensional numerical simulation, analysis of hydrate exploitation results, optimization of exploitation scheme and improvement of recovery rate, including the following steps: Figure 1 S1, obtaining the geological parameters of gas saturation, liquid saturation, hydrate saturation, porosity and permeability of the natural gas hydrate reservoir through the real-time monitoring device.

[0025] The real-time monitoring device includes exploration and drilling equipment, reservoir core pressure-maintaining sampling equipment and sample testing equipment. The exploration and drilling equipment is configured and built according to the seventh generation ultra-deep water semi-submersible drilling platform of "Blue Whale No. 1", the reservoir core pressure-maintaining sampling equipment selects the MSCL-PCATS testing system produced by Geotek Company in the United Kingdom, and the sample testing equipment selects the nuclear magnetic resonance equipment with model number of Bruker Minispec Q4, the CT scanner with model number of Zeiss Xradia520 Versa, the High Definition Induction Logging cable logging produced by Schlumberger Company and the permeability testing device with model number of CoreTest Systems PCD-2000.

[0026] S2, simulating the exploitation process of the hydrate reservoir by using the heterogeneous geological model established by color recognition to obtain the mean value and median value of the parameters of gas saturation, liquid saturation, hydrate saturation, porosity and permeability.

[0027] The corresponding heterogeneous geological model is established by using the color recognition technology, including the following steps: ​S2.1, Collecting information of natural gas hydrate reservoir and geological picture information from published authoritative literature, such as the literature of Ye et al. (Ye J.-l., Qin X.-w., Xie W.-w., et al. The second natural gas hydrate production test in the South China Sea[J]. China Geology, 2020, 3(2): 197-209.), and the average saturation of hydrate and the hydrate saturation distribution cloud diagram of the region can be obtained.

[0028] The collected information of natural gas hydrate reservoir includes the average saturation of hydrate , the average saturation of water , the average saturation of gas , and the average value of thermal conductivity , the average value of porosity , the average value of permeability , the average value of Young's modulus , the average value of solubility and other parameters of natural gas hydrate reservoir information, including the saturation information of different substances; the hydrate reservoir information is saved as a geological reference parameter text, and the geological picture information is saved as a file of spatial distribution cloud diagram.

[0029] S2.2, Input the collected information into the established geological module respectively, and obtain the corresponding heterogeneous geological parameters.

[0030] The geological module is the geological module of the hydrate simulation software, and the establishment of the geological module includes the following steps: S2.2.1, Record and save the spatial distribution cloud diagram of the natural gas hydrate reservoir information in the public information at different depths and positions.

[0031] S2.2.2, Load the spatial distribution cloud diagram using image processing software.

[0032] S2.2.3, Determine the working area in the spatial distribution cloud diagram, and divide the working area into grids in horizontal and vertical directions.

[0033] S2.2.4, Noise signal of the working area is processed, non-local mean denoising is adopted, and finally the quality of the denoised image is checked, including comparing the images before and after denoising, confirming the preservation of key details and the effectiveness of noise removal, to achieve the best effect.

[0034] S2.2.5, generate the identifier text of matrix distribution by inputting or automatically capturing the RGB value of each grid area color in the image processing software, each RGB value marks an identifier, a total of RGB values are selected, which are sequentially recorded as 0, 1, 2, …, -1.

[0035] First, select the white color scale, and white is used as the background color; then select multiple colors corresponding to the image working area for color identification and marking, if the color scale has no corresponding color in the image, directly supplement the RGB value corresponding to the color scale in the image.

[0036] S2.2.6, map the color characteristics to the reservoir parameters, convert the identifiers in the identifier text to the corresponding reservoir parameters, and form the reservoir text, specifically: calculate the weight of each identifier in the identifier text, the weight corresponding to the identifier is recorded as , The weight corresponding to the first identifier is 0, which is not counted in the statistics, and the weight corresponding to the th identifier is

[0037] Set the maximum hydrate saturation to , there are identifiers, the first identifier is 0 and is not counted in the statistics, and the th identifier corresponds to the saturation , and , then according to , it is obtained that , wherein is the average value of the hydrate saturation in the geological reference parameter text, is the weight of the th identifier.

[0038] Convert each identifier, convert the first identifier to , and convert the th identifier to specific value, so as to convert all identifiers in the identifier text to reservoir parameters, and save as a reservoir text.

[0039] S2.2.7, combine the natural gas hydrate reservoir information collected from the published authoritative literature in step S2.1 with the spatial distribution cloud map of the natural gas hydrate reservoir information saved in step S2.2.1 at different depths and positions, repeat steps S2.2.2 to S2.2.6, obtain various geological parameters of the required heterogeneous hydrate reservoir, and form the corresponding reservoir text.

[0040] S2.3, input the obtained heterogeneous geological parameters into the hydrate simulation software for numerical simulation.

[0041] Input all the reservoir text data in steps S2.2.6 and S2.2.7 into the hydrate simulation software, import all the reservoir parameters in the reservoir text into the corresponding positions in the geological module of the hydrate simulation software, perform numerical simulation and fitting with the production numerical simulation results disclosed in the literature, and adjust to obtain the corresponding heterogeneous geological model, thereby completing the geological modeling of the heterogeneous hydrate reservoir parameters.

[0042] The hydrate simulation software can be selected from CMG, TOUGH+HYDRATE, MH21-HYDRES, STOMP-HYDT or HYDRATE ResSim, etc.

[0043] S3, the measured monitoring device obtains the mean and median values of the gas saturation, liquid saturation, hydrate saturation, porosity, and permeability parameters obtained by the geophysical exploration method, the geochemical method, and the geological parameters obtained by the fidelity coring technology, and corrects the heterogeneous geological model to obtain a production model. The production model is used for numerical simulation, and the optimal production scheme is selected for production.

[0044] The mean and median values of the gas saturation, liquid saturation, hydrate saturation, porosity, and permeability parameters obtained by the measured monitoring device and the numerical simulation of the heterogeneous geological model are compared, and the heterogeneous geological model is corrected. When the error is less than 1%, the numerical simulation distribution cloud diagram of the heterogeneous geological model at this time is used as the model for production. The production model is numerically simulated, and the optimal production scheme is selected for actual production.

[0045] As an embodiment, the modeling of the Shenhu hydrate reservoir in the South China Sea is taken as an example, which specifically includes the following steps: S2.1, collect natural gas hydrate reservoir information, and collect information such as geological pictures disclosed in the literature.

[0046] The natural gas hydrate reservoir information includes thermal, hydraulic, mechanical, and chemical related geological parameters in the thermal-hydraulic-mechanical-chemical (THMC) process of the hydrate, including key material parameters such as average hydrate saturation , average water saturation , average gas saturation , average thermal conductivity , average porosity , average permeability , average Young's modulus , and average solubility Important parameters; geological picture information includes the distribution cloud of reservoir parameter information. Save the hydrate reservoir information as a geological reference parameter text, and save the geological picture information as a picture file.

[0047] Take the South China Sea Shenhu hydrate reservoir as an example, first model the heterogeneous hydrate saturation, collect comprehensive information on heterogeneous hydrate saturation, especially related geological logging data and cloud distribution images, according to the literature of Ye et al. (Ye J.-l., Qin X.-w., Xie W.-w., et al. The second natural gas hydrate production test in the South China Sea[J]. China Geology, 2020, 3(2): 197-209.) the average saturation of hydrate in this area is 0.233, and the hydrate saturation distribution is as shown in Figure 2 (a) of.

[0048] S2.2.1, record and save the spatial distribution cloud of natural gas hydrate reservoir information at different depths and positions in the literature and other published information. As shown in Figure 2 (a) of, download and save the heterogeneous hydrate saturation cloud.

[0049] S2.2.2, load the spatial distribution cloud using image processing software, such as inserting pictures using office software such as PowerPoint.

[0050] S2.2.3, determine the working area in the cloud and divide the grid: determine the effective recognition working area in the image software by selecting four boundary lines, the four boundary lines are as shown in the rectangular frame in Figure 2 (a), along the frame to crop and save as a new picture as the working area, the area outside the boundary line is not identified. Divide the grid unit according to the horizontal and vertical directions in the working area, first coarse division, then smaller grid division.

[0051] S2.2.4, noise reduction processing is performed on the identified noise signals and selected noise signals.

[0052] There are noise signals in the working area, these noise signal elements include irrelevant marks, messy lines or redundant text, these elements are often irrelevant to the main information to be analyzed, and will affect the accuracy of the results; for noise that is difficult to eliminate, non-local mean denoising is used to preserve the grid edges and key details. Finally, quality check is performed on the denoised image, including comparing the images before and after denoising, confirming the preservation of key details and the effectiveness of noise removal, to achieve the best effect. The effect after automatic recognition and denoising is as followsFigure 3 As shown.

[0053] S2.2.5. Identify and capture RGB values ​​within the working area. RGB values ​​can be obtained through input or automatically captured by selecting a location in the graphic, and then assigned a corresponding identifier. For ease of operation, natural numbers are used as identifiers. Obtain the RGB values ​​of the color of each grid area through software input or automatic capture, and mark each RGB value with an identifier, generating a matrix-distributed identifier text. The first selected RGB value is marked as 0, the second selected RGB value is marked as 1, and so on. The selected RGB values ​​are marked as -1; the identifier text of the generated matrix distribution is as follows Figure 4 As shown, this is denoted as the hydrate saturation identifier text.

[0054] In one embodiment, this embodiment first selects a white color stop, whose RGB value is (255, 255, 255). White is usually used as the background color. Then, a color corresponding to the working area is selected. For example, the working area's color is a gradient from red to blue, with 20 color stops. The RGB values ​​of these 20 color stops are (255, 15, 0), (255, 46, 0), (255, 78, 0), (255, 109, 0), and (255, 140, 0). (255, 172, 0), (255, 203, 0), (255, 235, 0), (243, 255, 0), (212, 255, 0), (180, 255, 0), (149, 255, 0), (117, 255, 0), (86, 255, 0), (55, 255, 0), (23, 255, 0), (0, 255, 0), (0, 255, 39), (0, 255, 70), (0, 255, 102). After color recognition, if some color marks do not have corresponding colors in the image, two color marks are directly added to the image. For example, the RGB values ​​of the two color marks are (68, 181, 73) and (0, 168, 78). These 23 color marks are represented by 0~22 respectively.

[0055] S2.2.6 Map color features to reservoir parameters, statistically analyze the probability distribution of reservoir parameters corresponding to different color features, and replace the corresponding identifiers with the corresponding reservoir parameters.

[0056] The grid within the working area is labeled with identifiers 0 through 22. The weight of each identifier in the hydrate saturation identifier text is calculated, with identifier 0 excluded from the statistics as it represents a hydrate saturation of 0. The weights corresponding to identifiers 1 through 22 are denoted as... ,in Average hydrate saturation The value is 0.233 (this value is obtained from step S2.1), and the maximum saturation of the hydrate is set to... Then the first Each identifier corresponds to saturation. identifier The weight is ,So , and thus The specific saturation values ​​of identifiers 1 through 22 are calculated. Each identifier in the hydrate saturation identifier text is converted into a hydrate saturation value using color recognition. The final hydrate saturation value is then generated as shown below. Figure 2 As shown in (b), cross-sectional views with depths of 1477m and 1515m and longitudinal distances of 603m and 848m were drawn and recorded as hydrate saturation files.

[0057] S2.2.7. Collect the natural gas hydrate reservoir information (including the average hydrate saturation) from publicly available authoritative literature in step S2.1. Average water saturation Average saturation of gas Key material parameters, and the average thermal conductivity. Average porosity Average penetration rate Average value of Young's modulus Average solubility (and other important parameters), combined with the spatial distribution cloud map of natural gas hydrate reservoir information at different depths and locations saved in step S2.2.1, repeat steps S2.2.2 to S2.2.6 to obtain various geological parameters of the required heterogeneous hydrate reservoir and form the corresponding reservoir text.

[0058] S2.3 Input the various geological parameters of the heterogeneous hydrate reservoir obtained through color recognition into the hydrate simulation software. The software can select CMG, TOUGH+HYDRATE, MH21-HYDRES, STOMP-HYDT or HYDRATEResSim, etc.

[0059] The parameters in the corresponding reservoir text are imported into the corresponding positions in the geological module of the hydrate simulation software, numerical simulation is carried out, and the simulation results are fitted with the test production numerical simulation results disclosed in the literature. According to the literature of Ye et al. (Ye J.-l., Qin X.-w., Xie W.-w., et al. The second natural gas hydrate production test in the South China Sea[J]. China Geology, 2020, 3(2): 197-209.), the second hydrate test was carried out in Shenhu sea area, and the average daily gas production was 2.87x10 4 m 3 , and the total gas production in 30 days reached 86.14x10 4 m 3 Taking the gas production as the fitting target, the corresponding heterogeneous geological model is obtained after adjustment, and the geological modeling of the heterogeneous hydrate reservoir parameters is completed.

[0060] The color recognition is used to establish the heterogeneous geological model, simulate the hydrate reservoir exploitation process, and obtain the mean and median values of the gas saturation, liquid saturation, hydrate saturation, porosity, permeability and other parameters.

[0061] S3, the distribution of the gas saturation, liquid saturation, hydrate saturation, porosity, permeability and other parameters obtained by using the heterogeneous geological model at different time steps is obtained, the mean and median values are counted, and the geological parameters obtained by the measured monitoring device are compared, the heterogeneous geological model is corrected, and thus the exploitation model is obtained. The exploitation model is numerically simulated, the optimal exploitation scheme is selected, and is used for actual exploitation.

[0062] The preferred embodiments disclosed above are only used to help explain the present application. The preferred embodiments do not describe all the details, nor limit the application to the specific embodiments described. Obviously, many modifications and changes can be made according to the content of the present application. The present application selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can well understand and utilize the present application.

Claims

1. A method for extracting natural gas hydrates based on color recognition, characterized in that, Includes the following steps: S1. Geological parameters are obtained through actual measurement and monitoring devices. These geological parameters include gas saturation, liquid saturation, hydrate saturation, porosity, and permeability parameters of the natural gas hydrate reservoir. S2. A heterogeneous geological model established using color recognition was used to simulate the mining process of hydrate reservoirs, and the mean and median values ​​of geological parameters were obtained. S3. By comparing the mean and median values ​​of the geological parameters obtained from the actual monitoring device with those obtained from the numerical simulation, the heterogeneous geological model is corrected to obtain the mining model. The mining model is then used for numerical simulation to select the optimal mining scheme for mining.

2. The method for natural gas hydrate extraction based on color recognition according to claim 1, characterized in that: In step S2, a heterogeneous geological model is established using color recognition, including the following steps: S2.1 Collect natural gas hydrate reservoir information and geological image information, and save the natural gas hydrate reservoir information as geological reference parameter text. The natural gas hydrate reservoir information includes thermal, chemical, mechanical and hydraulic geological parameters in the four-field coupling process of thermal-hydraulic-mechanical-chemical, wherein the hydraulic geological parameters include saturation parameters of different substances. S2.2 Input the collected information into the established geological module and use the geological module to obtain the corresponding heterogeneous geological parameters; S2.

3. Numerical simulation is performed on the obtained heterogeneous geological parameters, and the results are fitted with the existing numerical simulation results of pilot mining. After adjustment, the corresponding heterogeneous geological model is obtained.

3. The method for natural gas hydrate extraction based on color recognition according to claim 2, characterized in that: In step S2.1, the natural gas hydrate reservoir information specifically includes the average hydrate saturation. Average water saturation Average saturation of gas and the average thermal conductivity Average porosity Average penetration rate Average value of Young's modulus Average solubility Parameters; The geological image information includes cloud maps showing the distribution of reservoir parameter information.

4. The method for natural gas hydrate extraction based on color recognition according to claim 2, characterized in that: In step S2.2, the establishment of the geological module includes the following steps: S2.2.1 Record and store spatial distribution cloud maps of natural gas hydrate reservoirs at different depths and locations; S2.2.2 Load the spatial distribution cloud map using image processing software; S2.2.3 Determine the working area in the spatial distribution cloud map and divide the working area into grids; S2.2.

4. Noise reduction processing is performed on the noise signals in the working area; S2.2.5 Identify and capture the RGB values ​​of each grid within the working area, label each RGB value with an identifier, and generate identifier text for matrix distribution; S2.2.6 Map color features to reservoir parameters, convert the identifier of the identifier text to the corresponding reservoir parameter, and form the reservoir text; S2.2.

7. Combine the natural gas hydrate reservoir information collected in step S2.1 with the spatial distribution cloud map of the natural gas hydrate reservoir information at different depths and locations saved in step S2.2.1, and repeat steps S2.2.2 to S2.2.6 to obtain various geological parameters of the required heterogeneous hydrate reservoir and form the corresponding reservoir text.

5. The method for natural gas hydrate extraction based on color recognition according to claim 4, characterized in that: In step S2.2.3, the effective recognition working area is determined by selecting four boundary lines in the image processing software, and the area outside the boundary lines is not recognized.

6. The method for natural gas hydrate extraction based on color recognition according to claim 4, characterized in that: In step S2.2.5, the RGB values ​​of the color of each grid region are obtained by inputting into or automatically capturing the image in the image processing software, and each RGB value is labeled with an identifier; a total of [number missing] are selected. Each of the RGB values ​​is sequentially labeled 0, 1, 2, ... -1; First, a white color mark is selected and used as the background color. Then, multiple colors corresponding to the working area of ​​the image are selected for color recognition and marking. If the color mark does not have a corresponding color in the image, the RGB value corresponding to the color mark is directly added to the image.

7. The method for natural gas hydrate extraction based on color recognition according to claim 4, characterized in that: In step S2.2.6, the weight of each identifier in the identifier text is calculated, and the weight corresponding to the identifier is denoted as . , ... …… , For the first The weights corresponding to each identifier are then: The first identifier is the background color and is not included in the statistics; Convert the identifier text to the corresponding reservoir parameter, as follows: The maximum saturation of hydrates is set to There are a total of There are 10 identifiers, where the first identifier is 0 and is not counted in the statistics. Each identifier corresponds to saturation. , and According to , and thus ,in This represents the average hydrate saturation value given in the geological reference parameter text. For the first The weight of each identifier; Transform each identifier, converting the first identifier to... , will the Each identifier is converted into The specific values ​​are then used to convert all the identifiers in the identifier text into reservoir parameters, which are then saved as reservoir text.

8. The method for natural gas hydrate extraction based on color recognition according to claim 4, characterized in that: In step S2.3, the acquired heterogeneous geological parameters are input into the hydrate simulation software. This includes inputting all reservoir text data from steps S2.2.6 and S2.2.7 into the hydrate simulation software, importing all reservoir parameters from the reservoir text into the corresponding positions in the geological module of the hydrate simulation software for numerical simulation, fitting the results with the existing pilot production numerical simulation results, and adjusting them to obtain the corresponding heterogeneous geological model.

9. A system for implementing the color recognition-based natural gas hydrate extraction method of claim 1, characterized in that: This includes field monitoring devices for acquiring geological parameters and geological models; The actual measurement and monitoring device includes exploration and drilling equipment, reservoir core pressure-maintaining sampling equipment, and sample testing equipment. The geological model includes a heterogeneous geological model. The obtained geological parameters are used to correct the heterogeneous geological model to obtain the mining model. The mining model is then used to select the optimal mining scheme.

10. A computer device, characterized in that, include: A memory and a processor, and a computer program stored in the memory, which, when executed on the processor, implements the method as described in any one of claims 1 to 8.

Images