Prediction method and device for coal seam gas residual volume and gas content monitoring system

Abstract

The invention provides a prediction method and device for coal seam gas residual amount and a gas content monitoring system. The method comprises the steps that a gas-solid coupling model of working face gas extraction is constructed; performing sensitivity analysis on preset influence factors in the influence factor group according to a gas-solid coupling model to obtain importance degrees corresponding to the preset influence factors; determining the preset influence factors of which the corresponding importance degrees are greater than a first threshold value as target factors; extracting each target factor and gas content from historical data of gas extraction to obtain a training data set; training a gas quantity prediction model according to the training data set to obtain a target model, predicting gas content distribution of the working face through the target model and drawing the gas content distribution into an image to obtain a gas distribution prediction map. The method solves the problems that in the prior art, input and output of a gas content prediction model depend on manual setting, the prediction model is sensitive to input parameters, and the prediction accuracy is easily reduced due to improper setting of the input parameters.

Description

Technical Field

[0001] This invention relates to the field of gas monitoring technology, and more specifically, to a method, apparatus, computer-readable storage medium, and gas content monitoring system for predicting residual gas in coal seams. Background Technology

[0002] Gas drainage is the fundamental measure to solve coal mine gas disasters. For coal seams after pre-drainage, the residual gas after drainage exhibits uneven distribution in time and space at the working face due to various factors. Traditional methods often rely on on-site measurement of residual gas content, which not only provides static residual gas distribution at the current working face but also fails to capture dynamic residual gas distribution information. Furthermore, these methods face problems such as high engineering costs, long processing times, and reliance on experience to select the appropriate time point for residual gas content measurement to determine if drainage meets standards.

[0003] To address the aforementioned shortcomings, existing technologies propose using a gas content prediction model to predict the gas content of coal seams on a unit basis. However, the parameter settings and selection of the prediction model in existing technologies rely on manual settings, which often leads to inaccurate prediction results due to improper parameter settings. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, computer-readable storage medium, and gas content monitoring system for predicting residual gas in coal seams, so as to at least solve the problem that the input and output of the gas content prediction model in the prior art depend on manual settings, and the prediction model is sensitive to input parameters, which can easily lead to a decrease in prediction accuracy due to improper setting of input parameters.

[0005] To achieve the above objectives, according to one aspect of this application, a method for predicting residual gas in coal seams is provided, comprising: constructing a gas-solid coupling model for gas drainage at the working face, wherein the gas-solid coupling model is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas; performing sensitivity analysis on preset influencing factors in the influencing factor group based on the gas-solid coupling model to obtain the importance of each preset influencing factor; identifying preset influencing factors whose corresponding importance is greater than a first threshold as target factors; extracting each target factor and gas content from historical data of gas drainage to obtain a training dataset; training a gas quantity prediction model based on the training dataset to obtain a target model; predicting the gas content distribution at the working face using the target model and plotting it as an image to obtain a gas distribution prediction map.

[0006] Optionally, a gas-solid coupling model for gas drainage at the working face is constructed, including: constructing a continuity equation, which is used to simulate coal displacement and gas flow caused by changes in gas pressure. The continuity equation is as follows: Where G is the shear modulus of the coal, u i,jj and u j,ji Let v be the displacement component, v be Poisson's ratio, and β be the displacement component. f p is the effective stress coefficient corresponding to the fractures in the coal body. f For the fissure gas pressure, β m p is the effective stress coefficient corresponding to the pores of the coal body. m F is the matrix gas pressure of the coal seam. i Let be the volume force of the coal body in the i-th direction; construct a porosity variation equation, which is used to simulate the porosity change of the coal body caused by coal displacement and gas flow. The porosity variation equation is: in, Porosity refers to the percentage of material in a matrix, which includes fracture porosity and matrix porosity. Where p is the initial porosity, M is the constrained axial modulus, and p is the initial porosity. m0 p is the initial matrix gas pressure. f0 ε is the initial fracture gas pressure. L P represents the limiting adsorption expansion deformation of the coal body. L Let K be the Langmuir pressure constant and K be the bulk modulus. A permeability variation equation is constructed to simulate the permeability changes in the coal body caused by coal displacement and gas flow. The permeability variation equation is: Where k is the permeability and k1 is the initial permeability; a matrix gas pressure variation equation is constructed to simulate the change of matrix gas pressure with extraction time. The matrix gas pressure variation equation is as follows: Among them, V M V is the molar volume of gas under standard conditions. L ρ is the Langmuir volume constant. c Let be the pseudo density of the coal, R be the gas constant, T be the temperature, τ be the adsorption time, and t be the extraction time. Given the matrix porosity; a fracture gas flow equation is constructed, which is used to simulate the gas flow in fractures as a function of porosity and permeability. The fracture gas flow equation is as follows: in, The porosity is the fracture porosity. Let μ be the gradient and μ be the dynamic viscosity of the gas. By combining the continuity equation, the porosity variation equation, the permeability variation equation, the matrix gas pressure variation equation, and the fracture gas flow rate equation, a gas-solid coupling model is obtained.

[0007] Optionally, sensitivity analysis is performed on the preset influencing factors in the influencing factor group based on the gas-solid coupling model to obtain the importance of each preset influencing factor. This includes: setting any one preset influencing factor as a variable and setting other preset influencing factors in the influencing factor group as constants; determining multiple variable values ​​for the variable, and using the gas-solid coupling model to simulate each variable value and calculate the gas content of the working face based on the simulation results; calculating the ratio of the difference between the values ​​of each variable to obtain a first ratio, calculating the ratio of the difference in gas content corresponding to each variable value to obtain a second ratio; and determining the ratio of the second ratio to the first ratio as the importance.

[0008] Optionally, a gas-solid coupling model is used to perform simulations based on the values ​​of each variable, including: when the variable is any one of extraction time, permeability, and porosity, the variable is substituted into the gas-solid coupling model for simulation; when the variable is coal seam thickness, the volume forces of the coal body in each direction are calculated based on the variable and substituted into the gas-solid coupling model for simulation; when the variables are overburden load, borehole diameter, borehole negative pressure, borehole length, and borehole spacing, the variables are used as boundary conditions of the gas-solid coupling model for simulation; when the variable is the original gas pressure, the variable is decomposed into initial matrix gas pressure and initial fracture gas pressure, and the initial matrix gas pressure and initial fracture gas pressure are substituted into the gas-solid coupling model for simulation.

[0009] Optionally, the gas content of the working face is calculated based on the simulation results, including: calculating the amount of adsorbed gas based on porosity, matrix gas pressure and fracture gas pressure to obtain the first gas content; calculating the amount of free gas based on fracture gas pressure to obtain the second gas content; and calculating the sum of the first gas content and the second gas content to obtain the gas content.

[0010] Optionally, the amount of adsorbed gas is calculated based on porosity, matrix gas pressure, and fracture gas pressure to obtain the first gas content, including: determining the absolute gas pressure of the coal body based on matrix gas pressure and fracture gas pressure; and calculating the first gas content based on porosity and absolute gas pressure, wherein the calculation formula is as follows: Where W is the first gas content, a is the maximum adsorption capacity, b is the adsorption strength, and A is the adsorption capacity. ad M represents the moisture content of the coal. ad γ represents the ash content of the coal, P represents the absolute gas pressure, and γ represents the apparent density of the coal.

[0011] Optionally, a gas quantity prediction model is trained based on the training dataset to obtain a target model, including: constructing a gas quantity prediction model using PSO-XGBoost to obtain a candidate model; determining multiple hyperparameter values ​​based on the hyperparameters of the candidate model, and assigning each hyperparameter value to a particle in the particle swarm optimization algorithm, with a one-to-one correspondence between the hyperparameter values ​​and particles; optimizing the target particles using the PSO algorithm based on the particles to obtain target particles; configuring candidate models based on the hyperparameter values ​​corresponding to the target particles; and training the configured candidate models based on the training dataset to obtain the target model.

[0012] According to another aspect of this application, a device for predicting residual coal seam gas is provided. The device includes: a construction unit for constructing a gas-solid coupling model for gas extraction at the working face, wherein the gas-solid coupling model is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas; a first calculation unit for performing sensitivity analysis on preset influencing factors in the influencing factor group according to the gas-solid coupling model to obtain the importance of each preset influencing factor; a determination unit for determining preset influencing factors whose corresponding importance is greater than a first threshold as target factors; an acquisition unit for extracting each target factor and gas content from historical data of gas extraction to obtain a training dataset; and a prediction unit for training a gas quantity prediction model according to the training dataset to obtain a target model, predicting the gas content distribution of the working face through the target model and plotting it as an image to obtain a gas distribution prediction map.

[0013] According to another aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform any of the methods described.

[0014] According to another aspect of this application, a gas content monitoring system is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include methods for performing any one of the methods described.

[0015] Applying the technical solution of this application, in the above-mentioned method for predicting residual coal seam gas, firstly, a gas-solid coupling model for gas extraction at the working face is constructed. The gas-solid coupling model is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas. Then, based on the gas-solid coupling model, a sensitivity analysis is performed on the preset influencing factors in the influencing factor group to obtain the importance of each preset influencing factor. After that, the preset influencing factors with an importance greater than a first threshold are determined as target factors. Then, each target factor and gas content are extracted from the historical data of gas extraction to obtain a training dataset. Finally, a gas quantity prediction model is trained based on the training dataset to obtain a target model. The gas content distribution at the working face is predicted by the target model and plotted as an image to obtain a gas distribution prediction map. This application constructs a gas-solid coupling model to simulate the environmental state changes of the working face during the extraction process. Then, based on the gas-solid coupling model, the influence of different influencing factors on the gas content change is determined by the control variable method to identify the more important target factors. Then, a prediction model is constructed using the target factors as input parameters. Compared with manually setting the input parameters of the model, the accuracy of the prediction model is improved. This solves the problem that the input and output of the gas content prediction model in the prior art depend on manual setting, and the prediction model is very sensitive to the input parameters. Improper setting of the input parameters can easily lead to a decrease in prediction accuracy. Attached Figure Description

[0016] Figure 1 A hardware structure block diagram of a mobile terminal for a method of predicting coal seam gas residue provided in an embodiment of this application is shown.

[0017] Figure 2 A flowchart illustrating a method for predicting residual coalbed methane according to an embodiment of this application is shown.

[0018] Figure 3 A structural block diagram of a coal seam gas residual prediction device provided according to an embodiment of this application is shown.

[0019] The above figures include the following reference numerals:

[0020] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device. Detailed Implementation

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0023] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0024] As described in the background section, the parameter settings and selection of existing prediction models rely on manual settings. This often leads to inaccurate prediction results due to improper parameter settings. To address the issue that the input and output of existing gas content prediction models rely on manual settings, and that prediction models are sensitive to input parameters and prone to decreased prediction accuracy due to improper input parameter settings, embodiments of this application provide a method that at least...

[0025] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0026] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a method of predicting residual gas in coal seams according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0027] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the coal seam gas residual prediction method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the aforementioned networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0028] This embodiment provides a method for predicting residual coal seam gas that runs on a mobile terminal, computer terminal, or similar computing device. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0029] Figure 2 This is a flowchart of a method for predicting residual coal seam gas according to an embodiment of this application. Figure 2 As shown, the method includes the following steps:

[0030] Step S201: Construct a gas-solid coupling model for gas extraction at the working face. The gas-solid coupling model is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas.

[0031] Specifically, firstly, a continuity equation, a porosity variation equation, a permeability variation equation, a matrix gas pressure variation equation, and a fracture gas flow equation are constructed to cover the evolution of coal deformation, pore structure changes, permeability changes, and gas pressure and flow state within the coal body.

[0032] Step S202: Perform sensitivity analysis on the preset influencing factors in the influencing factor group based on the gas-solid coupling model to obtain the importance of each preset influencing factor.

[0033] Step S203: Determine the preset influencing factors whose importance is greater than the first threshold as target factors;

[0034] Specifically, sensitivity analysis is performed on preset influencing factors. By changing these factors one by one (such as extraction time, permeability, etc.) while keeping other factors constant, the changes in model output (such as gas content) are observed. This helps identify which factors are most critical for predicting gas content. By calculating the proportion of gas content change caused by variable changes, the importance of each factor can be quantified, and a threshold (first threshold) can be set to screen factors with significant impact as target factors.

[0035] Step S204: Extract the target factors and gas content from the historical data of gas extraction to obtain the training dataset;

[0036] Specifically, data related to the identified target factors are extracted from historical gas extraction data, along with actual observed gas content values, to form a training dataset. This dataset includes records of coal seam gas content under different mining conditions (such as different extraction times and borehole parameters), providing rich training material for machine learning algorithms.

[0037] Step S205: Train the gas quantity prediction model based on the training dataset to obtain the target model. Predict the gas content distribution of the working face using the target model and plot it as an image to obtain the gas distribution prediction map.

[0038] Specifically, the PSO-XGBoost algorithm is used to construct a gas quantity prediction model. By training the model with a training dataset, a high-precision target model capable of predicting the gas content distribution at the working face can be obtained. The trained target model is then used to predict the gas content distribution at the working face, and the prediction results are visualized as an image, i.e., a gas distribution prediction map. This visually illustrates the distribution of gas in the coal seam, aiding in the formulation of mining plans and risk assessment.

[0039] The PSO (Particle Swarm Optimization) algorithm is used to optimize the hyperparameters of the XGBoost (Extreme Gradient Boosting) model to find the optimal model configuration. XGBoost is an efficient machine learning algorithm, particularly suitable for handling large amounts of data and high-dimensional feature problems. It constructs a series of decision trees and sums the predictions of the trees in a weighted manner to form the final prediction model.

[0040] In this embodiment, firstly, a gas-solid coupling model for gas drainage at the working face is constructed. This model is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas. Then, based on the gas-solid coupling model, a sensitivity analysis is performed on the preset influencing factors in the influencing factor group to obtain the importance of each preset influencing factor. Next, preset influencing factors with an importance greater than a first threshold are identified as target factors. Then, each target factor and gas content are extracted from historical gas drainage data to obtain a training dataset. Finally, a gas quantity prediction model is trained based on the training dataset to obtain a target model. The gas content distribution at the working face is predicted using the target model and plotted as an image to obtain a gas distribution prediction map. This application constructs a gas-solid coupling model to simulate the environmental state changes of the working face during the extraction process. Then, based on the gas-solid coupling model, the influence of different influencing factors on the gas content change is determined by the control variable method to identify the more important target factors. Then, a prediction model is constructed using the target factors as input parameters. Compared with manually setting the input parameters of the model, the accuracy of the prediction model is improved. This solves the problem that the input and output of the gas content prediction model in the prior art depend on manual setting, and the prediction model is very sensitive to the input parameters. Improper setting of the input parameters can easily lead to a decrease in prediction accuracy.

[0041] To construct the above-mentioned gas-solid coupling model, in one optional implementation, step S201 includes:

[0042] Step S2011: Construct a continuity equation. This equation is used to simulate coal displacement and gas flow caused by changes in gas pressure. The continuity equation is as follows: Where G is the shear modulus of the coal, u i,jj and u j,ji Let v be the displacement component, v be Poisson's ratio, and β be the displacement component. f p is the effective stress coefficient corresponding to the fractures in the coal body. f For the fissure gas pressure, β m p is the effective stress coefficient corresponding to the pores of the coal body. m F is the matrix gas pressure of the coal seam. i Let be the volume force of the coal body in the i-th direction;

[0043] Specifically, the continuity equation, based on fundamental principles of solid-state physics and fluid mechanics, describes the interaction between coal displacement and gas flow caused by changes in gas pressure within the coal seam. During coal seam gas extraction, the decrease in gas pressure inside the coal seam causes coal shrinkage and changes in pore structure, thus affecting gas flowability and the mechanical properties of the coal. This equation considers the elastic behavior of the coal, using shear modulus (G) and Poisson's ratio (v) to describe its mechanical properties, and the effective stress coefficient (β)... f and β m This reflects the effective stress influence of gas pressure on coal seam fractures and pores. The F in the equation... i It is the external volume force of the coal body in the i-direction.

[0044] Step S2012: Construct the porosity variation equation. This equation is used to simulate the porosity changes in the coal body caused by coal displacement and gas flow. The porosity variation equation is as follows: in, Porosity refers to the percentage of material in a matrix, which includes fracture porosity and matrix porosity. Where p is the initial porosity, M is the constrained axial modulus, and p is the initial porosity. m0 p is the initial matrix gas pressure. f0 ε is the initial fracture gas pressure. L P represents the limiting adsorption expansion deformation of the coal body. L Let K be the Langmuir pressure constant and K be the bulk modulus.

[0045] Specifically, the porosity variation equation, based on the physicochemical properties of coal and adsorption theory, describes the influence of coal displacement and gas flow on coal porosity. The porosity variation of the coal body directly relates to the gas storage space and flow path. The equation includes the initial state of the coal body's pore structure. Its response to pressure changes (M, β) m β f ), and Langmuir adsorption model parameters (ε) L P L This allows for a comprehensive assessment of the dynamic evolution of porosity as a function of gas pressure.

[0046] Step S2013: Construct a permeability variation equation. This equation is used to simulate the permeability changes in the coal body caused by coal displacement and gas flow. The permeability variation equation is as follows: Where k is the permeability and k1 is the initial permeability;

[0047] Specifically, the permeability variation equation is based on the influence of changes in the microstructure of the coal seam on its macroscopic permeability. Permeability (k) is a key parameter for measuring the ease of gas flow in coal seams. The equation predicts changes in permeability through changes in porosity, reflecting the influence of the coal seam's pore structure on the resistance to gas flow. The direct proportionality between permeability and porosity is further amplified to a cubic relationship, indicating that small changes in porosity can lead to significant changes in permeability.

[0048] Step S2014: Construct the matrix gas pressure variation equation. This equation is used to simulate the change in matrix gas pressure with extraction time. The matrix gas pressure variation equation is as follows: Among them, V M V is the molar volume of gas under standard conditions. L ρ is the Langmuir volume constant. c Let be the pseudo density of the coal, R be the gas constant, T be the temperature, τ be the adsorption time, and t be the extraction time. The matrix porosity;

[0049] Specifically, the matrix gas pressure variation equation, based on the Langmuir adsorption isotherm theory, describes the matrix gas pressure (p... m The variation of gas molar volume (V) with extraction time (t) under standard conditions is considered. M ), Langmuir volume constant (V L ), coal body false density (ρ) c Parameters such as gas constant (R), temperature (T), and adsorption time (τ) collectively affect the adsorption and desorption rates of gas, as well as the accumulation and release process of gas in coal.

[0050] Step S2015: Construct the fracture gas flow equation. This equation is used to simulate the gas flow in fractures as porosity and permeability change. The fracture gas flow equation is as follows: in, The porosity is the fracture porosity. Let μ be the gradient, and μ be the dynamic viscosity of the gas.

[0051] Specifically, the fracture gas flow equation is based on Darcy's law and describes the gas flow velocity in fractures as a function of porosity. The law governing the variation of permeability (k) is as follows. Dynamic viscosity (μ) reflects the flow characteristics of gas, while the time derivative and gradient terms in the equations reflect the transient characteristics and spatial distribution effects of gas flow in fractures. The pressure gradient of gas in the fracture is also considered. Driven by gas flow, and simultaneously by the pressure difference (p) between the fractures and the coal matrix. m -p fThis facilitated the transfer of gas.

[0052] In step S2016, the continuity equation, porosity change equation, permeability change equation, matrix gas pressure change equation, and fracture gas flow equation are combined to obtain the gas-solid coupling model.

[0053] Specifically, the gas-solid coupling model combines the above five equations to form a comprehensive gas-solid coupling model that considers factors such as coal mechanical behavior, porosity, permeability, gas pressure, temperature, and time. This allows for a comprehensive simulation of the dynamic response of the coal seam during gas extraction, including changes in coal deformation, porosity, and permeability, as well as the flow and pressure distribution of gas within the coal seam.

[0054] Through the above embodiments, the construction of a gas-solid coupling model can more accurately predict the coal seam gas extraction process and reduce errors caused by model simplification, including but not limited to using Langmuir adsorption isotherm theory to describe the adsorption behavior of gas in coal, which greatly improves the accuracy and reliability of prediction.

[0055] In order to determine the importance of each preset influencing factor, in one optional implementation, step S202 above includes:

[0056] Step S2021: Set any one preset influencing factor as a variable, and set the other preset influencing factors in the influencing factor group as constants;

[0057] Specifically, firstly, one of the preset influencing factors is set as a variable, while all other factors are set as constants. That is, in each simulation, only one factor is changed, while all other factors remain unchanged. For example, in the first simulation, the extraction time can be changed while keeping other factors such as permeability and porosity constant.

[0058] Step S2022: Determine the values ​​of multiple variables, and use a gas-solid coupling model to perform simulations based on the values ​​of each variable, and calculate the gas content of the working face based on the simulation results;

[0059] Specifically, for the defined variables, multiple different values ​​are selected, and a gas-solid coupling model is used for simulation. Each variable value corresponds to one simulation, and the gas content at the working face is calculated based on the simulation results.

[0060] Step S2023: Calculate the ratio of the difference between the values ​​of each variable to obtain the first ratio; calculate the ratio of the difference in gas content corresponding to the values ​​of each variable to obtain the second ratio.

[0061] Specifically, for each variable's value, the ratio of its difference to a benchmark value (e.g., initial value or average value) is calculated to obtain a first ratio. Simultaneously, the ratio of the difference between the gas content corresponding to each variable's value and the benchmark gas content is calculated to obtain a second ratio.

[0062] Step S2024: Determine the importance level by the ratio of the second ratio to the first ratio.

[0063] Specifically, the ratio of the second proportion to the first proportion is defined as the importance of the pre-defined influencing factor. The ratio reflects the sensitivity of changes in gas content to changes in the variable. A large ratio indicates that the gas content is very sensitive to changes in this variable, and that the variable has a significant impact on the gas content.

[0064] Through the above embodiments, sensitivity analysis can help identify the factors that have the greatest impact on gas content during gas drainage, thereby guiding parameter optimization. By identifying important influencing factors, the risks during gas drainage, such as the risk of gas explosion, can be assessed more accurately.

[0065] In order to simulate different preset influencing factors, in an optional implementation, step S2022 above includes:

[0066] Step S20221: When the variable is any one of extraction time, permeability and porosity, substitute the variable into the gas-solid coupling model for simulation;

[0067] Specifically, since extraction time, permeability, and porosity directly affect gas flow and the physical properties of coal, these factors can be directly substituted into the model for simulation.

[0068] Step S20222: With the coal seam thickness as the variable, calculate the volume forces of the coal body in each direction based on the variable and substitute them into the gas-solid coupling model for simulation.

[0069] Specifically, the thickness of the coal seam affects the pressure-bearing capacity of the coal body, thereby affecting the flow of gas. Therefore, by changing the volume forces of the coal body in various directions, the simulation results of the model can be indirectly affected.

[0070] Step S20223: Simulation is performed with the variables being overburden load, borehole diameter, borehole negative pressure, borehole length, and borehole spacing, using the boundary conditions of the gas-solid coupling model as the variables.

[0071] Specifically, overburden load, borehole diameter, borehole negative pressure, borehole length, and borehole spacing reflect the impact of the mining environment and borehole design on gas extraction efficiency, serving as boundary conditions.

[0072] Step S20224: With the original gas pressure as the variable, the variable is decomposed into the initial matrix gas pressure and the initial fracture gas pressure, and the initial matrix gas pressure and the initial fracture gas pressure are substituted into the gas-solid coupling model for simulation.

[0073] Specifically, the original gas pressure has different effects on the stress and flow characteristics of the coal body due to the different forms in which gas exists in the coal body, and needs to be decomposed into two parts: matrix and fractures.

[0074] Through the above embodiments, the simulation strategy is refined to more realistically reflect the various complex phenomena in the gas extraction process, thereby improving the accuracy and reliability of subsequent predictions.

[0075] In an optional implementation, to calculate the gas content at the working face, step S2022 further includes:

[0076] Step S20225: Calculate the amount of adsorbed gas based on porosity, matrix gas pressure and fracture gas pressure to obtain the first gas content;

[0077] Step S20226: Calculate the amount of free-state gas based on the fracture gas pressure to obtain the second gas content;

[0078] Step S20227: Calculate the sum of the first gas content and the second gas content to obtain the gas content.

[0079] Specifically, the existence of gas in coal seams can be mainly divided into two forms: adsorbed state and free state. Adsorbed gas refers to gas molecules that are attracted by the molecular forces on the surface of the coal seam and are tightly adsorbed onto the pore surface of coal particles; free gas refers to gas that exists in a free state in the pores of the coal seam.

[0080] The above embodiments not only consider the distribution of gas in the coal seam but also integrate the physical properties of the coal body (such as porosity, matrix gas pressure, and fracture gas pressure), making the calculation results closer to reality. For example, increased porosity increases the flow space for free gas, thereby increasing the amount of free gas; while a decrease in matrix gas pressure leads to the release of adsorbed gas, which is converted into free gas, ultimately affecting the total gas content. The calculation method based on model prediction can help mines more accurately assess the effectiveness of gas drainage, adjust drainage strategies in a timely manner, effectively prevent gas accidents, and ensure safe production in the mine.

[0081] In order to calculate the aforementioned first gas content, in an optional embodiment, step S20225 includes:

[0082] The absolute gas pressure of the coal body is determined based on the matrix gas pressure and the fracture gas pressure.

[0083] The first gas content is calculated based on porosity and absolute gas pressure, where the calculation formula is as follows: Where W is the first gas content, a is the maximum adsorption capacity, b is the adsorption strength, and A is the adsorption capacity. ad M represents the moisture content of the coal. ad γ represents the ash content of the coal, P represents the absolute gas pressure, and γ represents the apparent density of the coal.

[0084] Specifically, the calculation of adsorbed gas quantity is based on the Langmuir adsorption isotherm theory, which describes the adsorption behavior of gas on the coal surface. By combining the matrix gas pressure and fracture gas pressure to calculate the absolute gas pressure, and then using factors such as porosity, moisture, and ash to correct the adsorption calculation, the amount of adsorbed gas can be predicted more accurately. For example, the presence of moisture and ash occupies the pore space of the coal body, reducing the area available for gas adsorption, thereby reducing the amount of adsorbed gas. This calculation method not only considers the physicochemical properties of the coal body but also integrates the influence of gas pressure, making the prediction results more reliable and helping mines to formulate reasonable gas drainage plans and reduce the risk of gas accumulation. Through calculation, we found that changes in coal body components, including but not limited to moisture and ash, and key parameters such as absolute gas pressure, significantly affect the amount of adsorbed gas. Therefore, precise management of these parameters plays an important role in improving gas drainage efficiency and safety.

[0085] To ensure the accuracy of the prediction model, in one optional implementation, step S205 includes:

[0086] Step S2051: Use PSO-XGBoost to construct a gas quantity prediction model to obtain alternative models;

[0087] Specifically, historical gas extraction data are collected, including extraction time, permeability, porosity, coal seam thickness, overlying load, borehole parameters, and corresponding gas content. A gas quantity prediction model is constructed using PSO-XGBoost to obtain alternative models.

[0088] Step S2052: Determine multiple hyperparameter values ​​based on the hyperparameters of the candidate models, and assign each hyperparameter value to a particle in the particle swarm algorithm, with each hyperparameter value corresponding to a particle.

[0089] Specifically, the hyperparameters of the XGBoost model, such as tree depth, learning rate, and regularization parameters, are determined, and different values ​​of these hyperparameters are used as "particles" in PSO. Each hyperparameter value represents a particle, and the initial position and velocity of the particles are randomly initialized.

[0090] Step S2053: Based on the particle, the target particle is obtained by optimizing the PSO algorithm;

[0091] Specifically, the model performance on the training dataset (such as the root mean square error of cross-validation) is used as the fitness metric for the particles. Based on the particle's current position, its position and velocity are updated through social learning (learning from the swarm's best particle) and individual learning (learning from its own historical best position) to find the optimal combination of hyperparameters. The PSO algorithm iterates repeatedly until a preset number of iterations is reached or the swarm's best particle is found. The hyperparameter values ​​corresponding to this best particle constitute the configuration of the target model, i.e., the target particle mentioned above.

[0092] Step S2054: Configure alternative models according to the hyperparameter values ​​corresponding to the target particle;

[0093] Specifically, the optimal hyperparameters found by PSO are used to configure the XGBoost model.

[0094] Step S2055: Train the configured alternative models based on the training dataset to obtain the target model.

[0095] Specifically, the configured XGBoost model is trained using the complete training dataset to obtain the target model.

[0096] Through the above embodiments, the PSO algorithm finds the optimal solution, i.e., the optimal combination of model hyperparameters, by simulating the foraging behavior of bird flocks; the XGBoost algorithm, on the other hand, improves prediction accuracy by constructing multiple decision trees to progressively optimize the model. Setting up combined algorithms not only enables rapid convergence and the finding of optimal model parameters, but also effectively handles high-dimensional and nonlinear data, improving the adaptability and accuracy of the prediction model. For example, the optimal hyperparameter values ​​obtained through the PSO algorithm enable the XGBoost model to better fit historical gas extraction data, improving prediction accuracy.

[0097] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0098] This application also provides a device for predicting residual coal seam gas. It should be noted that this device can be used to execute the method for predicting residual coal seam gas provided in this application. This device is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0099] The following describes the device for predicting residual coalbed methane provided in the embodiments of this application.

[0100] Figure 3 This is a structural block diagram of a coal seam gas residual prediction device according to an embodiment of this application. Figure 3 As shown, the device includes:

[0101] Building unit 10 is used to build a gas-solid coupling model for gas extraction in the working face. The gas-solid coupling model is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas.

[0102] Specifically, firstly, a continuity equation, a porosity variation equation, a permeability variation equation, a matrix gas pressure variation equation, and a fracture gas flow equation are constructed to cover the evolution of coal deformation, pore structure changes, permeability changes, and gas pressure and flow state within the coal body.

[0103] The first calculation unit 20 is used to perform sensitivity analysis on the preset influencing factors in the influencing factor group according to the gas-solid coupling model, and to obtain the importance of each preset influencing factor.

[0104] The determination unit 30 is used to determine the preset influencing factors whose importance is greater than the first threshold as target factors;

[0105] Specifically, sensitivity analysis is performed on preset influencing factors. By changing these factors one by one (such as extraction time, permeability, etc.) while keeping other factors constant, the changes in model output (such as gas content) are observed. This helps identify which factors are most critical for predicting gas content. By calculating the proportion of gas content change caused by variable changes, the importance of each factor can be quantified, and a threshold (first threshold) can be set to screen factors with significant impact as target factors.

[0106] The acquisition unit 40 is used to extract various target factors and gas content from historical data of gas extraction to obtain a training dataset.

[0107] Specifically, data related to the identified target factors are extracted from historical gas extraction data, along with actual observed gas content values, to form a training dataset. This dataset includes records of coal seam gas content under different mining conditions (such as different extraction times and borehole parameters), providing rich training material for machine learning algorithms.

[0108] Prediction unit 50 is used to train a gas quantity prediction model based on the training dataset to obtain a target model. The target model is used to predict the gas content distribution of the working face and plot it as an image to obtain a gas distribution prediction map.

[0109] Specifically, the PSO-XGBoost algorithm is used to construct a gas quantity prediction model. By training the model with a training dataset, a high-precision target model capable of predicting the gas content distribution at the working face can be obtained. The trained target model is then used to predict the gas content distribution at the working face, and the prediction results are visualized as an image, i.e., a gas distribution prediction map. This visually illustrates the distribution of gas in the coal seam, aiding in the formulation of mining plans and risk assessment.

[0110] The PSO (Particle Swarm Optimization) algorithm is used to optimize the hyperparameters of the XGBoost (Extreme Gradient Boosting) model to find the optimal model configuration. XGBoost is an efficient machine learning algorithm, particularly suitable for handling large amounts of data and high-dimensional feature problems. It constructs a series of decision trees and sums the predictions of the trees in a weighted manner to form the final prediction model.

[0111] In this embodiment, the construction unit constructs a gas-solid coupling model for gas extraction at the working face. The gas-solid coupling model is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas. The first calculation unit performs sensitivity analysis on the preset influencing factors in the influencing factor group according to the gas-solid coupling model to obtain the importance of each preset influencing factor. The determination unit determines the preset influencing factors whose corresponding importance is greater than a first threshold as target factors. The acquisition unit extracts each target factor and gas content from the historical data of gas extraction to obtain a training dataset. The prediction unit trains a gas quantity prediction model according to the training dataset to obtain a target model. The target model is used to predict the gas content distribution of the working face and plot it as an image to obtain a gas distribution prediction map. This application constructs a gas-solid coupling model to simulate the environmental state changes of the working face during the extraction process. Then, based on the gas-solid coupling model, the influence of different influencing factors on the gas content change is determined by the control variable method to identify the more important target factors. Then, a prediction model is constructed using the target factors as input parameters. Compared with manually setting the input parameters of the model, the accuracy of the prediction model is improved. This solves the problem that the input and output of the gas content prediction model in the prior art depend on manual setting, and the prediction model is very sensitive to the input parameters. Improper setting of the input parameters can easily lead to a decrease in prediction accuracy.

[0112] To construct the above-mentioned gas-solid coupling model, in one optional implementation, the building blocks include:

[0113] The first module is used to construct the continuity equation, which is used to simulate coal displacement and gas flow caused by changes in gas pressure. The continuity equation is as follows: Where G is the shear modulus of the coal, u i,jj and u j,ji Let v be the displacement component, v be Poisson's ratio, and β be the displacement component. f p is the effective stress coefficient corresponding to the fractures in the coal body. f For the fissure gas pressure, β m p is the effective stress coefficient corresponding to the pores of the coal body. m F is the matrix gas pressure of the coal seam. i Let be the volume force of the coal body in the i-th direction;

[0114] Specifically, the continuity equation, based on fundamental principles of solid-state physics and fluid mechanics, describes the interaction between coal displacement and gas flow caused by changes in gas pressure within the coal seam. During coal seam gas extraction, the decrease in gas pressure inside the coal seam causes coal shrinkage and changes in pore structure, thus affecting gas flowability and the mechanical properties of the coal. This equation considers the elastic behavior of the coal, using shear modulus (G) and Poisson's ratio (v) to describe its mechanical properties, and the effective stress coefficient (β)... f and βm This reflects the effective stress influence of gas pressure on coal seam fractures and pores. The F in the equation... i It is the external volume force of the coal body in the i-direction.

[0115] The second construction module is used to construct the porosity variation equation. This equation is used to simulate the porosity changes in coal caused by coal displacement and gas flow. The porosity variation equation is as follows: in, Porosity refers to the percentage of material in a matrix, which includes fracture porosity and matrix porosity. Where p is the initial porosity, M is the constrained axial modulus, and p is the initial porosity. m0 p is the initial matrix gas pressure. f0 ε is the initial fracture gas pressure. L P represents the limiting adsorption expansion deformation of the coal body. L Let K be the Langmuir pressure constant and K be the bulk modulus.

[0116] Specifically, the porosity variation equation, based on the physicochemical properties of coal and adsorption theory, describes the influence of coal displacement and gas flow on coal porosity. The porosity variation of the coal body directly relates to the gas storage space and flow path. The equation includes the initial state of the coal body's pore structure. Its response to pressure changes (M, β) m β f ), and Langmuir adsorption model parameters (ε) L P L This allows for a comprehensive assessment of the dynamic evolution of porosity as a function of gas pressure.

[0117] The third module is used to construct the permeability variation equation. This equation is used to simulate the permeability changes in the coal body caused by coal displacement and gas flow. The permeability variation equation is as follows: Where k is the permeability and k1 is the initial permeability;

[0118] Specifically, the permeability variation equation is based on the influence of changes in the microstructure of the coal seam on its macroscopic permeability. Permeability (k) is a key parameter for measuring the ease of gas flow in coal seams. The equation predicts changes in permeability through changes in porosity, reflecting the influence of the coal seam's pore structure on the resistance to gas flow. The direct proportionality between permeability and porosity is further amplified to a cubic relationship, indicating that small changes in porosity can lead to significant changes in permeability.

[0119] The fourth module is used to construct the matrix gas pressure variation equation. This equation is used to simulate the change in matrix gas pressure over extraction time. The matrix gas pressure variation equation is as follows: Among them, V M V is the molar volume of gas under standard conditions. L ρ is the Langmuir volume constant. c Let be the pseudo density of the coal, R be the gas constant, T be the temperature, τ be the adsorption time, and t be the extraction time. The matrix porosity;

[0120] Specifically, the matrix gas pressure variation equation, based on the Langmuir adsorption isotherm theory, describes the matrix gas pressure (p... m The variation of gas molar volume (V) with extraction time (t) under standard conditions is considered. M ), Langmuir volume constant (V L ), coal body false density (ρ) c Parameters such as gas constant (R), temperature (T), and adsorption time (τ) collectively affect the adsorption and desorption rates of gas, as well as the accumulation and release process of gas in coal.

[0121] The fifth module is used to construct the fracture gas flow equation. This equation is used to simulate the gas flow in fractures as porosity and permeability change. The fracture gas flow equation is as follows: in, The porosity is the fracture porosity. Let μ be the gradient, and μ be the dynamic viscosity of the gas.

[0122] Specifically, the fracture gas flow equation is based on Darcy's law and describes the gas flow velocity in fractures as a function of porosity. The law governing the variation of permeability (k) is as follows. Dynamic viscosity (μ) reflects the flow characteristics of gas, while the time derivative and gradient terms in the equations reflect the transient characteristics and spatial distribution effects of gas flow in fractures. The pressure gradient of gas in the fracture is also considered. Driven by gas flow, and simultaneously by the pressure difference (p) between the fractures and the coal matrix. m -p f This facilitated the transfer of gas.

[0123] The sixth building module is used to simultaneously solve the continuity equation, porosity variation equation, permeability variation equation, matrix gas pressure variation equation, and fracture gas flow equation to obtain the gas-solid coupling model.

[0124] Specifically, the gas-solid coupling model combines the above five equations to form a comprehensive gas-solid coupling model that considers factors such as coal mechanical behavior, porosity, permeability, gas pressure, temperature, and time. This allows for a comprehensive simulation of the dynamic response of the coal seam during gas extraction, including changes in coal deformation, porosity, and permeability, as well as the flow and pressure distribution of gas within the coal seam.

[0125] Through the above embodiments, the construction of a gas-solid coupling model can more accurately predict the coal seam gas extraction process and reduce errors caused by model simplification, including but not limited to using Langmuir adsorption isotherm theory to describe the adsorption behavior of gas in coal, which greatly improves the accuracy and reliability of prediction.

[0126] To determine the importance of each preset influencing factor, in one optional implementation, the first calculation unit includes:

[0127] The settings module is used to set any one preset influencing factor as a variable and set other preset influencing factors in the influencing factor group as constants;

[0128] Specifically, firstly, one of the preset influencing factors is set as a variable, while all other factors are set as constants. That is, in each simulation, only one factor is changed, while all other factors remain unchanged. For example, in the first simulation, the extraction time can be changed while keeping other factors such as permeability and porosity constant.

[0129] The first determination module is used to determine the values ​​of multiple variables, and uses a gas-solid coupling model to perform simulations based on the values ​​of each variable and calculate the gas content of the working face based on the simulation results;

[0130] Specifically, for the defined variables, multiple different values ​​are selected, and a gas-solid coupling model is used for simulation. Each variable value corresponds to one simulation, and the gas content at the working face is calculated based on the simulation results.

[0131] The first calculation module is used to calculate the ratio of the difference between the values ​​of each variable to obtain the first ratio, and to calculate the ratio of the difference in gas content corresponding to the values ​​of each variable to obtain the second ratio.

[0132] Specifically, for each variable's value, the ratio of its difference to a benchmark value (e.g., initial value or average value) is calculated to obtain a first ratio. Simultaneously, the ratio of the difference between the gas content corresponding to each variable's value and the benchmark gas content is calculated to obtain a second ratio.

[0133] The second determining module is used to determine the ratio of the second ratio to the first ratio as the degree of importance.

[0134] Specifically, the ratio of the second proportion to the first proportion is defined as the importance of the pre-defined influencing factor. The ratio reflects the sensitivity of changes in gas content to changes in the variable. A large ratio indicates that the gas content is very sensitive to changes in this variable, and that the variable has a significant impact on the gas content.

[0135] Through the above embodiments, sensitivity analysis can help identify the factors that have the greatest impact on gas content during gas drainage, thereby guiding parameter optimization. By identifying important influencing factors, the risks during gas drainage, such as the risk of gas explosion, can be assessed more accurately.

[0136] In order to simulate different preset influencing factors, in one optional implementation, the first determining module mentioned above includes:

[0137] The first simulation submodule is used to simulate the gas-solid coupling model when the variable is any one of extraction time, permeability and porosity.

[0138] Specifically, since extraction time, permeability, and porosity directly affect gas flow and the physical properties of coal, these factors can be directly substituted into the model for simulation.

[0139] The second simulation submodule is used to calculate the volume forces of the coal body in each direction based on the coal seam thickness as the variable and substitute them into the gas-solid coupling model for simulation.

[0140] Specifically, the thickness of the coal seam affects the pressure-bearing capacity of the coal body, thereby affecting the flow of gas. Therefore, by changing the volume forces of the coal body in various directions, the simulation results of the model can be indirectly affected.

[0141] The third simulation submodule is used to simulate the boundary conditions of the gas-solid coupling model when the variables are overburden load, borehole diameter, borehole negative pressure, borehole length and borehole spacing.

[0142] Specifically, overburden load, borehole diameter, borehole negative pressure, borehole length, and borehole spacing reflect the impact of the mining environment and borehole design on gas extraction efficiency, serving as boundary conditions.

[0143] The fourth simulation submodule is used to decompose the variable into initial matrix gas pressure and initial fracture gas pressure when the variable is the original gas pressure, and then substitute the initial matrix gas pressure and initial fracture gas pressure into the gas-solid coupling model for simulation.

[0144] Specifically, the original gas pressure has different effects on the stress and flow characteristics of the coal body due to the different forms in which gas exists in the coal body, and needs to be decomposed into two parts: matrix and fractures.

[0145] Through the above embodiments, the simulation strategy is refined to more realistically reflect the various complex phenomena in the gas extraction process, thereby improving the accuracy and reliability of subsequent predictions.

[0146] In an optional implementation, to calculate the gas content at the working face, the first determining module further includes:

[0147] The first calculation submodule is used to calculate the amount of adsorbed gas based on porosity, matrix gas pressure and fracture gas pressure, and obtain the first gas content.

[0148] The second calculation submodule is used to calculate the amount of free-state gas based on the fracture gas pressure, and obtain the second gas content.

[0149] The third calculation submodule is used to calculate the sum of the first gas content and the second gas content to obtain the gas content.

[0150] Specifically, the existence of gas in coal seams can be mainly divided into two forms: adsorbed state and free state. Adsorbed gas refers to gas molecules that are attracted by the molecular forces on the surface of the coal seam and are tightly adsorbed onto the pore surface of coal particles; free gas refers to gas that exists in a free state in the pores of the coal seam.

[0151] The above embodiments not only consider the distribution of gas in the coal seam but also integrate the physical properties of the coal body (such as porosity, matrix gas pressure, and fracture gas pressure), making the calculation results closer to reality. For example, increased porosity increases the flow space for free gas, thereby increasing the amount of free gas; while a decrease in matrix gas pressure leads to the release of adsorbed gas, which is converted into free gas, ultimately affecting the total gas content. The calculation method based on model prediction can help mines more accurately assess the effectiveness of gas drainage, adjust drainage strategies in a timely manner, effectively prevent gas accidents, and ensure safe production in the mine.

[0152] In order to calculate the aforementioned first gas content, in an optional implementation, the aforementioned first calculation submodule is further configured to:

[0153] The absolute gas pressure of the coal body is determined based on the matrix gas pressure and the fracture gas pressure.

[0154] The first gas content is calculated based on porosity and absolute gas pressure, where the calculation formula is as follows: Where W is the first gas content, a is the maximum adsorption capacity, b is the adsorption strength, and A is the adsorption capacity. ad M represents the moisture content of the coal. ad γ represents the ash content of the coal, P represents the absolute gas pressure, and γ represents the apparent density of the coal.

[0155] Specifically, the calculation of adsorbed gas quantity is based on the Langmuir adsorption isotherm theory, which describes the adsorption behavior of gas on the coal surface. By combining the matrix gas pressure and fracture gas pressure to calculate the absolute gas pressure, and then using factors such as porosity, moisture, and ash to correct the adsorption calculation, the amount of adsorbed gas can be predicted more accurately. For example, the presence of moisture and ash occupies the pore space of the coal body, reducing the area available for gas adsorption, thereby reducing the amount of adsorbed gas. This calculation method not only considers the physicochemical properties of the coal body but also integrates the influence of gas pressure, making the prediction results more reliable and helping mines to formulate reasonable gas drainage plans and reduce the risk of gas accumulation. Through calculation, we found that changes in coal body components, including but not limited to moisture and ash, and key parameters such as absolute gas pressure, significantly affect the amount of adsorbed gas. Therefore, precise management of these parameters plays an important role in improving gas drainage efficiency and safety.

[0156] To ensure the accuracy of the prediction model, in one optional implementation, the prediction unit includes:

[0157] The module is used to build a gas quantity prediction model using PSO-XGBoost and obtain alternative models.

[0158] Specifically, historical gas extraction data are collected, including extraction time, permeability, porosity, coal seam thickness, overlying load, borehole parameters, and corresponding gas content. A gas quantity prediction model is constructed using PSO-XGBoost to obtain alternative models.

[0159] The third determination module is used to determine multiple hyperparameter values ​​based on the hyperparameters of the candidate models, and to determine each hyperparameter value as a particle in the particle swarm algorithm, with a one-to-one correspondence between the hyperparameter value and the particle.

[0160] Specifically, the hyperparameters of the XGBoost model, such as tree depth, learning rate, and regularization parameters, are determined, and different values ​​of these hyperparameters are used as "particles" in PSO. Each hyperparameter value represents a particle, and the initial position and velocity of the particles are randomly initialized.

[0161] The optimization module is used to find the target particle based on the PSO algorithm.

[0162] Specifically, the model performance on the training dataset (such as the root mean square error of cross-validation) is used as the fitness metric for the particles. Based on the particle's current position, its position and velocity are updated through social learning (learning from the swarm's best particle) and individual learning (learning from its own historical best position) to find the optimal combination of hyperparameters. The PSO algorithm iterates repeatedly until a preset number of iterations is reached or the swarm's best particle is found. The hyperparameter values ​​corresponding to this best particle constitute the configuration of the target model, i.e., the target particle mentioned above.

[0163] The configuration module is used to configure alternative models based on the hyperparameter values ​​corresponding to the target particle.

[0164] Specifically, the optimal hyperparameters found by PSO are used to configure the XGBoost model.

[0165] The training module is used to train the configured candidate models based on the training dataset to obtain the target model.

[0166] Specifically, the configured XGBoost model is trained using the complete training dataset to obtain the target model.

[0167] Through the above embodiments, the PSO algorithm finds the optimal solution, i.e., the optimal combination of model hyperparameters, by simulating the foraging behavior of bird flocks; the XGBoost algorithm, on the other hand, improves prediction accuracy by constructing multiple decision trees to progressively optimize the model. Setting up combined algorithms not only enables rapid convergence and the finding of optimal model parameters, but also effectively handles high-dimensional and nonlinear data, improving the adaptability and accuracy of the prediction model. For example, the optimal hyperparameter values ​​obtained through the PSO algorithm enable the XGBoost model to better fit historical gas extraction data, improving prediction accuracy.

[0168] The aforementioned coal seam gas residual prediction device includes a processor and a memory. The aforementioned construction unit, first calculation unit, determination unit, acquisition unit, and prediction unit are all stored as program units in the memory. The processor executes these program units stored in the memory to achieve the corresponding functions. All of the above modules are located in the same processor; alternatively, the modules may be located in different processors in any combination.

[0169] The processor contains a kernel, which retrieves the corresponding program unit from memory. One or more kernels can be configured, and adjusting kernel parameters can improve the accuracy of gas content prediction.

[0170] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0171] This invention provides a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the method for predicting residual coal seam gas.

[0172] This invention provides a processor for running a program, wherein the program executes the method for predicting residual coal seam gas.

[0173] This invention provides a gas content monitoring system, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements at least the steps of the above-described method for predicting residual coal seam gas.

[0174] This application also provides a computer program product that, when executed on a data processing device, is adapted to perform the steps of initializing a prediction method having at least the aforementioned residual amount of coal seam gas.

[0175] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0176] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0177] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0178] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0179] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0180] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0181] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0182] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0183] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0184] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0185] As can be seen from the above description, the embodiments of this application achieve the following technical effects:

[0186] 1) The method for predicting residual coal seam gas in this application firstly constructs a gas-solid coupling model for gas drainage at the working face. The gas-solid coupling model is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas. Then, based on the gas-solid coupling model, a sensitivity analysis is performed on the preset influencing factors in the influencing factor group to obtain the importance of each preset influencing factor. After that, the preset influencing factors with an importance greater than a first threshold are determined as target factors. Then, each target factor and gas content are extracted from the historical data of gas drainage to obtain a training dataset. Finally, a gas quantity prediction model is trained based on the training dataset to obtain a target model. The gas content distribution at the working face is predicted by the target model and plotted as an image to obtain a gas distribution prediction map. This application constructs a gas-solid coupling model to simulate the environmental state changes of the working face during the extraction process. Then, based on the gas-solid coupling model, the influence of different influencing factors on the gas content change is determined by the control variable method to identify the more important target factors. Then, a prediction model is constructed using the target factors as input parameters. Compared with manually setting the input parameters of the model, the accuracy of the prediction model is improved. This solves the problem that the input and output of the gas content prediction model in the prior art depend on manual setting, and the prediction model is very sensitive to the input parameters. Improper setting of the input parameters can easily lead to a decrease in prediction accuracy.

[0187] 2) The coal seam gas residual quantity prediction device of this application comprises: a construction unit constructing a gas-solid coupling model for gas extraction at the working face, which is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas; a first calculation unit performing sensitivity analysis on the preset influencing factors in the influencing factor group according to the gas-solid coupling model to obtain the importance of each preset influencing factor; a determination unit determining the preset influencing factors with an importance greater than a first threshold as target factors; an acquisition unit extracting each target factor and gas content from the historical data of gas extraction to obtain a training dataset; and a prediction unit training a gas quantity prediction model according to the training dataset to obtain a target model, predicting the gas content distribution of the working face through the target model and plotting it as an image to obtain a gas distribution prediction map. This application constructs a gas-solid coupling model to simulate the environmental state changes of the working face during the extraction process. Then, based on the gas-solid coupling model, the influence of different influencing factors on the gas content change is determined by the control variable method to identify the more important target factors. Then, a prediction model is constructed using the target factors as input parameters. Compared with manually setting the input parameters of the model, the accuracy of the prediction model is improved. This solves the problem that the input and output of the gas content prediction model in the prior art depend on manual setting, and the prediction model is very sensitive to the input parameters. Improper setting of the input parameters can easily lead to a decrease in prediction accuracy.

[0188] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for predicting residual gas in coal seams, characterized in that, include: A gas-solid coupling model for gas extraction at the working face is constructed. The gas-solid coupling model is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas. Sensitivity analysis is performed on the preset influencing factors in the influencing factor group based on the gas-solid coupling model to obtain the importance of each preset influencing factor. The preset influencing factors whose importance level is greater than the first threshold are identified as target factors; The target factors and gas content are extracted from historical data of gas extraction to obtain a training dataset. A gas quantity prediction model is trained based on the training dataset to obtain a target model. The gas content distribution of the working face is predicted by the target model and plotted as an image to obtain a gas distribution prediction map.

2. The method according to claim 1, characterized in that, Constructing a gas-solid coupling model for gas extraction at the working face, including: A continuity equation is constructed to simulate coal seam displacement and gas flow caused by changes in gas pressure. The continuity equation is as follows: Where G is the shear modulus of the coal body, u i,jj and u j,ji Let v be the displacement component, v be Poisson's ratio, and β be the displacement component. f p is the effective stress coefficient corresponding to the fractures in the coal body. f For the fissure gas pressure, β m p is the effective stress coefficient corresponding to the pores of the coal body. m F is the matrix gas pressure of the coal body. i Let be the volume force of the coal body in the i-th direction; A porosity variation equation is constructed to simulate the porosity change of the coal body caused by the displacement of the coal body and the flow of gas. The porosity variation equation is as follows: in, The porosity includes fracture porosity and matrix porosity. Where p is the initial porosity, M is the constrained axial modulus, and p is the initial porosity. m0 p is the initial matrix gas pressure. f0 ε is the initial fracture gas pressure. L P represents the limiting adsorption expansion deformation of the coal body. L K is the Langmuir pressure constant, and K is the bulk modulus. A permeability variation equation is constructed to simulate the permeability change of the coal body caused by the displacement of the coal body and the flow of gas. The permeability variation equation is as follows: Where k is the permeability and k1 is the initial permeability; A matrix gas pressure variation equation is constructed to simulate the change of matrix gas pressure with extraction time. The matrix gas pressure variation equation is as follows: Among them, V M V is the molar volume of gas under standard conditions. L ρ is the Langmuir volume constant. c Let R be the pseudo density of the coal, T be the gas constant, τ be the temperature, and t be the adsorption time. The matrix porosity; A fracture gas flow equation is constructed to simulate the gas flow in fractures as a function of porosity and permeability. The fracture gas flow equation is as follows: in, The fracture porosity is... Let μ be the gradient, and μ be the dynamic viscosity of the gas. By combining the continuity equation, the porosity variation equation, the permeability variation equation, the matrix gas pressure variation equation, and the fracture gas flow rate equation, the gas-solid coupling model is obtained.

3. The method according to claim 2, characterized in that, Sensitivity analysis is performed on the preset influencing factors in the influencing factor group based on the gas-solid coupling model to obtain the importance of each preset influencing factor, including: Set any one of the preset influencing factors as a variable, and set the other preset influencing factors in the influencing factor group as constants; Determine multiple variable values ​​for the variable, and use the gas-solid coupling model to perform simulations based on each variable value, and calculate the gas content of the working face based on the simulation results; Calculate the ratio of the difference between the values ​​of each variable to obtain a first ratio, and calculate the ratio of the difference in the gas content corresponding to the values ​​of each variable to obtain a second ratio; The ratio of the second ratio to the first ratio is determined as the importance level.

4. The method according to claim 3, characterized in that, Simulations were performed using the aforementioned gas-solid coupling model based on the values ​​of each variable, including: When the variable is any one of the extraction time, the permeability, and the porosity, the variable is substituted into the gas-solid coupling model for simulation; When the variable is the coal seam thickness, the volume forces of the coal body in each direction are calculated based on the variable and substituted into the gas-solid coupling model for simulation. Simulations were performed using the variables as boundary conditions for the gas-solid coupling model, with the variables being the overburden load, borehole diameter, borehole negative pressure, borehole length, and borehole spacing. When the variable is the original gas pressure, the variable is decomposed into the initial matrix gas pressure and the initial fracture gas pressure, and the initial matrix gas pressure and the initial fracture gas pressure are substituted into the gas-solid coupling model for simulation.

5. The method according to claim 3, characterized in that, The gas content of the working face is calculated based on the simulation results, including: The amount of adsorbed gas is calculated based on the porosity, the matrix gas pressure, and the fracture gas pressure to obtain the first gas content; The amount of free-state gas is calculated based on the fracture gas pressure to obtain the second gas content; The gas content is obtained by calculating the sum of the first gas content and the second gas content.

6. The method according to claim 5, characterized in that, The amount of adsorbed gas is calculated based on the porosity, the matrix gas pressure, and the fracture gas pressure to obtain the first gas content, including: The absolute gas pressure of the coal body is determined based on the matrix gas pressure and the fracture gas pressure. The first gas content is calculated based on the porosity and the absolute gas pressure, wherein the calculation formula is as follows: Where W is the first gas content, a is the maximum adsorption amount and b is the adsorption strength, and A ad M represents the moisture content of the coal body. ad γ represents the ash content of the coal, P represents the absolute gas pressure, and γ represents the apparent density of the coal.

7. The method according to claim 1, characterized in that, A gas quantity prediction model is trained based on the training dataset to obtain the target model, including: The gas quantity prediction model was constructed using PSO-XGBoost to obtain alternative models. Based on the hyperparameters of the candidate models, multiple hyperparameter values ​​are determined, and each hyperparameter value is identified as a particle in the particle swarm optimization algorithm, with each hyperparameter value corresponding to a particle in a one-to-one manner. The target particle is obtained by optimizing the particle using the PSO algorithm. Configure the alternative models according to the hyperparameter values ​​corresponding to the target particle; The target model is obtained by training the configured candidate model based on the training dataset.

8. A device for predicting residual gas in coal seams, characterized in that, The device includes: The building unit is used to build a gas-solid coupling model for gas extraction in the working face. The gas-solid coupling model is used to simulate the interaction between the mechanical state of the coal body and the flow state of the gas. The first calculation unit is used to perform sensitivity analysis on the preset influencing factors in the influencing factor group according to the gas-solid coupling model, and to obtain the importance of each preset influencing factor. A determining unit is used to determine the preset influencing factors whose importance is greater than a first threshold as target factors; The acquisition unit is used to extract the target factors and gas content from historical data of gas extraction to obtain a training dataset. The prediction unit is used to train a gas quantity prediction model based on the training dataset to obtain a target model, and then use the target model to predict the gas content distribution of the working face and plot it as an image to obtain a gas distribution prediction map.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method according to any one of claims 1 to 7.

10. A gas content monitoring system, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising methods for performing any one of claims 1 to 7.

Images