A coal seam thickness and coal body structure spatial distribution combination type classification method

Abstract

The application discloses a coal seam thickness and coal body structure spatial distribution combination type classification method, and belongs to the technical field of coal reservoir reconstruction. The method first collects well logging data of a coalbed methane well in a target area, and establishes a coal body structure well logging interpretation model based on coring samples, and then identifies the coal body structure of a well without coring. Subsequently, GSI quantitative classification criteria are established, sensitive well logging parameters are selected to establish a GSI prediction model, and the quantitative identification of the coal body structure type is realized by calculating the GSI value. Then, the coal seam thickness and the coal body structure spatial distribution combination type are classified in combination with the thickness, proportion and spatial combination mode of different coal body structures. The application can indicate the direction of efficient reservoir reconstruction of the coalbed methane well through accurate identification of the coal body structure and division of the combination type, and then ensure the reconstruction effect of the coalbed methane well.

Description

Technical Field

[0001] This invention relates to the field of coal reservoir stimulation technology, specifically to a classification method based on the combination of coal seam thickness and spatial distribution of coal body structure. Background Technology

[0002] Coal body structure is the product of coal reservoir deformation under stress, and different coal body structures have different reservoir stimulation effects. Generally speaking, hard coal (natively structured coal and fractured coal) is suitable for conventional fracturing technology, and its reservoir stimulation effect is better; conversely, the stimulation effect of soft coal (granular coal and mylonite) is not satisfactory. The influence of coal body structure on fracturing effect is mainly reflected in the following two aspects: ① When the stress is uniform, the more severe the degree of coal body structure fragmentation, the less likely it is to form a new fracture surface at the fracture tip, and the shorter the outward extension length of the fracture is; ② The more severe the degree of coal body structure fragmentation, the easier it is for the borehole to collapse during fracturing, and a large amount of coal dust and coal slime are generated, causing serious reservoir damage and affecting the production capacity and development of coalbed methane wells.

[0003] To improve the effectiveness of coal seam stimulation, accurate identification and classification of coal body structure and its combination types are necessary. Currently, the focus is mainly on the impact of a single coal body structure on fracturing, treating the coal reservoir as a whole. The influence of coal body structure on reservoir stimulation is reflected by changes in permeability during stress-strain processes, neglecting the impact of different coal body thicknesses and combinations on fracturing. This leads to unsatisfactory reservoir stimulation results in coalbed methane wells. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a classification method based on the combination of coal seam thickness and coal body structure spatial distribution types, in order to solve the problem that the existing coal reservoir classification and stratification is too simple and affects the coal reservoir stimulation effect.

[0005] The technical solution adopted in this invention is: A classification method based on the combination of coal seam thickness and spatial distribution of coal body structure includes the following steps: S1. Collect logging data of each coalbed methane well in the target area; S2. Core samples are taken from the coal seam within the target area, and a well logging interpretation model of the coal body structure is established based on the core samples and the well logging series data from S1. S3. Based on the logging series data of S1 and the well logging interpretation model of coal body structure established by S2, the coal body structure of the non-cored well is identified, and then the coal body structure type is determined well by well and point by point. S4. Establish the GSI quantitative classification criteria and classify coal body structure types according to the criteria; S5. Select the sensitive logging parameters of the logging curve, and establish a GSI prediction model by combining the sensitive logging parameters with the GSI value; S6. Obtain the logging curves of the coalbed methane wells to be classified, obtain sensitive logging parameters based on the logging curves, calculate the GSI value using the GSI prediction model established in S5, and identify the coal body structure type based on the GSI value; then, in combination with the determination of different coal body structure thicknesses, classify the coal seam thickness and the spatial distribution combination type of coal body structure.

[0006] The beneficial technical effects of the present invention are as follows: This invention quantitatively identifies coal bodies based on well logging curves and GSI values, providing support for the identification of longitudinal coal body structures in coal seam segments. It also classifies and analyzes coal body structure combination types by combining coal seam thickness, the proportion of coal seam thickness in different coal body structures, and spatial combination patterns. This establishes a classification method based on coal seam thickness and the spatial distribution of coal body structure combinations, providing a theoretical basis for the classification of longitudinal coal reservoir combination types and pointing the way for precise coal reservoir stimulation, significantly improving the stimulation effect of coal reservoirs. Attached Figure Description

[0007] Figure 1 This is a flowchart illustrating the classification method of the present invention based on the combination of coal seam thickness and spatial distribution of coal body structure. Figure 2 A diagram showing the correspondence between typical well coal body structure and different well logging curves; Figure 3 GSI quantification template diagram for different coal body structures; Figure 4 This diagram shows different coal body structure combinations. Detailed Implementation

[0008] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be described in detail below with reference to the accompanying drawings, in order to explain the core points of the present invention.

[0009] To accurately stimulate coal seams and improve their effectiveness, this invention proposes a classification method based on the combination of coal seam thickness and spatial distribution of coal body structure. This method quantitatively and accurately identifies the coal body structure using coal seam logging curves and GSI values. Based on the coal seam thickness, the identification results of the coal body structure, and its spatial distribution, the method finely classifies the vertical coal body structure combination types, deriving different reservoir combination types and laying the foundation for selecting coal seam fracturing stimulation process parameters.

[0010] This invention provides a classification method based on the combination of coal seam thickness and spatial distribution of coal body structure, as described below. Figures 1-4 To elaborate in detail.

[0011] like Figure 1 As shown, this classification method includes the following steps: A classification method based on the combination of coal seam thickness and spatial distribution of coal body structure includes the following steps: S1. Collect logging data from each coalbed methane well within the target area.

[0012] The specific steps are as follows: Collect logging series data from each coalbed methane well in the target area, including natural gamma ray logging (GR), density compensated logging (DEN), acoustic transit time logging (AC), and deep lateral resistivity logging (RT), such as... Figure 2 As shown, all well logging curves and interpretation results are imported into Petrel geological modeling software. Using GR and AC curves as the main references and DEN curves as a supplement, the top and bottom interfaces of the coal seam are picked up interactively well by well to generate well stratification data tables.

[0013] Align the interpreted coal body structure type curves with the depth of the layered data, and export a structured dataset containing depth, thickness grade, and coal body structure category at 0.1-meter sampling intervals.

[0014] Due to factors such as drill string tension and core breakage during drilling, there is a systematic error (typically 0.5 to 2.0 meters) between the core recording depth and the actual logging depth. Depth alignment is necessary to establish an accurate correspondence between coal seam structure and logging response.

[0015] S2. Core samples are taken from the coal seam within the target area, and a well logging interpretation model of the coal body structure is established based on the core samples and the well logging series data from S1.

[0016] Specifically, the following steps are included: S21. Deploy core wells within the target area according to the principle of control, continuously core the target coal seam, and directly identify and determine the coal body structure type (Type I primary structure coal, Type II fractured coal, Type III granular coal, and Type IV mylonite) of each core section through macroscopic description and auxiliary testing.

[0017] S22. Accurately relocate the coring depth to the logging depth, extract the logging series data corresponding to the relocated coring section, such as the logging curve values ​​of natural gamma (GR), compensated density (DEN), sonic transit time (AC), and deep lateral resistivity (RT), and establish a sample set of coal body structure type-logging response values.

[0018] S23. Perform statistical analysis on the sample set to determine the range of characteristic values ​​for various coal body structures on each logging curve.

[0019] Further analysis using cross plots such as DEN-AC and GR-RT was conducted to determine the clustering regions and discrimination boundaries of various coal body structures on the two-dimensional plane. At the same time, the morphological characteristics of logging curves corresponding to various coal body structures (such as the degree of toothing, box-shaped features, and fluctuation amplitude) were summarized.

[0020] S24. Based on the range of comprehensive characteristic values, the boundary of the intersection diagram, and the morphological characteristics of the logging curves, a multi-parameter fusion logging interpretation model for coal body structure is established.

[0021] S3. Based on the logging series data of S1 and the well logging interpretation model of coal body structure established by S2, the coal body structure of the non-cored well is identified, and then the coal body structure type is determined well by well and point by point.

[0022] Specifically, the following steps are included: S31. Obtain the logging curves of the target area without coring wells, and perform preprocessing and depth alignment.

[0023] S32. Input the preprocessed logging curves into the coal body structure logging interpretation model established in S2, and use the following comprehensive identification method to determine the coal body structure type well by well and point by point: (1) Threshold preliminary judgment: Compare the logging parameter values ​​with the characteristic value range determined by S23 to preliminarily delineate the possible structural types.

[0024] (2) Cross-plot precise judgment: For well sections falling into the overlapping area, cross-plots such as DEN-AC or GR-RT are used to accurately assign them based on the clustering regions and discrimination boundaries determined in S23.

[0025] (3) Morphological correction: Combine the morphological characteristics such as the degree of toothing and box-shaped features of the logging curve to perform consistency verification and correction on the identification results.

[0026] S33. Output the interpretation results of coal body structure types at continuous depths of each non-coring well, forming a coal body structure classification data body for the entire region.

[0027] S4. To transform discrete coal body structure types into continuous and quantifiable mechanical indicators, such as... Figure 3 As shown, the GSI quantitative classification criterion is introduced, and the coal body structure type is classified according to this criterion.

[0028] Specifically, the criteria classify coal body structure into four levels, corresponding to the following GSI value ranges: GSI of 70-100 indicates primary structure coal, 50-70 indicates fractured coal, 25-50 indicates granular coal, and 0-25 indicates mylonite.

[0029] S5. Select the sensitive logging parameters of the logging curve and establish a GSI prediction model by combining the sensitive logging parameters with the GSI value.

[0030] Specifically, the following steps are included: S51. For each core sample obtained in S2, based on the structural characteristics description and surface condition evaluation according to the GSI criteria, assign a specific GSI value (not a range value) to each sample. The assignment methods include: a semi-quantitative lookup method based on the GSI description table, a quantitative determination method based on core mechanics testing, or a simplified assignment method using the median of various types of GSI intervals.

[0031] S52. Match the GSI assignment results of each core sample with the logging curve values ​​(GR, DEN, AC, RT) to form a GSI-logging response sample set.

[0032] S53. Calculate the single correlation coefficient between GSI and each logging parameter, and select parameters with an absolute value of correlation coefficient greater than 0.7 as sensitive logging parameters, including but not limited to DEN, AC, and RT.

[0033] S54. Using GSI as the dependent variable and the selected sensitive logging parameters as independent variables, establish a GSI prediction model using multiple linear regression or partial least squares regression methods, as follows: ; In the formula, M is the GSI value corresponding to different coal body structures; a, b, c, d, and e are the regression coefficients of each logging parameter; and DEN is the logging bulk density, g / cm³. 3 ; GR is the logging gamma value, API; CALX is the logging diameter in the X direction, cm; CALY is the logging diameter in the Y direction, cm.

[0034] S55. The GSI prediction model is validated using the reserved core samples. Once validated, the GSI prediction model can be applied to all coreless wells in the region to achieve continuous prediction of GSI values. Furthermore, it can be converted back into coal body structure type based on the range of GSI values.

[0035] S6. Obtain the logging curves of the coalbed methane wells to be classified, obtain sensitive logging parameters based on the logging curves, calculate the GSI value using the GSI prediction model established in S5, and identify the coal body structure type based on the GSI value; then, in combination with the determination of different coal body structure thicknesses, classify the coal seam thickness and the spatial distribution combination type of coal body structure.

[0036] Specifically, the following steps are included: S61. Obtain logging curve data of coalbed methane wells to be classified within the target area, including at least the sensitive logging parameters (DEN, AC, RT, etc.) selected in S5, and perform preprocessing.

[0037] S62. Substitute the preprocessed logging parameters point by point into the GSI prediction model established in S5, calculate the GSI prediction value at each depth point, and form a continuous curve of depth-GSI value.

[0038] S63. Based on the correspondence between GSI values ​​and coal body structure types established in S4 (GSI≥70 is primary structure coal, 50-70 is fractured coal, 25-50 is granular coal, <25 is mylonite), the continuous GSI values ​​are converted into discrete coal body structure types to achieve quantitative characterization of the longitudinal distribution of coal body structure.

[0039] S64. Based on the converted coal body structure type, calculate the cumulative thickness of various coal body structures within the target coal seam segment, and calculate the percentage of each structure thickness to the total coal seam thickness.

[0040] S65. Based on the thickness ratio and vertical sequence characteristics, the coal body structure combination is finely divided.

[0041] S66. Output the detailed subdivision results of the coal body structure of a single well, including the thickness, proportion, combination type, and vertical sequence of each structure type, providing basic data for subsequent three-dimensional spatial modeling and engineering applications.

[0042] In step S1, depth alignment can be performed using conventional methods. For example, in this embodiment, depth alignment can be performed using a combination of natural gamma scanning and marker layer contrast methods. (1) Perform continuous natural gamma scans on the core samples to obtain the core gamma curves; (2) Match the core gamma curve with the well logging natural gamma curve to calculate the optimal depth shift; (3) Use stable limestone and mudstone layers in the region as auxiliary marker layers for secondary verification; (4) After the depth alignment is completed, the alignment error of the main marker layer is required to be ≤0.2 meters and the error of the coal seam top and bottom interface is required to be ≤0.3 meters.

[0043] The aligned core depth serves as the sole reference depth for extracting logging responses from all subsequent core sections.

[0044] In step S23 above, cross-plot analysis is further employed because the value ranges for different coal body structure types overlap. For example, in a specific instance, the natural gamma value of primary structure coal is 45~108 API, and the density value is 1.35~1.87 g / cm³. 3 The well diameter X is 21~26cm, and the well diameter Y is 20~24cm; the natural gamma value of the fractured coal is 27~66 API, and the density is 1.24~1.43g / cm³. 3 The well diameter X is 27~35cm, and the well diameter Y is 28~34cm; the natural gamma value of the granular coal to mylonite is 18~46 API, and the density is 1.18~1.34g / cm³. 3 The well diameter X is 38~45cm, and the well diameter Y is 35~42cm.

[0045] Furthermore, step S31 above includes the following steps: S311. Obtain LAS format logging curve data for non-cored wells within the target area, including at least GR, DEN, AC, and RT.

[0046] S312. Preprocess the raw logging data, including: data loading and format conversion, environmental correction, outlier handling, and curve standardization.

[0047] S313. Perform depth alignment on the pre-processed logging curves, including: (1) Depth matching between curves: Check and correct the depth drift between different logging curves to ensure that the GR, DEN, AC and RT curves correspond strictly at the depth sampling points; (2) Formation depth correction: Using stable marker layers in the region (such as limestone and mudstone layers), the logging depth is shifted as a whole or corrected in segments to the true formation depth benchmark.

[0048] S314, Output the standard logging curve data body after depth alignment.

[0049] Furthermore, step S65 above includes the following steps: S651: Based on the thickness and proportion of each coal body structure type statistically analyzed in S64, the thickness ratio characteristics of a single well are determined.

[0050] S652: Analyze the vertical sequence characteristics of a single well, including the following five elements: (1) Arrangement order: Record the order of appearance of coal body structure types from top to bottom, such as fractured coal → granular coal → mylonite; (2) Thickness variation trend: Analyze the vertical evolution law of the thickness of each type of structure, and divide it into increasing type, decreasing type, stable type and fluctuating type; (3) Interface characteristics: Identify the contact relationship between different types, and divide them into abrupt interface (transition zone <0.1m), gradual interface (transition zone 0.1-0.5m), and serrated interface (alternating appearance); (4) Location of dominant structures: Determine the distribution range of the structure type with the highest proportion in the coal seam, and divide it into top dominance, middle dominance, bottom dominance, and whole-seam dominance; (5) Complexity: The number of vertical type changes is counted and divided into simple type (1 type), compound type (2 types, 1-2 changes), and complex type (≥3 types or ≥3 changes).

[0051] S653: Combining thickness ratio and vertical sequence characteristics, a refined division is performed according to the following rules: (1) Complete type: The proportion of primary structure coal is ≥80%, and the vertical sequence is single type; (2) Fragmented type: The proportion of primary + fragmented coal is ≥80% and fragmented coal is the main component. The vertical sequence is either single type or binary composite type (mainly whole above and fragmented below). (3) Transitional type: primary + fractured + fragmented proportion ≥ 80% and fragmented coal proportion > 20%, vertical sequence is binary composite or ternary composite; (4) Crushed type: Mylonite accounts for ≥30%, and the vertical sequence is binary composite type (mainly crushed above and whole below) or complex type; (5) Complex alternating type: No dominant structure (each type accounts for <50%), and the vertical sequence is complex (type changes ≥3 times).

[0052] S654: Outputs the fine-grained subdivision results of the coal body structure combination of a single well, including combination type, vertical sequence pattern, interface features, and dominant structure location, providing a basis for subsequent three-dimensional spatial modeling and engineering applications.

[0053] Based on the above classification method, this invention also provides a specific application example. This application example studies and collects data from several coalbed methane wells with complete logging data, but most lack core data. Using the GSI quantitative characterization method for coal body structure, the coal body structure of the selected well coal seams is identified and stratified. Vertically, the coal seams are mainly composed of 1-6 different coal body structures, with significant variations in vertical variation. Simultaneously, the proportion of primary structure coal ranges from 0-43%, with an average thickness of 11%; the proportion of fractured coal ranges from 0-97%, with an average thickness of 43%. The total thickness of primary structure coal and fractured structure coal (hard coal) ranges from 0-100%, with an average thickness of 54%, while the thickness of fragmented and mylonite coal (soft coal) ranges from 0-100%, with an average thickness of 46%. From the perspective of coal body structure classification, using the hard coal and soft coal identification method, different spatial combination types of coal body structures are refined into three types: binary, ternary, and multi-element, such as... Figure 4 As shown. Among them, the binary type: the upper part is hard coal, and the lower part is soft coal. This type of combination can be divided into three types, with hard coal accounting for more than 80%, 60%~80%, and 40%~60% respectively; the upper part is soft coal, and the lower part is hard coal. This type of combination can also be divided into three types, with soft coal accounting for more than 80%, 60%~80%, and 40%~60% respectively. The ternary type: "two soft coals sandwiching one hard coal" with hard coal accounting for 70%~100% and 0~30% respectively; "two hard coals sandwiching one soft coal" with soft coal accounting for 70%~100% and 0~30% respectively. The multi-element type: hard coal and soft coal alternate.

[0054] In summary, existing technologies often overlook the proportion of coal seam thickness and spatial combination characteristics, leading to blind coal seam stimulation and delaying the stimulation effect of coalbed methane wells. This invention uses well logging curves and GSI values ​​to quantitatively identify and classify coal body structure, and considers the proportion of coal seam thickness and spatial combination characteristics to classify coal reservoir combination types. This classification method can more accurately guide reservoir stimulation.

[0055] Specifically, this invention combines multiple regression analysis with the introduction of GSI values ​​to accurately and quantitatively identify and analyze the coal body structure of coal seams, improving the accuracy and effectiveness of coal body structure identification. During the research process, a large amount of logging data from coalbed methane wells was screened and analyzed. The logging curves that showed strong responses to coal body structure were identified as natural gamma ray logging (GR), density logging (DEN), and caliper logging (CAL), laying the foundation for establishing an accurate mathematical model for coal body structure identification. Using the logging identification model, the coal body structure of a large number of coal seams can be accurately identified. Based on the identification results, the proportion of coal seam thickness for different coal body structures is statistically analyzed. Simultaneously, combined with the spatial combination characteristics of the coal seams, the coal body structure combination types are accurately classified and analyzed.

[0056] The method of this invention is universally applicable to the identification of coal body structures in coalbed methane wells or underground coal mines. Through accurate identification of coal body structures and classification of combination types, this invention can provide direction for efficient reservoir stimulation of coalbed methane wells, thereby ensuring the effectiveness of the stimulation.

[0057] Specifically, this invention provides methodological support for the precise description and identification of coal reservoirs, pointing the way for efficient coalbed methane development. By accurately identifying coal body structure and coal seam combination types, it can provide a basis for selecting fracturing process parameters, while significantly improving the effectiveness of reservoir stimulation, thereby ensuring increased coalbed methane production.

[0058] The above descriptions are merely preferred embodiments of the present invention. The present invention is not limited to the embodiments listed above. It should be noted that any equivalent substitutions or obvious modifications made by those skilled in the art under the guidance of this specification fall within the scope of this specification and should be protected by the present invention.

Claims

1. A classification method based on the combination of coal seam thickness and spatial distribution of coal body structure, characterized in that... Includes the following steps: S1. Collect logging data from each coalbed methane well within the target area; S2. Core samples are taken from the coal seam within the target area, and a well logging interpretation model of the coal body structure is established based on the core samples and the well logging series data from S1. S3. Based on the logging series data of S1 and the well logging interpretation model of coal body structure established by S2, the coal body structure of the non-cored well is identified, and then the coal body structure type is determined well by well and point by point. S4. Establish the GSI quantitative classification criteria and classify coal body structure types according to the criteria; S5. Select the sensitive logging parameters of the logging curve, and establish a GSI prediction model by combining the sensitive logging parameters with the GSI value; S6. Obtain the logging curves of the coalbed methane wells to be classified, obtain sensitive logging parameters based on the logging curves, calculate the GSI value using the GSI prediction model established in S5, and identify the coal body structure type based on the GSI value; then, in combination with the determination of different coal body structure thicknesses, classify the coal seam thickness and the spatial distribution combination type of coal body structure.

2. The classification method based on the combination of coal seam thickness and spatial distribution of coal body structure according to claim 1, characterized in that, S1 includes the following steps: collecting logging series data of each coalbed methane well in the target area, including natural gamma logging curves, compensated density logging curves, sonic transit time logging curves, and deep lateral resistivity logging curves; importing all logging curves and interpretation results into Petrel geological modeling software; interactively picking the top and bottom interfaces of the coal seam well by well; and generating a well stratification data table.

3. The classification method based on the combination of coal seam thickness and coal body structure spatial distribution according to claim 2, characterized in that, S2 includes the following steps: S21. Core the target coal seam within the target area to directly identify and determine the coal body structure type of each core section; S22. Reset the coring depth to the logging depth, extract the logging series data corresponding to the coring section after resetting, and establish a sample set of coal body structure type-logging response value. S23. Perform statistical analysis on the sample set to determine the range of characteristic values ​​for various coal body structures on each logging curve. S24. Establish a well logging interpretation model for coal body structure based on the range of eigenvalues.

4. The classification method based on the combination of coal seam thickness and coal body structure spatial distribution according to claim 3, characterized in that: In S21, the coal body structure types are classified as follows: Type I primary structure coal, Type II fractured coal, Type III granular coal, and Type IV mylonite; S23 also includes the following steps: Since the value ranges of different coal body structure types overlap, cross plot analysis is used to determine the clustering regions and discrimination boundaries of various coal body structures on the two-dimensional plane; at the same time, the logging curve morphology characteristics corresponding to various coal body structures are summarized. In S24, a multi-parameter fusion well logging interpretation model for coal body structure is established by integrating the range of eigenvalues, the boundary of cross plot discrimination, and the morphological characteristics of logging curves.

5. The classification method based on the combination of coal seam thickness and spatial distribution of coal body structure according to claim 4, characterized in that, S3 includes the following steps: S31. Obtain logging curves for wells without coring within the target area and perform preprocessing. S32. Input the preprocessed logging curves into the coal body structure logging interpretation model established in S2, and use the following comprehensive identification method to determine the coal body structure type well by well and point by point: (1) Initial threshold judgment: The logging parameter values ​​are compared with the characteristic value range determined by S23 to preliminarily determine the structure type; (2) Cross-plot precise judgment: For well sections falling into the overlapping area, use DEN-AC or GR-RT cross-plots to accurately assign them according to the clustering regions and discrimination boundaries determined in S23; (3) Morphological correction: Based on the morphological characteristics of the logging curves, the identification results are checked and corrected for consistency. S33. Output the interpretation results of coal body structure types at continuous depths of each non-coring well, forming a coal body structure classification data body for the entire region.

6. The classification method based on the combination of coal seam thickness and spatial distribution of coal body structure according to claim 5, characterized in that, In S4: The GSI quantitative classification criterion divides the coal body structure into four levels, with the following corresponding GSI value ranges: 70≤GSI<100 is primary structure coal, 50≤GSI<70 is fractured coal, 25≤GSI<50 is granular coal, and GSI<25 is mylonite.

7. The classification method based on the combination of coal seam thickness and spatial distribution of coal body structure according to claim 6, characterized in that, The GSI prediction model established in S5 is as follows: ； In the formula, M is the GSI value corresponding to different coal body structures; a, b, c, d, and e are the regression coefficients of each logging parameter; and DEN is the logging bulk density, g / cm³. 3 ; GR is the logging gamma value, API; CALX is the logging diameter in the X direction, cm; CALY is the logging diameter in the Y direction, cm.

8. The classification method based on the combination of coal seam thickness and spatial distribution of coal body structure according to claim 6, characterized in that, S5 includes the following steps: S51. Assign a specific GSI value to each core sample obtained in S2; S52. Align the GSI assignment results of each core sample with the logging curve values ​​to form a GSI-logging response sample set; S53. Calculate the single correlation coefficient between GSI and each logging parameter, and select parameters with an absolute value of correlation coefficient greater than 0.7 as sensitive logging parameters; S54. Using GSI as the dependent variable and the selected sensitive logging parameters as independent variables, establish a GSI prediction model using multiple linear regression or partial least squares regression methods, as follows: ； S55. The GSI prediction model is validated using the reserved core samples. Once validated, the GSI prediction model can be applied to all coreless wells in the region to achieve continuous prediction of GSI values. Furthermore, it can be converted back into coal body structure type based on the range of GSI values.

9. The classification method based on the combination of coal seam thickness and spatial distribution of coal body structure according to claim 8, characterized in that, In S51, the assignment methods used include: a semi-quantitative lookup method based on the GSI description table, a quantitative determination method based on core mechanics testing, or a simplified assignment method using the median of each type of GSI interval.

10. A classification method based on the combination of coal seam thickness and spatial distribution of coal body structure according to claim 9, characterized in that, S6 includes the following steps: S61. Obtain logging curve data of coalbed methane wells to be classified within the target area, including at least the sensitive logging parameters selected in S5, and perform preprocessing. S62. Substitute the preprocessed logging parameters point by point into the GSI prediction model established in S5, calculate the GSI prediction value at each depth point, and form a continuous curve of depth-GSI value. S63. Based on the correspondence between GSI values ​​and coal body structure types established in S4, the continuous GSI values ​​are converted into discrete coal body structure types to achieve a quantitative characterization of the longitudinal distribution of coal body structure. S64. Based on the converted coal body structure type, calculate the cumulative thickness of various coal body structures within the target coal seam segment, and calculate the percentage of each structure thickness to the total coal seam thickness. S65. Based on the thickness ratio and vertical sequence characteristics, the coal body structure combination is finely divided; S66. Output the detailed subdivision results of the coal body structure of a single well, including the thickness, proportion, combination type, and vertical sequence of each structure type, providing basic data for subsequent applications.

Images