A coal mine underground gas leakage source positioning system and method

Abstract

The present application provides a kind of coal mine underground gas leakage source positioning system and method, comprising: using multiple gas sensors set in the coal mine underground area to be monitored to collect gas concentration, when detecting that gas concentration exceeds the set threshold, the collected gas concentration is input to revise Gaussian plume model, Gaussian plume model is solved, the position information of the maximum gas concentration is obtained, that is, the position of gas leakage source;Wherein, the revised Gaussian plume model is obtained by revising Gaussian plume model using the height and width of the coal mine underground area to be monitored and the reflection coefficient when gas collides with the side wall, roof and ground of the coal mine underground area to be monitored.The size of limited space and the collision factor of gas and wall are introduced into Gaussian plume model, so that a new revised Gaussian plume model is obtained, and the revised Gaussian plume model obtained after revision can be used for gas leakage source positioning in limited space.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of coal mine gas detection, and particularly relates to a coal mine underground gas leakage source positioning system and method. BACKGROUND

[0002] With the development of the semiconductor industry and the application of artificial intelligence, Internet of Things and other technologies, high-precision sensors combined with artificial intelligence algorithms are increasingly widely used in industry. By integrating disciplines such as artificial intelligence and sensor detection technology, research on gas mobile detection can effectively improve the level of gas management and ensure the safety of coal mine production. The positioning of leakage sources in existing literature is mainly focused on open environments, such as natural gas pipeline leakage positioning. In the positioning of open leakage sources, the Gaussian plume model is commonly used to describe the gas diffusion process.

[0003] The Gaussian plume model is a gas diffusion model based on turbulent flow statistics theory, and is generally used for continuous point source gas diffusion prediction. The establishment of the Gaussian plume model requires that: a. the size and direction of the wind speed are constant in time and space; b. the intensity of the leakage source is consistent during gas release; c. the effect of gravity is ignored; and d. the diffusion area is fixed. The Gaussian plume model is as follows:

[0004]

[0005] C(x,y,z,H) is the gas concentration, mg / m 3 ; Q is the gas leakage rate, mg / s; H is the height of the leakage source; μ is the wind speed; σ y and σ z are the gas diffusion coefficients in the y and z axis directions, respectively.

[0006] However, the Gaussian plume model cannot be used for finite space gas diffusion simulation calculation, and therefore cannot be applied to the positioning of gas leakage sources in coal mine underground finite spaces. SUMMARY

[0007] To solve the problem of positioning gas leakage sources in coal mine underground finite spaces, the present application provides a coal mine underground gas leakage source positioning system and method.

[0008] The present application is implemented by the following technical solutions:

[0009] A coal mine underground gas leakage source positioning method, comprising:

[0010] A plurality of gas sensors arranged in a coal mine underground area to be monitored are used to collect gas concentrations, and when the detected gas concentration exceeds a set threshold value, the collected gas concentration is input into a revised Gaussian plume model, the revised Gaussian plume model is solved, and the position information of the place with the maximum gas concentration is obtained, which is the position of the gas leakage source.

[0011] Wherein, the revised Gaussian plume model is obtained by revising the Gaussian plume model with the height and width of the underground coal mine area to be monitored and the reflection coefficient of the gas colliding with the side wall, roof and ground of the underground coal mine area to be monitored.

[0012] Preferably, the revised Gaussian plume model is:

[0013]

[0014] C(x, y, z) = C(x, y, z, H, L) + C(x, y, z, H, D) (2)

[0015] Wherein, C(x, y, z, H, L), C(x, y, z, H, D) and C(x, y, z) are all gas concentrations, mg / m 3 ; Q is the gas leakage speed, mg / s; H is the height of the gas leakage source; μ is the wind speed; σ y and σ z are the gas diffusion coefficients in the y-axis direction and z-axis direction respectively; α is the reflection coefficient of the gas colliding with the side wall, roof and ground of the underground coal mine area to be monitored; n is the total collision times of the gas with the side wall, roof and ground of the underground coal mine area to be monitored; L is the height of the underground coal mine area to be monitored, the height direction is consistent with the z-axis direction; D is the width of the underground coal mine area to be monitored, the width direction is consistent with the y-axis direction.

[0016] Preferably, the solving of the revised Gaussian plume model is specifically: taking the gas sensor as a particle and using a particle swarm optimization algorithm to solve.

[0017] Preferably, the solving of the revised Gaussian plume model obtains the position information of the maximum gas concentration, which is specifically:

[0018] The fitness function is constructed:

[0019]

[0020] Wherein, C i represents the gas concentration detected by the i th gas sensor, C d represents the gas concentration of the i th gas sensor calculated by the revised Gaussian plume model;

[0021] The revised Gaussian plume model is solved to obtain the optimal solution that minimizes the fitness function, and the coordinates corresponding to the optimal solution are the position information of the maximum gas concentration.

[0022] Preferably, the method for constructing the revised Gaussian plume model comprises:

[0023] 1) collecting gas concentration by using multiple gas sensors arranged in the region to be monitored in the coal mine;

[0024] 2) constructing the revised Gaussian plume model as follows:

[0025]

[0026] C(x, y, z) = C(x, y, z, H, L) + C(x, y, z, H, D) (2)

[0027] wherein C(x, y, z, H, L), C(x, y, z, H, D) and C(x, y, z) are all gas concentrations, mg / m 3 ; Q is the gas leakage speed, mg / s; H is the height of the gas leakage source; μ is the wind speed; σ y and σ z are the gas diffusion coefficients in the y-axis direction and the z-axis direction, respectively; α is the reflection coefficient of the gas when colliding with the side wall, the roof and the ground of the region to be monitored in the coal mine; n is the total collision times of the gas with the side wall, the roof and the ground of the region to be monitored in the coal mine; L is the height of the region to be monitored in the coal mine, the height direction being consistent with the z-axis direction; D is the width of the region to be monitored in the coal mine, the width direction being consistent with the y-axis direction;

[0028] 3) solving the model parameters α, σ z and σ y in the revised Gaussian plume model according to the collected gas concentration, and substituting the solved model parameters into the revised Gaussian plume model to obtain the revised Gaussian plume model.

[0029] Further, in step 3), the model parameters in the revised Gaussian plume model are solved according to the collected gas concentration, specifically:

[0030] constructing a fitness function:

[0031]

[0032] wherein C i represents the gas concentration detected by the i-th gas sensor, C d represents the gas concentration at the i-th gas sensor calculated by using the revised Gaussian plume model;

[0033] solving the revised Gaussian plume model to obtain α, σ z , σ y that make the fitness function minimum, i.e. to obtain the model parameters.

[0034] Further, in step 3), the Pasquill method is used to measure the atmospheric stability, the reference range of the gas diffusion coefficient is obtained according to the atmospheric stability, and the model parameters in the revised Gaussian plume model are solved according to the reference range of the gas diffusion coefficient and the collected gas concentration.

[0035] Further, the gas sensor is taken as a particle, and the particle swarm optimization algorithm is used to solve the model parameters in the revised Gaussian plume model.

[0036] Preferably, the coal mine underground area to be monitored is a coal mine tunnel.

[0037] A coal mine underground gas leakage source positioning system comprises a plurality of gas sensors arranged in a coal mine underground area to be monitored and a central processor.

[0038] The gas sensor is used to collect the gas concentration and transmit the gas concentration to the central processor.

[0039] The central processor comprises a revised Gaussian plume model, when the gas concentration detected exceeds a set threshold value, the received gas concentration is input into the revised Gaussian plume model, the revised Gaussian plume model is solved, the position information at the maximum gas concentration is obtained, and the gas leakage source position is obtained.

[0040] Compared with the prior art, the present application has the following beneficial effects:

[0041] The present application is used to solve the problem of positioning the gas leakage source in a limited space in a coal mine underground area, the existing Gaussian plume model is revised, the size of the limited space and the collision factors of the gas with the side wall, the roof and the ground are introduced into the Gaussian plume model, thereby obtaining a new revised Gaussian plume model, and the revised Gaussian plume model obtained after the revision can be used for positioning the gas leakage source in the limited space.

[0042] Further, the traditional multi-objective optimization method is used to convert the multi-objective problem into a single-objective problem by weighted summation, which has the problems of great difficulty in solving, low precision and difficult convergence. Compared with the traditional optimization algorithm, the particle swarm optimization algorithm used in the present application is easy to implement, has high precision and fast convergence speed.

[0043] The present application provides a system capable of positioning the gas leakage source, the coal mine underground gas can be monitored through simple gas sensor arrangement, and the gas leakage source can be quickly positioned when the gas leakage occurs. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 Module diagram of the gas detection and leak source location system of the present invention;

[0045] Figure 2 This invention relates to a diagram showing the arrangement of gas sensors in a U-shaped roadway of a coal mine.

[0046] Figure 3 The calculation process of the particle swarm optimization algorithm of this invention. Detailed Implementation

[0047] To further understand the present invention, the present invention will be described below with reference to embodiments. These descriptions are only for further explaining the features and advantages of the present invention and are not intended to limit the claims of the present invention.

[0048] like Figure 1 As shown, the underground gas leak source location system of the present invention consists of N gas sensors and 1 central processing unit. The gas sensors are responsible for detecting the gas concentration in the monitored area of ​​the underground coal mine, and the central processing unit performs data processing and calculation to obtain the gas leak source location information.

[0049] This invention discloses a method for locating gas leak sources in underground coal mines. First, a revised Gaussian plume model is constructed. Then, the model parameters of the revised Gaussian plume model are solved using an optimization algorithm to obtain a revised Gaussian plume model suitable for the area to be measured in the underground coal mine. Using the obtained revised Gaussian plume model, the coordinates of the point of maximum gas concentration are searched in the search space using another optimization algorithm to determine the location of the gas leak source. The preferred optimization algorithm is particle swarm optimization.

[0050] The present invention provides a method for locating underground gas leak sources in coal mines, which specifically includes the following steps:

[0051] (1) Constructing a revised Gaussian plume model

[0052] 1) Multiple gas sensors installed in the monitoring area underground in the coal mine are used to collect the concentration of methane gas.

[0053] 2) Constructing a revised Gaussian plume model

[0054] The monitored area in a coal mine is a confined space. During the diffusion of methane gas, collisions and reflections occur at the sidewalls, roof, and ground. When the methane gas collides with these surfaces, it undergoes specular reflection, with a reflection coefficient of α. This specular reflection exhibits a certain periodicity, which is positively correlated with the distance from the leak source to the boundary of the confined space, where n represents the number of gas collisions.

[0055] The following explanation uses a coal mine roadway as an example of the area to be monitored underground. The diffusion of methane gas in a coal mine roadway is correlated with the roadway height L and the roadway width D.

[0056] Therefore, when the gas spreads in the limited space such as the coal mine tunnel, the Gaussian plume model is revised as follows:

[0057]

[0058] The gas concentration at any point in the coal mine tunnel is:

[0059] C(x,y,z) = C(x,y,z,H,L) + C(x,y,z,H,D) (2)

[0060] Wherein, C(x,y,z,H,L), C(x,y,z,H,D) and C(x,y,z) are the gas concentrations, mg / m 3 ; Q is the gas leakage speed, mg / s; H is the height of the gas leakage source; μ is the wind speed; σ y and σ z are the gas diffusion coefficients in the y-axis direction and z-axis direction respectively; α is the reflection coefficient of the gas when colliding with the side wall, roof and ground of the coal mine tunnel; n is the total collision times of the gas with the side wall, roof and ground of the coal mine tunnel; L is the height of the underground coal mine monitoring area, the height direction is consistent with the z-axis direction; D is the width of the underground coal mine monitoring area, the width direction is consistent with the y-axis direction.

[0061] 3) Particle swarm optimization algorithm is used to solve the model parameters in the revised Gaussian plume model

[0062] For the revised Gaussian plume model obtained above, the model parameters need to be solved, which are specifically introduced as follows.

[0063] Firstly, Pasquill method is used to measure atmospheric stability, and gas diffusion coefficient reference is given through solar radiation, wind speed, weather and cloud cover. According to table 1, solar radiation grade is obtained, according to table 2, atmospheric stability grade is obtained, and according to table 3, gas diffusion coefficients σ y and σ z in y-axis direction and z-axis direction are obtained.

[0064] Table 1 Solar radiation grade

[0065]

[0066] Table 2 Atmospheric stability grade

[0067]

[0068] Table 3 Diffusion coefficient

[0069]

[0070] Secondly, the gas diffusion coefficient σ in the y-axis direction and z-axis direction obtained from Table 3 y and σ z , using the position information of the gas sensor and the gas concentration collected by it, the particle swarm optimization algorithm is used to solve the model parameters in the revised Gaussian plume model: reflection coefficient α, diffusion coefficient σ z , diffusion coefficient σ y . The particle swarm optimization algorithm flow is as Figure 3 .

[0071] Specifically: in the gas leakage search space O, the gas sensors distributed in various places are regarded as particles, each particle represents the concentration value at its location, then:

[0072]

[0073]

[0074] Wherein:

[0075] x id =(x i1 ,x i2 ,···,x iO ),i∈(1,N), the position of the i-th particle (representing the concentration value at a certain position in the coal mine tunnel space);

[0076] V id =(v i1 ,v i2 ,···,v iO ),i∈(1,N), the velocity of the i-th particle (particle movement direction and size, represented as the change of position information that satisfies the fitness function);

[0077] P id,gbest =(p i1 ,p i2 ,···,v iO ),i∈(1,N), the optimal position searched by the i-th particle;

[0078] P d,gbest =(p 1,gbest ,P 2,gbest ,···,P O,gbest ),i∈(1,N), the optimal position searched by the group;

[0079] r1,r2 are random numbers in [0,1], which increase the randomness of the search;

[0080] K, the number of iterations, commonly used value range [50,100], which can avoid too small to effectively converge, and can avoid too large to waste time;

[0081] ω, is an inertial weight, indicating the influence degree of the velocity of the last generation particle on the velocity of the current generation particle, that is, the inertia coefficient between two generations of particles; the greater the value of ω, the stronger the ability to explore new areas and the stronger the global optimization ability, but the weaker the local optimization ability, and vice versa; the global optimization ability is weaker, the local optimization ability is strong, a larger ω is conducive to global search and jumping out of local extreme value, so as not to fall into local optimum; and a smaller ω is conducive to local search, so as to quickly converge to the optimal solution;

[0082] c1, c2 are learning factors; c1 represents the weight of the part of the next action of the particle derived from the experience of itself, which is the acceleration weight of pushing the particle to the individual optimal position; and c2 represents the weight of the part of the next action of the particle derived from the experience of other particles, which is the acceleration weight of pushing the particle to the group optimal position.

[0083] The fitness function is constructed as follows:

[0084]

[0085] Wherein, C i represents the gas concentration detected by the i-th gas sensor, C d represents the gas concentration at the i-th gas sensor calculated by using the revised Gaussian plume model, that is, the value of x id in formula (4), the optimization parameters are the reflection coefficient α, the diffusion coefficient σ x , and the diffusion coefficient σ y , the optimization goal is to minimize the fitness function, and the reflection coefficient α, the diffusion coefficient σ x , and the diffusion coefficient σ y are substituted into the revised Gaussian plume model to obtain the final revised Gaussian plume model of the application, which is used for subsequent gas leakage source positioning.

[0086] (2) Gas concentration monitoring and gas leakage source positioning

[0087] The gas sensor is used to collect the gas concentration, and when the gas concentration exceeds the set threshold value, the collected gas concentration is input into the revised Gaussian plume model obtained by the application to construct the fitness function as follows.

[0088]

[0089] Wherein, C i represents the gas concentration detected by the i-th gas sensor, C dThe formula represents the gas concentration at the i-th gas sensor calculated by the revised Gauss plume model of the application, and the position coordinates of the maximum gas concentration are solved by the particle swarm optimization algorithm. That is, when the fitness function is minimized, the optimal solution obtained is the position coordinates of the gas leakage source.

[0090] Embodiment

[0091] The gas concentration is measured by the gas sensor installed in the coal mine roadway, as shown in Figure 2 Taking a coal mining face with U-shaped ventilation in a certain coal mine as an example, one gas sensor is arranged every 10 m to detect the gas concentration, and a total of 15 sensors are arranged.

[0092] In the initialization, a Gaussian rectangular coordinate system is established with the ground near the roadway wall below the gas sensor C1 as the coordinate (0, 0, 0) point. The x-axis direction is the downwind direction of the roadway, the y-axis direction is the width direction of the roadway, and the z-axis direction is the height direction of the roadway. Taking C1 as an example, the coordinates are (0, 0.3, 3.5) at this time, and the collision times n are taken as an empirical value of 4. The particle dimension to be solved is O=(α, σ x , σ y ), the learning factor c1=1.6, c2=2; in order to make the search space quickly converge, the inertia weight ω adopts an improved linear variation strategy, that is, (N max =200, the maximum number of iterations, N is the current number of iterations, ω max =1.5, ω min =0.7).

[0093] The model parameters are calculated: α=0.87, σ y =0.51x / (1+0.0001x), σ z =0.021x / (1+0.0003x)

[0094] When the gas leakage occurs, the gas sensor is regarded as a particle, a search space is constructed, and the coordinates (x, y, z) of the highest concentration point are solved. The fitness function is constructed as follows:

[0095]

[0096] When the fitness function is minimized, the optimal solution obtained is the leakage point coordinates.

Claims

1. A method for locating underground gas leak sources in coal mines, characterized in that, include: Multiple gas sensors installed in the monitoring area underground in the coal mine are used to collect gas concentration. When the gas concentration exceeds the set threshold, the collected gas concentration is input into the revised Gaussian plume model. The revised Gaussian plume model is solved to obtain the location information of the maximum gas concentration, which is the location of the gas leakage source. The revised Gaussian plume model is obtained by revising the Gaussian plume model using the height and width of the monitored area in the underground coal mine and the reflection coefficient of the gas when it collides with the sidewalls, roof and ground of the monitored area in the underground coal mine. The revised Gaussian plume model is as follows: （2） in, All values ​​are methane gas concentrations, mg / m³ 3 ; The gas leakage velocity is expressed in mg / s; H is the height of the gas leakage source. Wind speed; These are the gas diffusion coefficients along the y-axis and z-axis, respectively. denoted as , where is the reflection coefficient of the gas when it collides with the sidewalls, roof, and ground of the area to be monitored in the underground coal mine; denoted as n, where is the total number of collisions between the gas and the sidewalls, roof, and ground of the area to be monitored in the underground coal mine; denoted as L, where is the height of the area to be monitored in the underground coal mine, with the height direction aligned with the z-axis; and denoted as D, where is the width of the area to be monitored in the underground coal mine, with the width direction aligned with the y-axis.

2. The method for locating underground gas leakage sources in coal mines according to claim 1, characterized in that, The specific solution to the revised Gaussian plume model involves treating the gas sensor as a particle and using a particle swarm optimization algorithm.

3. The method for locating underground gas leakage sources in coal mines according to claim 1, characterized in that, Solving the revised Gaussian plume model yields the location information of the point of maximum methane concentration, specifically: Construct the fitness function: Among them, C i C represents the concentration of methane gas detected by the i-th methane sensor. d This indicates that the methane concentration of the i-th methane sensor was calculated using the revised Gaussian plume model. The revised Gaussian plume model is solved to obtain the optimal solution that minimizes the fitness function. The coordinates of the optimal solution are the location information of the point where the gas concentration is the maximum.

4. The method for locating underground gas leakage sources in coal mines according to claim 1, characterized in that, The method for constructing the revised Gaussian plume model includes: 1) Multiple gas sensors installed in the monitored area underground in the coal mine are used to collect the concentration of methane gas; 2) The revised Gaussian plume model is constructed as follows: （2） in, All values ​​are methane gas concentrations, mg / m³ 3 ; The gas leakage velocity is expressed in mg / s; H is the height of the gas leakage source. Wind speed; These are the gas diffusion coefficients along the y-axis and z-axis, respectively. denoted as , where is the reflection coefficient of methane gas when it collides with the sidewalls, roof, and ground of the area to be monitored in the underground coal mine; denoted as n, where is the total number of collisions between methane gas and the sidewalls, roof, and ground of the area to be monitored in the underground coal mine; denoted as L, where is the height of the area to be monitored in the underground coal mine, with the height direction aligned with the z-axis; and denoted as D, where is the width of the area to be monitored in the underground coal mine, with the width direction aligned with the y-axis. 3) Based on the collected methane concentration, solve for the model parameters in the revised Gaussian plume model. , Substitute the solved model parameters into the revised Gaussian plume model to obtain the revised Gaussian plume model.

5. The method for locating underground gas leakage sources in coal mines according to claim 4, characterized in that, In step 3), based on the collected methane concentration, the model parameters in the revised Gaussian plume model are solved, specifically as follows: Construct the fitness function: Among them, C i C represents the concentration of methane gas detected by the i-th methane sensor. d This indicates that the methane concentration at the i-th methane sensor was calculated using a revised Gaussian plume model. Solving the revised Gaussian plume model yields the result that minimizes the fitness function. , , This yields the model parameters.

6. The method for locating underground gas leakage sources in coal mines according to claim 4, characterized in that, In step 3), the Pasquill method is used to measure atmospheric stability. The reference range of the gas diffusion coefficient is obtained based on the atmospheric stability. Based on the reference range of the gas diffusion coefficient and the collected gas concentration, the model parameters in the revised Gaussian plume model are solved.

7. The method for locating underground gas leakage sources in coal mines according to claim 4, characterized in that, By treating the gas sensor as a particle, the model parameters in the revised Gaussian plume model are solved using the particle swarm optimization algorithm.

8. The method for locating underground gas leakage sources in coal mines according to claim 1, characterized in that, The area to be monitored underground in the coal mine is the coal mine roadway.

9. A coal mine underground gas leak source location system, characterized in that, It includes multiple gas sensors installed in the monitored area underground in the coal mine, as well as a central processing unit; A gas sensor is used to collect gas concentration data and transmit it to a central processing unit. The central processing unit includes a revised Gaussian plume model. When the detected methane gas concentration exceeds a set threshold, the received methane gas concentration is input into the revised Gaussian plume model, and the model is solved to obtain the location information of the point of maximum methane gas concentration, thus determining the location of the methane leak source. The revised Gaussian plume model is obtained by modifying the height and width of the monitored area in the coal mine and the reflection coefficients of the methane gas colliding with the sidewalls, roof, and ground of the monitored area. The revised Gaussian plume model is as follows: （2） in, All values ​​are methane gas concentrations, mg / m³ 3 ; The gas leakage velocity is expressed in mg / s; H is the height of the gas leakage source. Wind speed; These are the gas diffusion coefficients along the y-axis and z-axis, respectively. denoted as , where is the reflection coefficient of the gas when it collides with the sidewalls, roof, and ground of the area to be monitored in the underground coal mine; denoted as n, where is the total number of collisions between the gas and the sidewalls, roof, and ground of the area to be monitored in the underground coal mine; denoted as L, where is the height of the area to be monitored in the underground coal mine, with the height direction aligned with the z-axis; and denoted as D, where is the width of the area to be monitored in the underground coal mine, with the width direction aligned with the y-axis.

Images