Dynamic visualization method of coal seam gas basic parameters based on monitoring data calculation

Abstract

A method for dynamic visualization of coal seam gas basic parameters based on monitoring data calculation is characterized by comprising the following steps: (1) constructing a fluid-solid coupling model of gas-containing coal; (2) collecting relevant parameters of the target coal seam; (3) establishing the specific dimensions of length, width and height, boundary conditions and initial conditions of the geometric model; (4) collecting the extraction monitoring data of the target coal seam pre-extraction drilling hole; (5) using SURFER software to draw a contour map of the calculated coal seam in-situ gas pressure to realize the visualization of coal seam gas pressure; (6) predicting key parameters in the gas extraction process through a deep learning algorithm, and realizing the dynamic visualization of coal seam gas pressure in combination with the coal seam in-situ gas pressure calculation method. The present invention can realize the dynamic visualization of coal seam gas extraction parameters.

Description

Technical Field

[0001] The invention relates to the technical field of underground ventilation safety in coal mines, and in particular to a method for dynamically visualizing basic parameters of coal seam gas based on calculation of monitoring data. Background Art

[0002] During the pre-extraction of coal seam gas, the calculation method of coal seam gas pressure and content is mainly manual monitoring, and the effect of pre-extraction of coal seam gas is tested by directly measuring parameters such as coal seam residual gas pressure or residual gas content. Although this method is direct and easy to operate, it is difficult to control the time stage after gas extraction when it is suitable to directly measure the residual gas pressure or residual gas content of the coal seam. Repeated drilling and sampling measurements on site have caused a great waste of manpower, material resources and time in the entire measurement process, and it is impossible to provide timely feedback on the pre-extraction effect, resulting in an increase in the cost of gas control; it may also cause the coal body state to be damaged due to multiple construction drilling, resulting in inaccurate measurement results. In addition, there are a large number of blank areas in the selection of measurement holes during the measurement process, and the measurement results are not refined.

[0003] In the process of coal seam gas pre-extraction, the existing methods for monitoring key gas parameters and judging the extraction effect are mainly manual monitoring, which directly measures the residual gas pressure or residual gas content of the coal seam to test the effect of pre-extraction of coal seam gas. Although this method is direct and easy to operate, it is difficult to control the time stage after gas extraction when it is suitable to directly measure the residual gas pressure or residual gas content of the coal seam. Repeated drilling and sampling on site will greatly waste manpower, material resources and time in the entire evaluation process, and it is impossible to provide timely feedback on the pre-extraction effect, resulting in an increase in the cost of gas control; it may also cause the coal body state to be damaged due to multiple construction drilling, resulting in inaccurate measurement results. Summary of the invention

[0004] In order to overcome the above technical problems, the purpose of the present invention is to provide a method for dynamic visualization of basic parameters of coal seam gas calculated based on monitoring data, by constructing an inversion model of coal seam gas parameters for monitoring data in the process of coal seam gas extraction, using the monitoring data to invert the gas pressure in the process of coal seam gas extraction, establishing a data processing method based on deep learning to predict the coal seam gas monitoring data, using the predicted data to invert the distribution and change law of future coal seam gas pressure, and using the SURFER software to realize dynamic visualization of coal seam gas extraction parameters.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A method for dynamic visualization of coal seam gas basic parameters based on monitoring data calculation includes the following steps;

[0007] (1) Through the gas state equation, mass conservation law, coalbed methane flow continuity equation and coalbed methane motion equation, the mixed unsteady flow characteristics of coalbed methane seepage-diffusion in the "pore-fracture" dual medium of coal are obtained, the coalbed gas flow field control equation is established, the boundary conditions and initial conditions of the equation are determined, the analytical solution and numerical solution of the equation are obtained, and the relationship between the coalbed methane extraction flow rate and the coalbed methane pressure in the coal seam is established; a fluid-solid coupling model of gas-containing coal is constructed;

[0008] (2) Collect relevant parameters of the target coal seam;

[0009] (3) Based on the target coal seam, determine the specific dimensions of the length, width, and height of the geometric model, boundary conditions, and initial conditions;

[0010] (4) Collecting the extraction monitoring data of the target coal seam pre-extraction borehole;

[0011] (5) The calculated in-situ gas pressure of the coal seam is used to draw a contour map of the gas pressure through SURFER software, thereby realizing the visualization of the coal seam gas pressure.

[0012] (6) The key parameters in the gas extraction process are predicted through deep learning algorithms, and the dynamic visualization of coal seam gas pressure is achieved by combining the in-situ gas pressure calculation method of coal seams.

[0013] The fluid-solid coupling model of gas-containing coal in step (1) is specifically:

[0014]

[0015] The definitions of the parameters are as follows:

[0016] u i，j , i, j = 1, 2, 3---- displacement vector of the particle

[0017] ε ij ----Strain component

[0018] K m ----Bulk modulus of coal matrix, Pa

[0019] β M ----Wet expansion coefficient, m 3 / m 3

[0020] α sg ----Volume strain coefficient caused by adsorption and desorption, kg·m -3

[0021] V sg ----Adsorbed gas content, m 3 kg-1

[0022] c1----pressure coefficient, MPa -1

[0023] c2----temperature coefficient, K -1

[0024] V L ---- is the Langmuir volume constant, m 3 kg -1

[0025] p L ----Langmuir pressure constant, Pa

[0026] T——Coal seam temperature, K

[0027] T t ----The reference temperature in the adsorption and desorption test, K, is 298

[0028] E----elastic modulus, Pa

[0029] υ----Poisson's ratio

[0030] λ,G----Lame constant

[0031] φ m0 ----Initial porosity

[0032] ε m ----Coal matrix volume strain

[0033] K Y ----Volume compression coefficient, MPa -1

[0034] K s ----Skeleton bulk modulus

[0035] φ m ----Matrix porosity

[0036] φ f ----Matrix crack rate

[0037] k m ----Matrix pore permeability

[0038] k f ----Matrix fracture permeability

[0039] p m ----Matrix gas pressure, Pa

[0040] p f ----Fissure gas pressure, Pa

[0041] τ----gas desorption and diffusion time, s

[0042] M g ----Molar mass of gas, kg / mol

[0043] V L ----Langmuir volume constant, m 3 / kg

[0044] P L ----Langmuir pressure constant, Pa

[0045] ρ s ----Skeleton density, kg / m 3

[0046] Among them, formula 1 is the stress field control equation of gas-containing coal, which is the constitutive equation of coal body stress deformation; formula 2 is the gas matrix porosity control equation, formula 3 is the pore permeability control equation, formula 4 is the gas fracture rate control equation, and formula 5 is the fracture permeability control equation; formula 6 is the gas diffusion control equation in the matrix, and formula 7 is the gas seepage control equation in the fracture. These two equations express the seepage diffusion motion equation of gas in the coal seam; the seven equations together constitute the fluid-solid coupling model of gas-containing coal;

[0047] Coal is a dual medium with pores and fractures. Porosity, fracture ratio and its corresponding permeability are important factors that determine gas content and gas migration, and are also key factors affecting the fluid-solid coupling process of gas-containing coal. During the drilling gas extraction process, the flow of gas will simultaneously affect the gas pressure and ground stress inside the coal seam, causing the effective stress on the coal skeleton to change, causing the coal skeleton to deform accordingly, resulting in dynamic changes in porosity and permeability. Therefore, considering the dynamic change characteristics of the porosity and permeability of the coal body under the action of gas-solid coupling is a prerequisite for studying the deformation of the disturbed coal body and the gas-solid coupling mechanism that causes gas migration.

[0048] The step (2) is specifically:

[0049] Collect the parameters in the fluid-solid coupling model of gas-containing coal:

[0050] According to coal safety regulations, coal mines will conduct relevant experiments and field tests before coal mining and gas extraction, and will issue corresponding reports; all the above parameters can be collected by collecting data from the coal mine's "Coal Seam Coal and Gas Outburst Hazard Assessment Report", "Mine Gas Emission Prediction Method", "Coal Mine Gas Extraction Engineering Design Standard", "Coal Mine Safety Facility Design" and "Coal Seam Gas Parameter Basic Parameter Test Report";

[0051] The step (3) is specifically:

[0052] The boundary conditions and initial conditions (initial conditions and boundary conditions are different for different target coal seams, and the boundaries include: boundary load, boundary displacement, boundary flow, etc. The initial conditions include: initial displacement of coal rock, initial stress of coal rock, and initial gas pressure) are brought into the fluid-solid coupling model of gas-containing coal, and the control equations are solved using COMSOL or FLUENT computational fluid dynamics software; the data of formulas 2 and 3 can be obtained by solving the coupling model of formula 1 with numerical calculation software, and the data can be fitted to obtain formulas 2 and 3, thereby calculating the change law of the gas pressure of the coal seam at a distance x from the borehole for the coal seam extraction borehole; the functional relationship between the coal seam gas pressure and the borehole flow rate and the attenuation law of the borehole flow rate along the borehole length are fitted through the change law data diagram, that is:

[0053] Q L =f(L) (2)

[0054]

[0055] Where P x ----In-situ gas pressure of the coal seam at distance x from the borehole, Q L ----Monitor the flow rate of the gas pre-extraction borehole at the orifice L along the long distance of the hole, L----the distance from the orifice along the long distance of the hole.

[0056] The gas borehole flow rate in the collected gas data of a single borehole is substituted into formulas (2) and (3) to obtain the gridded values ​​of the in-situ residual gas pressure of the target coal seam during the extraction process. By substituting each value into the SURFER software, the visualization effect of the target coal seam gas during the extraction process can be achieved.

[0057] The step (4) is specifically:

[0058] That is, the borehole extraction condition flow rate, gas concentration, pure gas flow rate, and extraction negative pressure; the data collected from each borehole is preprocessed, and the data of each borehole is used to calculate the in-situ gas pressure of the coal seam at the total distance x from the borehole using the inversion formula fitted in step (3).

[0059] The step (5) is specifically as follows:

[0060] The GRU model in the recurrent neural network is used to train the historical monitoring data of coal seam gas pre-extraction, and the preprocessed data set is divided into a training set, a validation set, and a test set. The model is trained using the training set, the training results are verified using the validation set, and finally the model accuracy is tested using the test set. Finally, a prediction model and prediction data for key parameters of coal seam gas extraction are formed.

[0061] The predicted data is combined with the coal seam gas inversion model to calculate the in-situ gas pressure during coal seam gas extraction. The SURFER software is used to achieve visualization effects, thereby realizing the dynamic prediction and inversion of the in-situ gas pressure of the coal seam.

[0062] A coal seam gas comprehensive parameter measurement system, including a drainage system, gas drainage negative pressure, flow, gas concentration, temperature sensors and an industrial ring network and a data server;

[0063] The pumping system is connected to the sensor, the sensor is connected to the industrial ring network, and the industrial ring network transmits the received data to the data server.

[0064] By adjusting the single borehole switch valve on the pumping system, the data of the single pre-drilling borehole can be monitored online;

[0065] The extraction system is connected to the coal seam gas pre-extraction borehole. The single borehole is switched on and off by adjusting the switch valve of the single borehole on the extraction system. When the extraction system and the sensor need to measure the gas data in a certain borehole, the switch valve is used to shut down other boreholes in the line and the borehole that needs to be measured is retained to achieve the measurement of a single borehole. The measured data is transmitted between the industrial ring networks and finally saved to the data server.

[0066] The drilled pipe is connected to the extraction pump, which is then connected to the gas extraction negative pressure, flow, gas concentration and temperature sensors.

[0067] Beneficial effects of the present invention.

[0068] The method provided by the present invention for finely dividing coal seam gas pressure high and low areas and realizing dynamic visualization can make up for the shortcomings of traditional methods for measuring coal seam gas pressure, such as breadth, fuzziness and staticity.

[0069] The method provided by the present invention is to finely divide the high and low gas pressure areas of coal seams and realize dynamic visualization. By monitoring the data of a single borehole on site with fine monitoring equipment, a fluid-solid coupling model of gas-containing coal is established to solve the inversion model of the borehole extraction flow for the coal seam gas residual pressure and the attenuation law of the borehole flow along the hole length. Combined with the multi-sensor fusion gas parameter multi-step prediction method based on deep learning, it can make up for the shortcomings of traditional methods for measuring coal seam gas pressure, such as breadth, fuzziness and staticity. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 This is a schematic diagram of the layout of pre-drainage drilling holes in this coal seam.

[0071] Figure 2 It is an on-site monitoring system for gas pre-extraction drilling parameters in coal seam working faces.

[0072] Figure 3 This is the flow chart of wavelet denoising.

[0073] Figure 4 Schematic diagram of outlier data processing.

[0074] Figure 5 Schematic diagram of noise value data processing.

[0075] Figure 6 This is the GRU structure diagram.

[0076] Figure 7 The figure is a flow chart of the dynamic visualization method of coal seam gas parameters based on monitoring data.

[0077] Figure 8 This is a schematic diagram of the original gas distribution area of ​​a mine.

[0078] Fig. 9 This is a schematic diagram of the target coal seam gas distribution of the present invention. DETAILED DESCRIPTION

[0079] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0080] like Figure 1-Figure 9 As shown in the figure, the pre-drainage boreholes of coal seams are generally densely distributed, usually with a spacing of 3-6m. The present invention can use the monitoring data of a single borehole to achieve a refined parameterized calculation of the gas pressure and content of the target coal seam through an inversion model, thereby achieving a precise measurement that is not achievable with current technology. This is equivalent to gridding the target coal seam, such as Figure 1 shown.

[0081] In addition, the present invention establishes a prediction model through a deep learning recurrent neural network to train the historical data of gas pre-extraction monitoring, so as to predict the key parameters of future gas monitoring data, and realizes the dynamic visualization of gas pressure and content in the process of coal seam gas extraction by combining the predicted value with the inversion model.

[0082] In the related art, the calibration of the in-situ gas pressure during the pre-extraction of coal seam gas is determined by manual measurement. The measurement results are very broad and imprecise in guiding the entire coal seam. In addition, the measurement process consumes a lot of manpower and material resources. The technical problem solved by the present invention is to solve the problem that the in-situ gas pressure of the coal seam during the pre-extraction of coal seam gas in the mine is unclear and opaque, and there is a lack of dynamic visualization method for the coal seam gas pressure.

[0083] The first aspect of the present invention provides a method for inverting the in-situ gas pressure of a coal seam based on coal seam pre-extraction gas monitoring data, and the specific operations are as follows:

[0084] (1) Through the gas state equation, mass conservation law, coalbed methane flow continuity equation and coalbed methane motion equation, the mixed unsteady flow characteristics of coalbed methane seepage-diffusion in the "pore-fracture" dual medium of coal are derived, the coalbed gas flow field control equation is established, the boundary conditions and initial conditions of the equation are established, the analytical solution and numerical solution of the equation are obtained, and the relationship between the coalbed methane extraction flow rate and the coalbed methane pressure in the coal seam is established. The fluid-solid coupling model of gas is shown in Equation 1

[0085]

[0086] The above formula includes the stress deformation model of gas coal body, the dynamic change model of porosity and permeability, the dynamic change model of fracture ratio and permeability and the seepage and diffusion model of gas in coal seam.

[0087] (2) Collect relevant parameters of the target coal seam. The parameter collection table is shown below. The values ​​of each parameter are derived from the actual measured values ​​or material fixed parameters of the target mine.

[0088] Parameter Value

[0089]

[0090]

[0091] (3) Combined with the target coal seam, establish the specific dimensions of the length, width, and height of the geometric model, boundary conditions, and initial conditions. Bring the boundary conditions and initial conditions into the coupling model, and use computational fluid dynamics software such as COMSOL or FLUENT to solve the control equations, so as to calculate the change law of the gas pressure at the distance x from the coal seam to the coal seam extraction borehole. The functional relationship between the coal seam gas pressure and the borehole flow rate is fitted through the change law data graph, that is:

[0092]

[0093] Where P x ----In-situ gas pressure of the coal seam at a distance x from the borehole, Q----gas pre-extraction borehole monitoring flow.

[0094] (4) Collect the extraction monitoring data of the target coal seam pre-extraction borehole, i.e., the borehole extraction condition flow rate, gas concentration, pure gas flow rate, and extraction negative pressure. The data collected from each borehole are pre-processed, and the in-situ gas pressure of the coal seam at the total distance x from the borehole is calculated using the inversion formula fitted in step (3).

[0095] (5) The calculated in-situ gas pressure of the coal seam is used to draw a contour map of the gas pressure through SURFER software, thereby realizing the visualization of the coal seam gas pressure.

[0096] The fifth step realizes visualization, but it is only a static visualization result. The visualization of the residual gas pressure of the coal seam at the current moment is realized through the monitoring data at a certain moment. The following steps are to predict the real-time online monitoring data, animate the static visualization, and realize dynamic visualization. The dynamic change trend of the residual gas pressure of the coal seam can be predicted through the existing monitoring data.

[0097] (6) In order to better realize the dynamic inversion of coal seam gas parameters, the present invention also provides a coal seam gas comprehensive parameter measurement system, which is composed of an extraction pipeline, gas extraction negative pressure, flow, gas concentration, temperature and other sensors, an industrial ring network and a data server. By adjusting the switch valve of a single borehole on the extraction system, online monitoring of the data of a single pre-extraction borehole can be achieved.

[0098] (7) Preprocess the collected gas drilling monitoring data. The missing values, abnormal values ​​and noise values ​​of the data are processed by using the triple exponential smoothing method, autoregressive model and wavelet denoising method respectively.

[0099] The processing method is as follows:

[0100] A: The triple exponential smoothing method can make the supplemented missing values ​​smooth and maintain the change trend of the original time series. t The missing value at a certain moment in (t=1,2,3…,N) is subjected to triple exponential smoothing. The formula is as follows:

[0101]

[0102] Where: l is the number of inserted data points and the step size of smoothing; X t+l is the missing value to be requested; a t ,b t ,c t are the prediction parameters of the triple exponential smoothing method; is the exponential smoothing value at time t, is the quadratic exponential smoothing index at time t, is the cubic exponential smoothing index at time t; α is the smoothing coefficient. The calculation formula is as follows:

[0103]

[0104]

[0105] Where: X t is the missing value of coalbed methane concentration; is the exponential smoothing value at time t-1; is the quadratic exponential smoothing value at time t-1; is the triple exponential smoothing value at time t-1. By calculating through the above formula, we can finally get the missing value X to be supplemented. t+l .

[0106] B: When dealing with outlier data, the method of elimination or replacement is generally used. This paper uses the auto-regressive model (AR) of time series to deal with a small number of outliers. The specific processing steps are as follows:

[0107] Determine the samples that need to be processed. Find the outliers according to the threshold setting, select the modeling samples from the monitoring data before the outliers appear, and try to avoid the data that needs to be replaced in the selected samples;

[0108] Standardization. In order to improve the operational efficiency of the model, the data is standardized using the data standardization method mentioned above;

[0109] Model parameter estimation and order determination. The parameters of the estimated model are determined using the Burg method, and the model order is determined by the Akaike Information Criterion (AIC), as shown in the following formula:

[0110] AIC(p)=-2ln L+2p (10)

[0111] Among them, p is the model order, L is the likelihood function, and the formula of L can be expressed as:

[0112]

[0113] Replacement of abnormal data. According to the number of abnormal values ​​i, the AR model is used to make i predictions respectively, and each predicted value is standardized and inversely transformed to replace each abnormal value.

[0114] C: Wavelet transform is to transform the basic wavelet After the function is shifted by τ, the inner product is made with the signal to be analyzed x(t) at different scales α. Since the time series data taken is discrete, this application mainly introduces the threshold denoising method combined with discrete wavelet transform. The wavelet threshold denoising process is as follows Figure 3 As shown:

[0115] ① Wavelet transform multi-scale decomposition. By determining the optimal wavelet function, the decomposition layer number L, the time series x t Perform multi-layer wavelet decomposition to obtain the wavelet coefficients α of each layer j,k , x(k) is x t The sampling data, x t The orthogonal wavelet transform decomposition formula is:

[0116]

[0117] Among them, α j,k is the wavelet coefficient, α j,k is the scale factor, α j,k and g are a pair of orthogonal mirror filter banks, j is the number of decomposition layers, and n is the discrete sampling point.

[0118] ②Threshold processing. Usually, the threshold processing methods include hard threshold and soft threshold. The hard threshold denoising method is to use the wavelet coefficient α j,k When the absolute value of is less than the given threshold, it is set to 0; when it is greater than the threshold, it is kept unchanged. is the new wavelet coefficient, and its expression is:

[0119]

[0120] The soft threshold denoising method is to use the wavelet coefficient α j,k When the absolute value of is less than the given threshold, it is set to 0; when it is greater than the threshold, it is subtracted from the threshold. The expression is:

[0121]

[0122] ③ Inverse wavelet transform to reconstruct the signal. The new wavelet coefficients after threshold processing and the scale factor β j,k The expression of reconstructing the denoised signal is:

[0123]

[0124] (8) The GRU model in the recurrent neural network is used to train the historical monitoring data of coal seam gas pre-extraction, and the training steps are as follows:

[0125] The Gated Recurrent Unit (GRU) is a type of recurrent neural network. The GRU model structure consists of two logic gates, namely the update gate and the reset gate. The update gate in the GRU determines whether the information of the previous moment is saved to the current moment; the reset gate controls whether the historical information is saved to the next moment. The GRU model structure is as follows Figure 6 shown.

[0126] where r t To reset the gate, Z t is the update gate, U t To reset the information, h t and h t-1 are input information and output information respectively, X t is the input sequence at time t, y t The output sequence at time t. The operating principle formula of the GRU model is as follows.

[0127] Update gate calculation formula:

[0128] Z t =σ(W z ·[h t-1 ,x t ])

[0129] =σ(W hz ·h t-1 +W xz ·x t ) (17)

[0130] Reset gate calculation formula:

[0131] r t =σ(W r ·[h t-1 ,x t ])

[0132] =σ(W hr ·h t-1 +W xr ·x t ) (18)

[0133] Update parameters:

[0134]

[0135] (9) Combine the predicted data with the coal seam gas inversion model, collect the gas flow and extraction negative pressure parameters in the coal seam pre-extraction borehole, train and predict the data using the deep learning prediction model of coal seam gas key parameters, and then bring the predicted data into the inversion model to calculate the in-situ gas pressure during coal seam gas extraction. Use SURFER software to achieve visualization effects, thereby realizing the dynamic prediction and inversion of the in-situ gas pressure of the coal seam.

[0136] Example:

[0137] The method of the present invention can realize the refined representation of the coal seam gas pressure distribution of the 215 working face. Taking the coal seam pre-drainage boreholes No. 316 to No. 416 of the 215 working face of a certain mine as an example, the coal seam pre-drainage boreholes in this area are arranged in parallel, with a borehole spacing of 6m, and a target coal seam length of 600m. By using the monitoring data of the gas flow and concentration of the coal seam pre-drainage boreholes, combined with the inversion model and the prediction model, the following refined visualization of the gas pressure during coal seam extraction can be obtained.

[0138] From the front and back Fig. 9 It can be seen that the coal seam gas pressure calculation method of the present invention is more refined and can realize dynamic display, which can better guide the regulation of the coal seam gas pre-extraction system; for example:

[0139] 1. It can reduce the negative pressure of the extraction boreholes in gas-bearing areas and increase the extraction negative pressure of the extraction boreholes in high gas-bearing areas, thereby reducing the pre-extraction time for the overall elimination of coal seams.

[0140] 2. It can better estimate the specific time when coal seam gas pre-extraction meets the standards.

[0141] 3. It can better give a specific evaluation of the extraction effect of a single borehole.

Claims

1. A method for dynamic visualization of basic parameters of coal seam gas based on monitoring data calculation, characterized in that: The steps include: (1) Through the gas state equation, mass conservation law, coalbed methane flow continuity equation and coalbed methane motion equation, the mixed unsteady flow characteristics of coalbed methane seepage-diffusion in the "pore-fracture" dual medium of coal are obtained, the coalbed gas flow field control equation is established, the boundary conditions and initial conditions of the equation are determined, the analytical solution and numerical solution of the equation are obtained, and the relationship between the coalbed methane extraction flow rate and the coalbed methane pressure in the coal seam is established; thus, a fluid-solid coupling model of gas-containing coal is constructed; (2) Collect relevant parameters of the target coal seam; (3) Based on the target coal seam, determine the specific dimensions of the length, width, and height of the geometric model, boundary conditions, and initial conditions; (4) Collecting the extraction monitoring data of the target coal seam pre-extraction borehole; (5) The calculated in-situ gas pressure of the coal seam is used to draw a contour map of the gas pressure through SURFER software, thereby realizing the visualization of the gas pressure of the coal seam; (6) The key parameters in the gas extraction process are predicted through deep learning algorithms, and the dynamic visualization of coal seam gas pressure is realized by combining the in-situ gas pressure calculation method of coal seams; The fluid-solid coupling model of gas-containing coal in step (1) is specifically: The definitions of the parameters are as follows: u i，j Where i, j = 1, 2, 3 - displacement vector of the particle ε ij ----Strain component K m ----Bulk modulus of coal matrix, Pa β M ----Wet expansion coefficient, m 3 / m 3 α sg ----Volume strain coefficient caused by adsorption and desorption, kg·m -3 V sg ----Adsorbed gas content, m 3 kg -1 c1----pressure coefficient, MPa -1 c2----temperature coefficient, K -1 V L ---- is the Langmuir volume constant, m 3 kg -1 p L ----Langmuir pressure constant, Pa T——Coal seam temperature, K T t ----Reference temperature in adsorption and desorption test, K, take 298E----Elastic modulus, Pa υ----Poisson's ratio λ,G----Lame constant φ m0 ----Initial porosity ε m ----Coal matrix volume strain K Y ----Volume compression coefficient, MPa -1 K s ----Skeleton bulk modulus φ m ----Matrix porosity φ f ----Matrix crack rate k m ----Matrix pore permeability k f ----Matrix fracture permeability p m ----Matrix gas pressure, Pa p f ----Fissure gas pressure, Pa τ----gas desorption and diffusion time, s M g ----Molar mass of gas, kg / mol V L ----Langmuir volume constant, m 3 / kg P L ----Langmuir pressure constant, Pa ρ s ----Skeleton density, kg / m 3 Among them, Formula 1 is the stress field control equation of gas-containing coal, which is the constitutive equation of coal body stress deformation; Formula 2 is the gas matrix porosity control equation, Formula 3 is the pore permeability control equation, Formula 4 is the gas fissure rate control equation, and Formula 5 is the fissure permeability control equation; Formula 6 is the diffusion control equation of gas in the matrix, and Formula 7 is the seepage control equation of gas in the fissure. These two equations express the seepage and diffusion motion equations of gas in the coal seam; the seven equations together constitute the fluid-solid coupling model of gas-containing coal.

2. The method for dynamic visualization of basic parameters of coal seam gas based on monitoring data calculation according to claim 1 is characterized in that: The step (2) is specifically: Collect the parameters in the fluid-solid coupling model of gas-containing coal: According to coal safety regulations, coal mines will conduct relevant experiments and on-site tests before coal mining and gas extraction, and will issue corresponding reports; all of the above parameters can be collected by collecting information from the coal mine's "Coal Seam Coal and Gas Outburst Hazard Assessment Report", "Mine Gas Emission Prediction Method", "Coal Mine Gas Extraction Engineering Design Standard", "Coal Mine Safety Facility Design" and "Coal Seam Gas Parameter Basic Parameter Test Report".

3. The method for dynamic visualization of basic parameters of coal seam gas based on monitoring data calculation according to claim 1 is characterized in that: The step (3) is specifically: Bring boundary conditions and initial conditions into the fluid-solid coupling model of gas-containing coal, and use computational fluid dynamics software such as COMSOL or FLUENT to solve the control equations; By solving the coupling model of Formula 1 with numerical calculation software, the data required by Formulas 2 and 3 can be obtained. Formulas 2 and 3 can be obtained by fitting the data, thereby calculating the change law of the gas pressure at the distance x from the coal seam to the coal seam extraction borehole; the functional relationship between the coal seam gas pressure and the borehole flow rate and the attenuation law of the borehole flow rate along the borehole length are fitted through the change law data diagram, that is: Q L =f(L)(2) P x =f x (Q L ),in Where P x ----In-situ gas pressure of the coal seam at distance x from the borehole, Q L ----Monitor the flow rate of the gas pre-extraction borehole at the orifice L along the long distance of the hole, L----the distance from the orifice along the long distance of the hole.

4. The method for dynamic visualization of basic parameters of coal seam gas based on monitoring data calculation according to claim 3 is characterized in that: The gas borehole flow rate in the collected gas data of a single borehole is substituted into formulas (2) and (3) to obtain the gridded values ​​of the in-situ residual gas pressure of the target coal seam during the extraction process. By substituting each value into the SURFER software, the visualization effect of the target coal seam gas during the extraction process can be achieved.

5. The method for dynamic visualization of basic parameters of coal seam gas based on monitoring data calculation according to claim 1 is characterized in that: The step (4) is specifically: That is, the borehole extraction condition flow rate, gas concentration, pure gas flow rate, and extraction negative pressure; the data collected from each borehole is preprocessed, and the data of each borehole is used to calculate the in-situ gas pressure of the coal seam at the total distance x from the borehole using the inversion formula fitted in step (3).

6. The method for dynamic visualization of coal seam gas basic parameters based on monitoring data calculation according to claim 3 is characterized in that: The GRU model of the deep learning algorithm is used to train and predict the coal seam gas monitoring data. The trained prediction model is combined with the coal seam in-situ residual gas pressure calculation model, and the static visualization model can be converted into a dynamic visualization model.

7. A coal seam gas comprehensive parameter measurement system according to any one of claims 1 to 6, characterized in that: Including drainage system, gas drainage negative pressure, flow, gas concentration, temperature sensor and industrial ring network and data server; The pumping system is connected to sensors, the sensors are connected to an industrial ring network, and the industrial ring network transmits received data to a data server.

8. A coal seam gas comprehensive parameter measurement system according to claim 7, characterized in that: By adjusting the single borehole switch valve on the pumping system, the data of the single pre-draining borehole can be monitored online; The extraction system is connected to the coal seam gas pre-extraction borehole. The single borehole is switched on and off by adjusting the switch valve of the single borehole on the extraction system. When the extraction system and the sensor need to measure the gas data in a certain borehole, the switch valve is used to shut down other boreholes and the borehole that needs to be measured is retained to achieve the measurement of a single borehole. The measured data is transmitted between the industrial ring network and finally saved to the data server.

9. A coal seam gas comprehensive parameter measurement system according to claim 8, characterized in that: The drilled pipe is connected to the extraction pump, which is then connected to the gas extraction negative pressure, flow, gas concentration and temperature sensors.

Images