Mine ventilation control method, electronic device, and program product

Abstract

This invention discloses a mine ventilation control method, electronic equipment, and program product. The mine ventilation control method uses a built-in miniature sensor in the breathing mask to collect air quality parameters in the breathing zone of the breathing mask in real time, and simultaneously obtains the current location coordinates of the miner. The air quality parameters are used to calculate the real-time hazard level of the breathing zone by combining the corresponding safe thresholds for carbon monoxide concentration, fine particulate matter concentration, oxygen concentration, and the minimum safe wind speed when the particulate matter concentration exceeds the standard. When the preset hazard threshold is exceeded, the method generates and issues control commands for the rotation speed, damper angle, and / or airflow direction of the ventilation fans installed in the mine roadway, based on the miner's location coordinates, so as to minimize the hazard level in subsequent moments.

Description

Technical Field

[0001] This disclosure belongs to the field of mine safety technology, and specifically relates to a mine ventilation control method, electronic equipment and program products. Background Technology

[0002] Mine ventilation is a crucial aspect of ensuring safety in underground operations. Its primary purpose is to deliver fresh air underground and dilute and remove harmful gases and dust. Traditional mine ventilation systems have several shortcomings. First, existing sensors are typically installed at the top or side of the tunnel, monitoring average air quality but failing to accurately reflect the actual concentration of pollutants in the miners' breathing zone (around their faces) in real time. By the time harmful substances are detected to exceed safe levels, miners may already be exposed to a hazardous environment. Second, most existing systems rely on simple control logic that adjusts ventilation based on fixed thresholds. This approach is ill-suited to the complex and ever-changing underground environment, easily leading to insufficient ventilation. Finally, they lack predictive and adaptive capabilities. Traditional systems cannot predict pollutant diffusion patterns based on historical data and real-time trends for preventative control. Especially in the event of catastrophic accidents such as mine disasters, existing ventilation control systems cannot provide accurate guidance for rescue and evacuation. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art and provide a mine intelligent ventilation method based on breathing zone monitoring and AI decision-making, so as to achieve accurate real-time monitoring of air quality in the breathing zone of each miner and dynamically optimize ventilation strategy using AI model.

[0004] One aspect of this disclosure is a method for predicting and controlling mine ventilation based on respiratory zone monitoring and intelligent AI decision-making. The method includes the following steps:

[0005] A1, using miniature sensors built into the breathing masks worn by miners to collect air quality parameters in the breathing zone of the masks in real time, including at least carbon monoxide concentration. Fine particulate matter concentration Oxygen concentration and breathing zone wind speed And simultaneously obtain the miner's current location coordinates;

[0006] A2, compare the air quality parameters with the corresponding safe threshold for carbon monoxide concentration in the tunnel. Safety threshold for fine particulate matter concentration Oxygen concentration safety threshold and the minimum safe wind speed when particulate matter concentration exceeds the standard. Substitute into the following formula to calculate the real-time hazard level of the respiratory zone. ,

[0007]

[0008] Here, It refers to the specific location of a miner in the mine. It is time. , , , The weights of carbon monoxide, fine particulate matter, oxygen, and safe wind speed are respectively determined. Take any minimum positive value;

[0009] A3, when When the preset danger threshold is exceeded, ventilation optimization is initiated. Using H(x,t) as the objective function and combining the miner's location coordinates, control commands are generated and issued to the ventilation fans installed in the mine roadway regarding the speed, damper angle, and / or airflow direction, thereby reducing the danger level in the breathing zone. Minimize in subsequent time steps.

[0010] The miners are equipped with RFID positioning trackers. The data collected by their breathing masks, along with positioning signal data, are transmitted to a central ventilation control server located on the ground via wireless data transceivers with fixed coordinates deployed in the tunnels.

[0011] The central ventilation control server is equipped with a mine ventilation control model based on machine learning. This model is trained and optimized based on historical fan control parameters. The data collected from miners' breathing masks, miners' positioning data, and environmental data obtained from fixed collection points in the roadway are processed using the aforementioned method, and control commands are output for each ventilation fan in the roadway. Attached Figure Description

[0012] The above and other objects, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description taken in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated in the drawings by way of example and not limitation, wherein:

[0013] Figure 1 A schematic diagram of a ventilation control method according to one embodiment of the present invention. Detailed Implementation

[0014] According to one or more embodiments, this disclosure proposes a mine ventilation prediction and control system, which, from the perspective of the composition of the control system, includes the following different layers.

[0015] (1) Monitoring level, which includes different data sources:

[0016] Intelligent breathing mask - Miners wear intelligent protective masks with built-in miniature sensors that monitor key air quality parameters in the breathing zone of the roadway in real time, including but not limited to dust concentration (PM2.5, PM10) and concentration of toxic and harmful gases, and periodically transmit the parameters to the mine air quality monitoring layer of the system.

[0017] Fixed monitoring points – Fixed environmental monitoring sensors are deployed at key nodes such as mine roadways and working faces to collect overall environmental data at the monitoring points, including pollutant concentration, ambient temperature, humidity, wind direction, wind force, etc.

[0018] Personnel / Equipment Positioning Module – Tracks the location of each miner in real time using RFID locators worn by the miners.

[0019] (2) Data transmission layer, including the following processing procedures:

[0020] Data collected by miners' smart masks and fixed monitoring points is transmitted in real-time with low latency to a central control server on the surface via the mine's internal wireless communication network. Specific data includes: mine temperature and humidity (fixed-point monitoring), dust concentration (fixed-point monitoring and mask monitoring), harmful gas concentration (fixed-point monitoring and mask monitoring), and wind speed and direction (fixed-point monitoring). The data processing methods include:

[0021] Outlier removal using IQR / Isolation Forest methods, and / or,

[0022] Standardize the time frame, for example, to a baseline of 1 second; use linear interpolation / spline fitting for missing data, and / or...

[0023] The data is spatially mapped. The mapping process involves mapping and matching the data collected from the miner's mask and the data collected from fixed detection points to specific roadway cross-sections based on the location data, forming roadway data cross-sections within the mine. These data cross-sections are then mapped to specific three-dimensional mine roadway grid units, forming a comprehensive perception and display of the roadway environment data by the system.

[0024] (3) Intelligent control level, including:

[0025] The central control server has a built-in AI intelligent mine ventilation control model for outputting ventilation control decisions in the mine.

[0026] The model includes a data fusion and anomaly diagnosis module, which integrates data from breathing masks, fixed monitoring points, and positioning systems to generate a real-time dynamic pollution cloud map of the entire mine area and intelligently diagnose pollution sources and diffusion trends.

[0027] The mine ventilation control model is trained based on historical ventilation data, environmental data, and expert experience to learn the relationship between pollutant diffusion patterns and ventilation effectiveness under different working conditions (such as tunneling, blasting, and transportation). Its real-time decision-making process involves the model receiving real-time environmental data and considering not only the current pollutant concentration but also factors such as personnel distribution and equipment operating status. Using machine learning algorithms, it calculates the optimal ventilation control command. Based on CFD simulation of pollutant diffusion and flow field evolution, it predicts the aerodynamic distribution and concentration in the breathing zone within the mine in real time over a short period.

[0028] A mine ventilation control method includes the following steps:

[0029] S101 monitors the breathing zone in real time. It collects carbon monoxide, fine particulate matter, oxygen concentration and micro-wind speed in the breathing zone every second through the smart mask worn by the miner, and integrates the miner's RFID location information, which is then aggregated to the ground server through underground wireless nodes.

[0030] S102, the server substitutes the above four types of parameters into the weighted normalization formula to obtain the respiratory zone hazard H(x,t) of each miner as a function of location and time; the weights are set according to CO>O2≥wind speed≫dust, reflecting the priority of acute poisoning;

[0031] S103, if the H(x,t) at a certain location in the mine exceeds the set threshold, the dual-line optimization threads of "ventilation optimization" and "escape assessment" are immediately started. The optimization thread takes H(x,t) as the main objective and adds the maximum exposure of all escape paths to form a multi-objective function J(u). The escape thread quickly generates n paths to the nearest safe exit for the miner's location and calculates the integral exposure E(γ) of each path using the same hazard field. Upwind escape is automatically penalized.

[0032] S104 combines adjustable parameters such as fan speed, damper opening, and flow direction into a control vector u, and uses a genetic algorithm or particle swarm optimization to search in the three-dimensional space of safety-escape-energy consumption to minimize J(u); CFD calculations are performed to provide a prediction of the pollutant field in the next 5–10 minutes as an assessment.

[0033] S105, the optimal individual u is broken down into specific values: the speed of each fan, the angle of each damper, and the direction of each guide port, which are written into the fan control PLC via the Internet of Things link;

[0034] S106 simultaneously pushes the escape route γ* with the least exposure and the real-time danger level to the miner's headlamp or voice terminal, realizing the coordination between ventilation adjustment and personnel escape direction.

[0035] Furthermore, the monitoring-optimization-execution-feedback cycle is repeated every 30–60 seconds, and the actual effects (concentration decrease curve, escape time, power consumption) are fed back into the historical database each time to update the initial population or weights of the algorithm offline, forming a continuously evolving intelligent ventilation strategy.

[0036] According to one or more embodiments, a mine ventilation prediction and control method based on respiratory zone monitoring and intelligent AI model decision-making includes the following steps:

[0037] Step 1: Collect and process data from the sensors. The acquired data includes PM2.5, PM10, SO2, CO, CO2, and location coordinates from the respirator sensor; PM2.5, PM10, SO2, CO, CO2, and location coordinates from fixed sampling points in the mine roadway; and wind speed, wind direction, temperature, and humidity data. The data sampling frequency is maintained at once every 10 seconds. For data transmission, the data collected from the respirator and fixed detection points is transmitted to the fixed-point data collector via Bluetooth Low Energy or a wireless communication unit. The data collector then transmits the data via wired transmission to the central control center server on the ground. The server automatically cleans, fuses, and analyzes all the data to monitor for anomalies.

[0038] The ground-based central control center server will match the monitoring data to a corresponding 3D geometric model of the mine roadway. This model will then be used for CFD simulation calculations to predict particulate matter movement trends within the roadway and to dynamically adjust the ventilation system based on the optimal ventilation scheme. When the pollutant concentration exceeds a set threshold, the system will issue a warning and automatically initiate the ventilation system's adjustment procedure. If the pollutant concentration exceeds the standard and the temperature rises abnormally while the humidity decreases, the system will activate the fire suppression mode.

[0039] Step 2: During the system ventilation control execution, components including variable frequency fans, intelligent dampers, and air purification devices are connected to the central control server via an IoT communication link. The server precisely adjusts the operating status to achieve dynamic and on-demand control of the mine ventilation network. The ventilation control model uses fan speed, damper angle, and pollutant concentration as decision variables to achieve dynamic constraints.

[0040] Decision variables:

[0041] Objective function:

[0042] The constraints include two items:

[0043] (1) Safety aspects: (2) Operating conditions: .

[0044] here, The wind speeds of fans 1 through m are given. The angles of the first to nth dampers, This represents the status of air purifiers 1 through k. For monitoring pollutant concentrations, Pollutant concentration, For energy consumption, These are the standards for limiting pollutants in mines.

[0045] Step 3: After receiving air quality data from the masks, the breathing zone, and the mine roadway, the system will automatically generate an early warning-evacuation command based on an optimized objective function when an anomaly is detected. This will be done by combining the pollutant distribution prediction model, the miner distribution area, and the geometric model parameters of the mine roadway.

[0046]

[0047] in, It is a collection of all monitoring points and breathing zone points in the mine roadway (the system can collect the detected gas concentration at multiple locations or breathing zones simultaneously). In time Time, location The concentration of harmful gases at monitoring points in the breathing zone or roadway. These are national safety limits for gases in mines, such as safety limits for dust levels.

[0048] In emergency situations involving fires in mine roadways, the system model will introduce fire control constraints based on signals of abnormally high temperatures to limit the further spread of the fire caused by unreasonable airflow and direction in the ventilation system. Specifically, the method is...

[0049] The system model first calculates the difference between adjacent monitoring points based on temperature data obtained from sensors in the mine, then derives a rough velocity field, and jointly determines the direction of fire propagation based on airflow distribution.

[0050] (1)

[0051] in, The direction of the temperature gradient This is the airflow velocity vector.

[0052] At the same time, define the normal vector of the protected area boundary. Its direction points towards the normal direction outside the boundary of the safe zone. The protected area, which is the interface between the escape route and the fire-contaminated zone, is determined by real-time dynamic data from airflow monitoring. The protected area includes clean airflow areas, escape areas, and areas not yet invaded by high-temperature pollutants; the interface refers to the dynamic boundary between the protected area and the contaminated zone. For example, in a mine... , , A defined plane serves as the boundary, with one side representing the fire and its spread, and the other a region with relatively better air quality. It's important to note that the location of this protected area will change depending on the fire's intensity and the spread of pollutants. Therefore, the normal vector of the protected area boundary... The calculation method is as follows:

[0053] (2)

[0054] in, , , There are three fixed monitoring points.

[0055] (3)

[0056] in, Let the normal vector of the protected area boundary be . The velocity vector represents the direction of fire propagation. This constraint is used to ensure that the fire does not spread further due to the airflow from the ventilation system.

[0057] It has (4)

[0058] in, This represents the highest concentration of harmful gases at the monitoring point in the breathing zone. To help the system predict the evacuation time for everyone, For the energy consumption of the ventilation system, , and For weighting coefficients. Formula (4) means that the system automatically selects the ventilation strategy with the lowest overall risk J based on the current spread trend of the accident (e.g., fire), ventilation status, and personnel distribution. Here, the priority of the objective function is: human safety > air quality improvement rate > ventilation energy consumption. Furthermore, at the same time,

[0059] It has (5)

[0060] in Indicate evacuation routes, This indicates the distribution of pollutant concentrations in the respiratory zone along the path. For path The corresponding evacuation time. Ventilation strategies are considered simultaneously, and the value of formula (5) is minimized.

[0061] The system ventilation control model can obtain the optimal value based on the genetic algorithm (GA) or particle swarm optimization algorithm (PSO) and output control commands such as ventilation rate, damper angle and airflow direction, so as to achieve the effect of real-time monitoring and decision-making.

[0062] According to one or more embodiments, a ventilation control method based on a typical coal mine tunneling face includes the following steps:

[0063] S201, Coal mine tunneling face operation begins. During the production process, a large amount of dust and harmful particulate matter are generated. The ventilation control system controls the fan equipment to operate at a medium air volume.

[0064] S202, smart masks worn by miners at the production site monitor the concentration of dust and harmful substances in the breathing zone, and fixed points in the mine roadway are monitored in real time.

[0065] In S203, data monitored by the smart mask and fixed monitoring points is wirelessly transmitted to a wireless data collector, which then uploads the data to the central control server on the surface via an underground wired network.

[0066] S204: After the data is transmitted to the central ground control server, the server automatically performs data verification and cleaning operations.

[0067] If a decline in air quality is detected, the latest monitoring data and ventilation conditions are used as initial conditions. Based on these initial conditions, the ventilation conditions are adjusted, and a computational fluid dynamics model is used to perform simulation and prediction to evaluate the optimal ventilation scheme in order to improve the air quality in the mine roadways.

[0068] If an abnormally high temperature is detected and the CO content in the air rises sharply, the ventilation control system model switches to fire suppression mode. The algorithm assesses the direction of the fire and locates the fire area through fixed-point monitoring. A series of data are used as initial conditions, and multi-directional adjustment boundary conditions (such as ventilation volume, airflow direction, etc.) are imported into the CFD numerical model for simulation. Based on the optimal calculation results (i.e. the results most conducive to fire prevention and control), the ventilation mode is quickly adjusted to prevent the fire from spreading further due to the ventilation system and to minimize the pollutants generated by the fire in the air, thus providing favorable conditions for personnel to escape.

[0069] The AI-powered intelligent control model calculates the optimal ventilation scheme based on data, personnel location information, and tunnel models, adjusting parameters such as ventilation volume and angle. Control commands are then sent to specific ventilation equipment. Once the dust concentration in the miners' breathing zone and the mine has decreased to a safe level, the fan power is gradually adjusted to a more energy-efficient and stable operating state. Furthermore, all data from each pollutant diffusion event is used for the model's self-learning and iterative training of the AI ​​model.

[0070] When the most serious safety accident in a coal mine—an explosion at the tunneling face—is triggered, a large amount of smoke and harmful gases are produced. Under such circumstances, workers are highly susceptible to inhaling excessive amounts of carbon monoxide (CO) and suffocating. In severe cases, there is also a risk of explosion. In such situations, if dust and harmful gas levels exceed safe limits within a short period, the smart breathing masks worn by miners at the forefront of the tunneling operation, equipped with monitoring functions, and fixed monitoring points deployed near the working face, will quickly detect data such as CO, CH4, dust, wind speed, and temperature in the roadway cross-section near the working face. Simultaneously, this data is matched with the roadway spatial coordinates of the monitoring points and transmitted back to the central control server.

[0071] In the central control server, the system model maps monitoring data onto a CFD mesh according to the location of monitoring points in the three-dimensional geometric model of the tunnel. This is used to set the initial concentration field and boundary conditions, specifically:

[0072] At the inlet boundary on the air intake side, the velocity inlet boundary and inlet concentration are set according to the wind speed and gas composition at the fixed monitoring points in the air intake tunnel.

[0073] A pressure outlet boundary is set at the outlet boundary on the return air side;

[0074] Within a certain length of area near the working face, based on CO and dust data from breathing masks and fixed monitoring points during the last sampling cycle after blasting, an initial high-concentration zone is established to reflect the dust status. The hazard function for the breathing zone is defined for an individual miner as follows:

[0075] (6)

[0076] in, , , Represent , , The safety threshold; , , These represent the concentrations of carbon monoxide, fine particulate matter, and oxygen, respectively. It refers to the wind speed near the breathing zone of the working face. This refers to the minimum wind speed at which the concentration of particulate matter exceeds the standard. , , , These are the weights of each factor. Here, the weights are set to... Here Take any extremely small positive value to prevent the denominator from being zero. First term This is because CO is a lethal risk item, ranking first and second in many mine-related safety accidents. The risk of oxygen deprivation is because it can temporarily impair a person's ability to concentrate on escaping, and may even lead to loss of consciousness; the third point... The first item is wind speed, because low wind speeds can lead to insufficient air circulation, which in turn can cause poisoning; the fourth item... Dust is a risk because the damage it causes is often chronic, and is less severe and slower than the previous three hazards.

[0077] In addition, the breathing mask needs to be precisely positioned. Here, the height range of miners is set between 1.5 and 2 meters, therefore:

[0078] (7)

[0079] Then, a combined steady-state and transient CFD model was used to make short-time predictions of the diffusion process of fumes and dust. The transport equation for pollutant concentration C can be expressed as:

[0080] (8)

[0081] in, The wind speed field within the tunnel (determined by the operating conditions of the ventilation system at that time). For the effective diffusion coefficient, This represents the source term of the residual blasting fumes after the explosion. Calculations can yield the concentration distribution of CO, dust, and other substances at various locations along the tunnel (especially near the miners' breathing height) over the next 10 minutes.

[0082] After obtaining the calculation results, the system model optimizes the ventilation parameters based on these results. Under extreme conditions of rapidly increasing smoke and dust, the adjustable parameters of the ventilation system mainly include the air volume or air pressure of the main fan or local fans (such as speed or air volume setting), the angle of the duct outlet, the opening degree of the air curtain, and the opening and closing status of temporary dampers. The system records the adjustable ventilation parameters (such as fan air volume, duct distance and angle, local fan on / off status, etc.) as control vectors. Subsequently, the candidate parameters are combined, and a short-term prediction is performed using a CFD model to evaluate the efficiency of dust and particulate matter removal within the tunnel.

[0083] Miners' exposure at the front-line work site is closely related not only to the ventilation system but also to evacuation routes. In the event of an accident, the system can match the location of the accident with the nearest escape exit. This route defines the evacuation path. The exposure levels are:

[0084] (9)

[0085] in, Indicates by path Exposure levels during evacuation For along The time from the start to the end of the evacuation. At time t, on the path The position above, In order to be in The hazard function for the breathing zone at that time. Because the degree of danger of escaping with the wind and against the wind is different, with the headwind being more dangerous, therefore, a setting is used. The wind direction penalty coefficient increases the risk of escaping against the wind. Let be the unit vector representing the miner's movement. Here, the wind speed vector is... The significance lies in determining whether one escapes with or against the wind; when their product is less than zero, that is... At this point, the escape route is against the wind, triggering a penalty. This penalty means that the exposure of the escape route in the windward direction will be amplified, making the route numerically "poor," and the system will not select this route as the best escape route.

[0086] When the escape route The system assesses exposure using the exposure assessment formula and provides safety escape prompts based on the path with the least exposure.

[0087] (10)

[0088] Meanwhile, the following multi-objective optimization function is defined to optimize and evaluate the scheme, so that a rapid response can be made based on empirical parameters in the event of a secondary accident.

[0089] (11)

[0090] in, The ventilation parameters to be optimized include some actively adjustable parameters, such as air volume and ventilation angle. This was a gathering of all the people in the mine at that time. For the first The location of the miner. The most dangerous number The danger level of the location of each miner. In order to control the amount Down, along the path The amount of pollutant exposure. In the above formula, the weighting principle is... The first term in the formula is the escape time. The most dangerous miner is considered the most dangerous, and the second is the evacuation route with the highest exposure. The purpose of the calculation is to minimize both as much as possible.

[0091] It should be understood that in the embodiments of the present invention, the term "and / or" is merely a description of the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, the character " / " in this document generally indicates that the preceding and following associated objects have an "or" relationship.

[0092] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0093] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for controlling mine ventilation, characterized in that, The method includes the following steps: A1, using miniature sensors built into the breathing masks worn by miners to collect air quality parameters in the breathing zone of the masks in real time, including at least carbon monoxide concentration. Fine particulate matter concentration Oxygen concentration and breathing zone wind speed And simultaneously obtain the miner's current location coordinates; A2, compare the air quality parameters with the corresponding safe threshold for carbon monoxide concentration in the tunnel. Safety threshold for fine particulate matter concentration Oxygen concentration safety threshold and the minimum safe wind speed when particulate matter concentration exceeds the standard. Substitute into the following formula to calculate the real-time hazard level of the respiratory zone. , Here, It refers to the specific location of a miner in the mine. It is time. , , , These are the weights for the carbon monoxide, fine particulate matter, oxygen, and safe wind speed items, respectively. Take any minimum positive value, , , , The current concentrations of carbon monoxide, fine particulate matter, oxygen, and wind speed at the miner's location are given, with the weights set as follows: ; A3, when When the preset danger threshold is exceeded, ventilation optimization is initiated. Using H(x,t) as the objective function and combining the miner's location coordinates, control commands are generated and issued to the ventilation fans installed in the mine roadway regarding the speed, damper angle, and / or airflow direction, thereby reducing the danger level in the breathing zone. Minimize in subsequent time steps; A4, when When the danger threshold is exceeded and evacuation is required, the system is designed based on the miners' location and the nearest escape exit. When calculating evacuation routes, the evacuation path is determined. Exposure level: in, Indicates by path Exposure levels during evacuation For along the path The time from the start to the end of the evacuation. At time t, on the path The position above, In order to be in The respiratory zone hazard function at that time To calculate the initial time of the time period, This is the wind direction penalty coefficient. Let be the unit vector representing the miner's movement. For wind speed vector, The result determines whether the escape was made with the wind or against the wind; the product of these factors is less than zero. This means entering the penalty phase for escaping against the wind. When evacuation route Calculate the exposure level An assessment was conducted, and miners were prompted to evacuate via the least exposed evacuation route.

2. The method according to claim 1, characterized in that, The method also includes the steps, A5. Based on the obtained air quality parameters of the breathing zones of multiple miners, the comprehensive objective function was further optimized and calculated. in, For the ventilation parameters to be optimized, This was a gathering of all the people in the mine at that time. For the first The location of the miners. For escape time, Indicate evacuation routes, The most dangerous number The danger level of the location of each miner. In order to control the amount Down, along the path Pollutant exposure levels in, and These are the danger levels of the most dangerous miners. The weight of the path with the highest exposure in the evacuation path, by To minimize the optimal combination of ventilation parameters, we seek the best combination of ventilation parameters. .

3. The method according to claim 2, characterized in that, The weight .

4. The method according to claim 3, characterized in that, In step A1, the miner is equipped with a positioning RFID tracker. The data collected by the breathing mask he wears, as well as the positioning signal data, are transmitted to the central ventilation control server located on the ground through a wireless data transceiver with fixed coordinates deployed in the tunnel.

5. The method according to claim 4, characterized in that, The central ventilation control server is equipped with a machine learning-based mine ventilation control model. This model is trained and optimized based on historical fan control parameters. The method is used to process data collected from miners' breathing masks, miners' positioning data, and environmental data obtained from fixed collection points in the roadway, and output control commands for each ventilation fan in the roadway.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the method as described in any one of claims 1 to 5.

7. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 5.

8. A computer program product, comprising a computer program, characterized in that, The computer program is executed by a processor to implement the method according to any one of claims 1 to 5.

Images