Coal face gas concentration dynamic distribution model determination method

Abstract

The invention relates to a coal face gas concentration dynamic distribution model determination method, and belongs to the technical field of coal mine safety monitoring. The technical problems that traditional single-point static monitoring cannot reflect working face gas dynamic distribution and prediction precision is low are solved. According to the technical scheme, the method comprises the steps of deploying air inlet corner, working face middle and air return corner sensor networks to collect data in real time; a gas distribution function model y = ae (1-x) is built, a is the return air corner concentration value, and b is the dynamic attenuation coefficient; fitting measured data through a least square method to solve parameters; and setting an update period t to dynamically correct the model. The technical effects are that working face gas concentration full-space dynamic mapping is realized, ventilation condition changes are responded in real time, disaster early warning capability is improved in combination with a visual interface, gas explosion risk is obviously reduced, and safe production of a coal mine is guaranteed.

Description

Technical Field

[0001] The invention belongs to the technical field of coal mine safety monitoring and relates to a method for determining a dynamic distribution model of gas concentration in a coal mining working face. Background Art

[0002] During coal mining, gas, a major hazardous gas, has a concentration distribution that directly impacts the safety of workers and the smooth operation of production. Coal faces, key areas for gas outbursts, are subject to highly dynamic changes in gas concentrations at different locations, influenced by a variety of complex factors, including coal seam conditions, mining processes, and ventilation system operation. Currently, the most commonly used gas concentration monitoring methods in the industry are single-point static monitoring, which only captures instantaneous gas concentration values ​​at specific locations and fails to fully reflect the overall distribution and dynamic evolution of gas concentration at the working face.

[0003] While some attempts have been made based on point-by-point field measurements or numerical simulations, the resulting gas concentration distribution models are static and difficult to adapt to the dynamic interactions of multiple factors, such as ventilation speed, airflow turbulence, and coal seam permeability. Consequently, such models are unable to dynamically predict the spatial distribution of gas concentration. This has led to coal mining companies remaining in a passive state of response to gas prevention and control, preventing them from formulating effective prevention and control measures in advance. This has led to frequent safety incidents such as gas explosions and gas over-limit, seriously threatening coal mine production safety.

[0004] Therefore, the industry urgently needs to develop a method that can accurately grasp the dynamic distribution of gas concentration in coal mining faces in real time. Through high-precision data collection and intelligent modeling technology, it can break through the limitations of traditional monitoring methods and provide a reliable basis for ventilation optimization, disaster warning and safe production decision-making. Summary of the Invention

[0005] In view of this, an object of the present invention is to provide a method for determining a dynamic distribution model of gas concentration in a coal mining face.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] A method for determining a dynamic distribution model of gas concentration in a coal mining face comprises the following steps:

[0008] S1: Deploy at least three gas concentration sensors in the air inlet area, upper corner, and middle of the coal mining face, and collect gas concentration data at each location in real time within a time period of t to form a monitoring equipment location subset X = (x1, x2, ..., x n ) and gas concentration data set P = (P1, P2, ..., P n ), where x1 to x n Indicates the distance between the sensor and the air inlet channel and satisfies x1 <x2<…<x n, P i =[(t i1 ,c i1 ),(t i2 ,c i2 ),…,(t im ,c im )] T represents the concentration sequence of the i-th sensor, c im is time t im The concentration value of

[0009] S2: Process the data set P and calculate the average concentration at each sensor location Generate gas concentration data set W=(u1,u2,...,u n ), and merge it with the position subset X to obtain the basic data set Z;

[0010] S3: Set the working face gas concentration distribution function to an exponential model:

[0011] y=ae b(l-x)

[0012] Wherein, y represents the gas concentration at position x, in %; a represents the gas concentration value at the return air corner, in %; b is an undetermined coefficient and satisfies b<0; l is the inclined length of the working face, in m;

[0013] S4: Based on the basic data set Z, the measured data are fitted with the distribution function using the least squares method to solve the unknown coefficients a and b;

[0014] S5: Substitute the coefficients a, b and parameter l into the distribution function to construct a static gas concentration distribution model;

[0015] S6: Set the update cycle t and repeat S1 to S4 to dynamically update the coefficients a and b to achieve real-time correction of the gas concentration distribution model.

[0016] Furthermore, the sensor deployment positions in S1 include the wind inlet corner, the middle of the working surface and the return air corner, and the middle sensor is located within ±25% of the geometric center of the inclined length of the working surface.

[0017] Furthermore, in S4, the least squares fitting process includes an abnormal data processing mechanism: when the concentration value exceeds a preset fluctuation threshold, data correction is performed by interpolation, trend extrapolation or manual correction.

[0018] Furthermore, the abnormal data processing mechanism specifically includes: comparing adjacent time periods, adjacent location data and historical mining data, and making correction decisions based on changes in geological structure and ventilation status.

[0019] Furthermore, in S6, when updating coefficients a and b, a deep learning model is used to predict the trend of gas concentration changes and dynamically adjust the data collection frequency.

[0020] Furthermore, it also includes:

[0021] S7: Build a visual interactive interface to render the dynamic gas concentration distribution model in real time as two-dimensional / three-dimensional graphics, where different color areas represent different concentration levels, and support interface zooming, location query and historical data backtracking functions.

[0022] Furthermore, the visual interactive interface realizes data visualization through a large monitoring screen or a mobile application (Application, APP), and the color grading threshold is dynamically set according to the coal mine safety regulations.

[0023] A system for determining a dynamic distribution model of gas concentration in a coal mining face, used to implement the method, comprises:

[0024] Sensor network module: consists of at least three anti-interference infrared / laser gas sensors deployed at the air inlet corner, the middle of the working face, and the return air corner;

[0025] Data transmission module: uses wireless transmission device to send sensor data to the monitoring center in real time;

[0026] Data processing module: performs data mean calculation, anomaly correction and least squares fitting operations;

[0027] Dynamic modeling module: reconstructs the distribution function coefficients according to the update period t;

[0028] Visual output module: Generates spatial distribution map of gas concentration and early warning information of the working face.

[0029] The beneficial effects of the present invention are:

[0030] (1) Based on the coordinated monitoring data of the air inlet corner, the middle of the working face and the return air corner, a continuous distribution model of the working face gas concentration is constructed through the exponential distribution function. Only a limited number of monitoring points are needed to dynamically restore the gas concentration at any position in the inclination direction, which completely changes the fragmented cognitive mode of single-point static monitoring and realizes the overall control of the gas distribution law in two-dimensional space.

[0031] (2) By periodically updating the distribution function coefficients, the model can respond in real time to changes in ventilation conditions, mining disturbances, and geological structure changes, ensuring that the model continuously adapts to the actual working conditions of the working face. This adaptive dynamic modeling mechanism effectively overcomes the lag inherent in traditional static models and provides timely and reliable data support for disaster warnings.

[0032] (3) In the process of parameter solution, the abnormal data identification and correction mechanism is integrated, and the decision-making is assisted by multi-dimensional data cross-validation and expert experience, which significantly improves the model's anti-interference ability and output stability, ensuring that the gas concentration prediction results are not affected by local instantaneous fluctuations.

[0033] (4) The dynamic model is converted into an intuitive two-dimensional or three-dimensional visualization interface, which displays a real-time gas concentration distribution heat map through color grading to assist management personnel in quickly identifying high-risk areas. Combined with historical data backtracking and location precision query functions, it provides a scientific basis for ventilation system tuning and emergency evacuation route planning.

[0034] (5) Fundamentally reverse the passive situation of responding to excessive gas, promote coal mine safety management from post-processing to pre-prevention by predicting the evolution trend of concentration distribution in advance, significantly reduce the risk of gas explosion accidents, and ensure the safe and efficient operation of the coal industry.

[0035] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings, in which:

[0037] Figure 1 This is a schematic diagram of the overall process of the method for determining the dynamic distribution model of gas concentration in a coal mining face according to the present invention;

[0038] Figure 2 This is the sensor layout diagram;

[0039] Figure 3 The distribution law of gas concentration along the working face of 3205 working face is shown;

[0040] Figure 4 This is the distribution pattern of gas concentration along the working face of 1206 working face. DETAILED DESCRIPTION

[0041] The following describes the embodiments of the present invention by means of specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0042] Among them, the accompanying drawings are only for illustrative purposes and represent only schematic diagrams rather than actual pictures, and should not be understood as limiting the present invention. In order to better illustrate the embodiments of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the dimensions of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions may be omitted in the accompanying drawings.

[0043] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", "front", "back", etc. indicating directions or positional relationships, they are based on the directions or positional relationships shown in the drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operate in a specific direction. Therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0044] Figure 1 This is a schematic diagram of the overall process of the method for determining the dynamic distribution model of gas concentration in a coal mining face according to the present invention;

[0045] Figure 2 Figure 1 is the sensor layout diagram; T1 is the gas sensor at the air inlet corner, T2 is the gas sensor in the middle of the working face, and T3 is the gas sensor at the return air corner.

[0046] Figure 3 The distribution law of gas concentration along the working face of 3205 working face is shown;

[0047] Figure 4 This is the distribution pattern of gas concentration along the working face of 1206 working face.

[0048] A method for determining a dynamic distribution model of gas concentration in a coal mining face comprises the following steps:

[0049] S1: Using gas concentration monitoring equipment, continuously collect methane concentration data at the air inlet area, upper corner, and middle of the coal mining face to obtain a gas concentration dataset P at different locations during time period t.

[0050] S2: Process the measured data set P of gas concentration at different locations within time t obtained based on S1 to obtain the basic data set Z of the gas concentration distribution model.

[0051] S3: Set the working face gas concentration distribution function expression to y = ae b(l-x) , where y is the gas concentration at different positions on the working face, in %; a represents the gas concentration at the return air corner, in %, which is the maximum value in the gas concentration distribution described by the function, reflecting the peak concentration characteristic of the upper corner as a key area prone to gas accumulation; b is an undetermined coefficient, and b is strictly limited to b<0. This coefficient is related to the combined effects of gas migration, diffusion and ventilation flow field on the working face, and determines the changing trend of the function curve; l is the inclined length of the working face, in m, which is determined based on the actual geological mining design parameters of the working face;

[0052] S4: Based on the basic data set Z of the gas concentration distribution model at different locations on the working face obtained in S2, the least squares method is used to match the measured gas concentration data set Z with the distribution function set in S3 to solve the unknown coefficients a and b in the gas distribution fitting function;

[0053] S5: Substitute the undetermined coefficients a and b and the working face length l obtained by S4 into the working face gas distribution fitting function set in S3. Based on the working face position variable x, calculate the predicted value of the gas concentration at the corresponding position to achieve static preliminary modeling of the working face gas concentration distribution.

[0054] S6: Set the working face gas concentration distribution model update cycle to t. As the mining process progresses, the working face ventilation conditions and coal seam gas emission volume change dynamically. Repeat S1 to S3 continuously, and update the methane concentration monitoring data at different locations of the working face in real time. At intervals of t, substitute the data into the data fitting algorithm in S4 again and re-solve the undetermined coefficients. This allows the undetermined coefficients to track and reflect changes in the working face gas environment in real time, ensuring that the model closely matches the actual working conditions.

[0055] S7: Leveraging the dynamic update mechanism for the undetermined coefficients in S6 and combining the function calculation results from S5, a dynamic working face methane concentration distribution model is constructed. This model can quickly and in real time visualize the dynamic distribution of gas concentration at different locations and during different mining periods within the coal face, providing a key decision-making basis for gas disaster warnings and ventilation control.

[0056] At least three gas concentration monitoring devices are deployed at the air inlet area, upper corner and middle of the coal mining face; the location subset of the monitoring devices is recorded as X = (x1, x2, ..., x n ), where x1 is the distance between the first monitoring device and the air inlet tunnel, x2 is the distance between the second monitoring device and the air inlet tunnel, and x n is the distance between the nth monitoring device and the air inlet tunnel, n is the number of installed sensors, and x1 <x2<…<x n ;

[0057] By monitoring the oxygen concentration data, a set of gas concentration monitoring data sets P = (P1, P2, ..., P n ), where P1 is the gas concentration sequence subset of the first device; P2 is the gas concentration sequence subset of the second device; P n is the subset of gas concentration sequence of the nth equipment. Where, P1=[(t 11 ,c 11 ),(t 12 ,c 12 ),…,(t 1m ,c 1m )] T ; P2=[(t 21 ,c 21 ),(t 22 ,c 22 ),…,(t 2n ,c 2n )] T

[0058] ;P n =[(t n1 ,c n1 ),(t n2 ,c n2 ),...,(t nm ,c nm )] T , where m is the number of gas concentration samples in time period t, t nm is the monitoring time of the mth gas concentration value of the nth device, c nm is the monitoring time t nm The corresponding gas concentration value.

[0059] The processing formula of the gas concentration monitoring data set P in S2 is as follows: n ).

[0060]

[0061] Where, i = 1, 2, , n;

[0062] The method for obtaining Z in S2 is to combine the gas concentration data set W with the location subset X of the monitoring equipment to obtain the basic data set Z of the gas concentration distribution model;

[0063] Taking the 3205 working face of a coal mine as an example, the specific process of applying the present invention in the working face is introduced.

[0064] S1: Accurate data collection. Use advanced infrared gas sensors or laser gas sensors with strong anti-interference ability and high sensitivity. According to the working face ventilation system structure and stope layout, scientific layout is made for the working face air inlet, upper corner and middle positions. Gas concentration sensors are installed at key positions such as the air inlet tunnel entrance, upper corner and middle of the working face (110# frame). Figure 2 As shown, the monitoring equipment location set X = (0, 140, 200) is obtained. The sensor installation locations are carefully designed to avoid ventilation blind spots and interference sources to ensure representative data. The sensors also feature automatic calibration, which regularly corrects measurement deviations, and fault self-detection, which immediately issues an alarm and uploads fault information upon failure, ensuring continuous data collection. Collected gas concentration data is transmitted to the ground monitoring center in real time via a stable and efficient wireless transmission module and stored in a specially constructed high-capacity, high-reliability database, providing ample data resources for subsequent model construction and analysis. The collection frequency is not fixed but dynamically adjusted based on the progress of the working face. For example, during the initial mining phase or when the ventilation system is being adjusted, the collection frequency is appropriately increased to capture subtle changes in gas concentration. During periods of relative stability, the frequency is appropriately reduced based on historical data fluctuations, conserving resources while ensuring data accuracy.

[0065] S2: Start the data acquisition system and collect data every minute according to the preset initial collection frequency. The collected data will be transmitted to the ground monitoring center in real time and stored. After t = 10 minutes of data accumulation, a set of gas concentration monitoring data sets P = (P1, P2, P3) is formed, where P1 is the gas concentration sequence subset of T1 equipment; P2 is the gas concentration sequence subset of T2 equipment; P3 is the gas concentration sequence subset of T3 equipment.

[0066] P1=[(0,0.04),(1,0.02),(2,0.03),(3,0.04),(4,0.05),(5,0.05),(6,0.04),(7,0.06),(8,0.04),(9,0.03),(10,0

[0067] .03)] T ;

[0068] P2=[(0,0.08),(1,0.07),(2,0.05),(3,0.04),(4,0.05),(5,0.05),(6,0.06),(7,0.06),(8,0.06),(9,0.07),(10,0

[0069] .06)] T ;

[0070] P3=[(0,0.41),(1,0.41),(2,0.43),(3,0.42),(4,0.4),(5,0.39),(6,0.4),(7,0.38),(8,0.39),(9,0.38),(10,0.3

[0071] 7)] T ;

[0072] The gas concentration monitoring data set P is processed using the following formula to obtain the gas concentration data set W = (0.04, 0.06, 0.4);

[0073]

[0074] Where, i = 1, 2, , n;

[0075] Merge the gas concentration data set W with the location subset X of the monitoring equipment to obtain the basic data set Z of the gas concentration distribution model = [(0, 0.04), (140, 0.06), (200, 0.4)];

[0076] S3: Set the working face gas concentration distribution function to y = ae b(l-x) . Among them, y is the gas concentration at different positions of the working face, %; a is the gas concentration at the return air corner, which is the maximum value in the gas concentration distribution represented by this function. This is based on the key position of the upper corner in the gas accumulation characteristics of the working face. Usually, gas tends to reach the highest concentration here. b is an undetermined coefficient, and b is limited to b<0. It comprehensively reflects the diffusion and attenuation characteristics of gas in the working face under the action of wind flow, and is closely related to many factors such as ventilation wind speed, wind flow turbulence, and coal seam permeability. l accurately corresponds to the working face dip length based on the actual geological mining design of the working face, and the unit is meter. This parameter provides a benchmark scale for the subsequent calculation of gas concentration based on position. Since the dip length of the 3205 working face is 200m, l = 200m. This function builds a basic framework for the distribution of gas concentration along the working face dip from a theoretical mathematical level, laying the foundation for the subsequent data-driven solution of the undetermined coefficients and fitting of the actual distribution law.

[0077] S4: Solve for the undetermined coefficients. Based on the basic data set Z of the gas concentration distribution model obtained in S2, a specialized data fitting algorithm, such as the high-precision least squares method or a machine learning-based regression algorithm, is applied. When using the least squares method, the basic data set Z of the gas concentration distribution model is rigorously matched with the distribution function set in S3. A sum-of-squares error function is constructed and minimized to determine the undetermined coefficient b. When using a machine learning-based regression algorithm, a pre-trained model is used as input, along with the measured data and corresponding location information, to output an estimate of the undetermined coefficient. During this process, if abnormal fluctuations in the measured gas concentration data are detected, exceeding the preset normal gas concentration fluctuation range, an abnormal data identification and correction mechanism is immediately activated. By comparing data from adjacent time periods and locations with historical mining data from the same period, and taking into account factors such as changes in the working face's geological structure and the operating status of ventilation facilities, interpolation, trend extrapolation, or manual correction based on expert experience are used to identify and appropriately correct abnormal data. This ensures the reliability of the data involved in the fitting calculation. After multiple iterations of optimization and solution, an accurate undetermined coefficient is ultimately obtained.

[0078] By using a data fitting algorithm based on the least squares method and combining it with the gas concentration sequence W, the unknown coefficients a=0.4 and b=-0.0173 in the distribution function were solved. During the solution process, it was found that some data had abnormal fluctuations due to temporary support operations in the stope. By comparing with data from adjacent time periods and locations and referring to geological structure information, the abnormal data were corrected using the interpolation method, and finally more accurate coefficient values ​​of a=0.4 and b=-0.0175 were obtained.

[0079] S5: Substitute the coefficients to improve the model. Substitute the undetermined coefficients obtained by S4 into the working face gas distribution fitting function set in S3, and preliminarily build the static model of gas concentration in the 3205 working face as follows:

[0080] y=0.4e -0.0175(200-x)

[0081] Where: y is the gas concentration at different positions of the working face, in %, and x is the distance from the air inlet tunnel along the working face, in m. According to the gas concentration distribution model, the gas concentration distribution law at different positions along the working face is drawn as follows: Figure 3 As shown. Figure 3 It can be seen that gas concentration gradually increases along the working face from the air intake corner to the return air corner, with the gas concentration increasing more rapidly as it approaches the upper corner. By constructing a static gas concentration model, based solely on measured gas concentration values ​​at a limited number of key nodes, it is possible to efficiently calculate gas concentration values ​​at any location along the entire working face's inclination, achieving comprehensive, multi-level, and precise control of gas concentration. This model initially achieves a quantitative description of the gas concentration distribution in the working face, providing a core computational module for the subsequent construction of dynamic models.

[0082] S6: Dynamically update the model. As mining progresses, ventilation conditions at the working face change due to wear and tear of ventilation facilities and adjustments, and coal seam gas outflow changes dynamically due to mining disturbances and geological structural changes. At this point, S1-S2 are continuously repeated, updating the methane concentration monitoring data at the working face's air intake, upper corners, and center in real time. This data is then fed back into the data fitting algorithm in S3 at a fixed interval, t, to ensure that the dynamic model promptly reflects changes in gas concentration at the working face.

[0083] To quickly and accurately update the undetermined coefficients, a deep learning-based gas concentration prediction model is employed as an aid. This model, pre-trained using extensive historical gas concentration data and working face mining parameter data, intelligently predicts trends in gas concentration within the working face. When rapid changes in working face gas concentration are detected, the monitoring frequency is adjusted in advance to proactively collect more data for updating the undetermined coefficients. The deep learning model's predictions are then integrated and verified with those based on distribution function fitting, further enhancing the reliability and adaptability of the dynamic model and ensuring that it consistently closely matches actual gas concentration fluctuations within the working face.

[0084] Finally, coal mine managers and front-line workers access the visual interactive interface through the monitoring screen and mobile APP, view the gas concentration distribution of the working face in real time, intuitively judge the gas concentration level based on different color areas, and use query and backtracking functions to provide a basis for ventilation control and safe production decision-making. In the subsequent mining process of the working face, gas over-limit accidents were effectively avoided, ensuring safe and efficient production.

[0085] Example 1: Basic model construction and dynamic update

[0086] Scenario: 1206 fully mechanized mining face in a coal mine, with a dip length of 220 meters

[0087] 1. Sensor deployment: Install laser gas sensors at the air inlet corner, 0 meters away from the air inlet tunnel, 110 meters from the middle of the working face, and 220 meters from the return air corner;

[0088] 2. Data collection: Concentration data is collected every 1 minute for 10 minutes. The average concentration in the inlet corner is 0.03%, in the middle is 0.08%, and in the return corner is 0.65%;

[0089] 3. Model construction:

[0090] Set the distribution function: y = ae b(l-x) ;

[0091] The least squares fitting results are: a = 0.65, b = -0.0185;

[0092] Generate static model: y = 0.65e-0.0185(220-x) ;

[0093] 4. Dynamic Updates:

[0094] Set the update cycle t = 30 minutes;

[0095] In the next cycle, the return air corner concentration rises to 0.69%, and the refitting results are a = 0.69, b = -0.0191;

[0096] The model is corrected in real time to y = 0.69e -0.0191(220-x) The gas concentration distribution of the working face after dynamic update can be obtained as follows: Figure 4 shown.

[0097] Example 2: Abnormal data processing and visual warning

[0098] Scenario: Abnormal concentration fluctuations in the middle of the work surface during the rack transfer process

[0099] 1. Anomaly Identification:

[0100] The concentration of the middle sensor suddenly increased from 0.06% to 0.15% within 5 minutes;

[0101] The system compares the data of adjacent time periods with the historical data of the same period and determines whether the data exceeds the preset fluctuation threshold of ±50%;

[0102] 2. Data correction:

[0103] Using the trend extrapolation method: the normal value is predicted to be 0.065% based on the slope of the data in the first 3 minutes;

[0104] Manually confirm dust interference caused by stope support adjustment and adopt the correction value;

[0105] 3. Visual linkage:

[0106] The modified model generates a thermal map of gas concentration in the working face;

[0107] The return air corner area displays a red warning (concentration ≥ 0.8%);

[0108] The mobile app pushes over-limit alarms to management personnel.

[0109] Example 3: Dynamic Regulation of Multi-Factor Coupling

[0110] Scenario: Accelerated mining progress leads to a sharp increase in gas outbursts

[0111] 1. Adaptive acquisition:

[0112] Deep learning models analyze historical data to predict gas outburst trends;

[0113] Automatically increase the collection frequency to once every 30 seconds;

[0114] 2. Cross-system collaboration:

[0115] Real-time acquisition of ventilation system wind speed data (reduced from 2.5m / s to 2.0m / s);

[0116] The refitting coefficient b of the integrated wind speed change changes from -0.0175 to -0.019;

[0117] 3. Active prevention and control:

[0118] The model shows that the gas concentration in the area from rack 136# to the upper corner of the return air is over 0.8%;

[0119] Automatically trigger linkage instructions: increase ventilation volume to 110% of the rated value; cut off power supply to the exceeded area; activate the sound and light alarm device.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for determining a dynamic distribution model of gas concentration in a coal mining face, characterized by: The following steps are involved: S1: Deploy at least three gas concentration sensors in the air inlet area, upper corner, and middle of the coal mining face, and collect gas concentration data at each location in real time within a time period of t to form a monitoring equipment location subset X = (x1, x2, ..., x n ) and gas concentration data set P = (P1, P2, ..., P n ), where x1 to x n Indicates the distance between the sensor and the air inlet channel and satisfies x1 <x2<…<x n , P i =[(t i1 ,c i1 ),(t i2 ,c i2 ),...,(t im ,c im )] T represents the concentration sequence of the i-th sensor, c im is time t im The concentration value of S2: Process the data set P and calculate the average concentration at each sensor location Generate gas concentration data set W=(u1,u2,...,u n ), and merge it with the position subset X to obtain the basic data set Z; S3: Set the working face gas concentration distribution function to an exponential model: y=ae b(l-x) Wherein, y represents the gas concentration at position x, in %; a represents the gas concentration value at the return air corner, in %; b is an undetermined coefficient and satisfies b<0; l is the inclined length of the working face, in m; S4: Based on the basic data set Z, the measured data are fitted with the distribution function using the least squares method to solve the unknown coefficients a and b; S5: Substitute the coefficients a, b and parameter l into the distribution function to construct a static gas concentration distribution model; S6: Set the update cycle t and repeat S1 to S4 to dynamically update the coefficients a and b to achieve real-time correction of the gas concentration distribution model.

2. The method for determining a dynamic distribution model of gas concentration in a coal mining face according to claim 1, wherein: In S1, the sensor deployment positions include the wind inlet corner, the middle of the working surface and the return air corner, and the middle sensor is located within ±25% of the geometric center of the inclined length of the working surface.

3. The method for determining a dynamic distribution model of gas concentration in a coal mining face according to claim 1, wherein: In S4, the least squares fitting process includes an abnormal data processing mechanism: when the concentration value exceeds a preset fluctuation threshold, data correction is performed through interpolation, trend extrapolation or manual correction.

4. The method for determining a dynamic distribution model of gas concentration in a coal mining face according to claim 3, wherein: The abnormal data processing mechanism specifically includes: comparing adjacent time periods, adjacent location data and historical mining data, and making correction decisions based on geological structure changes and ventilation facility status.

5. The method for determining a dynamic distribution model of gas concentration in a coal mining face according to claim 1, wherein: In S6, when updating coefficients a and b, a deep learning model is used to predict the trend of gas concentration changes and dynamically adjust the data collection frequency.

6. The method for determining a dynamic distribution model of gas concentration in a coal mining face according to claim 1, wherein: Also includes: S7: Build a visual interactive interface to render the dynamic gas concentration distribution model in real time as two-dimensional / three-dimensional graphics, where different color areas represent concentration levels, and support interface zooming, location query and historical data backtracking functions.

7. The method for determining a dynamic distribution model of gas concentration in a coal mining face according to claim 6, characterized in that: The visual interactive interface realizes data visualization through a large monitoring screen or a mobile application APP, and the color grading threshold is dynamically set according to the coal mine safety regulations.

8. A system for determining a dynamic distribution model of gas concentration in a coal mining face, for implementing the method according to any one of claims 1 to 7, characterized in that: include: Sensor network module: consists of at least three anti-interference infrared / laser gas sensors deployed at the air inlet corner, the middle of the working face, and the return air corner; Data transmission module: uses wireless transmission device to send sensor data to the monitoring center in real time; Data processing module: performs data mean calculation, anomaly correction and least squares fitting operations; Dynamic modeling module: reconstructs the distribution function coefficients according to the update period t; Visual output module: Generates spatial distribution map of gas concentration and early warning information of the working face.

Images