Mine safety risk early warning method and system

Abstract

The invention discloses a mine safety risk early warning method and system, and relates to the technical field of mine safety early warning, and the method comprises the following steps: in the operation process of a fresh air system, obtaining various parameter information generated in the operation process of the fresh air system, and the parameter information comprises air flow characteristic information and gas diffusion characteristic information; according to the method, the airflow turbulence index and the gas diffusion delay index generated by the fresh air system are intelligently analyzed, the system automatically generates an operation state coefficient based on a machine learning model, high and low risk states are accurately divided, corresponding countermeasures are taken for burst type, unstable type and continuous type high risk states, and early warning or records are given out. According to the grading early warning mechanism, the risk identification precision is effectively improved, man-made misjudgment is reduced, it is ensured that serious risks are handled in time, and the safety and reliability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of mine safety early warning, and in particular to a mine safety risk early warning method and system. Background Art

[0002] Mine safety risk warning is to predict, evaluate and monitor potential safety hazards in the mining process, and issue early warning signals in time to prevent accidents such as mine collapse and gas explosion, and protect the lives and property of miners. The system collects environmental data in the mine (such as gas concentration, geological changes, etc.) in real time through sensors, and uses big data and artificial intelligence to analyze and predict risks and identify potential hazards. Based on the analysis results, the system issues early warnings in a timely manner and notifies relevant personnel to take measures, such as emergency evacuation or equipment maintenance, thereby effectively reducing the accident rate.

[0003] The role of the fresh air system in mining operations is mainly to ensure air circulation in the mine and provide a clean and safe working environment. First, it can transport fresh air from the surface to the depths of the mine to ensure that miners have sufficient oxygen supply and prevent health problems caused by lack of oxygen. Secondly, the fresh air system helps to discharge harmful gases accumulated in the mine, such as gas, carbon monoxide, sulfur dioxide, etc., and reduce safety hazards such as explosions and poisoning. The system can also regulate the temperature and humidity in the mine, improve the working environment, and improve the comfort and work efficiency of miners, while extending the service life of equipment and reducing the risk of accidents. Therefore, the fresh air system is an important facility to ensure the safety and health of mining operations.

[0004] The prior art has the following deficiencies:

[0005] One of the main functions of the fresh air system is to deliver enough oxygen to the mine. If the fresh air system is not properly operated and is not detected, the mine will be short of oxygen. Miners may experience dizziness, fatigue, confusion and other symptoms due to lack of oxygen. In severe cases, it may lead to collective suffocation and death. The underground environment lacks natural ventilation, so the supply of oxygen is particularly important. If oxygen cannot be replenished for a long time, the lives of miners will be directly threatened.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0007] The purpose of the present invention is to provide a mine safety risk early warning method and system, which can intelligently analyze the airflow turbulence index and gas diffusion delay index generated by the fresh air system, automatically generate the operating state coefficient based on the machine learning model, accurately divide the high and low risk states, and take corresponding response measures for sudden, unstable and continuous high-risk states, and issue early warnings or records. This hierarchical early warning mechanism effectively improves the accuracy of risk identification, reduces human misjudgment, ensures timely response to serious risks, and improves system safety and reliability to solve the problems in the above-mentioned background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solution: a mine safety risk early warning method, comprising the following steps:

[0009] During the operation of the fresh air system, various parameter information generated during the operation of the fresh air system is obtained, including air flow characteristic information and gas diffusion characteristic information;

[0010] Perform abnormal analysis on the collected air flow characteristic information and gas diffusion characteristic information, input the air flow characteristic information and gas diffusion characteristic information after abnormal analysis into the pre-learned machine model, generate the operation status coefficient through the machine model, and perform intelligent perception of the operation process of the fresh air system based on the operation status coefficient generated by the machine learning model;

[0011] Based on the results generated by the machine learning model, the operation process of the fresh air system is divided into a high-risk operation state or a low-risk operation state;

[0012] For the high-risk operating status assessed, continuously obtain various parameter information generated during the operation of the fresh air system, and further divide the high-risk operating status into sudden, unstable and continuous types;

[0013] Take different countermeasures for different operating conditions.

[0014] Preferably, the air flow characteristic information includes airflow stability, which refers to whether the air flows smoothly and evenly in the ventilation system, and measures whether there is turbulence and fluctuation in the airflow. The gas diffusion characteristic information includes gas mixing diffusion velocity, which refers to the velocity at which two or more gases are mixed and evenly distributed in space.

[0015] Preferably, an airflow turbulence index is generated after an abnormal analysis of the airflow stability, and a gas diffusion delay index is generated after an abnormal analysis of the gas mixing diffusion. The airflow turbulence index and the gas diffusion delay index are input into a pre-learned machine model, and an operating status coefficient is generated through the machine model. The operating process of the fresh air system is intelligently perceived based on the operating status coefficient generated by the machine learning model.

[0016] Preferably, the logic of performing abnormal analysis on airflow stability and generating an airflow turbulence index is as follows:

[0017] Under the detection window, the velocity field vector of the airflow is collected and expressed in the form of a three-dimensional velocity vector. The expression is as follows: v(x, y, z, t) = (v x (x,y,z,t),v y (x,y,z,t),v z (x,y,z,t)), v(x,y,z,t) represents the velocity field vector of the airflow at different positions x,y,z and time t in the mine, where x,y,z are spatial coordinates, representing the position of the airflow in the mine, and v x 、v y 、v z are the velocity components of the airflow in the x, y, and z directions, representing the instantaneous velocity of the airflow in the x, y, and z directions respectively;

[0018] Calculate the local velocity gradient tensor of the airflow. The local velocity gradient tensor is a 3*3 matrix expressed as follows:

[0019] , In the formula, They represent the velocity components v x For the rate of change in the x, y, and z directions, They represent the velocity components v y For the rate of change in the x, y, and z directions, They represent the velocity components v z For the rate of change in the x, y, and z directions;

[0021] The turbulent characteristics of the airflow are analyzed based on the vorticity of the airflow. The vorticity calculation expression is as follows: In the formula, is the curl operator, v is the velocity field vector, ω is the vorticity, and the three-dimensional components of the vorticity ω are as follows:

[0022] Calculate the energy dissipation rate in turbulence. The energy dissipation rate in turbulence reflects how the turbulent energy in the airflow gradually decays on a small scale. The calculation expression is as follows: ε = 2v Where ε is the energy dissipation rate, Represent the gradient vector of velocity in the x, y, and z directions respectively;

[0023] In order to quantify the turbulence intensity, the turbulence intensity coefficient is calculated. The calculation expression is as follows:

[0024] , In the formula, I T is the turbulence intensity coefficient, which indicates the intensity of disturbance in the airflow;

[0026] Based on the vorticity ω, energy dissipation rate ε and turbulence intensity coefficient I T , generate the airflow turbulence index, the calculation expression is as follows: Where, T i represents the airflow turbulence index, α is the adjustment coefficient of vorticity ω, β is the adjustment coefficient of energy dissipation rate ε, and γ is the adjustment coefficient of turbulence intensity coefficient.

[0027] Preferably, the logic of performing abnormal analysis on the gas mixture diffusion velocity and generating the gas diffusion delay index is as follows:

[0028] Under the detection window, calculate the diffusion speed of gas in the mine. The gas diffusion speed is related to the gas viscosity coefficient, gas temperature, gas pressure in the mine and the geometric characteristics of the airflow path. The calculation expression is as follows:

[0029] , Where V d represents the gas diffusion rate, k1 is the diffusion constant, which is set according to the specific gas type and mine conditions, T is the gas temperature, η is the viscosity coefficient of the gas, P is the gas pressure in the mine, and D is the characteristic length of the gas flow path;

[0031] The effectiveness of gas diffusion is further evaluated by calculating the mixing efficiency of the gas during the diffusion process. The calculation expression is as follows:

[0032] , In the formula, E m represents the gas mixing efficiency, k2 is the mixing constant, L is the effective path length of the air flow in the mine, R is the rotation radius of the airflow during the diffusion process, and t d is the gas diffusion time;

[0034] Based on the gas diffusion velocity and gas mixing efficiency, the gas diffusion delay index is finally generated. The calculation expression is as follows:

[0035] , Where GDI represents the gas diffusion delay index and k3 is the delay correction factor.

[0037] Preferably, the airflow turbulence index T generated during the operation of the fresh air system is i The gas diffusion delay index GDI is input into the pre-learned machine learning model, and the operating state coefficient OCC is generated by the machine model according to the following formula:

[0038] , Where f1 and f2 are the airflow turbulence index T i And the preset proportional coefficient of the gas diffusion delay index GDI, and both f1 and f2 are greater than 0.

[0040] Preferably, the operating state coefficient generated during the operation of the fresh air system is compared and analyzed with a preset operating state coefficient reference threshold, and the analysis results are as follows:

[0041] If the operating state coefficient is greater than or equal to a preset operating state coefficient reference threshold, the fresh air system operation process is classified as a high-risk operating state;

[0042] If the operating status coefficient is less than a preset operating status coefficient reference threshold, the fresh air system operation process is classified as a low-risk operating state.

[0043] Preferably, the average value and standard deviation of several operating state coefficients in the data set are obtained, and the obtained average value and standard deviation of the operating state coefficient are calibrated as OCC 平均 and OCC 标准 , and the calculated average operating state coefficient OCC 平均 The standard deviation of the operating state coefficient OCC 标准 The preset operating state coefficient average reference threshold OCC 参考平均 And the operating state coefficient standard deviation reference threshold OCC 参考标准 The comparative analysis results are as follows:

[0044] If OCC 平均 ≥OCC 参考平均 , then the operation process of the fresh air system is further divided into continuous type;

[0045] like Then the operation process of the fresh air system is further divided into unstable type;

[0046] The operation process of the fresh air system is further divided into sudden types.

[0047] A mine safety risk early warning system, comprising a parameter acquisition module, an abnormal analysis and model input module, a risk assessment module, a high-risk state classification module and a response strategy module;

[0048] The parameter acquisition module obtains various parameter information generated during the operation of the fresh air system, including air flow characteristic information and gas diffusion characteristic information;

[0049] The regular analysis and model input module performs abnormal analysis on the collected air flow characteristic information and gas diffusion characteristic information, inputs the air flow characteristic information and gas diffusion characteristic information after abnormal analysis into the pre-learned machine model, generates the operation status coefficient through the machine model, and performs intelligent perception of the operation process of the fresh air system based on the operation status coefficient generated by the machine learning model;

[0050] The risk assessment module divides the operation process of the fresh air system into a high-risk operation state or a low-risk operation state based on the results generated by the machine learning model;

[0051] The high-risk state classification module continuously obtains various parameter information generated during the operation of the fresh air system for the assessed high-risk operation state, and further divides the high-risk operation state into sudden, unstable and continuous types;

[0052] The response strategy module takes different response measures for different operating states.

[0053] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0054] The present invention can generate an operating state coefficient representing the operating state of the system by intelligently analyzing the airflow turbulence index and gas diffusion delay index generated during the operation of the fresh air system, and inputting the results into a pre-learned machine learning model. Based on this coefficient, the system can automatically sense the operating status of the fresh air system and intelligently divide it into high-risk or low-risk states. This intelligent perception process not only improves the accuracy of risk identification, but also greatly reduces the error of human judgment. At the same time, for high-risk operating states, it is further divided into sudden, unstable and continuous types. For different risk types, the system can take corresponding countermeasures. For example, for unstable and continuous high-risk states, the system issues a first-level and a second-level warning to prompt relevant staff to carry out maintenance to ensure that timely action is taken before the problem worsens. This hierarchical warning system can effectively improve the risk response speed and reduce the possibility of accidents.

[0055] The present invention achieves accurate distinction between different high-risk operating states by carefully analyzing the mean value and standard deviation of the operating state coefficient. In terms of response strategy, for sudden high-risk operating states, the system does not immediately issue an early warning, but records it, which helps to reduce unnecessary interference and avoid overreaction to short-term fluctuations. However, for unstable and continuous high-risk states, the system will issue level one and level two warnings respectively. Such a hierarchical early warning mechanism can not only prevent minor fluctuations from causing misjudgments, but also ensure that effective response measures are taken in a timely manner when serious risks arise. In addition, the system's automatic recording and analysis of these high-risk states also provides an important basis for subsequent review, summary and optimization of the fresh air system. This optimized risk response strategy significantly improves the reliability and safety of the system and reduces the complexity of human intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0057] Figure 1 The present invention is a method flow chart of a mine safety risk early warning method.

[0058] Figure 2 This is a module schematic diagram of a mine safety risk early warning system of the present invention. DETAILED DESCRIPTION

[0059] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0060] The present invention provides Figure 1 A mine safety risk early warning method shown includes the following steps:

[0061] During the operation of the fresh air system, various parameter information generated during the operation of the fresh air system is obtained, including air flow characteristic information and gas diffusion characteristic information

[0062] The air flow characteristic information includes air flow stability, which refers to whether the air flows smoothly and evenly in the ventilation system, and measures whether there is turbulence and fluctuation in the air flow. The gas diffusion characteristic information includes gas mixing diffusion speed, which refers to the speed at which two or more gases are mixed and evenly distributed in space.

[0063] Perform abnormal analysis on the collected air flow characteristic information and gas diffusion characteristic information, input the air flow characteristic information and gas diffusion characteristic information after abnormal analysis into the pre-learned machine model, generate the operation status coefficient through the machine model, and perform intelligent perception of the operation process of the fresh air system based on the operation status coefficient generated by the machine learning model;

[0064] After abnormal analysis of airflow stability, an airflow turbulence index is generated. After abnormal analysis of gas mixture diffusion, a gas diffusion delay index is generated. The airflow turbulence index and the gas diffusion delay index are input into a pre-learned machine model, and an operation status coefficient is generated through the machine model. Based on the operation status coefficient generated by the machine learning model, the operation process of the fresh air system is intelligently perceived.

[0065] Changes in air flow stability during the operation of the fresh air system may cause the system to operate abnormally, resulting in insufficient oxygen supply in the mine. Air flow stability is an important factor in the effective air delivery of the fresh air system. When the air flow stability decreases, turbulence or irregular fluctuations will occur in the air flow, resulting in uneven distribution of fresh air inside the mine. In a turbulent state, the air flow becomes irregular, which may lead to insufficient fresh air delivery in some areas, while other areas may have excessive air concentration. This uneven air distribution directly affects the effective supply of oxygen in various working areas of the mine, resulting in a decrease in the oxygen concentration in the miners' working areas. In addition, air flow disturbances may also affect the timely discharge of harmful gases, such as methane and carbon monoxide cannot be quickly diluted and discharged, further exacerbating the risk of miners' hypoxia. Over time, if the fresh air system cannot effectively provide enough fresh air and maintain air quality, miners may experience health problems such as dizziness, fatigue, and even suffocation caused by hypoxia.

[0066] The logic of abnormal analysis of airflow stability and generation of airflow turbulence index is as follows:

[0067] Under the detection window, the velocity field vector of the airflow is collected and expressed in the form of a three-dimensional velocity vector. The expression is as follows: v(x, y, z, t) = (v x (x,y,z,t),v y (x,y,z,t),v z (x,y,z,t)), v(x,y,z,t) represents the velocity field vector of the airflow at different positions x,y,z and time t in the mine, where x,y,z are spatial coordinates, representing the position of the airflow in the mine, and v x 、v y 、v z are the velocity components of the airflow in the x, y, and z directions, representing the instantaneous velocity of the airflow in the x, y, and z directions respectively;

[0068] It should be noted that velocity field data refers to the velocity distribution of each point in a certain spatial region. It describes the flow state of a fluid (such as air or water) in the region as it changes with time and position. Velocity field data is usually represented by a three-dimensional velocity vector, because the flow of a fluid involves not only the magnitude of the velocity but also the direction of the flow. A three-dimensional velocity vector is used to represent the velocity at each position in three-dimensional space, and each velocity component v x 、v y 、v z Respectively represent the magnitude of velocity in the x, y, and z directions. This representation method can fully describe the motion characteristics of the fluid at each point in space and is suitable for describing complex flow phenomena such as turbulence.

[0069] Calculate the local velocity gradient tensor of the airflow. The local velocity gradient tensor is a 3*3 matrix expressed as follows:

[0070] , where They represent the velocity components v x For the rate of change in the x, y, and z directions, They represent the velocity components v y For the rate of change in the x, y, and z directions, They represent the velocity components v z For the rate of change in the x, y, and z directions;

[0072] It should be noted that the local velocity gradient tensor is a mathematical tool that describes the rate of change of velocity of a fluid (such as gas or liquid) in all directions in space. It is a 3*3 matrix that represents how the velocity components of the fluid in the three coordinate directions change with the change of position. Specifically, each element of the velocity gradient tensor represents the rate of change of the velocity of the fluid in a certain direction with respect to the spatial coordinate. It is used to describe the motion characteristics of the fluid such as stretching, compression, shearing and rotation in a local area, and helps to understand the deformation and turbulent behavior of the fluid.

[0073] The turbulent characteristics of the airflow are analyzed based on the vorticity of the airflow. The vorticity calculation expression is as follows: In the formula, is the curl operator, v is the velocity field vector, ω is the vorticity, and the three-dimensional components of the vorticity ω are as follows:

[0074] It should be noted that vorticity refers to the intensity and direction of rotation of a fluid (such as air or water) at a certain point. It reflects the rotation or vortex phenomenon of the fluid in a local area. Specifically, vorticity is obtained by calculating the curl of the velocity field, which indicates the rotation speed and direction of the fluid at a certain point. The magnitude of the vorticity describes the intensity of the fluid's rotation, while the direction of the vorticity indicates the axis of rotation. The larger the vorticity, the stronger the rotation of the fluid, which is very common in turbulent or complex fluid flows.

[0075] Calculate the energy dissipation rate in turbulence. The energy dissipation rate in turbulence reflects how the turbulent energy in the airflow gradually decays on a small scale. The calculation expression is as follows: Where ε is the energy dissipation rate, Represent the gradient vector of velocity in the x, y, and z directions respectively;

[0076] In order to quantify the turbulence intensity, the turbulence intensity coefficient is calculated. The calculation expression is as follows:

[0077]

[0078] , where I T is the turbulence intensity coefficient, which indicates the intensity of disturbance in the airflow;

[0079] Based on the vorticity ω, energy dissipation rate ε and turbulence intensity coefficient I T , generate the airflow turbulence index, the calculation expression is as follows: Where, T i represents the airflow turbulence index, α is the adjustment coefficient of the vorticity ω, T is the adjustment coefficient of the energy dissipation rate ε, and γ is the adjustment coefficient of the turbulence intensity coefficient;

[0080] It can be seen from the calculation expression of the airflow turbulence index that, under the detection window, the larger the airflow turbulence index performance value generated after abnormal analysis of airflow stability, the higher the turbulence intensity in the airflow and the greater the instability of the airflow. This means that the operation of the fresh air system may be abnormal, because turbulence will affect the uniform distribution of air, resulting in the inability to effectively deliver fresh air to various working areas in the mine, thereby increasing the risk of insufficient oxygen supply. Conversely, the smaller the airflow turbulence index performance value, the more stable the airflow, the air can flow smoothly and be evenly distributed to various areas in the mine, reducing the possibility of insufficient oxygen supply.

[0081] During the operation of the fresh air system, if the gas mixing and diffusion rate changes, it may cause the fresh air system to operate abnormally, resulting in insufficient oxygen supply in the mine. The gas mixing and diffusion rate refers to the mutual mixing, dilution and diffusion rate of fresh air and air in the mine (including harmful gases). When the diffusion rate slows down, fresh air cannot be quickly and evenly distributed to various areas of the mine, and the oxygen delivery efficiency is greatly reduced, which may lead to insufficient oxygen concentration in some working areas. At the same time, harmful gases in the mine (such as gas and carbon monoxide) cannot be diluted and discharged in time, which will further aggravate the lack of oxygen. Changes in gas diffusion rate are often related to air turbulence, uneven pressure difference or blockage of ventilation ducts. These factors will cause poor air flow in the mine, making it impossible for fresh air to quickly cover the entire mine, especially in some deep places or places where the air flow is slow, the decrease in oxygen concentration will directly threaten the safety of miners. Therefore, the reduction in gas mixing and diffusion rate will seriously affect the air quality and safety in the mine, leading to insufficient oxygen supply, thereby increasing the risk of mine accidents.

[0082] The logic of abnormal analysis of gas mixture diffusion velocity and generation of gas diffusion delay index is as follows:

[0083] Under the detection window, calculate the diffusion speed of gas in the mine. The gas diffusion speed is related to the gas viscosity coefficient, gas temperature, gas pressure in the mine and the geometric characteristics of the airflow path. The calculation expression is as follows:

[0084]

[0085] , where V d represents the gas diffusion rate, k1 is the diffusion constant, which is set according to the specific gas type and mine conditions, T is the gas temperature, η is the viscosity coefficient of the gas, P is the gas pressure in the mine, and D is the characteristic length of the gas flow path;

[0086] It should be noted that the viscosity coefficient of a gas is a physical quantity that measures the strength of the interaction between gas molecules and the size of the resistance generated when the gas flows internally. When the gas flows, collisions and friction occur between the gas molecules, resulting in resistance in the flow process. The larger the viscosity coefficient, the greater the resistance encountered when the gas flows, and the flow speed slows down accordingly. The viscosity coefficient is closely related to factors such as the temperature and pressure of the gas. Generally, as the temperature increases, the viscosity coefficient of the gas increases. This parameter plays an important role in fluid dynamics, especially in ventilation systems, where the viscosity coefficient affects the diffusion rate and flow efficiency of the gas.

[0087] The effectiveness of gas diffusion is further evaluated by calculating the mixing efficiency of the gas during the diffusion process. The calculation expression is as follows:

[0088]

[0089] , where E m represents the gas mixing efficiency, k2 is the mixing constant, L is the effective path length of the air flow in the mine, R is the rotation radius of the airflow during the diffusion process, and t d is the gas diffusion time;

[0090] Based on the gas diffusion velocity and gas mixing efficiency, the gas diffusion delay index is finally generated. The calculation expression is as follows:

[0091]

[0092] , where GDI represents the gas diffusion delay index and k3 is the delay correction factor;

[0093] The delay correction factor is a parameter used to correct or adjust the delay effect caused by complex environmental conditions in the system. It takes into account the impact of environmental factors (such as temperature changes, path complexity, resistance, turbulence, etc.) on system response or material transmission speed in actual operation, which may cause longer or shorter delays than expected under ideal conditions. By introducing the delay correction factor, these uncertainties can be more accurately reflected and compensated, ensuring that the calculation or model is closer to the actual situation, thereby improving the accuracy of the prediction and the stability of the system.

[0094] From the calculation expression of the gas diffusion delay index, it can be seen that the larger the gas diffusion delay index performance value generated after the abnormal analysis of the gas mixture diffusion velocity, the more serious the delay of the gas mixture diffusion velocity, indicating that the possibility of abnormal operation of the fresh air system is higher. This delay causes fresh air to be unable to diffuse quickly and evenly to various areas of the mine, and the risk of insufficient oxygen supply increases. The oxygen concentration in the mine may decrease, endangering the lives of miners. On the contrary, the smaller the gas diffusion delay index performance value, the smoother the diffusion process, the effective circulation of gas, the normal operation of the fresh air system, and the stability of the oxygen supply is guaranteed, so the probability of insufficient oxygen in the mine is small.

[0095] The airflow turbulence index T generated during the operation of the fresh air system i The gas diffusion delay index GDI is input into the pre-learned machine learning model, and the operating state coefficient OCC is generated by the machine model according to the following formula:

[0096]

[0097] , where f1 and f2 are the airflow turbulence index T i And the preset proportional coefficient of the gas diffusion delay index GDI, and both f1 and f2 are greater than 0.

[0098] From the calculation expression of the operating state coefficient, it can be seen that the larger the airflow turbulence index expression value generated after the abnormal analysis of the airflow stability, the larger the gas diffusion delay index expression value generated after the abnormal analysis of the gas mixing diffusion velocity, which means that the larger the operating state coefficient expression value, the greater the probability of abnormality in the operation process of the fresh air system, and vice versa, the smaller the probability of abnormality in the operation process of the fresh air system.

[0099] Based on the results generated by the machine learning model, the operation process of the fresh air system is divided into a high-risk operation state or a low-risk operation state;

[0100] The operating status coefficient generated during the operation of the fresh air system is compared and analyzed with the preset operating status coefficient reference threshold. The analysis results are as follows:

[0101] If the operating state coefficient is greater than or equal to a preset operating state coefficient reference threshold, the fresh air system operation process is classified as a high-risk operating state;

[0102] If the operating state coefficient is less than the preset operating state coefficient reference threshold, the fresh air system operation process is classified as a low-risk operating state;

[0103] For the high-risk operating status assessed, continuously obtain various parameter information generated during the operation of the fresh air system, and further divide the high-risk operating status into sudden, unstable and continuous types;

[0104] When the operation process of the fresh air system is classified as a high-risk operation state, a number of operation state coefficients generated in the subsequent operation process of the fresh air system are continuously obtained to establish a data set, and a comprehensive analysis is performed on the operation state coefficients in the data set;

[0105] Calculate the mean and standard deviation of several operating state coefficients in the data set, and mark the calculated mean and standard deviation of the operating state coefficients as OCC 平均 and OCC 标准 , and the calculated average operating state coefficient OCC 平均 The standard deviation of the operating state coefficient OCC 标准 The preset operating state coefficient average reference threshold OCC 参考平均 And the operating state coefficient standard deviation reference threshold OCC 参考标准 The comparative analysis results are as follows:

[0106] If OCC 平均 ≥OCC 参考平均 , then the operation process of the fresh air system is further divided into continuous type;

[0107] like Then the operation process of the fresh air system is further divided into unstable type;

[0108] like The operation process of the fresh air system is further divided into sudden types.

[0109] Take different countermeasures for different operating states;

[0110] For sudden high-risk operating conditions, given that they may recover on their own in a short period of time, no warning information will be issued, but records need to be made for later review and analysis. For unstable and continuous high-risk operating conditions, a first-level warning (more serious warning) and a second-level warning (more serious warning) will be issued to prompt relevant staff to inspect the operating status of the fresh air system.

[0111] The present invention can generate an operating state coefficient representing the operating state of the system by intelligently analyzing the airflow turbulence index and gas diffusion delay index generated during the operation of the fresh air system, and inputting the results into a pre-learned machine learning model. Based on this coefficient, the system can automatically sense the operating status of the fresh air system and intelligently divide it into high-risk or low-risk states. This intelligent perception process not only improves the accuracy of risk identification, but also greatly reduces the error of human judgment. At the same time, for high-risk operating states, it is further divided into sudden, unstable and continuous types. For different risk types, the system can take corresponding countermeasures. For example, for unstable and continuous high-risk states, the system issues a first-level and a second-level warning to prompt relevant staff to carry out maintenance to ensure that timely action is taken before the problem worsens. This hierarchical warning system can effectively improve the risk response speed and reduce the possibility of accidents.

[0112] The present invention achieves accurate distinction between different high-risk operating states by carefully analyzing the mean value and standard deviation of the operating state coefficient. In terms of response strategy, for sudden high-risk operating states, the system does not immediately issue an early warning, but records it, which helps to reduce unnecessary interference and avoid overreaction to short-term fluctuations. However, for unstable and continuous high-risk states, the system will issue level one and level two warnings respectively. Such a hierarchical early warning mechanism can not only prevent minor fluctuations from causing misjudgments, but also ensure that effective response measures are taken in a timely manner when serious risks arise. In addition, the system's automatic recording and analysis of these high-risk states also provides an important basis for subsequent review, summary and optimization of the fresh air system. This optimized risk response strategy significantly improves the reliability and safety of the system and reduces the complexity of human intervention.

[0113] The present invention provides Figure 2As shown, a mine safety risk early warning system includes a parameter acquisition module, an abnormal analysis and model input module, a risk assessment module, a high-risk state classification module and a response strategy module;

[0114] The parameter acquisition module obtains various parameter information generated during the operation of the fresh air system, including air flow characteristic information and gas diffusion characteristic information;

[0115] The regular analysis and model input module performs abnormal analysis on the collected air flow characteristic information and gas diffusion characteristic information, inputs the air flow characteristic information and gas diffusion characteristic information after abnormal analysis into the pre-learned machine model, generates the operation status coefficient through the machine model, and performs intelligent perception of the operation process of the fresh air system based on the operation status coefficient generated by the machine learning model;

[0116] The risk assessment module divides the operation process of the fresh air system into a high-risk operation state or a low-risk operation state based on the results generated by the machine learning model;

[0117] The high-risk state classification module continuously obtains various parameter information generated during the operation of the fresh air system for the assessed high-risk operation state, and further divides the high-risk operation state into sudden, unstable and continuous types;

[0118] The response strategy module takes different response measures for different operating states.

[0119] A mine safety risk warning method provided in an embodiment of the present invention is implemented through the above-mentioned mine safety risk warning system. The specific method and process of a mine safety risk warning system are detailed in the embodiment of the above-mentioned mine safety risk warning method, which will not be repeated here.

[0120] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A mine safety risk early warning method, characterized in that: The following steps are involved: During the operation of the fresh air system, various parameter information generated during the operation of the fresh air system is obtained, including air flow characteristic information and gas diffusion characteristic information; Perform abnormal analysis on the collected air flow characteristic information and gas diffusion characteristic information, input the air flow characteristic information and gas diffusion characteristic information after abnormal analysis into the pre-learned machine model, generate the operation status coefficient through the machine model, and perform intelligent perception of the operation process of the fresh air system based on the operation status coefficient generated by the machine learning model; Based on the results generated by the machine learning model, the operation process of the fresh air system is divided into a high-risk operation state or a low-risk operation state; For the high-risk operating status assessed, continuously obtain various parameter information generated during the operation of the fresh air system, and further divide the high-risk operating status into sudden, unstable and continuous types; Take different countermeasures for different operating conditions.

2. A mine safety risk early warning method according to claim 1, characterized in that: The air flow characteristic information includes airflow stability, which refers to whether the air flows smoothly and evenly in the ventilation system. It measures whether there is turbulence and fluctuation in the airflow. The gas diffusion characteristic information includes the gas mixing diffusion rate, which refers to the speed at which two or more gases are mixed and evenly distributed in space.

3. A mine safety risk early warning method according to claim 2, characterized in that: After abnormal analysis of airflow stability, an airflow turbulence index is generated. After abnormal analysis of gas mixing diffusion, a gas diffusion delay index is generated. The airflow turbulence index and the gas diffusion delay index are input into a pre-learned machine model, and an operating status coefficient is generated through the machine model. The operating process of the fresh air system is intelligently perceived based on the operating status coefficient generated by the machine learning model.

4. A mine safety risk early warning method according to claim 3, characterized in that: The logic of abnormal analysis of airflow stability and generation of airflow turbulence index is as follows: Under the detection window, the velocity field vector of the airflow is collected and expressed in the form of a three-dimensional velocity vector. The expression is as follows: v(x, y, z, t) = (v x (x,y,z,t),v y (x,y,z,t),v z (x,y,z,t)), v(x,y,z,t) represents the velocity field vector of the airflow at different positions x,y,z and time t in the mine, where x,y,z are spatial coordinates, representing the position of the airflow in the mine, and v x 、v y 、v z are the velocity components of the airflow in the x, y, and z directions, representing the instantaneous velocity of the airflow in the x, y, and z directions respectively; Calculate the local velocity gradient tensor of the airflow. The local velocity gradient tensor is a 3*3 matrix expressed as follows: , where They represent the velocity components v x For the rate of change in the x, y, and z directions, They represent the velocity components v y For the rate of change in the x, y, and z directions, They represent the velocity components v z For the rate of change in the x, y, and z directions; The turbulent characteristics of the airflow are analyzed based on the vorticity of the airflow. The vorticity calculation expression is as follows: In the formula, is the curl operator, v is the velocity field vector, ω is the vorticity, and the three-dimensional components of the vorticity ω are as follows: Calculate the energy dissipation rate in turbulence. The energy dissipation rate in turbulence reflects how the turbulent energy in the airflow gradually decays on a small scale. The calculation expression is as follows: Where ε is the energy dissipation rate, Represent the gradient vector of velocity in the x, y, and z directions respectively; In order to quantify the turbulence intensity, the turbulence intensity coefficient is calculated. The calculation expression is as follows: , where I T is the turbulence intensity coefficient, which indicates the intensity of disturbance in the airflow; Based on the vorticity ω, energy dissipation rate ε and turbulence intensity coefficient I T , generate the airflow turbulence index, the calculation expression is as follows: Where, T i represents the airflow turbulence index, α is the adjustment coefficient of vorticity ω, β is the adjustment coefficient of energy dissipation rate ε, and γ is the adjustment coefficient of turbulence intensity coefficient.

5. A mine safety risk early warning method according to claim 3, characterized in that: The logic of abnormal analysis of gas mixture diffusion velocity and generation of gas diffusion delay index is as follows: Under the detection window, calculate the diffusion speed of gas in the mine. The gas diffusion speed is related to the gas viscosity coefficient, gas temperature, gas pressure in the mine and the geometric characteristics of the airflow path. The calculation expression is as follows: ， Where V d represents the gas diffusion rate, k1 is the diffusion constant, which is set according to the specific gas type and mine conditions, T is the gas temperature, η is the viscosity coefficient of the gas, P is the gas pressure in the mine, and D is the characteristic length of the gas flow path; The effectiveness of gas diffusion is further evaluated by calculating the mixing efficiency of the gas during the diffusion process. The calculation expression is as follows: ， In the formula, E m represents the gas mixing efficiency, k2 is the mixing constant, L is the effective path length of the air flow in the mine, R is the rotation radius of the airflow during the diffusion process, and t d is the gas diffusion time; Based on the gas diffusion velocity and gas mixing efficiency, the gas diffusion delay index is finally generated. The calculation expression is as follows: ， Where GDI represents the gas diffusion delay index and k3 is the delay correction factor.

6. A mine safety risk early warning method according to claim 3, characterized in that: The airflow turbulence index T generated during the operation of the fresh air system i The gas diffusion delay index GDI is input into the pre-learned machine learning model, and the operating state coefficient OCC is generated by the machine model according to the following formula: ， Where f1 and f2 are the airflow turbulence index T i And the preset proportional coefficient of the gas diffusion delay index GDI, and both f1 and f2 are greater than 0.

7. A mine safety risk early warning method according to claim 6, characterized in that: The operating status coefficient generated during the operation of the fresh air system is compared and analyzed with the preset operating status coefficient reference threshold. The analysis results are as follows: If the operating state coefficient is greater than or equal to a preset operating state coefficient reference threshold, the fresh air system operation process is classified as a high-risk operating state; If the operating status coefficient is less than a preset operating status coefficient reference threshold, the fresh air system operation process is classified as a low-risk operating state.

8. A mine safety risk early warning method according to claim 7, characterized in that: Calculate the mean and standard deviation of several operating state coefficients in the data set, and mark the calculated mean and standard deviation of the operating state coefficients as OCC 平均 and OCC 标准 , and the calculated average operating state coefficient OCC 平均 The standard deviation of the operating state coefficient OCC 标准 The preset operating state coefficient average reference threshold OCC 参考平均 And the operating state coefficient standard deviation reference threshold OCC 参考标准 The comparative analysis results are as follows: If OCC 平均 ≥OCC 参考平均 , then the operation process of the fresh air system is further divided into continuous type; like Then the operation process of the fresh air system is further divided into unstable type; like The operation process of the fresh air system is further divided into sudden types.

9. A mine safety risk early warning system, used to implement a mine safety risk early warning method as described in any one of claims 1 to 8, characterized in that: It includes parameter acquisition module, abnormal analysis and model input module, risk assessment module, high-risk status classification module and response strategy module; The parameter acquisition module obtains various parameter information generated during the operation of the fresh air system, including air flow characteristic information and gas diffusion characteristic information; The regular analysis and model input module performs abnormal analysis on the collected air flow characteristic information and gas diffusion characteristic information, inputs the air flow characteristic information and gas diffusion characteristic information after abnormal analysis into the pre-learned machine model, generates the operation status coefficient through the machine model, and performs intelligent perception of the operation process of the fresh air system based on the operation status coefficient generated by the machine learning model; The risk assessment module divides the operation process of the fresh air system into a high-risk operation state or a low-risk operation state based on the results generated by the machine learning model; The high-risk state classification module continuously obtains various parameter information generated during the operation of the fresh air system for the assessed high-risk operation state, and further divides the high-risk operation state into sudden, unstable and continuous types; The response strategy module takes different response measures for different operating states.

Images