Early warning method, device, medium and equipment for coal spontaneous combustion and gas coupling in goaf

Abstract

The invention discloses a goaf coal spontaneous combustion and gas coupling early warning method, device, medium and equipment, and relates to the technical field of coal mine safety production. The method comprises the following steps: constructing a disaster characteristic parameter data set about coal spontaneous combustion and gas coupling in a goaf; for each disaster characteristic parameter, training an inversion model corresponding to the disaster characteristic parameter by using a plurality of alternative training algorithms to obtain a plurality of trained inversion models, and determining an optimal inversion model based on the plurality of trained inversion models; performing inversion on the disaster characteristic parameters by using the optimal inversion model to obtain an inversion result of each disaster characteristic parameter; and according to an inversion result, carrying out graded early warning on the risk of coal spontaneous combustion and gas coupling in the goaf. According to the scheme, the accuracy of early warning of the coal spontaneous combustion and gas coupling disaster risk in the goaf can be improved.

Description

Technical Field

[0001] The invention relates to the technical field of coal mine production safety, and in particular to a method, device, medium and equipment for early warning of spontaneous combustion of coal and gas coupling in goaf areas. Background Art

[0002] In the process of coal mining, gas explosion accidents caused by spontaneous combustion of coal occur from time to time, seriously endangering the safety of workers. Goaf is a high-incidence area of ​​coal spontaneous combustion and the area most prone to gas accumulation. Therefore, it is a key area for the prevention and control of the risk of coupled disasters of coal spontaneous combustion and gas. The coupling of spontaneous combustion of coal in goaf and gas disasters is characterized by concealment, dynamism, complexity and suddenness. Its disaster-causing mechanism is extremely complex, with high risks, difficult identification and difficult early warning.

[0003] At present, the most commonly used methods for early warning of the risk of spontaneous combustion of coal and gas coupling disasters in goafs are temperature index method and gas index method (including multi-gas combination index method). The core of risk early warning is to infer the risk level based on the monitoring value of the index. According to the location of the indicator monitoring point, the risk monitoring method of spontaneous combustion of coal and gas coupling disasters in goafs can be divided into two types: internal monitoring of goafs and external monitoring of goafs. "Internal monitoring" refers to setting up measuring points inside the goaf to continuously monitor the temperature or gas component changes at specific points. Although this method can directly obtain information inside the goaf and then directly determine the risk of spontaneous combustion of coal and gas disasters near the measuring points, the technical difficulty and cost of setting and maintaining the measuring points are high. Since there is no guarantee that high-temperature points or explosive points will definitely appear near the measuring points, especially when the measuring points are far away from the actual high-temperature points or explosive points, it is easy to underestimate the risk of spontaneous combustion of coal and gas coupling disasters in goafs, resulting in low accuracy in early warning of the risk of spontaneous combustion of coal and gas coupling disasters in goafs. "External monitoring" refers to setting up measuring points outside the goaf, such as the inlet and return air lanes or inlet and return air corners of the mining face. Since the technical difficulty and cost of setting and maintaining measuring points are low, it is the main method for daily monitoring of coal spontaneous combustion and gas disaster risks in goafs. However, the external monitoring method cannot directly obtain information inside the goaf. Since the flow field inside the goaf, the transportation process of various gas components, and the evolution mechanism of coal spontaneous combustion and gas coupling disasters are very complex, the existing early warning methods based on external monitoring of goafs can only infer the general trend of oxidation and self-heating of residual coal in the goaf or the general range of goaf temperature, and the accuracy of early warning of coal spontaneous combustion and gas coupling disaster risks in goafs is not high.

[0004] In summary, the existing methods cannot provide accurate early warning for the risk of spontaneous combustion of coal and gas coupling disasters in goaf areas. Summary of the invention

[0005] Based on this, it is necessary to provide early warning methods, devices, media and equipment for coal spontaneous combustion and gas coupling in goafs to address the technical problem that existing methods are unable to make accurate early warnings for the disaster risks of coal spontaneous combustion and gas coupling in goafs.

[0006] The present invention adopts the following technical solutions: In a first aspect, the present invention provides a method for early warning of spontaneous combustion of coal and gas coupling in goaf, the method comprising: Constructing a disaster characteristic parameter data set about the coupling of spontaneous combustion of coal and gas in goaf, wherein the disaster characteristic parameter data set includes disaster characteristic parameters and multiple environmental parameters, wherein the disaster characteristic parameters include characteristic parameters of high temperature points in goaf and characteristic parameters of gas explosion hazard areas; For each disaster characteristic parameter, the multiple environmental parameters are used as input and the disaster characteristic parameter is used as output, and the inversion model corresponding to the disaster characteristic parameter is trained using multiple alternative training algorithms to obtain multiple trained inversion models, and the optimal inversion model corresponding to the disaster characteristic parameter is determined based on the multiple trained inversion models; the disaster characteristic parameter is inverted using the optimal inversion model corresponding to the disaster characteristic parameter to obtain the inversion result of each disaster characteristic parameter; According to the inversion results, a graded early warning is provided for the risk of spontaneous combustion of coal and gas coupling in the goaf.

[0007] Furthermore, the characteristic parameters of the high-temperature points in the goaf include: the temperature value and three-dimensional coordinates of the high-temperature points in the air intake half area, and the temperature value and three-dimensional coordinates of the high-temperature points in the return air half area; the characteristic parameters of the gas explosion hazard area include: the volume of the gas explosion hazard area, the three-dimensional coordinates of the explosive point and the volume fraction of each type of gas in the mixed gas; Among them, the air intake half zone and the return air half zone are divided by taking the inclination center plane of the goaf as the dividing interface; the gas explosion hazardous area is the area in the goaf where the porosity is greater than the critical porosity and the mixed gas is explosive, and the mixed gas contains oxygen, methane, carbon monoxide, carbon dioxide and nitrogen; the explosive point is the location of the maximum methane concentration in the gas explosion hazardous area.

[0008] Furthermore, the multiple environmental parameters include coal seam parameters, roof parameters, mining parameters, face ventilation parameters, goaf gas emission parameters, coal residue distribution parameters and face environmental parameters of the goaf, and the construction method of the disaster characteristic parameter data set specifically includes: Determine the level numbers and level values ​​of the plurality of environmental parameters according to the field environment of the goaf, and determine the test plan and the number of tests according to the level numbers and level values; A three-dimensional numerical model of the goaf is established according to the test scheme, and a computer simulation test is performed on the oxidation self-heating process of the residual coal in the goaf based on the three-dimensional numerical model and the test times to obtain a simulation result; Based on the simulation results, extract the air inlet corner environmental parameters, the air return corner environmental parameters and the corresponding disaster characteristic parameter values ​​of the mining face; Based on the inlet wind corner environmental parameters, the return wind corner environmental parameters of the mining face, the disaster characteristic parameter values ​​and the values ​​of each environmental parameter, a data sample is obtained, and the disaster characteristic parameter data set is constructed based on the data sample; Among them, the coal seam parameters include coal seam thickness and coal seam inclination; the roof parameters include direct roof thickness and length of rock blocks falling from the old roof; the mining parameters include mining face length, mining face height and control top distance; the mining face ventilation parameters include mining face air volume, friction resistance coefficient and ventilation method; the goaf gas emission parameters include methane emission and carbon dioxide emission; the residual coal distribution parameters include width and thickness of each residual coal strip; the mining face environmental parameters include temperature and gas concentration of mining face air inlet tunnel, air inlet corner and return air corner.

[0009] Further, for each disaster characteristic parameter, the multiple environmental parameters are used as input and the disaster characteristic parameter is used as output, and the inversion model corresponding to the disaster characteristic parameter is trained using multiple alternative training algorithms to obtain multiple trained inversion models, and the optimal inversion model corresponding to the disaster characteristic parameter is determined based on the multiple trained inversion models; the disaster characteristic parameter is inverted using the optimal inversion model corresponding to the disaster characteristic parameter to obtain the inversion result of each disaster characteristic parameter, which specifically includes: The disaster characteristic parameter data set is j The first disaster characteristic parameter is used as the dependent variable, the coal seam parameter, the roof parameter, the mining parameter, the mining face ventilation parameter, the residual coal distribution parameter, the mining face air intake environment parameter, the mining face air intake corner environment parameter and the mining face return air corner environment parameter are used as independent variables, and each of the multiple alternative training algorithms is used to train the first disaster characteristic parameter respectively. j The disaster characteristic parameters are used to train the inversion model, and the various candidate training algorithms are obtained for the first j The inversion model of the disaster characteristic parameters is provided, and the candidate training algorithms include a lightweight gradient boosting algorithm, a fast tree algorithm and a fast forest algorithm. j is a positive integer, 1≦j≦N, N is the number of disaster characteristic parameters; Get each candidate training algorithm about the first j The determination coefficient of the inversion model of the disaster characteristic parameters is determined, and the inversion model with the largest determination coefficient is determined as the first jThe optimal inversion model of each disaster characteristic parameter; The actual value of the independent variable is input into the optimal inversion model for inversion to obtain the temperature value and three-dimensional coordinates of the high-temperature points in the air intake half of the goaf, the temperature value and three-dimensional coordinates of the high-temperature points in the return air half, the volume of the gas explosion hazardous area, the three-dimensional coordinates of the explosive point and the volume fraction of each type of gas in the mixed gas.

[0010] Furthermore, according to the inversion results, a graded early warning is performed on the risk of spontaneous combustion of coal and gas coupling in the goaf, specifically including: Compare the temperature value of the high temperature point of the air inlet half zone with the temperature value of the high temperature point of the return air half zone, determine the high temperature point with the higher temperature value as the high temperature point of the goaf according to the comparison result, and calculate the Euclidean distance between the high temperature point and the explosive point; The temperature value of the high temperature point of the goaf, the volume of the gas explosion hazard zone, the volume fraction of methane at the explosive point, and the Euclidean distance between the high temperature point and the explosive point are determined as a risk basic factor set, and a first-level fuzzy comprehensive evaluation is performed based on the risk basic factor set and the evaluation set to obtain a comprehensive evaluation vector; Based on the comprehensive evaluation vector, the current risk warning level of the goaf is determined using the maximum membership principle.

[0011] Further, the inversion model with the largest determination coefficient is determined as the first j After the optimal inversion model of the disaster characteristic parameters, it also includes: For the said j The performance of the optimal inversion model of each disaster characteristic parameter is tested and the test results are obtained; When the test result shows that the performance of the optimal inversion model is less than the preset performance index, j The inversion model of each disaster characteristic parameter is further optimized.

[0012] In a second aspect, the present invention provides a goaf area coal spontaneous combustion and gas coupling early warning device, comprising: A construction module is used to construct a disaster characteristic parameter data set about the coupling of spontaneous combustion of coal and gas in goaf, wherein the disaster characteristic parameter data set includes disaster characteristic parameters and multiple environmental parameters, wherein the disaster characteristic parameters include characteristic parameters of high temperature points in goaf and characteristic parameters of gas explosion hazard areas; An inversion module is used for, for each disaster characteristic parameter, using the multiple environmental parameters as input and the disaster characteristic parameter as output, training the inversion model corresponding to the disaster characteristic parameter using multiple alternative training algorithms to obtain multiple trained inversion models, and determining the optimal inversion model corresponding to the disaster characteristic parameter based on the multiple trained inversion models; inverting the disaster characteristic parameter using the optimal inversion model corresponding to the disaster characteristic parameter to obtain an inversion result for each disaster characteristic parameter; The early warning module is used to provide graded early warning for the risk of spontaneous combustion of coal and gas coupling in the goaf according to the inversion result.

[0013] The present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the goaf area coal spontaneous combustion and gas coupling early warning method is implemented.

[0014] The present invention provides a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the goaf area coal spontaneous combustion and gas coupling early warning method when executing the program.

[0015] At least one technical solution adopted by the present invention can achieve the following beneficial effects: the present invention constructs a disaster characteristic parameter data set about the spontaneous combustion of coal in goaf and the coupling of gas, and the disaster characteristic parameter data set includes disaster characteristic parameters and multiple environmental parameters; for each disaster characteristic parameter, multiple environmental parameters are used as input and the disaster characteristic parameter is used as output, and the inversion model corresponding to the disaster characteristic parameter is trained using multiple alternative training algorithms to obtain multiple trained inversion models, and the optimal inversion model corresponding to the disaster characteristic parameter is determined based on the multiple trained inversion models; the disaster characteristic parameter is inverted using the optimal inversion model corresponding to the disaster characteristic parameter to obtain the inversion result of each disaster characteristic parameter; according to the inversion result, the risk of spontaneous combustion of coal in goaf and the coupling of gas is graded and warned. Through the above scheme, multiple alternative training algorithms can be used to train the inversion model of each type of disaster characteristic parameter in the disaster characteristic parameter data set respectively, and the optimal inversion model with the best inversion effect associated with different types of disaster characteristic parameters is obtained, so that different types of disaster characteristic parameters can be accurately inverted, and accurate inversion results of high temperature point characteristic parameters and gas explosion hazard area characteristic parameters of goaf can be obtained. The inversion results of accurate characteristic parameters of high-temperature points in goaf and characteristic parameters of gas explosion hazard areas can not only reflect the state of coal oxidation self-heating and gas accumulation in goaf and the risk of coal spontaneous combustion and gas coupling disaster, but also reveal the evolution process of high-temperature points and gas explosion hazard areas and the changing trend of coal spontaneous combustion and gas coupling disaster risk. Based on the inversion results of accurate characteristic parameters of high-temperature points in goaf and characteristic parameters of gas explosion hazard areas, the risk of coal spontaneous combustion and gas coupling disaster in goaf can be graded and warned, which can improve the accuracy of warning of coal spontaneous combustion and gas coupling disaster risk in goaf. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 A schematic diagram of the process flow of the goaf area coal spontaneous combustion and gas coupling early warning method provided by the present invention; Figure 2 A comparison chart of the predicted values ​​and sample values ​​of the inversion model of the disaster characteristic parameters provided by the present invention; Figure 3 A schematic diagram of the goaf area coal spontaneous combustion and gas coupling early warning device provided by the present invention; Figure 4 A schematic diagram of a computer device for realizing a method for coupling early warning of spontaneous combustion of coal and gas in goaf provided by the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. 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 making creative work are within the scope of protection of the present invention.

[0018] The server mentioned in the present invention may be a server set on a business platform, or a device such as a desktop computer, a notebook computer, etc. that can execute the solution of the present invention. For the convenience of description, the following description is only based on the server as the execution subject.

[0019] Before further describing the embodiments of the present invention, the abbreviations of the terms used in the present invention are explained: "Goaf" refers to the goaf area of ​​the coal mining face, "mining face" refers to the coal mining face or the mining face, and "high temperature point" refers to the highest temperature point.

[0020] The technical solutions provided by various embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] Figure 1 The figure is a flow chart of the method for early warning of spontaneous combustion of coal and gas in goaf of the present invention, which specifically includes the following steps: S10: Construct a disaster characteristic parameter data set about the coupling of spontaneous combustion of coal and gas in goaf. The disaster characteristic parameter data set includes disaster characteristic parameters and multiple environmental parameters. The disaster characteristic parameters include characteristic parameters of high temperature points in goaf and characteristic parameters of gas explosion hazardous areas.

[0022] In this embodiment, the characteristic parameters of high-temperature points in the goaf include: the temperature value and three-dimensional coordinates of the high-temperature points in the air intake half area, and the temperature value and three-dimensional coordinates of the high-temperature points in the return air half area; the characteristic parameters of the gas explosion hazardous area include: the volume of the gas explosion hazardous area, the three-dimensional coordinates of the explosive points and the volume fraction of each type of gas in the mixed gas.

[0023] Among them, the air intake half zone and the return air half zone are divided by the inclination center plane of the goaf as the dividing interface; the gas explosion hazardous area is the area in the goaf where the porosity is greater than the critical porosity and the mixed gas is explosive, and the mixed gas contains oxygen, methane, carbon monoxide, carbon dioxide and nitrogen; the explosive point is the location of the maximum methane concentration in the gas explosion hazardous area.

[0024] In this embodiment, the critical void ratio refers to the minimum void ratio that allows explosion flame to propagate.

[0025] Specifically, the temperature value of the high temperature point in the air inlet half area T1 With three-dimensional coordinates ( x 1 , y 1 , z 1 ), the temperature value of the high temperature point in the return air half zone T 2 With three-dimensional coordinates ( x 2 , y 2 , z 2 ) A total of 8 parameters are determined as the characteristic parameters of the high-temperature points in the goaf. In this embodiment, the three-dimensional coordinates ( x 3 , y 3 , z 3 ) is determined as the explosive point. And the volume of the gas explosion hazard area V exp , the three-dimensional coordinates of the explosive point ( x 3 , y 3 , z 3 ) and the volume fraction of each type of gas (oxygen, methane, carbon monoxide, carbon dioxide and nitrogen) in the mixed gas, a total of 9 parameters are determined as characteristic parameters of the gas explosion hazardous area.

[0026] S20: For each disaster characteristic parameter, with multiple environmental parameters as input and the disaster characteristic parameter as output, use multiple alternative training algorithms to train the inversion model corresponding to the disaster characteristic parameter to obtain multiple trained inversion models, and determine the optimal inversion model corresponding to the disaster characteristic parameter based on the multiple trained inversion models; use the optimal inversion model corresponding to the disaster characteristic parameter to invert the disaster characteristic parameter to obtain the inversion result of each disaster characteristic parameter.

[0027] In this embodiment, the optimal inversion model refers to the inversion model with the best inversion effect for each type of disaster characteristic parameters, and multiple candidate training algorithms include but are not limited to lightweight gradient boosting algorithm, fast tree algorithm and fast forest algorithm. Based on the artificial intelligence inversion method, the inversion model training algorithm that is most suitable for each type of disaster characteristic parameters is selected from multiple candidate training algorithms, and the optimal inversion model corresponding to each type of disaster characteristic parameters is established using the selected most suitable inversion model training algorithm and the disaster characteristic parameter data set.

[0028] S30: Based on the inversion results, a graded warning is provided for the risk of spontaneous combustion of coal and gas coupling in the goaf.

[0029] In this embodiment, a graded early warning is performed for the risk of spontaneous combustion of coal and gas coupling in the goaf, specifically including: S301: Compare the temperature value of the high temperature point in the air intake half zone with the temperature value of the high temperature point in the return air half zone, determine the high temperature point with the higher temperature value as the high temperature point in the goaf according to the comparison result, and calculate the Euclidean distance between the high temperature point and the explosive point.

[0030] For example, if the temperature value of the high temperature point in the air inlet half area is T 1 Greater than the temperature value of the high temperature point in the return air half zone T 2 , then the temperature of the high temperature point in the goaf is T max = T 1 , the coordinates are ( x 1 , y 1 , z 1 ), the Euclidean distance between the high temperature point and the explosive point in the goaf L The calculation expression is: L = (1) in, L is the Euclidean distance between the high temperature point and the explosive point in the goaf, x 1 , y 1 , z 1 is the three-dimensional coordinate of the high temperature point in the goaf, x 3 , y 3 , z 3 are the three-dimensional coordinates of the explosive point.

[0031] S302: The temperature value of the high-temperature point in the goaf, the volume of the gas explosion hazardous area, the volume fraction of methane at the explosive point, and the Euclidean distance between the high-temperature point and the explosive point are determined as the basic risk factor set, and a first-level fuzzy comprehensive evaluation is performed based on the basic risk factor set and the evaluation set to obtain a comprehensive evaluation vector.

[0032] Specifically, the basic risk factors of spontaneous combustion of coal and gas coupling disaster in goaf are defined as: the maximum temperature of goaf T max , volume of gas explosion hazard zone Vexp , explosive point methane volume fraction X CH4 , the distance between the highest temperature point and the explosive point L ; Take the basic factor set U ={ T max , V exp , X CH4 , L}、Evaluation set V ={low risk, medium risk, medium-high risk, high risk, extremely high risk}, carry out the first-level fuzzy comprehensive evaluation and obtain the comprehensive evaluation vector.

[0033] S303: Based on the comprehensive evaluation vector, the current risk warning level of the goaf is determined using the maximum membership principle.

[0034] Specifically, according to the obtained comprehensive evaluation vector, the maximum membership principle is used to determine the current risk warning level of the goaf.

[0035] Exemplarily, the obtained comprehensive evaluation vector is B =[ B 1 , B 2 , B 3 , B 4 , B 5 ],if B 3 for B 1 ~ B 5 The current risk warning level of spontaneous combustion of coal and gas coupling disaster in goaf is “medium-high risk”.

[0036] based on Figure 1The method for early warning of spontaneous combustion of coal and gas coupling in goaf shown in the present invention constructs a disaster characteristic parameter data set about the coupling of spontaneous combustion of coal and gas in goaf, and the disaster characteristic parameter data set includes disaster characteristic parameters and multiple environmental parameters; for each disaster characteristic parameter, multiple environmental parameters are used as input and the disaster characteristic parameter is used as output, and the inversion model corresponding to the disaster characteristic parameter is trained using multiple alternative training algorithms to obtain multiple trained inversion models, and the optimal inversion model corresponding to the disaster characteristic parameter is determined based on the multiple trained inversion models; the disaster characteristic parameter is inverted using the optimal inversion model corresponding to the disaster characteristic parameter to obtain the inversion result of each disaster characteristic parameter; according to the inversion result, the risk of spontaneous combustion of coal and gas coupling in goaf is graded and early warned. Through the above scheme, multiple alternative training algorithms can be used to train the inversion model of each type of disaster characteristic parameter in the disaster characteristic parameter data set, and the optimal inversion model with the best inversion effect associated with different types of disaster characteristic parameters can be obtained, so that different types of disaster characteristic parameters can be accurately inverted, and accurate inversion results of high temperature point characteristic parameters and gas explosion hazard area characteristic parameters can be obtained. The accurate inversion results of high temperature point characteristic parameters and gas explosion hazard area characteristic parameters in the goaf can not only reflect the coal oxidation self-heating and gas accumulation state in the goaf and the risk of coal spontaneous combustion and gas coupling disaster, but also reveal the evolution process of high temperature points and gas explosion hazard areas and the changing trend of coal spontaneous combustion and gas coupling disaster risks. According to the inversion results of accurate high temperature point characteristic parameters and gas explosion hazard area characteristic parameters in the goaf, the risk of coal spontaneous combustion and gas coupling disaster in the goaf can be graded and warned, which can improve the accuracy of warning of coal spontaneous combustion and gas coupling disaster risks in the goaf.

[0037] When applying the goaf area coal spontaneous combustion and gas coupling early warning method provided by the present invention, it is not necessary to Figure 1 The steps are executed in the order shown. The specific execution order of the steps can be determined according to needs, and the present invention does not limit this.

[0038] In addition, in one or more embodiments of the present invention, the plurality of environmental parameters include coal seam parameters, roof parameters, mining parameters, face ventilation parameters, goaf gas emission parameters, coal remnant distribution parameters and face environmental parameters of the goaf, and the method for constructing the disaster characteristic parameter data set specifically includes: The level numbers and level values ​​of multiple environmental parameters are determined according to the on-site environment of the goaf, and the test plan and the number of tests are determined according to the level numbers and level values.

[0039] According to the test plan, a three-dimensional numerical model of the goaf was established, and based on the three-dimensional numerical model and the number of tests, a computer simulation test was carried out on the oxidation self-heating process of the residual coal in the goaf to obtain the simulation results.

[0040] Based on the simulation results, the environmental parameters of the air inlet corner, the air return corner and the corresponding disaster characteristic parameter values ​​of the mining face are extracted.

[0041] Based on the environmental parameters of the air inlet corner, the air return corner and the disaster characteristic parameter values ​​of the mining surface and the values ​​of each environmental parameter, a data sample is obtained, and a disaster characteristic parameter data set is constructed based on the data sample.

[0042] Among them, coal seam parameters include coal seam thickness and coal seam inclination; roof parameters include direct roof thickness and length of rock blocks falling from old roof; mining parameters include mining face length, mining face height and roof control distance; mining face ventilation parameters include mining face air volume, friction resistance coefficient and ventilation method; goaf gas emission parameters include methane emission and carbon dioxide emission; residual coal distribution parameters include width and thickness of each residual coal strip; mining face environmental parameters include temperature and gas concentration of mining face air inlet tunnel, air inlet corner and return air corner.

[0043] In this embodiment, coal seam parameters (thickness and inclination), roof parameters (immediate roof thickness and length of rock blocks falling from the old roof), mining parameters (mining face length, mining height and top control distance), mining face ventilation parameters (air volume, friction resistance coefficient and ventilation method), goaf gas emission parameters (methane emission, carbon dioxide emission), residual coal distribution parameters (width and thickness of each residual coal strip) and mining face environment parameters (temperature and gas concentration) are used as experimental variables, and the number of levels and level values ​​of each type of experimental variables are selected with reference to the coal mine field data.

[0044] For example, two levels are taken for coal seam thickness (9m and 10m), two levels are taken for coal seam inclination (25° and 30°), two levels are taken for direct roof thickness (3m and 6m), two levels are taken for the length of rock blocks falling from the old roof (15m and 20m), two levels are taken for the length of the mining face (140m and 150m), one level is taken for the height of the mining face (3m), one level is taken for the top control distance of the mining face (7.5m), and three levels are taken for the wind volume of the mining face (500 m 3 / min,1500 m 3 / min,2500 m 3 / min, and the friction coefficient of the mining surface is set at a level of 0.035 kg / m 3 ), two levels were taken for ventilation mode (upward and downward), and two levels were taken for methane emission in goaf (1.0 m 3 / min and 2.0 m 3 / min, and the carbon dioxide outflow in the goaf was measured at two levels (0.5 m 3 / min and 1.0 m 3 / min), the residual coal distribution parameters are taken at one level (multiple residual coal strips can be considered, the width and thickness of each strip remain unchanged, and the specific parameter examples are omitted), and the mining face inlet air environment parameters are taken at one level (20℃, the volume fractions of oxygen and nitrogen are 0.21 and 0.79 respectively).

[0045] For example, comprehensive tests are conducted on the four relatively important factors, namely, the length of the old roof caving block, the air volume of the mining face, the methane outflow of the goaf, and the carbon dioxide outflow of the goaf. A total of 2×3×2×2=24 tests are required. Orthogonal tests are conducted on other factors, such as L 8 (2 7 ) orthogonal table, 8 experiments are required; combining comprehensive experiments with orthogonal experiments, a total of n =24×8=192 trials.

[0046] Let the test number be i , first let i = 1, the test steps are as follows: 1) According to the i The experimental conditions of the first experiment (i.e. the i-th combination of the experimental factor level values) were used to establish a three-dimensional numerical model of the goaf, and a computer simulation experiment on the oxidation self-heating process of the residual coal in the goaf was carried out.

[0047] 2) Extract the environmental parameter values ​​(temperature and gas concentration) of the inlet and return air corners of the mining face and the corresponding 8 high-temperature point characteristic parameter values ​​and 9 gas explosion hazardous area characteristic parameter values ​​from the simulation results, combine them with the experimental factor values ​​to form a high-temperature point data sample and a gas explosion hazardous area data sample, and add them to the corresponding data sets respectively.

[0048] 3) Order i = i +1.

[0049] Repeat steps 1), 2), and 3) until i > n .

[0050] So far, a high-temperature point dataset in the goaf and a gas explosion hazardous area dataset containing 192 data samples have been established.

[0051] In addition, in one or more embodiments of the present invention, for each disaster characteristic parameter, multiple environmental parameters are used as input and the disaster characteristic parameter is used as output, and multiple candidate training algorithms are used to train the inversion model corresponding to the disaster characteristic parameter to obtain multiple trained inversion models, and the optimal inversion model corresponding to the disaster characteristic parameter is determined based on the multiple trained inversion models; the disaster characteristic parameter is inverted using the optimal inversion model corresponding to the disaster characteristic parameter to obtain the inversion result of each disaster characteristic parameter, which specifically includes: The disaster characteristic parameter data set is j The disaster characteristic parameters are taken as dependent variables, and the coal seam parameters, roof parameters, mining parameters, face ventilation parameters, coal distribution parameters, face air environment parameters, face air corner environment parameters and face return air corner environment parameters are taken as independent variables. Each of the multiple alternative training algorithms is used to train the first j The inversion model is trained based on the characteristic parameters of the disaster, and the various candidate training algorithms are obtained. j The inversion model of disaster characteristic parameters is used. The alternative training algorithms include lightweight gradient boosting algorithm, fast tree algorithm and fast forest algorithm. j is a positive integer, 1≦j≦N, where N is the number of disaster characteristic parameters.

[0052] Get the information about each candidate training algorithm j The determination coefficient of the inversion model of the disaster characteristic parameters is calculated, and the inversion model with the largest determination coefficient is determined as the first j The optimal inversion model of disaster characteristic parameters.

[0053] The actual values ​​of the independent variables are input into the optimal inversion model for inversion, and the temperature values ​​and three-dimensional coordinates of the high-temperature points in the air intake half of the goaf, the temperature values ​​and three-dimensional coordinates of the high-temperature points in the return air half, the volume of the gas explosion hazardous area, the three-dimensional coordinates of the explosive point and the volume fraction of each type of gas in the mixed gas are obtained.

[0054] Specifically, the disaster characteristic parameter inversion models established in this embodiment total 17, that is, N is 17. They are used to invert 17 types of disaster characteristic parameters. The 17 types of disaster characteristic parameters include: the temperature value of the high temperature point in the windward half zone; T 1 With three-dimensional coordinates ( x 1 , y 1 , z 1 ), the temperature value of the high temperature point in the return air half zone T 2 With three-dimensional coordinates ( x 2 , y2 , z 2 ), the volume of the gas explosion hazard zone V exp , the three-dimensional coordinates of the explosive point ( x 3 , y 3 , z 3 ), and oxygen concentration at the explosive point X O2 , explosive point methane concentration X CH4 , Carbon monoxide concentration at the explosive point X CO , Explosive point carbon dioxide concentration X CO2 , Nitrogen concentration at explosive point X N2 .

[0055] a) Exemplary, number of known disaster characteristic parameters m = 17, let parameter number j = 1, take the j The disaster characteristic parameters are taken as dependent variables. Based on the disaster characteristic parameter data set containing the taken dependent variables, the coal seam parameters, roof parameters, mining parameters, mining face ventilation parameters, residual coal distribution parameters, mining face air intake environment parameters, mining face air intake corner environment parameters and mining face return air corner environment parameters are taken as independent variables, and the alternative training algorithms are respectively used to train the inversion models corresponding to the dependent variables, and the inversion models corresponding to each training algorithm are obtained.

[0056] b) Obtain the determination coefficients of the inversion models corresponding to each training algorithm, and determine the inversion model with the largest determination coefficient as the first j The optimal inversion model of disaster characteristic parameters.

[0057] In this embodiment, the optimal inversion model refers to j The inversion model with the most suitable disaster characteristic parameters and the best inversion effect is the first j Specifically, the determination coefficients of the inversion models corresponding to each training algorithm are compared ( R 2 ), the training algorithm of the inversion model with the largest determination coefficient is taken as the optimal training algorithm, and the inversion model obtained based on the optimal training algorithm is taken as the j The optimal inversion model of disaster characteristic parameters.

[0058] c) Input the actual values ​​of the independent variables into the optimal inversion model to invert the temperature values ​​and three-dimensional coordinates of the high-temperature points in the air intake half of the goaf, the temperature values ​​and three-dimensional coordinates of the high-temperature points in the return air half, the volume of the gas explosion hazardous area, the three-dimensional coordinates of the explosive points and the volume fractions of each type of gas in the mixed gas.

[0059] In this embodiment, for the goaf area to be evaluated, the actual values ​​of the independent variables collected (including the monitoring values ​​of the environmental parameters of the inlet and return air corners of the mining face) are input into the first j The optimal inversion model of the disaster characteristic parameters is used to invert the temperature values ​​of the high-temperature points in the air intake half of the goaf. T 1 and three-dimensional coordinates ( x 1 , y 1 , z 1 ) and the temperature value of the high temperature point in the return air half area T 2 and three-dimensional coordinates ( x 2 , y 2 , z 2 ), volume of gas explosion hazard zone V exp , explosive point coordinates ( x 3 , y 3 , z 3 ) and the volume fraction of the five gases in the explosive point ( X O2 , X CH4 , X CO , X CO2 and X N2 ).

[0060] Through this embodiment, the inversion model with the best inversion effect can be constructed for different disaster characteristic parameters, and the accuracy of inversion of various types of disaster characteristic parameters can be improved, so as to obtain more accurate parameters related to spontaneous combustion of coal and gas coupling disasters in goaf areas, thereby improving the accuracy of early warning of the risk of spontaneous combustion of coal and gas coupling disasters in goaf areas. In addition, there is no need to set measurement points inside the goaf by pre-embedding or drilling. It is only necessary to monitor the environmental parameters of the air flow in the mining face and the air inlet and return corners on a daily basis, which is simple and easy.

[0061] In addition, in one or more embodiments of the present invention, the inversion model with the largest determination coefficient is determined as the firstj After the optimal inversion model of the disaster characteristic parameters, it also includes: For j The performance of the optimal inversion model of each disaster characteristic parameter is tested and the test results are obtained.

[0062] When the test result shows that the performance of the optimal inversion model is less than the preset performance index, j The inversion model of each disaster characteristic parameter is further optimized.

[0063] In this embodiment, reference Figure 2 , with the sample value of the dependent variable (taking the temperature value of the high temperature point in the air inlet half area as an example) as the horizontal axis x , with the predicted value of the dependent variable as the vertical axis y , draw a scatter plot, if most of the data points are distributed on the straight line y = x If the majority of data points are close to the straight line, it means that the predicted value of the dependent variable output by the inversion model is consistent with the sample value (actual value) of the dependent variable, indicating that the inversion model has good performance. y = x If the distance is far, it means that there is a large deviation between the predicted value of the dependent variable output by the inversion model and the sample value of the dependent variable, and the performance of the inversion model is poor. Figure 3 It can be seen that the triangles in the figure (the predicted values ​​of the dependent variable) are all on the straight line y = x This indicates that the performance of the inversion model is good.

[0064] If the optimal inversion model performs well, j = j +1, repeat step a), step b), step c), until j > m ; If the optimal inversion model performs poorly, return to step 1), consider more test conditions, and expand the disaster characteristic parameter data set.

[0065] The solution of this embodiment can ensure that the performance of the inversion model corresponding to each type of disaster characteristic parameter reaches a preset threshold, thereby guaranteeing the inversion effect of the disaster characteristic parameters.

[0066] The above is a method for early warning of spontaneous combustion of coal and gas coupling in goaf provided by one or more embodiments of the present invention. Based on the same idea, the present invention also provides a corresponding early warning device for spontaneous combustion of coal and gas coupling in goaf, such as Figure 3 As shown, including: A construction module is used to construct a disaster characteristic parameter data set about the spontaneous combustion of coal in goaf and the coupling of gas. The disaster characteristic parameter data set includes disaster characteristic parameters and multiple environmental parameters. The disaster characteristic parameters include characteristic parameters of high temperature points in goaf and characteristic parameters of gas explosion hazardous areas.

[0067] The inversion module is used to train the inversion model corresponding to each disaster characteristic parameter using multiple environmental parameters as input and the disaster characteristic parameter as output using multiple alternative training algorithms to obtain multiple trained inversion models, and determine the optimal inversion model corresponding to the disaster characteristic parameter based on the multiple trained inversion models; invert the disaster characteristic parameter using the optimal inversion model corresponding to the disaster characteristic parameter to obtain the inversion result of each disaster characteristic parameter.

[0068] The early warning module is used to provide graded early warning of the risk of spontaneous combustion of coal and gas coupling in goaf areas based on the inversion results.

[0069] For the specific definition of the goaf coal spontaneous combustion and gas coupling early warning device, please refer to the definition of the goaf coal spontaneous combustion and gas coupling early warning method above, which will not be repeated here. Each module in the goaf coal spontaneous combustion and gas coupling early warning device can be fully or partially implemented by software, hardware and their combination. Each module can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0070] The present invention also provides a computer-readable storage medium, which stores a computer program, which can be used to execute Figure 1 A method for early warning of spontaneous combustion of coal and gas coupling in goaf areas is provided.

[0071] The present invention also provides Figure 4 The structural diagram of the computer device shown in FIG. Figure 4 As shown in the figure, at the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include other hardware required for the business. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to achieve Figure 1 A method for early warning of spontaneous combustion of coal and gas coupling in goaf areas is provided.

[0072] A person of ordinary skill in the art can understand that all or part of the processes in the embodiment method can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the methods described. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0073] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present invention.

Claims

1. A method for early warning of spontaneous combustion of coal and gas coupling in goaf, characterized in that: include: Constructing a disaster characteristic parameter data set about the coupling of spontaneous combustion of coal and gas in goaf, wherein the disaster characteristic parameter data set includes disaster characteristic parameters and multiple environmental parameters, wherein the disaster characteristic parameters include characteristic parameters of high temperature points in goaf and characteristic parameters of gas explosion hazard areas; For each disaster characteristic parameter, the multiple environmental parameters are used as input and the disaster characteristic parameter is used as output, and the inversion model corresponding to the disaster characteristic parameter is trained using multiple alternative training algorithms to obtain multiple trained inversion models, and the optimal inversion model corresponding to the disaster characteristic parameter is determined based on the multiple trained inversion models; the disaster characteristic parameter is inverted using the optimal inversion model corresponding to the disaster characteristic parameter to obtain the inversion result of each disaster characteristic parameter; According to the inversion results, a graded early warning is provided for the risk of spontaneous combustion of coal and gas coupling in the goaf.

2. The method for early warning of spontaneous combustion of coal and gas coupling in goaf as claimed in claim 1, characterized in that: The characteristic parameters of the high-temperature points in the goaf include: the temperature value and three-dimensional coordinates of the high-temperature points in the air intake half area, and the temperature value and three-dimensional coordinates of the high-temperature points in the return air half area; the characteristic parameters of the gas explosion hazard area include: the volume of the gas explosion hazard area, the three-dimensional coordinates of the explosive point and the volume fraction of each type of gas in the mixed gas; Among them, the air intake half zone and the return air half zone are divided by taking the inclination center plane of the goaf as the dividing interface; the gas explosion hazardous area is the area in the goaf where the porosity is greater than the critical porosity and the mixed gas is explosive, and the mixed gas contains oxygen, methane, carbon monoxide, carbon dioxide and nitrogen; the explosive point is the location of the maximum methane concentration in the gas explosion hazardous area.

3. The method for early warning of spontaneous combustion of coal and gas coupling in goaf as claimed in claim 2, characterized in that: The multiple environmental parameters include coal seam parameters, roof parameters, mining parameters, face ventilation parameters, goaf gas emission parameters, coal residue distribution parameters and face environmental parameters of the goaf. The construction method of the disaster characteristic parameter data set specifically includes: Determine the level numbers and level values ​​of the plurality of environmental parameters according to the field environment of the goaf, and determine the test plan and the number of tests according to the level numbers and level values; A three-dimensional numerical model of the goaf is established according to the test scheme, and a computer simulation test is performed on the oxidation self-heating process of the residual coal in the goaf based on the three-dimensional numerical model and the test times to obtain a simulation result; Based on the simulation results, extract the air inlet corner environmental parameters, the air return corner environmental parameters and the corresponding disaster characteristic parameter values ​​of the mining face; Based on the inlet wind corner environmental parameters, the return wind corner environmental parameters of the mining face, the disaster characteristic parameter values ​​and the values ​​of each environmental parameter, a data sample is obtained, and the disaster characteristic parameter data set is constructed based on the data sample; Among them, the coal seam parameters include coal seam thickness and coal seam inclination; the roof parameters include direct roof thickness and length of rock blocks falling from the old roof; the mining parameters include mining face length, mining face height and control top distance; the mining face ventilation parameters include mining face air volume, friction resistance coefficient and ventilation method; the goaf gas emission parameters include methane emission and carbon dioxide emission; the residual coal distribution parameters include width and thickness of each residual coal strip; the mining face environmental parameters include temperature and gas concentration of mining face air inlet tunnel, air inlet corner and return air corner.

4. The method for early warning of spontaneous combustion of coal and gas coupling in goaf as claimed in claim 3, characterized in that: For each disaster characteristic parameter, the multiple environmental parameters are used as input and the disaster characteristic parameter is used as output, and the inversion model corresponding to the disaster characteristic parameter is trained using multiple candidate training algorithms to obtain multiple trained inversion models, and the optimal inversion model corresponding to the disaster characteristic parameter is determined based on the multiple trained inversion models; the disaster characteristic parameter is inverted using the optimal inversion model corresponding to the disaster characteristic parameter to obtain the inversion result of each disaster characteristic parameter, which specifically includes: The disaster characteristic parameter data set is j The first disaster characteristic parameter is used as the dependent variable, the coal seam parameter, the roof parameter, the mining parameter, the mining face ventilation parameter, the residual coal distribution parameter, the mining face air intake environment parameter, the mining face air intake corner environment parameter and the mining face return air corner environment parameter are used as independent variables, and each of the multiple alternative training algorithms is used to train the first disaster characteristic parameter respectively. j The disaster characteristic parameters are used to train the inversion model, and the various candidate training algorithms are obtained for the first j The inversion model of the disaster characteristic parameters is provided, and the candidate training algorithms include a lightweight gradient boosting algorithm, a fast tree algorithm and a fast forest algorithm. j is a positive integer, 1≦j≦N, N is the number of disaster characteristic parameters; Get each candidate training algorithm about the first j The determination coefficient of the inversion model of the disaster characteristic parameters is determined, and the inversion model with the largest determination coefficient is determined as the first j The optimal inversion model of each disaster characteristic parameter; The actual value of the independent variable is input into the optimal inversion model for inversion to obtain the temperature value and three-dimensional coordinates of the high-temperature points in the air intake half of the goaf, the temperature value and three-dimensional coordinates of the high-temperature points in the return air half, the volume of the gas explosion hazardous area, the three-dimensional coordinates of the explosive point and the volume fraction of each type of gas in the mixed gas.

5. The method for early warning of spontaneous combustion of coal and gas coupling in goaf as claimed in claim 2, characterized in that: According to the inversion results, a graded early warning is carried out for the risk of spontaneous combustion of coal and gas coupling in the goaf, specifically including: Compare the temperature value of the high temperature point of the air inlet half zone with the temperature value of the high temperature point of the return air half zone, determine the high temperature point with the higher temperature value as the high temperature point of the goaf according to the comparison result, and calculate the Euclidean distance between the high temperature point and the explosive point; The temperature value of the high temperature point of the goaf, the volume of the gas explosion hazard zone, the volume fraction of methane at the explosive point, and the Euclidean distance between the high temperature point and the explosive point are determined as a risk basic factor set, and a first-level fuzzy comprehensive evaluation is performed based on the risk basic factor set and the evaluation set to obtain a comprehensive evaluation vector; Based on the comprehensive evaluation vector, the current risk warning level of the goaf is determined using the maximum membership principle.

6. The method for early warning of spontaneous combustion of coal and gas coupling in goaf as claimed in claim 4, characterized in that: After the inversion model with the largest determination coefficient is determined as the j After the optimal inversion model of the disaster characteristic parameters, it also includes: For the said j The performance of the optimal inversion model of each disaster characteristic parameter is tested and the test results are obtained; When the test result shows that the performance of the optimal inversion model is less than the preset performance index, j The inversion model of each disaster characteristic parameter is further optimized.

7. The goaf area coal spontaneous combustion and gas coupling early warning device is characterized by: include: A construction module is used to construct a disaster characteristic parameter data set about the coupling of spontaneous combustion of coal and gas in goaf, wherein the disaster characteristic parameter data set includes disaster characteristic parameters and multiple environmental parameters, wherein the disaster characteristic parameters include characteristic parameters of high temperature points in goaf and characteristic parameters of gas explosion hazard areas; An inversion module is used for, for each disaster characteristic parameter, using the multiple environmental parameters as input and the disaster characteristic parameter as output, training the inversion model corresponding to the disaster characteristic parameter using multiple alternative training algorithms to obtain multiple trained inversion models, and determining the optimal inversion model corresponding to the disaster characteristic parameter based on the multiple trained inversion models; inverting the disaster characteristic parameter using the optimal inversion model corresponding to the disaster characteristic parameter to obtain an inversion result for each disaster characteristic parameter; The early warning module is used to provide graded early warning for the risk of spontaneous combustion of coal and gas coupling in the goaf according to the inversion result.

8. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the method for early warning of spontaneous combustion of coal and gas coupling in goaf as claimed in any one of claims 1 to 6 is implemented.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for early warning of spontaneous combustion of coal and gas coupling in goaf as claimed in any one of claims 1 to 6 is implemented.

Images