Coal rock gas reservoir compressibility evaluation method and system

Abstract

The invention relates to the technical field of coal rock gas reservoir evaluation, in particular to a coal rock gas reservoir compressibility evaluation method and system, and the method comprises the steps: obtaining and sorting multi-source reservoir comprehensive test data of a research area; constructing a prediction model between rock physical property and mechanical parameters; calculating main rock mechanical parameters of the stratum based on the prediction model; calculating the principal stress distribution of the coal rock reservoir, the top plate and the bottom plate by combining an experiment and a combined spring model; the method comprises the following steps: calculating mechanical and stress differences between a reservoir and a top and bottom plate stratum, revealing geomechanical difference characteristics, and determining main parameters influencing the compressibility of the reservoir by using Pearson correlation analysis and an XGBoost integrated learning algorithm based on gas production data of a coal and rock gas well; according to the method, the geomechanical difference between the coal rock and the top and bottom plates is comprehensively reflected, and the accuracy of the compressibility evaluation of the deep coal rock gas reservoir is improved.

Description

Technical Field

[0001] This invention relates to the field of coal gas reservoir evaluation technology, and in particular to a method and system for evaluating the compressibility of coal gas reservoirs. Background Technology

[0002] Deep coal and gas reservoirs, due to long-term geological processes, generally exhibit characteristics such as great burial depth, high in-situ stress, poor pore and fracture development, and extremely low permeability. This makes it difficult for conventional reservoir stimulation and evaluation methods to accurately reflect their true fracturing characteristics. Especially in deep environments, the rock structure is dense, the stress field is complex, and the mechanical differences between the top and bottom plates are significant. Conventional fracturing prediction methods often fail to accurately reflect the deformation and fracturing response of the reservoir under external pressure.

[0003] Currently, the evaluation of the fracturing capability of coalbed methane reservoirs mainly draws on the evaluation system for shallow coalbed methane or tight gas reservoirs, using a single indicator (such as rock brittleness index, horizontal in-situ stress, or Poisson's ratio) to determine fracturing capability. However, the fracturing behavior of deep coalbed methane reservoirs is influenced by a combination of factors, including differences in the mechanical properties of the reservoir and the top and bottom rocks, interlayer stress differences, reservoir tensile strength, elastic modulus, and formation tectonic strain characteristics. Relying on a single factor for judgment can easily lead to evaluation bias, resulting in problems such as inaccurate reservoir selection, limited fracture development, or ineffective fracturing in fracturing projects, seriously affecting production capacity and economic benefits.

[0004] Furthermore, the geological conditions of deep coal-rock gas reservoirs are complex, and blindly implementing high-pressure stimulation can easily lead to wellbore instability, reservoir damage, and even wellbore safety risks, further increasing development costs and technical difficulties. Current technologies have not yet developed a systematic evaluation method that comprehensively reflects the differences in geomechanics, stress states, and interlayer coupling between the coal-rock reservoir and its top and bottom plates, nor do they possess a quantitative compressibility evaluation system based on a combination of multi-source test data and machine learning models.

[0005] Therefore, a method for evaluating the compressibility of coal-gas reservoirs based on the geomechanical differences between coal and rock and the roof and floor is needed. This method should comprehensively utilize core experimental data, well logging data, and geological parameters. By establishing a rock mechanics prediction model, differential strain experiments, and stress field calculation model, and combining the data-driven XGBoost ensemble learning algorithm, it can achieve quantitative calculation and graded evaluation of reservoir compressibility indicators, thereby improving the accuracy of judging the fracturing capability of deep coal-gas reservoirs and providing a scientific basis for fracturing design and production capacity prediction. Summary of the Invention

[0006] This invention addresses the challenges of complex geological conditions in deep coal-gas reservoirs and the insufficient accuracy of existing fracturing capability assessment methods. It proposes a method and system for evaluating the fracturing capability of coal-gas reservoirs. This method establishes a predictive model of rock mechanics parameters between the coal and the roof and floor using comprehensive reservoir testing data, and calculates key mechanical parameters based on well logging data. It determines the formation stress state based on differential strain experiments and a combined spring model. Pearson correlation analysis and the XGBoost ensemble learning algorithm are used to screen key parameters affecting gas production, constructing a fracturing capability evaluation index, and clustering algorithms are employed to classify the reservoir. This method comprehensively reflects the geomechanical differences between the coal and the roof and floor, improving the accuracy of fracturing capability assessment for deep coal-gas reservoirs.

[0007] This invention is achieved through the following technical solution:

[0008] A method for evaluating the compressibility of coal-rock gas reservoirs includes the following steps:

[0009] S1. Collect comprehensive reservoir test data for the study area, including geological data, core test data, and logging parameters; based on the comprehensive reservoir test data, extract basic parameters such as porosity, permeability, density, and sonic velocity.

[0010] S2. Based on the comprehensive reservoir test data, establish a rock mechanical parameter prediction model for coal, rock and roof and floor. The prediction model establishes the relationship between rock density, acoustic velocity and clay content and rock mechanical parameters through multivariate nonlinear fitting.

[0011] S3. Calculate the compressive strength, elastic modulus, Poisson's ratio, tensile strength, cohesion, internal friction angle and fracture toughness of the formation based on the prediction model.

[0012] S4. Obtain the stratigraphic strain coefficient through differential strain experiments, and calculate the minimum and maximum horizontal principal stress of the coal reservoir and the top and bottom strata using a combined spring model in combination with elastic parameters and overlying strata pressure.

[0013] S5. Calculate the differences in rock mechanical parameters, horizontal stress, and interlayer horizontal stress of the coal reservoir and its top and bottom strata.

[0014] S6. Based on gas production data from coal-rock gas wells, Pearson correlation analysis and XGBoost ensemble learning algorithm are used to screen key parameters affecting gas production.

[0015] S7. Based on the selected key parameters, establish a coal and rock reservoir compressibility evaluation index and use a cluster analysis algorithm to achieve reservoir compressibility classification evaluation.

[0016] Furthermore, the steps for establishing the rock mechanical parameter prediction model include: using a multivariate nonlinear fitting method to establish a relationship model between rock density, sound wave velocity, and clay content and rock mechanical parameters, and constructing different prediction models for coal and rock layers and roof and floor layers respectively.

[0017] Furthermore, the differential strain experiment includes: processing the sampled rock core into a cubic sample, attaching strain gauges and consolidating it in a container, applying hydrostatic pressure in a triaxial testing machine and collecting multi-directional strain data; calculating the principal strain and the ratio of the three-dimensional effective principal stress based on the strain matrix, and then calculating the stratigraphic strain coefficient.

[0018] Furthermore, the step of screening key parameters affecting daily gas production includes: using the XGBoost ensemble learning algorithm to determine feature importance based on information gain and number of splits, and screening out the minimum horizontal stress difference between reservoir top layers, reservoir horizontal stress difference, reservoir tensile strength, and reservoir elastic modulus as key influencing parameters.

[0019] Furthermore, the clustering analysis algorithm uses Euclidean distance to measure the differences between samples and completes the reservoir compressibility classification by iteratively updating the cluster centers.

[0020] This invention also proposes a system for evaluating the compressibility of coal-rock gas reservoirs, comprising:

[0021] The data acquisition module is used to collect comprehensive reservoir test data and extract basic parameters such as porosity, permeability, density, and acoustic velocity.

[0022] The parameter prediction module is used to establish a rock mechanical parameter prediction model for coal and rock and roof and floor based on the comprehensive reservoir test data and to calculate the mechanical parameters.

[0023] The stress calculation module is used to calculate the horizontal principal stress of the formation and the stress difference between layers through differential strain experiments and combined spring models.

[0024] The feature selection module is used to select key parameters affecting gas production based on Pearson correlation analysis and the XGBoost ensemble learning algorithm.

[0025] The compressibility evaluation module is used to establish a reservoir compressibility evaluation index based on key parameters and to classify reservoir compressibility through cluster analysis algorithms.

[0026] Furthermore, the stress calculation module uses the formation elastic modulus, Poisson's ratio, overlying strata pressure, and tectonic strain coefficient to calculate the maximum and minimum horizontal principal stresses of the coal and rock and the top and bottom plates.

[0027] Furthermore, the compressibility evaluation module employs a clustering algorithm based on Euclidean distance to achieve reservoir classification evaluation based on the calculated reservoir compressibility evaluation index.

[0028] The beneficial effects of this invention are:

[0029] (1) This invention integrates geological data, core test data and logging parameters into comprehensive reservoir test data. It establishes a prediction model of rock mechanical parameters of coal and rock layers and top and bottom layers through multivariate nonlinear fitting, which overcomes the problem of prediction instability caused by scattered data sources and incompatible parameters in the existing technology, and realizes the integrated quantitative characterization of reservoir geology, logging and mechanical characteristics.

[0030] (2) This invention quantifies the stratigraphic strain coefficient by conducting differential strain tests under laboratory conditions, and calculates the minimum and maximum horizontal principal stresses of the reservoir and the top and bottom plates by combining the combined spring model, so that the acquisition of the stratigraphic stress is more consistent with the actual stress state of the deep coal and gas reservoir, and makes up for the shortcomings of traditional estimation based on empirical formulas or uniaxial stress.

[0031] (3) Based on the calculation of the difference in rock mechanical parameters between the coal and rock layers and the difference in interlayer stress, this invention reveals the influence law of the mechanical properties of the top and bottom plates on the compressibility of the reservoir. It establishes a new mechanism for compressibility evaluation from the perspective of stress transmission and rock mass coupling, breaking through the limitation of only considering the parameters of the reservoir itself in the past.

[0032] (4) By combining Pearson correlation analysis and XGBoost ensemble learning algorithm, this invention automatically selects key control factors such as minimum horizontal stress difference between reservoir tops, horizontal stress difference of reservoir, tensile strength of reservoir and elastic modulus of reservoir from multidimensional parameters, realizing intelligent identification of the main geomechanical parameters affecting gas production, and avoiding the uncertainty of manual experience in parameter selection.

[0033] (5) This invention establishes a reservoir compressibility evaluation index using the selected key parameters and uses a clustering analysis algorithm based on Euclidean distance to classify the reservoir, realizing the transformation from qualitative experience judgment to quantitative intelligent zoning, and providing a scientific basis for fracturing design.

[0034] (6) By comprehensively reflecting the differences in geomechanics between the coal and rock strata and the top and bottom plates, this invention can more accurately identify the compressible and incompressible areas of the reservoir, guide the optimization design of fracturing parameters, effectively improve the efficiency of fracture network formation and gas production, and reduce construction risks and transformation costs. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a schematic flowchart of a method for evaluating the compressibility of coal-rock gas reservoirs proposed in this invention.

[0037] Figure 2 This is a schematic diagram of the terminal equipment of a coal-rock gas reservoir compressibility evaluation system proposed in this invention;

[0038] Figure 3 This is a schematic diagram of a readable storage medium for a coal-rock gas reservoir compressibility evaluation system proposed in this invention;

[0039] In the diagram, 200 is the terminal device, 210 is the memory, 211 is the RAM, 212 is the cache, 213 is the ROM, 214 is the program / utility, 215 is the program module, 220 is the processor, 230 is the bus, 240 is the external device, 250 is the I / O interface, 260 is the network adapter, and 300 is the program product. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0041] Example 1

[0042] refer to Figure 1 This embodiment provides a method for evaluating the compressibility of coal-rock gas reservoirs. This method collects experimental core data, geological data, logging data, drilling and completion report data, fracturing operation data, and daily production data of single coal-rock gas wells in the study area. It then describes the obtained downhole experimental samples in the study area and obtains the porosity, permeability, and acoustic velocity of the original rock samples.

[0043] Specifically, firstly, standard rock samples are drilled, dried, and then core descriptions are performed. The core surface is observed using a magnifying glass, and the main descriptive content includes the core's apparent characteristics and lithological features. Secondly, conventional measurement methods are used to test the porosity, permeability, density, and acoustic velocity of the dried rock samples to obtain the original rock sample's porosity, permeability, density, and acoustic velocity. Porosity and permeability parameters are used to analyze the rock's pore characteristics, while density and acoustic velocity are used to construct subsequent rock mechanics models.

[0044] Based on the above experimental results, rock cores with intact apparent structures and similar test data were selected for rock mechanics testing. This testing mainly followed the relevant requirements of the "Standard for Engineering Rock Mass Testing Methods (GB / T 50266-2013)," obtaining the compressive strength of the rock through uniaxial triaxial compression testing. Static elastic modulus Static Poisson's ratio Cohesion internal friction angle Rock mechanical parameters were obtained, and the tensile strength of the rock was determined using the Brazilian splitting test. Three-point bending test to obtain the fracture toughness of rock .

[0045] Based on the test results of rock density, acoustic velocity and rock mechanical parameters, a prediction model for the rock mechanical parameters of the target deep coal-gas reservoir and its top and bottom strata is established.

[0046] Specifically, an Excel spreadsheet was used to analyze the correlation between parameters such as rock density, acoustic velocity, and clay content and rock mechanical parameters (compressive strength, elastic modulus, Poisson's ratio, cohesion, internal friction angle, tensile strength, and fracture toughness). A calculation model was established based on multivariate nonlinear fitting to establish the relationship between these parameters and the rock mechanical parameters. Because the rock mechanical properties and basic physical properties of coal differ significantly from those of conventional strata such as sandstone, limestone, and mudstone, different rock mechanical parameter prediction models need to be established for coal.

[0047] The prediction models for coal and rock mechanical parameters include:

[0048] Compressive strength: (1);

[0049] Tensile strength: (2);

[0050] Elastic modulus: (3);

[0051] Poisson's ratio: (4);

[0052] Fracture toughness: (5);

[0053] Cohesion: (6);

[0054] Angle of internal friction: (7);

[0055] The prediction models for rock mechanical parameters of coal and rock roof and floor include:

[0056] Compressive strength: (8);

[0057] Tensile strength: (9);

[0058] Elastic modulus: (10);

[0059] Poisson's ratio: (11);

[0060] Cohesion: (12);

[0061] Angle of internal friction: (13);

[0062] in The reservoir compressive strength is expressed in MPa. The tensile strength of the reservoir is expressed in MPa. Cohesion, measured in MPa; This refers to the reservoir's elastic modulus, expressed in GPa. The reservoir's Poisson's ratio; This is the internal friction angle of the reservoir, in degrees. Reservoir fracture toughness, in MPa.m 0.5 ; The longitudinal wave velocity is expressed in m / s. Density, unit: g / cm³ 3 ; Clay content, expressed as % The compressive strength of the top and bottom plates is expressed in MPa. The tensile strength of the top and bottom plates is expressed in MPa. The cohesion of the top and bottom plates is expressed in MPa. The elastic modulus of the top and bottom plates is expressed in GPa. Poisson's ratio for the top and bottom plates; The internal friction angle of the top and bottom plates is expressed in degrees (°).

[0063] Based on sonic transit time (AC), density logging (DEN), and gamma logging (GR) data from well logging, formation rock mechanical parameters are calculated. The actual parameters used in the prediction model are converted from the well logging data using the formulas described above. The prediction model is then input into an Excel spreadsheet. The AC, DEN, and GR data are imported and entered into the prediction model table using Excel formulas. AC, DEN, and GR are selected as input parameters to calculate rock mechanical parameters such as compressive strength, elastic modulus, Poisson's ratio, tensile strength, cohesion, internal friction angle, and fracture toughness.

[0064] (14);

[0065] (15);

[0066] (16).

[0067] The minimum and maximum horizontal principal stress tectonic strain coefficients of the strata were obtained based on indoor differential strain experiments.

[0068] Specifically, core samples were taken from the target strata in the study area. Based on regional geological data, drilling and completion report data, and experimental data, data such as the elastic modulus, Poisson's ratio, tensile strength, overlying formation pressure, pore pressure, and Biot coefficient of the sampled strata were obtained. The core samples were processed into approximate cubes with sides of 50-60 mm. Three sets of strain gauges were attached to the mutually perpendicular faces of the cube. The strain gauges were connected to a strain acquisition card. The rock sample was placed in a container slightly larger than its side length, and prepared silica gel was poured between the container and the rock sample. After the silica gel solidified, differential strain experiments were conducted. In the experiment, the solidified rock sample was placed in the vessel of a triaxial rock testing machine, and hydrostatic pressure was continuously applied at a rate less than 0.01 MPa / s until it exceeded 1.5 times the stress level of the stratum where the rock sample was located. Based on the collected strain in nine directions, according to the basic principles of elasticity, the normal strain and shear strain during the loading process of the sample can be calculated using the nine strains, as shown in the following formula:

[0069] (17);

[0070] in, to These are the nine strain values ​​measured in the experiment. It is a triaxial positive strain; It represents triaxial shear strain.

[0071] The strain matrix is ​​determined based on the calculated normal strain and shear strain, and the principal strains in the three directions are obtained using a ternary linear equation, as shown in the following formula:

[0072] (18);

[0073] in The tensor expression for the principal strain is given, where A, B, and C represent the three principal strains experienced by the rock sample.

[0074] Based on the stress-strain relationship in elasticity, the ratio of the three effective principal stresses is obtained from the three principal strains:

[0075] (19);

[0076] in These are the ratios of the three principal strains, and the magnitude of the ground stress can be calculated based on experiments.

[0077] The stratigraphic strain coefficient is calculated using the following formula:

[0078] (20);

[0079] in, The strain coefficients are constructed along the directions of maximum and minimum principal stresses, and the decimals are dimensionless.

[0080] This represents the maximum horizontal principal stress of the formation, expressed in MPa. This represents the minimum horizontal principal stress of the formation, expressed in MPa. The stress in the overlying strata is expressed in MPa. This represents the formation pore pressure, expressed in MPa. This refers to the reservoir's elastic modulus, expressed in GPa. The effective stress coefficient is dimensionless. Poisson's ratio, in MPa.

[0081] Based on rock mechanics parameters and tectonic strain coefficients, the minimum and maximum horizontal principal stresses of deep coal-gas reservoirs and their top and bottom strata are calculated using a combined spring model based on overlying strata stress, formation pore pressure, effective stress coefficient, and tectonic strain coefficient. The formulas are as follows:

[0082] (twenty one);

[0083] in, The strain coefficients are constructed along the directions of maximum and minimum principal stresses, and the decimals are dimensionless.

[0084] This represents the maximum horizontal principal stress of the formation, expressed in MPa. This represents the minimum horizontal principal stress of the formation, expressed in MPa. The stress in the overlying strata is expressed in MPa. This represents the formation pore pressure, expressed in MPa. This refers to the reservoir's elastic modulus, expressed in GPa. The effective stress coefficient is dimensionless. Poisson's ratio, in MPa.

[0085] Based on the established rock mechanics parameter prediction model, the rock mechanics parameters and their differences in deep coalbed methane reservoirs and their top and bottom strata were calculated. Simultaneously, the horizontal stress differences and inter-layer horizontal stress differences in deep coalbed methane reservoirs and their top and bottom strata were also calculated. The calculated rock mechanics differences are used as the basis for subsequent analysis of the impact on the daily gas production of single coalbed methane wells. The formula is as follows:

[0086] (twenty two);

[0087] (twenty three);

[0088] in, This represents the difference in rock mechanical parameters between the reservoir and the top and bottom plates, and the units are the same as the actual calculated parameters. These are reservoir rock mechanical parameters, and the units are the same as those used in actual calculations. These are the rock mechanical parameters of the reservoir's top and bottom plates, and are in the same units as the actual calculated parameters.

[0089] △σ is the difference in horizontal stress between the reservoir and the top and bottom plates, in MPa; The reservoir horizontal stress is expressed in MPa. The stress is the horizontal stress of the top and bottom plates, expressed in MPa.

[0090] The daily gas production of a single coal-rock gas well was collected from its daily production report, and Pearson correlation analysis was used.

[0091] Analyze reservoir parameters, including reservoir elastic modulus, reservoir internal friction angle, reservoir fracture toughness, reservoir tensile strength, reservoir compressive strength, reservoir Poisson's ratio, reservoir vertical stress, reservoir maximum horizontal principal stress, reservoir minimum horizontal principal stress, reservoir horizontal stress difference, and reservoir cohesion.

[0092] Top plate parameters include top plate elastic modulus, top plate internal friction angle, top plate tensile strength, top plate compressive strength, top plate Poisson's ratio, top plate vertical stress, top plate maximum horizontal principal stress, top plate minimum horizontal principal stress, top plate horizontal stress difference, top plate brittleness index, and top plate cohesion.

[0093] The base plate parameters include the base plate elastic modulus, the base plate internal friction angle, the base plate tensile strength, the base plate compressive strength, the base plate Poisson's ratio, the base plate vertical stress, the base plate maximum horizontal principal stress, the base plate minimum horizontal principal stress, the base plate horizontal stress difference, the base plate brittleness index, and the base plate cohesion.

[0094] Interlayer parameters include the minimum horizontal stress difference between reservoir top layers, the difference in compressive strength between reservoir top layers, the difference in tensile strength between reservoir top layers, the ratio of elastic modulus of reservoir top layers, the ratio of Poisson's ratio of reservoir top layers, the ratio of brittleness index of reservoir top layers, the minimum horizontal stress difference between reservoir bottom layers, the difference in compressive strength between reservoir bottom layers, the difference in tensile strength between reservoir bottom layers, the ratio of elastic modulus of reservoir bottom layers, the ratio of Poisson's ratio of reservoir bottom layers, and the ratio of brittleness index of reservoir bottom layers.

[0095] The influence of different parameters on the daily gas production of a single coal-rock gas well was investigated, clarifying the impact of various parameters on the daily gas production. Analysis revealed a negative correlation between daily gas production and reservoir mechanical parameters. Daily gas production showed a positive correlation with the Poisson's ratio, maximum horizontal stress, and minimum horizontal stress of the roof, but a negative correlation with other roof and floor mechanical parameters. Daily gas production also showed a positive correlation with the minimum horizontal stress difference between the top and bottom layers, but a negative correlation with other interlayer mechanical parameters. The correlation coefficients between daily gas production and reservoir elastic modulus, reservoir tensile strength, reservoir horizontal stress difference, and minimum horizontal stress difference between the top and bottom layers were all above 0.6.

[0096] There are 45 influencing parameters mentioned above. In this embodiment, the XGBoost ensemble learning model is used to establish a high-dimensional data key factor screening method to screen the key influencing parameters affecting daily gas production. The feature importance of the ensemble tree model is used to calculate the gain information gain and weight split number. Combined with preprocessing such as missing value imputation and category feature encoding, a specified number (4 by default) of controlling factors are screened out. The calculation results are shown in Table 1.

[0097] Table 1: Results of screening influencing factors based on the XGBoost ensemble learning model

[0098]

[0099] Based on the calculation results, the key influencing factors selected are the minimum horizontal stress difference between reservoir tops, the horizontal stress difference in the reservoir, the tensile strength of the reservoir, and the elastic modulus of the reservoir. Compared with common methods such as correlation analysis and univariate tests, the method in this embodiment uses CART trees as base learners and utilizes the hierarchical splitting of trees to fit nonlinear patterns. By combining multiple trees, the interaction between features is naturally encoded. Combined with regularization and gradient optimization, nonlinear correlations and feature interaction effects are simultaneously mined when selecting key factors in high-dimensional data. It can capture the nonlinear / interactive relationship between features and targets, is robust to noise and collinearity, and does not require manual threshold setting or manual integration processes.

[0100] Based on the screened key influencing parameters, a calculation equation for the quantitative evaluation index of the compressibility of deep coal-gas reservoirs is established, as follows:

[0101] (twenty four);

[0102] in It serves as an evaluation index for the fracturing capability of deep coal and rock gas reservoirs. This represents the minimum horizontal principal stress difference between reservoir layers, expressed in MPa. This refers to the reservoir's elastic modulus, expressed in GPa. The tensile strength of the reservoir is expressed in MPa. Horizontal stress difference in reservoir, measured in MPa.

[0103] This embodiment uses a clustering analysis algorithm to complete the reservoir compressibility classification by iteratively updating the cluster centers.

[0104] Specifically, the number of clusters and the distance metric, Euclidean distance, are determined using a formula, and the cluster centers are initialized.

[0105] (25);

[0106] Calculate the distance of each sample to the center, and assign the samples to the corresponding clusters according to the nearest principle; recalculate the mean of each cluster to update the center; repeat the assignment and update steps until the center change is less than the threshold or the maximum number of iterations is reached, and finally obtain the clustering result.

[0107] Where x and y represent two n-dimensional samples; k is the dimension index (1-n); and Let represent the feature values ​​of the two samples in the k-th dimension, respectively; The square root sign indicates the summation of the results across all dimensions; the square root sign is used to convert the sum of squares into a distance value with the same dimensions as the original data.

[0108] Based on the calculated fracturing evaluation index of coal-rock gas reservoirs and the daily gas production of a single coal-rock gas well, the reservoir classification results are calculated and shown in Table 2.

[0109] Table 2: Cluster Analysis Results

[0110]

[0111] The fracturing ability of deep coal-rock gas reservoirs was evaluated based on the calculation results. The fracturing ability of deep coal-rock gas reservoirs was divided into three categories: Category I reservoirs (FI>0.32) have good fracturing ability and are prone to forming complex fracture networks; Category II reservoirs (0.32≥FI>0.21) have moderate fracturing ability and are prone to forming relatively simple fracture networks; Category III reservoirs (FI<0.21) have poor fracturing ability and fracturing is not recommended.

[0112] Example 2

[0113] refer to Figure 2Based on Example 1, this example proposes a terminal device for a coal-rock gas reservoir compressibility evaluation system. The terminal device 200 includes at least one memory 210, at least one processor 220, and a bus 230 connecting different platform systems.

[0114] The memory 210 may include a readable medium in the form of volatile memory, such as RAM 211 and / or cache memory 212, and may further include ROM 213.

[0115] The memory 210 also stores a computer program that can be executed by the processor 220, causing the processor 220 to perform any of the above-described applications of the coal-rock gas reservoir compressibility evaluation system in this application. The specific implementation method and the achieved technical effects are consistent with those described in the above-described application embodiments, and some details will not be repeated here. The memory 210 may also include a program / utility 214 having a set (at least one) of program modules 215. Such program modules include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.

[0116] Accordingly, processor 220 can execute the aforementioned computer program, as well as executable program / utility 214.

[0117] Bus 230 can represent one or more of several types of bus structures, including a memory bus or memory controller, peripheral bus, graphics acceleration port, processor, or a local bus using any of the various bus structures.

[0118] Terminal device 200 can also communicate with one or more external devices 240, such as keyboards, pointing devices, Bluetooth devices, etc., and with one or more devices capable of interacting with it, and / or with any device that enables it to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication can be performed via I / O interface 250. Furthermore, terminal device 200 can communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 260. Network adapter 260 can communicate with other modules of terminal device 200 via bus 230. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with terminal device 200, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms.

[0119] Example 3

[0120] This embodiment proposes a readable storage medium for a coal-rock gas reservoir compressibility evaluation system. The computer-readable storage medium stores instructions that, when executed by a processor, implement any of the aforementioned coal-rock gas reservoir compressibility evaluation systems. The specific implementation method and the achieved technical effects are consistent with those described in the above-mentioned application embodiments, and some details will not be repeated.

[0121] Figure 3 The present embodiment illustrates a program product 300 for implementing the above-described applications. This product may employ a portable compact disc read-only memory (CD-ROM) and include program code, and may run on a terminal device, such as a personal computer. However, the program product 300 of the present invention is not limited thereto. In this embodiment, the readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device. The program product 300 may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0122] Computer-readable storage media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof. Program code for performing operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on a user computing device, partially on a user device, as a standalone software package, partially on a user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing devices can be connected to user computing devices via any type of network, including local area networks (LANs) or wide area networks (WANs), or they can be connected to external computing devices (e.g., via the Internet using an Internet service provider).

[0123] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the compressibility of coal-rock gas reservoirs, characterized in that, Includes the following steps: S1. Collect comprehensive reservoir test data for the study area, including geological data, core test data, and logging parameters; based on the comprehensive reservoir test data, extract basic parameters such as porosity, permeability, density, and sonic velocity. S2. Based on the comprehensive reservoir test data, establish a rock mechanical parameter prediction model for coal, rock and roof and floor. The prediction model establishes the relationship between rock density, acoustic velocity and clay content and rock mechanical parameters through multivariate nonlinear fitting. S3. Calculate the compressive strength, elastic modulus, Poisson's ratio, tensile strength, cohesion, internal friction angle and fracture toughness of the formation based on the prediction model. S4. Obtain the stratigraphic strain coefficient through differential strain experiments, and calculate the minimum and maximum horizontal principal stresses of the coal reservoir and the top and bottom strata using a combined spring model in combination with the predicted elastic parameters and the pressure of the overlying strata. S5. Calculate the differences in rock mechanical parameters, horizontal stress, and interlayer horizontal stress of the coal reservoir and its top and bottom strata. S6. Based on gas production data from coal-rock gas wells, Pearson correlation analysis and XGBoost ensemble learning algorithm are used to screen key parameters affecting gas production. S7. Based on the selected key parameters, establish a coal and rock reservoir compressibility evaluation index and use a cluster analysis algorithm to achieve reservoir compressibility classification evaluation.

2. The method for evaluating the compressibility of coal-rock gas reservoirs according to claim 1, characterized in that, The steps for establishing a rock mechanics parameter prediction model include: using a multivariate nonlinear fitting method to establish a relationship model between rock density, acoustic velocity, clay content and rock mechanics parameters, and constructing different prediction models for coal and rock layers and roof and floor layers respectively.

3. The method for evaluating the compressibility of coal-rock gas reservoirs according to claim 1, characterized in that, The differential strain experiment includes: processing the sampled rock core into a cubic sample, attaching strain gauges and consolidating it in a container, applying hydrostatic pressure in a triaxial testing machine and collecting multi-directional strain data; calculating the principal strain and the ratio of the three-dimensional effective principal stress based on the strain matrix, and then calculating the stratigraphic strain coefficient.

4. The method for evaluating the compressibility of coal-rock gas reservoirs according to claim 1, characterized in that, The steps for screening key parameters affecting daily gas production include: using the XGBoost ensemble learning algorithm to determine feature importance based on information gain and number of splits, and selecting the minimum horizontal stress difference between reservoir top layers, reservoir horizontal stress difference, reservoir tensile strength, and reservoir elastic modulus as key influencing parameters.

5. The method for evaluating the compressibility of coal-rock gas reservoirs according to claim 1, characterized in that, The clustering analysis algorithm uses Euclidean distance to measure the differences between samples and completes the reservoir compressibility classification by iteratively updating the cluster centers.

6. A coal-rock gas reservoir compressibility evaluation system, applicable to the method described in any one of claims 1-4, characterized in that, include: The data acquisition module is used to collect comprehensive reservoir test data and extract basic parameters such as porosity, permeability, density, and acoustic velocity. The parameter prediction module is used to establish a rock mechanical parameter prediction model for coal and rock and roof and floor based on the comprehensive reservoir test data and to calculate the mechanical parameters. The stress calculation module is used to calculate the horizontal principal stress of the formation and the stress difference between layers through differential strain experiments and combined spring models. The feature selection module is used to select key parameters affecting gas production based on Pearson correlation analysis and the XGBoost ensemble learning algorithm. The compressibility evaluation module is used to establish a reservoir compressibility evaluation index based on key parameters and to classify reservoir compressibility through cluster analysis algorithms.

7. The coal-rock gas reservoir compressibility evaluation system according to claim 6, characterized in that, The stress calculation module uses the formation elastic modulus, Poisson's ratio, overlying strata pressure, and tectonic strain coefficient to calculate the maximum and minimum horizontal principal stresses of coal, rock, and roof and floor.

8. The coal-rock gas reservoir compressibility evaluation system according to claim 6, characterized in that, The compressibility evaluation module uses a clustering algorithm based on Euclidean distance to achieve reservoir classification evaluation based on the calculated reservoir compressibility evaluation index.

Images