Intelligent feedback type tunnel catastrophe monitoring method

Abstract

The invention belongs to the technical field of tunnel catastrophe, and particularly relates to an intelligent feedback type tunnel catastrophe monitoring method, which comprises the following steps of: 1, deploying a tunnel monitoring system; a temperature sensor, a smoke sensor, an air velocity sensor and a gas concentration sensor are arranged in the tunnel; the temperature sensor is used for monitoring temperature changes in the tunnel. According to the intelligent feedback type tunnel catastrophe monitoring method, the safety and the emergency response capability of the tunnel are effectively improved through real-time data monitoring and analysis and an automatic control mechanism. Firstly, through temperature, smoke, gas concentration and air velocity sensors, the system can detect and identify potential catastrophe risks in a tunnel in time, such as fire, smoke diffusion and harmful gas leakage. Once an abnormality occurs, the system performs catastrophe prediction through an algorithm model, and performs accurate evaluation on a fire propagation path and smoke diffusion in combination with a simulation analysis result, thereby providing reliable data support for quick response.

Description

Technical Field

[0001] This invention relates to the field of tunnel disaster technology, and in particular to an intelligent feedback-based tunnel disaster monitoring method. Background Technology

[0002] With the acceleration of urbanization and the continuous growth of transportation demand, the number and scale of tunnels, as an important transportation infrastructure, are constantly increasing. However, as relatively enclosed and narrow spaces, tunnels are prone to serious consequences in the event of disasters such as fires, smoke spread, or leaks of harmful gases, including casualties, property damage, and traffic disruptions. Therefore, how to effectively monitor and prevent disaster risks within tunnels has become a crucial issue that urgently needs to be addressed in the field of tunnel safety.

[0003] Traditional tunnel disaster monitoring methods primarily rely on manual inspections and simple sensor monitoring, which have several limitations. First, manual inspections are inefficient and struggle to provide real-time and comprehensive information on environmental changes within the tunnel. Second, simple sensor monitoring often only provides single environmental parameter information, failing to comprehensively assess and predict disaster risks. Furthermore, traditional monitoring methods lack intelligent and automated control mechanisms; once a disaster occurs, manual intervention is often required to initiate emergency response measures, undoubtedly delaying optimal response times. Therefore, an intelligent, feedback-based tunnel disaster monitoring method is needed. Summary of the Invention

[0004] Based on existing technical problems, this invention proposes an intelligent feedback-based tunnel disaster monitoring method.

[0005] This invention proposes an intelligent feedback-based tunnel disaster monitoring method, comprising the following steps: Step 1: Deployment of a tunnel monitoring system; installing temperature sensors, smoke sensors, air velocity sensors, and gas concentration sensors inside the tunnel; the temperature sensors are used to monitor temperature changes inside the tunnel; the smoke sensors are used to detect smoke concentration to determine if a fire has occurred; the air velocity sensors monitor air velocity to ensure ventilation inside the tunnel; the gas concentration sensors monitor the concentrations of harmful gases CO and CO2; each sensor collects data in real time and transmits the data to the central control system via a communication network.

[0006] Step 2: Data Analysis and Disaster Identification; The central control system is used to analyze the data transmitted by the sensors in real time, and the algorithm model is used to identify whether any abnormalities have occurred in the tunnel. Anomaly detection model and disaster prediction model are adopted.

[0007] Step 3: Feedback control mechanism; Once a disaster is detected, the system will automatically activate the emergency fans and smoke exhaust system. When the smoke or temperature exceeds the threshold, the system will automatically calculate the required number of fans and wind speed, and start the corresponding number of fans. When the location of the fire source is detected, the system will automatically open or close the corresponding smoke exhaust outlets and guide evacuation vehicles through traffic lights and signs.

[0008] Step 4: Data Feedback and Optimization; The system collects data in real time through sensors and calculates the smoke concentration and temperature indices of the current fire area; Using the calculated data, the system optimizes the fan operation mode and smoke exhaust flow rate.

[0009] Step 5: Post-disaster assessment and improvement; After a disaster occurs and an emergency response is received, a post-disaster assessment is conducted, and the entire system is optimized and improved to ensure that it can better respond to similar disaster events in future monitoring.

[0010] Preferably, in step one, a temperature sensor and a smoke sensor are installed at the tunnel entrance to detect changes in the external environment; at the tunnel exit, a gas concentration sensor and a temperature sensor are installed to monitor exhaust gas emissions and temperature inside the tunnel; air velocity sensors are installed at tunnel intersections and in the middle section of the tunnel to monitor ventilation velocity inside the tunnel; and smoke sensors and temperature sensors are installed at the smoke exhaust outlets inside the tunnel to enable rapid exhaust of smoke.

[0011] Formula for calculating air velocity (V): Where V min For the minimum flow rate, Q flow For the output flow rate of the fan, A section This represents the cross-sectional area of ​​the tunnel.

[0012] Preferably, the anomaly detection model in step two involves collecting real-time data from sensors, such as temperature, smoke, gas concentration, and flow rate, and preprocessing this data, including noise reduction and standardization.

[0013] Randomly select a feature in the dataset, randomly select a split point on that feature, divide the data into two subsets based on that split point, and repeat the operation until each data point is completely isolated or the preset tree depth is reached;

[0014] For each data point, calculate its average path length across all isolated trees, using the outlier calculation method of the isolated forest: Where h(x) is the isolation depth of sample point x, and c(n) is a constant for the number of samples n; the closer the anomaly score S(x) is to 1, the higher the probability that the data point is an anomaly; when it is close to 0, it means that the point is normal.

[0015] Real-time data is input into the model for prediction to determine whether it is abnormal behavior; if it is abnormal, the system triggers an alarm.

[0016] Preferably, in step two, the disaster prediction model involves: determining the tunnel's geometry, specifically including its length, width, and height; dividing the tunnel into grids based on its geometry; and using the Navier-Stokes equations to describe airflow, with the following calculation formula: Among them, u 速度场 Let ρ be the velocity field, p be the density, v be the pressure, v be the dynamic viscosity, and F be the external force, such as gravity.

[0017] The temperature changes inside the tunnel are described using the heat conduction calculation formula: Where k is thermal conductivity, T is temperature, and Q is the heat source per unit volume;

[0018] The formula used to describe the spread of smoke and harmful gases produced by a fire is as follows: Where C is the gas concentration, u 流通速度 Where is the fluid velocity, D is the diffusion coefficient, and S is the source term, such as the gas produced by an ignition source.

[0019] The formula used to describe the heat radiation during a fire is as follows: Where E is the radiation energy density, k r For the radiation conductivity, S r For radiation source terms;

[0020] The heat release rate of a fire source, representing the heat released by the fire per unit time, is calculated using the following formula: in, denoted as , where is the combustion rate of the burning substance, and h is the calorific value per unit mass.

[0021] Simulations can yield results on temperature distribution, gas concentration distribution, and flow velocity field; the analysis results help assess the fire propagation path and smoke diffusion in the tunnel.

[0022] Preferably, in step four, the fan operation mode is optimized by automatically adjusting the fan's start / stop status and wind speed based on the current fire scale and smoke concentration; the fan start / stop decision formula is as follows: Among them, C smoke C represents the concentration of smoke. threshold T is the threshold for smoke concentration. fire For temperature, T threshold This is the threshold temperature.

[0023] Fan quantity optimization: The system dynamically adjusts the required number of fans based on real-time smoke concentration, using the following calculation formula: Where, N fanC is the required number of fans. fan It is the smoke concentration handled by each fan.

[0024] Preferably, in step four, the smoke exhaust system flow distribution is optimized. The smoke exhaust system optimizes the flow distribution according to the fire situation in different areas, using the following calculation formula: Among them, Q smoke,area It is necessary to calculate the smoke exhaust flow rate of the area, C smoke,area It is necessary to calculate the smoke concentration in the area, Q. total It is the total smoke exhaust flow rate, C max This is the maximum smoke concentration;

[0025] Machine learning is used to continuously optimize and adjust the operation mode and smoke exhaust flow of the fan based on historical data and feedback information.

[0026] Preferably, in step five, after the disaster occurs, the sensors continue to monitor and collect data in real time, and data analysis methods are used to conduct in-depth analysis of the real-time data during the disaster process;

[0027] Smoke diffusion model analysis: Based on smoke concentration and velocity data, the path and range of smoke diffusion are analyzed; the formula can be expressed as: Where C(x,t) is the smoke concentration at time t at a distance x from the source, Q is the release rate of the smoke source, D is the diffusion coefficient, x0 is the initial position of the smoke source, and e is the base of the natural logarithm.

[0028] Fire propagation prediction model analysis: By using temperature changes and smoke concentration, a fire propagation algorithm is employed to estimate the fire spread range. The formula is: T(t)=T0+(T1-T0)·(1-e -kt ); where T(t) is the temperature at fire time t, T0 is the initial temperature, T1 is the maximum temperature, and k is the fire spread rate;

[0029] The response of fans and smoke extraction systems during a disaster is evaluated, including the timing, number, and speed of fan startup, and whether the smoke extraction system effectively reduced smoke concentration and temperature. Fan efficiency is assessed using the following formula: Among them, E f For the efficiency of the wind turbine, V out V is the exhaust volume of the smoke extraction system. in This refers to the air intake volume;

[0030] In response to disaster events, the design will be optimized in terms of the number of fans, the configuration of the smoke exhaust system, and traffic evacuation. The number of fans can be increased or decreased in the area, and the wind speed of the fans can be adjusted. Through optimization and adjustment, the system can be made to cope with disaster risks in the tunnel more intelligently and efficiently in the future.

[0031] Preferably, the tunnel monitoring system deployment in step one includes the tunnel body, which consists of an arc-shaped entrance, a straight entrance, a submerged section, a first arc-shaped exit, a first straight exit, a second arc-shaped exit, and a second straight exit.

[0032] Preferably, the inner top wall of the tunnel body is also equipped with jet fans, and multiple jet fans are respectively arranged on the top of the arc-shaped entrance, straight entrance, immersed tube section, first arc-shaped exit, first straight exit, second arc-shaped exit and second straight exit of the tunnel body.

[0033] Preferably, the jet fan has a hub diameter of 0.386M, a blade tip diameter of 0.71M, an impeller width of 0.12M, an impeller speed of 2900r / s, a fan pressure step of 46pa, and a maximum flow rate limit of 14m³ / s. 3 / s.

[0034] The beneficial effects of this invention are as follows:

[0035] This intelligent feedback-based tunnel disaster monitoring method effectively improves tunnel safety and emergency response capabilities through real-time data monitoring, analysis, and automatic control mechanisms. First, using sensors for temperature, smoke, gas concentration, and air velocity, the system can promptly detect and identify potential disaster risks within the tunnel, such as fire, smoke spread, and hazardous gas leaks. Once an anomaly occurs, the system uses algorithmic models to predict the disaster and, combined with simulation analysis results, accurately assesses the fire propagation path and smoke diffusion, thus providing reliable data support for rapid response.

[0036] Secondly, the feedback control mechanism automatically adjusts the fans and smoke extraction system to ensure rapid fan activation and smoke discharge in the event of a disaster, significantly reducing the impact on personnel and equipment. Simultaneously, the dynamic optimization and adjustment of the number of fans and smoke extraction flow rate enhances the system's adaptability to different fire scales. Through machine learning algorithms, the system can also continuously optimize control strategies, improving disaster response efficiency.

[0037] Ultimately, post-disaster assessment and system optimization ensured that the monitoring system was better equipped to respond to future disasters, enhancing the tunnel's safety and reliability. This intelligent approach not only reduced human intervention and improved emergency response speed, but also effectively reduced losses after an accident, providing strong support for the tunnel's operation and management. Attached Figure Description

[0038] Figure 1 A schematic diagram of an intelligent feedback tunnel disaster monitoring method;

[0039] Figure 2 A three-dimensional view of the tunnel structure as a method for intelligent feedback-based tunnel disaster monitoring;

[0040] Figure 3 A three-dimensional diagram of an immersed tunnel section structure for an intelligent feedback-based tunnel disaster monitoring method;

[0041] Figure 4 A schematic diagram of airflow velocity at each end face of a tunnel at 50 km / s for a vehicle-to-tunnel-to-cath speed as part of an intelligent feedback-based tunnel disaster monitoring method.

[0042] Figure 5 A schematic diagram of airflow velocity at each end face of a tunnel at 40 km / s for a vehicle, illustrating an intelligent feedback-based tunnel disaster monitoring method.

[0043] Figure 6 A schematic diagram of airflow velocity at each end face of a tunnel at 30km / s for a vehicle to perform an intelligent feedback tunnel disaster monitoring method.

[0044] Figure 7 A schematic diagram of airflow velocity at each end face of a tunnel at 20km / s for a vehicle-to-tunnel-disaster-monitoring method based on intelligent feedback.

[0045] Figure 8 This is a schematic diagram of the airflow velocity at each end of a tunnel at a speed of 10 km / s, which is used to illustrate an intelligent feedback-based tunnel disaster monitoring method.

[0046] In the diagram: 1. Tunnel body; 2. Arc-shaped entrance; 3. Straight entrance; 4. Immersed tube section; 5. First arc-shaped exit; 6. First straight exit; 7. Second arc-shaped exit; 8. Second straight exit; 9. Jet fan. Detailed Implementation

[0047] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0048] Reference Figures 1-8 An intelligent feedback-based tunnel disaster monitoring method includes the following steps: Step 1: Deployment of the tunnel monitoring system; Temperature sensors, smoke sensors, air velocity sensors, and gas concentration sensors are installed in the tunnel; Temperature sensors are used to monitor temperature changes in the tunnel; Smoke sensors are used to detect smoke concentration to determine if a fire has occurred; Air velocity sensors monitor air velocity to ensure ventilation in the tunnel; Gas concentration sensors monitor the concentration of harmful gases CO and CO2; Each sensor collects data in real time and transmits the data to the central control system through a communication network.

[0049] In step one, temperature and smoke sensors are installed at the tunnel entrance to detect changes in the external environment; at the tunnel exit, gas concentration and temperature sensors are installed to monitor exhaust gas emissions and temperature inside the tunnel; air velocity sensors are installed at tunnel intersections and in the middle section of the tunnel to monitor ventilation velocity inside the tunnel; and smoke and temperature sensors are installed at the smoke exhaust outlets inside the tunnel to enable rapid exhaust of smoke.

[0050] Formula for calculating air velocity (V): Where V min For the minimum flow rate, Q flow For the output flow rate of the fan, A section This represents the cross-sectional area of ​​the tunnel.

[0051] Step one of the tunnel monitoring system deployment includes the tunnel body 1, which consists of an arc-shaped entrance 2, a straight entrance 3, a submerged section 4, a first arc-shaped exit 5, a first straight exit 6, a second arc-shaped exit 7, and a second straight exit 8. The inner top wall of the tunnel body 1 is also equipped with jet fans 9, with multiple jet fans 9 respectively located at the top of the arc-shaped entrance 2, straight entrance 3, submerged section 4, first arc-shaped exit 5, first straight exit 6, second arc-shaped exit 7, and second straight exit 8 of the tunnel body. The jet fans 9 have a hub diameter of 0.386M, a blade tip diameter of 0.71M, an impeller width of 0.12M, an impeller speed of 2900r / s, a fan pressure of 46pa, and a maximum flow rate limit of 14m³ / s. 3 / s.

[0052] Step 2: Data Analysis and Disaster Identification; The central control system is used to analyze the data transmitted by the sensors in real time, and the algorithm model is used to identify whether any abnormalities have occurred in the tunnel. Anomaly detection model and disaster prediction model are adopted.

[0053] Step 2 Anomaly detection model: Collect real-time data from sensors, such as temperature, smoke, gas concentration and flow rate, and preprocess this data, including noise reduction and standardization;

[0054] Randomly select a feature in the dataset, randomly select a split point on that feature, divide the data into two subsets based on that split point, and repeat the operation until each data point is completely isolated or the preset tree depth is reached;

[0055] For each data point, calculate its average path length across all isolated trees, using the outlier calculation method of the isolated forest: Where h(x) is the isolation depth of sample point x, and c(n) is a constant for the number of samples n; the closer the anomaly score S(x) is to 1, the higher the probability that the data point is an anomaly; when it is close to 0, it means that the point is normal.

[0056] Real-time data is input into the model for prediction to determine whether it is abnormal behavior; if it is abnormal, the system triggers an alarm.

[0057] Step two, the disaster prediction model, involves: determining the tunnel's geometry, specifically its length, width, and height; meshing the tunnel based on its geometry; and using the Navier-Stokes equations to describe airflow, with the following calculation formula: Among them, u 速度场 Let ρ be the velocity field, p be the density, v be the pressure, v be the dynamic viscosity, and F be the external force, such as gravity.

[0058] The temperature changes inside the tunnel are described using the heat conduction calculation formula: Where k is thermal conductivity, T is temperature, and Q is the heat source per unit volume;

[0059] The formula used to describe the spread of smoke and harmful gases produced by a fire is as follows: Where C is the gas concentration, u 流通速度 Where is the fluid velocity, D is the diffusion coefficient, and S is the source term, such as the gas produced by an ignition source.

[0060] The formula used to describe the heat radiation during a fire is as follows: Where E is the radiation energy density, k r For the radiation conductivity, S r For radiation source terms;

[0061] The heat release rate of a fire source, representing the heat released by the fire per unit time, is calculated using the following formula: in, denoted as , where is the combustion rate of the burning substance, and h is the calorific value per unit mass.

[0062] Simulations can yield results on temperature distribution, gas concentration distribution, and flow velocity field; the analysis results help assess the fire propagation path and smoke diffusion in the tunnel.

[0063] like Figure 4 As shown, the airflow velocity of the vehicle at 50km / h in each fault of the tunnel is insufficient to meet the ventilation requirements by means of piston air. The wind speed in the immersed tube section is 1.42m / s, and an additional jet fan is needed to reach 1.77m / s in the immersed tube section.

[0064] like Figure 5 As shown, the airflow velocity of the vehicle at 40km / h in each fault of the tunnel is insufficient to meet the ventilation requirements by means of piston air. The wind speed in the immersed tube section is 1.08m / s, and an additional jet fan is needed to reach 1.65m / s in the immersed tube section.

[0065] like Figure 6As shown, the airflow velocity of the vehicle at 30km / h in each fault of the tunnel is insufficient to meet the ventilation requirements by piston air. The wind speed in the immersed tube section is 1.02m / s, and two additional jet fans in the immersed tube section are needed to reach 1.87m / s.

[0066] like Figure 7 As shown, the airflow velocity of the vehicle at 20km / h in each fault of the tunnel is insufficient to meet the ventilation requirements by piston air. The wind speed in the immersed tube section is only 0.81m / s, and two jet fans in the immersed tube section are needed to reach 1.77m / s.

[0067] like Figure 8 As shown, the airflow velocity of the vehicle at 10km / h in each fault of the tunnel is not sufficient to meet the ventilation requirements by means of piston air. It is necessary to add two jet fans in the immersed tube section to reach 1.68m / s.

[0068] Step 3: Feedback control mechanism; Once a disaster is detected, the system will automatically activate the emergency fans and smoke exhaust system. When the smoke or temperature exceeds the threshold, the system will automatically calculate the required number of fans and wind speed, and start the corresponding number of fans. When the location of the fire source is detected, the system will automatically open or close the corresponding smoke exhaust outlets and guide evacuation vehicles through traffic lights and signs.

[0069] Step 4: Data Feedback and Optimization; The system collects data in real time through sensors and calculates the smoke concentration and temperature indices of the current fire area; Using the calculated data, the system optimizes the fan operation mode and smoke exhaust flow rate.

[0070] Step four involves optimizing the fan operation mode by automatically adjusting the fan's start / stop status and speed based on the current fire scale and smoke concentration. The fan start / stop decision formula is as follows: Among them, C smoke C represents the concentration of smoke. threshold T is the threshold for smoke concentration. fire For temperature, T threshold This is the threshold temperature.

[0071] Fan quantity optimization: The system dynamically adjusts the required number of fans based on real-time smoke concentration, using the following calculation formula: Where, N fan C is the required number of fans. fan It is the smoke concentration handled by each fan.

[0072] Step four involves optimizing the flow distribution of the smoke exhaust system. The flow distribution is optimized based on the fire conditions in different areas, using the following calculation formula: Among them, Q smoke,area It is necessary to calculate the smoke exhaust flow rate of the area, C smoke,area It is necessary to calculate the smoke concentration in the area, Q. totalIt is the total smoke exhaust flow rate, C max This is the maximum smoke concentration;

[0073] Machine learning is used to continuously optimize and adjust the operation mode and smoke exhaust flow of the fan based on historical data and feedback information.

[0074] Step 5: Post-disaster assessment and improvement; After a disaster occurs and an emergency response is received, a post-disaster assessment is conducted, and the entire system is optimized and improved to ensure that it can better respond to similar disaster events in future monitoring.

[0075] In step five, after the disaster occurs, the sensors continue to monitor and collect data in real time, and data analysis methods are used to conduct in-depth analysis of the real-time data during the disaster process.

[0076] Smoke diffusion model analysis: Based on smoke concentration and velocity data, the path and range of smoke diffusion are analyzed; the formula can be expressed as: Where C(x,t) is the smoke concentration at time t at a distance x from the source, Q is the release rate of the smoke source, D is the diffusion coefficient, x0 is the initial position of the smoke source, and e is the base of the natural logarithm.

[0077] Fire propagation prediction model analysis: By using temperature changes and smoke concentration, a fire propagation algorithm is employed to estimate the fire spread range. The formula is: T(t)=T0+(T1-T0)·(1-e -kt ); where T(t) is the temperature at fire time t, T0 is the initial temperature, T1 is the maximum temperature, and k is the fire spread rate;

[0078] The response of fans and smoke extraction systems during a disaster is evaluated, including the timing, number, and speed of fan startup, and whether the smoke extraction system effectively reduced smoke concentration and temperature. Fan efficiency is assessed using the following formula: Among them, E f For the efficiency of the wind turbine, V out V is the exhaust volume of the smoke extraction system. in This refers to the air intake volume;

[0079] In response to disaster events, the design will be optimized in terms of the number of fans, the configuration of the smoke exhaust system, and traffic evacuation. The number of fans can be increased or decreased in the area, and the wind speed of the fans can be adjusted. Through optimization and adjustment, the system can be made to cope with disaster risks in the tunnel more intelligently and efficiently in the future.

[0080] This intelligent feedback-based tunnel disaster monitoring method effectively improves tunnel safety and emergency response capabilities through real-time data monitoring, analysis, and automatic control mechanisms. First, using sensors for temperature, smoke, gas concentration, and air velocity, the system can promptly detect and identify potential disaster risks within the tunnel, such as fire, smoke spread, and hazardous gas leaks. Once an anomaly occurs, the system uses algorithmic models to predict the disaster and, combined with simulation analysis results, accurately assesses the fire propagation path and smoke diffusion, thus providing reliable data support for rapid response.

[0081] Secondly, the feedback control mechanism automatically adjusts the fans and smoke extraction system to ensure rapid fan activation and smoke discharge in the event of a disaster, significantly reducing the impact on personnel and equipment. Simultaneously, the dynamic optimization and adjustment of the number of fans and smoke extraction flow rate enhances the system's adaptability to different fire scales. Through machine learning algorithms, the system can also continuously optimize control strategies, improving disaster response efficiency.

[0082] Ultimately, post-disaster assessment and system optimization ensured that the monitoring system was better equipped to respond to future disasters, enhancing the tunnel's safety and reliability. This intelligent approach not only reduced human intervention and improved emergency response speed, but also effectively reduced losses after an accident, providing strong support for the tunnel's operation and management.

[0083] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An intelligent feedback-based tunnel disaster monitoring method, characterized in that: The process includes the following steps: Step 1, Deployment of the tunnel monitoring system; Install temperature sensors, smoke sensors, air velocity sensors, and gas concentration sensors inside the tunnel; Temperature sensors are used to monitor temperature changes inside the tunnel; Smoke sensors are used to detect smoke concentration to determine if a fire has occurred; Air velocity sensors monitor air velocity to ensure ventilation inside the tunnel; Gas concentration sensors monitor the concentration of harmful gases such as CO and CO2; Each sensor collects data in real time and transmits the data to the central control system via a communication network; Step 2: Data Analysis and Disaster Identification; The central control system is used to analyze the data transmitted by the sensors in real time, and the algorithm model is used to identify whether any abnormalities have occurred in the tunnel. Anomaly detection model and disaster prediction model are adopted. Step 3: Feedback control mechanism; Once a disaster is detected, the system will automatically activate the emergency fans and smoke exhaust system. When the smoke or temperature exceeds the threshold, the system will automatically calculate the required number of fans and wind speed, and start the corresponding number of fans. When the location of the fire source is detected, the system will automatically open or close the corresponding smoke exhaust outlets and guide evacuation vehicles through traffic lights and signs. Step 4: Data Feedback and Optimization; The system collects data in real time through sensors and calculates the smoke concentration and temperature indices of the current fire area; Using the calculated data, the system optimizes the fan operation mode and smoke exhaust flow rate; Step 5: Post-disaster assessment and improvement; After a disaster occurs and an emergency response is received, a post-disaster assessment is conducted, and the entire system is optimized and improved to ensure that it can better respond to disaster events in future monitoring.

2. The intelligent feedback-based tunnel disaster monitoring method according to claim 1, characterized in that: In step one, temperature and smoke sensors are installed at the tunnel entrance to detect changes in the external environment; at the tunnel exit, gas concentration and temperature sensors are installed to monitor exhaust gas emissions and temperature inside the tunnel; air velocity sensors are installed at tunnel intersections and in the middle section of the tunnel to monitor ventilation velocity inside the tunnel; and smoke and temperature sensors are installed at the smoke exhaust outlets inside the tunnel to enable rapid exhaust of smoke. Formula for calculating air velocity (V): Where V min For the minimum flow rate, Q flow For the output flow rate of the fan, A section This represents the cross-sectional area of ​​the tunnel.

3. The intelligent feedback-based tunnel disaster monitoring method according to claim 1, characterized in that: The anomaly detection model in step two involves collecting real-time data from sensors, such as temperature, smoke, gas concentration, and flow rate, and preprocessing this data, including noise reduction and standardization. Randomly select a feature in the dataset, randomly select a split point on that feature, divide the data into two subsets based on that split point, and repeat the operation until each data point is completely isolated or the preset tree depth is reached; For each data point, calculate its average path length across all isolated trees, using the outlier calculation method of the isolated forest: Where h(x) is the isolation depth of sample point x, and c(n) is a constant for the number of samples n; the closer the anomaly score S(x) is to 1, the higher the probability that the data point is an anomaly; when it is close to 0, it means that the point is normal. Real-time data is input into the model for prediction to determine whether it is abnormal behavior; if it is abnormal, the system triggers an alarm.

4. The intelligent feedback-based tunnel disaster monitoring method according to claim 1, characterized in that: The disaster prediction model in step two involves: determining the tunnel's geometry, specifically its length, width, and height; dividing the tunnel into grids based on its geometry; and using the Navier-Stokes equations to describe airflow, with the following calculation formula: Among them, u 速度场 Let ρ be the velocity field, p be the density, v be the pressure, v be the dynamic viscosity, and F be the external force, such as gravity. The temperature changes inside the tunnel are described using the heat conduction calculation formula: Where k is thermal conductivity, T is temperature, and Q is the heat source per unit volume; The formula used to describe the spread of smoke and harmful gases produced by a fire is as follows: Where C is the gas concentration, u 流通速度 Where is the fluid velocity, D is the diffusion coefficient, and S is the source term, such as the gas produced by an ignition source. The formula used to describe the heat radiation during a fire is as follows: Where E is the radiation energy density, k r For the radiation conductivity, S r For radiation source terms; The heat release rate of a fire source, representing the heat released by the fire per unit time, is calculated using the following formula: in, denoted as , where is the combustion rate of the burning substance, and h is the calorific value per unit mass. The simulation yielded results on temperature distribution, gas concentration distribution, and flow velocity field; the analysis results helped assess the fire propagation path and smoke diffusion in the tunnel.

5. The intelligent feedback-based tunnel disaster monitoring method according to claim 1, characterized in that: In step four, the fan operation mode is optimized by automatically adjusting the fan's start / stop status and wind speed according to the current fire scale and smoke concentration. Wind turbine start / stop decision formula: Among them, C smoke C represents the concentration of smoke. threshold T is the threshold for smoke concentration. fire For temperature, T threshold This is the threshold temperature. Fan quantity optimization: The system dynamically adjusts the required number of fans based on real-time smoke concentration, using the following calculation formula: Where, N fan C is the required number of fans. fan It is the smoke concentration handled by each fan.

6. The intelligent feedback-based tunnel disaster monitoring method according to claim 1, characterized in that: In step four, the smoke exhaust system flow distribution is optimized. The smoke exhaust system optimizes the flow distribution based on the fire conditions in different areas, using the following calculation formula: Among them, Q smoke,area It is necessary to calculate the smoke exhaust flow rate of the area, C smoke,area It is necessary to calculate the smoke concentration in the area, Q. total It is the total smoke exhaust flow rate, C max This is the maximum smoke concentration; Machine learning is used to continuously optimize and adjust the operation mode and smoke exhaust flow of the fan based on historical data and feedback information.

7. The intelligent feedback-based tunnel disaster monitoring method according to claim 1, characterized in that: In step five, after the disaster occurs, the sensors continue to monitor and collect data in real time, and data analysis methods are used to conduct in-depth analysis of the real-time data during the disaster process. Smoke diffusion model analysis: Based on smoke concentration and velocity data, the path and range of smoke diffusion are analyzed; the formula can be expressed as: Where C(x,t) is the smoke concentration at time t at a distance x from the source, Q is the release rate of the smoke source, D is the diffusion coefficient, x0 is the initial position of the smoke source, and e is the base of the natural logarithm. Fire propagation prediction model analysis: By using temperature changes and smoke concentration, a fire propagation algorithm is employed to estimate the fire spread range. The formula is: T(t)=T0+(T1-T0)·(1-e -kt ); where T(t) is the temperature at fire time t, T0 is the initial temperature, T1 is the maximum temperature, and k is the fire spread rate; The response of fans and smoke extraction systems during a disaster is evaluated, including the timing, number, and speed of fan startup, and whether the smoke extraction system effectively reduced smoke concentration and temperature. Fan efficiency is assessed using the following formula: Among them, E f For wind turbine efficiency, V out V is the exhaust volume of the smoke extraction system. in This refers to the air intake volume; In response to disaster events, the design will be optimized in terms of the number of fans, the configuration of the smoke exhaust system, and traffic evacuation. The number of fans can be increased or decreased in the area, and the wind speed of the fans can be adjusted. Through optimization and adjustment, the system can be made to cope with disaster risks in the tunnel more intelligently and efficiently in the future.

8. The intelligent feedback-based tunnel disaster monitoring method according to claim 1, characterized in that: The tunnel monitoring system deployment in step one includes the tunnel body (1), which consists of an arc-shaped entrance (2), a straight entrance (3), a submerged tube section (4), a first arc-shaped exit (5), a first straight exit (6), a second arc-shaped exit (7), and a second straight exit (8).

9. The intelligent feedback-based tunnel disaster monitoring method according to claim 8, characterized in that: The inner top wall of the tunnel body (1) is also equipped with jet fans (9), and multiple jet fans (9) are respectively located on the top of the arc-shaped entrance (2), straight entrance (3), immersed tube section (4), first arc-shaped exit (5), first straight exit (6), second arc-shaped exit (7) and second straight exit (8) of the tunnel body.

10. The intelligent feedback-based tunnel disaster monitoring method according to claim 9, characterized in that: The jet fan (9) has a hub diameter of 0.386M, a blade tip diameter of 0.71M, an impeller width of 0.12M, an impeller speed of 2900r / s, a fan pressure of 46pa, and a maximum flow rate of 14m³ / s. 3 / s.

Images