Disaster prevention emergency rescue method based on subsea tunnel construction period

Abstract

The invention discloses a disaster prevention emergency rescue method based on a subsea tunnel construction period. The disaster prevention emergency rescue method comprises the following steps that a flood prevention sensing system, a fire prevention sensing system and a collapse prevention sensing system are arranged in a tunnel, and data collection is carried out; uploading data obtained by the three disaster sensing systems to a central processing system; the central processing system sets a threshold value, judges a risk index, gives an alarm and notifies construction personnel to evacuate; through a comprehensive risk index algorithm and a machine learning algorithm, parameters such as water pressure, water flow velocity, water level, humidity, temperature, smoke concentration, flame signals, carbon monoxide concentration, displacement, vibration, inclination angle, sound wave intensity change and strain are monitored in real time, comprehensive assessment is carried out on flood, fire and collapse, and when a risk index exceeds a threshold value, the risk index is detected. The system triggers alarm and starts emergency response measures to ensure construction safety.

Description

Technical Field

[0001] The present invention relates to the technical field of underwater tunnel construction environment safety, and more specifically, to a disaster prevention and emergency rescue method based on underwater tunnel construction period. Background Art

[0002] During the construction of an undersea tunnel, various sudden disasters may occur, such as sudden water and mud, leakage, earthquake, tunnel boring machine accidents, fire and explosion, harmful gases, floods, landslides, etc. Among them, floods, fires (harmful gases such as methane and hydrogen sulfide have high energy and there is also the risk of fire) and landslides are more serious and cause greater losses to the construction. However, these disasters are difficult to prevent and respond to in the traditional construction process, resulting in low construction safety. The construction environment of undersea tunnels is complex and the risk of disasters is high. Traditional early warning and emergency systems are difficult to monitor and respond to emergencies in real time and accurately, which may cause serious personal injury and property loss. Therefore, how to improve early warning capabilities and rescue efficiency has become a technical problem that needs to be solved urgently. This paper proposes a disaster prevention and emergency rescue method based on the construction period of an undersea tunnel to solve the above problems. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a disaster prevention and emergency rescue method based on the construction period of an undersea tunnel. This patent provides a disaster prevention and emergency rescue method based on the construction period of an undersea tunnel. Through a comprehensive risk index algorithm and a machine learning algorithm, real-time monitoring of parameters such as water pressure, water flow velocity, water level, humidity, temperature, smoke concentration, flame signal, carbon monoxide concentration, methane concentration and hydrogen sulfide concentration, displacement, vibration, inclination, sound wave intensity change and strain, among which excessive harmful gases such as methane concentration and hydrogen sulfide concentration may cause fire. A comprehensive assessment is conducted on floods, fires and landslides. When the risk index exceeds the threshold, the system triggers an alarm and initiates emergency response measures to ensure construction safety.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a disaster prevention and emergency rescue method based on the construction period of an undersea tunnel, wherein the ventilation and cooling parameter adjustment method comprises the following steps: arranging a flood prevention sensor system, a fire prevention sensor system and a landslide prevention sensor system in the tunnel and collecting data; uploading the data obtained by the three disaster sensor systems to a central processing system; the central processing system sets a threshold, determines the risk index and issues an alarm, and notifies construction personnel to evacuate.

[0005] In a preferred embodiment: the central processing system receives the data measured by each sensor, and centrally processes the data through a machine learning algorithm: support vector machine, classification function: f(x)=sign(w·x+b), where f(x) is the classification function, w is the weight vector, x is the input sensor data vector, and b is the bias term;

[0006] Optimization goal: The constraints are: i (w·x i +b)≥1-ξ i ,ξ i ≥0, where C is the penalty parameter, ξ i is the slack variable, y i is the label data;

[0007] Forward and backward propagation and loss function calculation through neural network:

[0008] Forward propagation: a (l+1) =σ(W (l) a (l) +b (l) ), where a (l) is the activation value of the lth layer, W (l) is the weight matrix of the lth layer, b (l) is the bias vector of the lth layer, σ is the activation function;

[0009] Backward Propagation: δ (l) =(W (l+1) )Tδ (l+1) ⊙σ′(z (l) ), where δ (l) is the error term of layer l, σ′(z (l) ) is the derivative of the activation function;

[0010] Loss function: Among them, J(θ) is the loss function, m is the number of samples, and y (i) is the true label of the lth sample, h θ (x (i) ) is the predicted value of the model;

[0011] Finally, a comprehensive evaluation is conducted by combining the prediction results of the comprehensive risk index algorithm and the machine learning algorithm: total =α×R index +β×R ml , where R total is the comprehensive risk index, R index is the result of risk index algorithm calculation, R ml is the result predicted by the machine learning algorithm, and α and β are comprehensive weight coefficients.

[0012] In a preferred embodiment: the flood prevention sensing system includes a water pressure sensor, a flow rate sensor and a water level sensor.

[0013] Water pressure sensor: V = k × P, where V is the output voltage, k is the sensor sensitivity coefficient, and P is the water pressure;

[0014] Flow rate sensor: Among them, υ is the water flow velocity, Δf is the frequency offset, c is the speed of sound, f0 is the initial frequency, and θ is the incident angle;

[0015] Water level sensor: Among them, H is the water level, c is the speed of sound, and t is the time from the emission to the reception of the ultrasonic wave.

[0016] In a preferred embodiment, the central processor integrates the sensor parameters into a risk index by weighted average method. When the risk index exceeds the threshold, a flood alarm is triggered. The flood risk is determined as follows:

[0017]

[0018] ; Among them, R water is the flood risk index, P is the real-time water pressure value, v is the real-time water flow velocity, H is the real-time water level, RH is the real-time humidity, T is the real-time water temperature, and P max ,υ max , H max RH max 、T max is the maximum allowed value of each parameter, w1, w2, w3, w4, w5 are the weights of each parameter;

[0019] If R water >R water，threshold , a flood alarm is triggered to warn staff to evacuate in time.

[0020] In a preferred embodiment: the fire prevention sensor system includes a smoke sensor, a temperature sensor and a flame sensor.

[0021] Smoke sensor: I = I0e -σL , where I is the received light intensity, I0 is the initial light intensity, σ is the scattering coefficient, and L is the smoke concentration;

[0022] Temperature sensor: V = α × ΔT, where V is the output voltage, α is the temperature coefficient of the thermocouple material, and ΔT is the temperature difference;

[0023] Flame sensor: Where I is the received light intensity, I0 is the initial light intensity, h is Planck's constant, c is the speed of light, λ is the wavelength, k is the Boltzmann constant, and T is the flame temperature.

[0024] In a preferred embodiment, the central processor integrates the sensor parameters into a risk index by weighted average method. When the risk index exceeds the threshold, a fire alarm is triggered. The fire risk is determined as follows:

[0025] , where R fire is the fire risk index, I smoke is the real-time smoke concentration, T is the real-time temperature, I flame is the flame detection signal strength, CO is the real-time carbon monoxide concentration, RH is the real-time humidity, I smoke,max , T max ,I flame,max , CO max RH max is the maximum allowed value of each parameter, w1, w2, w3, w4, w5 are the weights of each parameter;

[0026] If R fire >R fire，threshold , the fire alarm is triggered to warn the staff to evacuate in time.

[0027] In a preferred embodiment: the landslide prevention sensor includes a displacement sensor, a vibration sensor and an inclination sensor;

[0028] Displacement Sensor: Where C is the capacitance, ∈ is the dielectric constant, A is the electrode area, and d is the distance between electrodes;

[0029] Vibration sensor: V = k × a, where V is the output voltage; k is the sensor sensitivity coefficient, and a is the vibration acceleration;

[0030] Tilt sensor: Where θ is the inclination angle, a x and a y are the gravitational acceleration components along the x and y axes.

[0031] In a preferred embodiment: the central processor integrates the sensor parameters into a risk index by weighted average method. When the risk index exceeds the threshold, a landslide alarm is triggered and the landslide risk is determined:

[0032] , where R collapse is the landslide risk index, d is the real-time displacement, a is the real-time vibration acceleration, θ is the real-time inclination angle, ΔI is the real-time sound wave intensity change, ε is the real-time strain, and d max 、a max,θ max , ΔI max , ε max is the maximum allowed value of each parameter, w1, w2, w3, w4, w5 are the weights of each parameter;

[0033] If R collapse >R collapse，threshold , a landslide alarm is triggered to warn staff to evacuate in time.

[0034] The technical effects and advantages of the present invention are as follows: through the comprehensive risk index algorithm and machine learning algorithm, the water pressure, water flow velocity, water level, humidity, temperature, smoke concentration, flame signal, carbon monoxide concentration, displacement, vibration, inclination, sound wave intensity change and strain and other parameters are monitored in real time to conduct a comprehensive assessment of floods, fires and landslides. When the risk index exceeds the threshold, the system triggers an alarm and initiates emergency response measures to ensure construction safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of the working steps of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] An example embodiment will now be described more fully with reference to the accompanying drawings. The disaster prevention and emergency rescue method includes the following steps: arranging a flood prevention sensor system, a fire prevention sensor system, and a landslide prevention sensor system in a tunnel and collecting data; uploading the data obtained by the three disaster sensor systems to a central processing system; the central processing system sets a threshold, determines the risk index, and issues an alarm to notify construction personnel to evacuate; the central processing system receives the data measured by each sensor, and centrally processes the data through a machine learning algorithm: support vector machine, classification function: f(x)=sign(w·x+b), where f(x) is the classification function, w is the weight vector, x is the input sensor data vector, and b is the bias term; optimization target: The constraints are: i (w·x i +b)≥1-ξ i ,ξ i ≥0, where C is the penalty parameter, ξ i is the slack variable, y iis the label data; forward and backward propagation and loss function calculation are performed through the neural network: forward propagation: a (l+1) =σ(W (l) a (l) +b (l) ), where a (l) is the activation value of the lth layer, W (l) is the weight matrix of the lth layer, b (l) is the bias vector of the lth layer, σ is the activation function; back propagation: δ (l) =(W (l+1) )Tδ (l+1) ⊙σ′(z (l) ), where δ (l) is the error term of layer l, σ′(z (l) ) is the derivative of the activation function; loss function: Among them, J(θ) is the loss function, m is the number of samples, and y (i) is the true label of the lth sample, h θ (x (i) ) is the predicted value of the model; finally, a comprehensive evaluation is performed by combining the prediction results of the comprehensive risk index algorithm and the machine learning algorithm: R total =α×R index +β×R ml , where R total is the comprehensive risk index, R index is the result of risk index algorithm calculation, R ml is the result predicted by the machine learning algorithm, α and β are comprehensive weight coefficients;

[0038] The flood prevention sensor system includes a water pressure sensor, a flow rate sensor and a water level sensor. The water pressure sensor: V = k × P, where V is the output voltage, k is the sensor sensitivity coefficient, and P is the water pressure; the flow rate sensor: Where v is the water flow velocity, Δf is the frequency offset, c is the speed of sound, f0 is the initial frequency, and θ is the incident angle; water level sensor: Among them, H is the water level, c is the speed of sound, and t is the time from the emission to the reception of the ultrasonic wave. The central processor uses the weighted average method to integrate the sensor parameters into a risk index. When the risk index exceeds the threshold, a flood alarm is triggered. The flood risk is determined as follows: ; Among them, R water is the flood risk index, P is the real-time water pressure value, υ is the real-time water flow velocity, H is the real-time water level, RH is the real-time humidity, T is the real-time water temperature, and P max 、v max , H max RH max 、T maxis the maximum allowable value of each parameter, ω1, ω2, ω3, ω4, ω5 are the weights of each parameter; if R water >R water，threshold, a flood alarm is triggered to warn workers to evacuate in time; flood risk is an important consideration during the construction of an undersea tunnel. In order to ensure the safety and smooth progress of construction, effective monitoring and timely warning of floods are necessary. Flood monitoring involves multiple key parameters, including water pressure, water flow velocity, water level, humidity and water temperature. Through real-time monitoring and comprehensive evaluation of these parameters, potential flood risks can be effectively predicted and warned. In flood risk assessment, water pressure is an important indicator. Changes in water pressure can reflect the pressure of the water inside and around the tunnel. When the water pressure is too high, the tunnel structure may be subjected to excessive pressure, causing leakage or structural damage. Therefore, real-time monitoring of water pressure changes is an important means of predicting floods. The monitoring system collects real-time water pressure data through water pressure sensors installed at key locations in the tunnel. These data are transmitted to the central processing system via wired or wireless networks for real-time analysis and evaluation. Water flow velocity is also a key parameter. Changes in water flow velocity can reflect the dynamic situation of water flow inside and outside the tunnel. When the water flow velocity increases abnormally, it may indicate a potential flood risk. For example, a sudden acceleration of the flow of water around a tunnel may mean an increase in water pressure or a leak in a certain area. To monitor the water flow rate, the system installs flow rate sensors inside and outside the tunnel to collect water flow rate data in real time. This data is also transmitted to the central processing system for analysis through the data transmission module. The water level is a parameter that directly reflects the change in the amount of water inside and outside the tunnel. When the water level suddenly rises, it may mean that water is pouring into the tunnel or that the surrounding water has changed abnormally. The monitoring system monitors the water level in real time through water level sensors installed at different locations in the tunnel. These sensors can accurately measure water level changes and transmit data to the central processing system for analysis. Through the changes in water level data, it can be determined whether there is an abnormality in the water body, and then corresponding emergency measures can be taken. Humidity is another important parameter. Changes in humidity can reflect the water vapor content inside the tunnel. When the humidity rises abnormally, it may mean that there is leakage or water intrusion inside the tunnel. To monitor humidity, the system installs multiple humidity sensors in the tunnel to monitor humidity changes in real time. These sensors can accurately detect small changes in humidity and transmit data to the central processing system for analysis. Through the changes in humidity data, potential flood risks can be discovered early and corresponding preventive measures can be taken. Water temperature is a supplementary parameter. Changes in water temperature can reflect the temperature dynamics of the water body inside and outside the tunnel. When the water temperature changes abnormally, it may mean that the water body has undergone thermodynamic changes or the external environment has changed. The monitoring system monitors the changes in water temperature in real time through temperature sensors installed inside and outside the tunnel. These sensors can accurately measure the water temperature and transmit the data to the central processing system for analysis. The changes in water temperature data can assist in judging the dynamic situation of the water body and further improve the accuracy of flood risk prediction. After collecting these key parameters, the central processing system will conduct a comprehensive analysis and evaluation of the data.First, the system will pre-process the data to filter out outliers and noise to ensure the accuracy of the data. Then, the system will use the comprehensive risk index algorithm to integrate various parameters into a risk index. This risk index reflects the overall risk level of the current flood. When the risk index exceeds the preset threshold, the system will trigger a flood alarm. After the alarm is triggered, the system will immediately start the preset emergency response measures. For example, closing the tunnel entrance, starting drainage equipment, evacuating construction personnel, etc. At the same time, the system will notify relevant personnel through sound and light alarms, text messages, emails, etc. to ensure timely action. In addition, the system will also record detailed information of the alarm event, including alarm time, parameter values, etc., for subsequent analysis and processing. Through real-time monitoring and comprehensive analysis of parameters such as water pressure, flow rate, water level, humidity and water temperature, the flood risk assessment system can effectively predict and warn of potential flood risks, thereby ensuring the safety and smooth progress of undersea tunnel construction.

[0039] The fire prevention sensing system includes smoke sensors, temperature sensors and flame sensors. Smoke sensors: I = I0e -σL , where I is the received light intensity, I0 is the initial light intensity, σ is the scattering coefficient, and L is the smoke concentration; Temperature sensor: V = α × ΔT, where V is the output voltage, α is the temperature coefficient of the thermocouple material, and ΔT is the temperature difference; Flame sensor: Where I is the received light intensity, I0 is the initial light intensity, h is Planck's constant, c is the speed of light, λ is the wavelength, k is the Boltzmann constant, and T is the flame temperature. When the risk index exceeds the threshold, a fire alarm is triggered, and the fire risk is determined as follows: , where R fire is the fire risk index, I smoke is the real-time smoke concentration, T is the real-time temperature, I flame is the flame detection signal strength, CO is the real-time carbon monoxide concentration, RH is the real-time humidity, I smoke，max , T max ,I flame，max , CO max RH max is the maximum allowable value of each parameter, w1, w2, w3, w4, w5 are the weights of each parameter; if R fire >R fire，threshold, a fire alarm is triggered to warn staff to evacuate in time; fire is another risk that requires high vigilance during the construction of an undersea tunnel. Due to the closed environment and limited space in the tunnel, once a fire occurs, it may cause serious casualties and property losses. Therefore, effective monitoring and timely warning of fires are the key to ensuring construction safety. Fire monitoring involves multiple key parameters, including smoke concentration, temperature, flame signal, carbon monoxide concentration, and humidity. Through real-time monitoring and comprehensive evaluation of these parameters, potential fire risks can be effectively predicted and warned. Smoke concentration is an important indicator for fire monitoring. When a fire occurs in a tunnel, the smoke produced by the combustion will spread rapidly, resulting in an increase in smoke concentration. The smoke sensor determines the change in smoke concentration by detecting the scattering or absorption of light by smoke particles. The system installs smoke sensors at key locations in the tunnel to monitor the changes in smoke concentration in real time. These data are transmitted to the central processing system for analysis through the data transmission module. When the smoke concentration exceeds the preset threshold, the system triggers a fire alarm. Temperature is another important monitoring parameter. When a fire occurs, the combustion process releases a large amount of heat, causing the temperature in the tunnel to rise rapidly. The temperature sensor can detect changes in ambient temperature and transmit data to the central processing system. The system installs multiple temperature sensors in the tunnel to monitor temperature changes in real time. When the temperature exceeds the preset safety threshold, the system triggers a fire alarm. Changes in temperature data can not only help detect fires early, but also assist in determining the intensity and spread of fires. Flame signals are parameters that directly reflect the occurrence of fires. Flame sensors determine whether there is a flame by detecting the specific wavelength of light radiated by the flame. The system installs flame sensors in flammable areas in the tunnel to monitor flame signals in real time. These sensors can quickly respond to the appearance of flames and transmit data to the central processing system. When a flame signal is detected, the system immediately triggers a fire alarm. Monitoring of flame signals can help the system quickly confirm the occurrence of fires and improve the timeliness of early warning. Carbon monoxide concentration is an indicator of harmful gases produced during the combustion process of a fire. When a fire occurs in a tunnel, combustion will release a large amount of carbon monoxide gas, resulting in an increase in carbon monoxide concentration. Carbon monoxide sensors can detect the carbon monoxide concentration in the environment and transmit data to the central processing system. The system installs carbon monoxide sensors in the tunnel to monitor changes in carbon monoxide concentration in real time. When the carbon monoxide concentration exceeds the preset threshold, the system triggers a fire alarm. By monitoring the concentration of carbon monoxide, it is possible to effectively warn of the occurrence of fire and protect the safety of personnel. Humidity is an auxiliary monitoring parameter. When a fire occurs, combustion will cause changes in the ambient humidity. The humidity sensor can detect changes in ambient humidity and transmit the data to the central processing system. The system installs multiple humidity sensors in the tunnel to monitor humidity changes in real time. Through the changes in humidity data, it can assist in judging the occurrence of fire and provide additional warning information.After collecting these key parameters, the central processing system will conduct a comprehensive analysis and evaluation of the data. First, the system will pre-process the data to filter out outliers and noise to ensure the accuracy of the data. Then, the system will use the comprehensive risk index algorithm to integrate the various parameters into a risk index. This risk index reflects the overall risk level of the current fire. When the risk index exceeds the preset threshold, the system will trigger a fire alarm. After the alarm is triggered, the system will immediately start the preset emergency response measures. For example, closing the tunnel entrance, starting the fire extinguishing equipment, evacuating the construction personnel, etc. At the same time, the system will notify the relevant personnel through sound and light alarms, text messages, emails, etc. to ensure timely action. In addition, the system will also record detailed information of the alarm event, including alarm time, parameter values, etc., for subsequent analysis and processing. Through real-time monitoring and comprehensive analysis of parameters such as smoke concentration, temperature, flame signal, carbon monoxide concentration and humidity, the fire risk assessment system can effectively predict and warn of potential fire risks, thereby ensuring the safety and smooth progress of undersea tunnel construction.

[0040] Landslide prevention sensors include displacement sensors, vibration sensors and inclination sensors; displacement sensors: Where C is capacitance, ε is dielectric constant, A is electrode area, and d is the distance between electrodes; Vibration sensor: V = k × a, where V is output voltage; k is sensor sensitivity coefficient, and a is vibration acceleration; Inclination sensor: Where θ is the inclination angle, a x and a y is the gravity acceleration component along the x and y axes; the central processor integrates the sensor parameters into a risk index through the weighted average method. When the risk index exceeds the threshold, the landslide alarm is triggered and the landslide risk is determined: , where R collapse is the landslide risk index, d is the real-time displacement, a is the real-time vibration acceleration, θ is the real-time inclination angle, ΔI is the real-time sound wave intensity change, ε is the real-time strain, and d max 、a max ,θ max , ΔI max , ε max is the maximum allowable value of each parameter, w1, w2, w3, w4, w5 are the weights of each parameter; if R collapse >R collapse，threshold, a landslide alarm is triggered to warn the staff to evacuate in time; landslide is another major risk that needs to be paid close attention to during the construction of undersea tunnels. Due to the complex structure of the tunnel and the harsh construction environment, once a landslide occurs, it may cause serious casualties and property losses. Therefore, effective monitoring and timely warning of landslides are the key to ensuring construction safety. The monitoring of landslides involves multiple key parameters, including displacement, vibration, inclination, changes in sound wave intensity and strain. Through real-time monitoring and comprehensive evaluation of these parameters, potential landslide risks can be effectively predicted and warned. Displacement is a key parameter reflecting the stability of the tunnel structure. When the tunnel structure is subjected to external forces, displacement or deformation may occur. The displacement sensor can detect the displacement changes of the tunnel structure and transmit the data to the central processing system. The system installs displacement sensors at key structural parts inside and outside the tunnel to monitor the displacement changes in real time. When the displacement exceeds the preset safety threshold, the system triggers a landslide alarm. Through the changes in displacement data, abnormalities in the tunnel structure can be discovered in time and corresponding preventive measures can be taken. Vibration is another important monitoring parameter. During tunnel construction, vibration may be caused by geological activities, mechanical equipment operation or other external forces. The vibration sensor can detect the vibration inside and outside the tunnel and transmit the data to the central processing system. The system installs vibration sensors inside and outside the tunnel to monitor the vibration changes in real time. When the vibration acceleration exceeds the preset threshold, the system will trigger a landslide alarm. The change in vibration data can help the system determine whether the tunnel structure is impacted by external forces and take timely measures to prevent landslide accidents. The inclination angle is an auxiliary parameter reflecting the stability of the tunnel structure. When the tunnel structure tilts or collapses, the inclination sensor can detect the change in inclination angle and transmit the data to the central processing system. The system installs inclination sensors at key structural parts in the tunnel to monitor the change in inclination angle in real time. When the inclination angle exceeds the preset safety threshold, the system will trigger a landslide alarm. Through the change in inclination angle data, it can assist in judging the stability of the tunnel structure and discover potential risks in time. The change in sound wave intensity is an effective means to detect changes in geological structures inside and outside the tunnel. Changes in geological structures may cause abnormal sound wave signals in the tunnel. The sound wave sensor can detect changes in sound wave intensity inside and outside the tunnel and transmit the data to the central processing system. The system installs sound wave sensors inside and outside the tunnel to monitor sound wave changes in real time. When the sound wave intensity changes beyond the preset threshold, the system will trigger a landslide alarm. By changing the sound wave data, it is possible to determine whether there is an abnormality in the geological structure around the tunnel and take preventive measures in time. Strain is a parameter that reflects the stress state of the tunnel structure. When the tunnel structure is subjected to external forces, strain may occur. The strain sensor can detect the strain of the tunnel structure and transmit the data to the central processing system. The system installs strain sensors at key structural parts inside and outside the tunnel to monitor strain changes in real time. When the strain exceeds the preset safety threshold, the system will trigger a landslide alarm.Through the change of strain data, it can be judged whether the tunnel structure is in a dangerous state, and timely measures can be taken to prevent the occurrence of landslide accidents. After collecting these key parameters, the central processing system will conduct a comprehensive analysis and evaluation of the data. First, the system will pre-process the data to filter out outliers and noise to ensure the accuracy of the data. Then, the system will integrate various parameters into a risk index through a comprehensive risk index algorithm. This risk index reflects the overall risk level of the current landslide. When the risk index exceeds the preset threshold, the system will trigger a landslide alarm. After the alarm is triggered, the system will immediately start the preset emergency response measures. For example, stop construction activities, evacuate construction personnel, and strengthen the tunnel structure. At the same time, the system will notify relevant personnel through sound and light alarms, text messages, emails, etc. to ensure timely action. In addition, the system will also record detailed information of the alarm event, including alarm time, parameter values, etc., for subsequent analysis and processing.

[0041] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0042] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0043] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0044] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0045] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0046] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0047] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A disaster prevention and emergency rescue method based on the construction period of an undersea tunnel, characterized by: The disaster prevention and emergency rescue method includes the following steps: arranging a flood prevention sensor system, a fire prevention sensor system and a landslide prevention sensor system in the tunnel and collecting data; uploading the data obtained by the three disaster sensor systems to a central processing system; the central processing system sets a threshold, determines the risk index and issues an alarm, and notifies construction personnel to evacuate.

2. A disaster prevention and emergency rescue method based on the construction period of an undersea tunnel according to claim 1, characterized in that: The central processing system receives the data measured by each sensor and processes the data centrally through the machine learning algorithm support vector machine. The classification function is: f(x) = sign(w·x+b), where f(x) is the classification function, w is the weight vector, x is the input sensor data vector, and b is the bias term; Optimization goal: The constraints are: i (w·x i +b)≥1-ξ i ,ξ i ≥0, where C is the penalty parameter, ξ i is the slack variable, y i is the label data; Forward and backward propagation and loss function calculation through neural network: Forward propagation: a (l+1) =σ(W (l) a (l) +b (l) ), where a (l) is the activation value of the lth layer, W (l) is the weight matrix of the lth layer, b (l) is the bias vector of the lth layer, σ is the activation function; Backward Propagation: δ (l) =(W (l+1) ) T δ (l+1) ⊙σ′(z (l) ), where δ (l) is the error term of layer l, σ′(z (l) ) is the derivative of the activation function; Loss function: Among them, J(θ) is the loss function, m is the number of samples, g (i) is the true label of the lth sample, h θ (x (i) ) is the predicted value of the model; Finally, a comprehensive evaluation is conducted by combining the prediction results of the comprehensive risk index algorithm and the machine learning algorithm: total =Q×R index +β×R ml , where R total is the comprehensive risk index, R index is the result calculated by the risk index algorithm, R ml is the result predicted by the machine learning algorithm, and α and β are comprehensive weight coefficients.

3. The method for disaster prevention and emergency rescue during the construction period of a submarine tunnel according to claim 1, characterized in that: The flood prevention sensing system includes water pressure sensors, flow rate sensors and water level sensors. Water pressure sensor: V = k × P, where V is the output voltage, k is the sensor sensitivity coefficient, and P is the water pressure; Flow rate sensor: Where v is the water flow velocity, Δf is the frequency offset, C is the speed of sound, f0 is the initial frequency, and θ is the incident angle; Water level sensor: Among them, H is the water level, C is the speed of sound, and t is the time from the emission to the reception of the ultrasonic wave.

4. The method for disaster prevention and emergency rescue during the construction period of a submarine tunnel according to claim 1, characterized in that: The central processor uses the weighted average method to integrate the sensor parameters into a risk index. When the risk index exceeds the threshold, a flood alarm is triggered. Flood risk determination: ; Among them, R water is the flood risk index, P is the real-time water pressure value, v is the real-time water flow velocity, H is the real-time water level, RH is the real-time humidity, T is the real-time water temperature, and P max 、v max , Hmax, RH max 、T max is the maximum allowed value of each parameter, w1, w2, w3, w4, w5 are the weights of each parameter; If R water >R water，threshold , a flood alarm is triggered to warn staff to evacuate in time.

5. The method for disaster prevention and emergency rescue during the construction period of a submarine tunnel according to claim 1, characterized in that: The fire prevention sensing system includes smoke sensors, temperature sensors and flame sensors. Smoke sensor: I = I0e -σL , where I is the received light intensity, I0 is the initial light intensity, σ is the scattering coefficient, and L is the smoke concentration; Temperature sensor: V = α × ΔT, where V is the output voltage, α is the temperature coefficient of the thermocouple material, and ΔT is the temperature difference; Flame sensor: Where I is the received light intensity, I0 is the initial light intensity, h is Planck's constant, C is the speed of light, λ is the wavelength, k is the Boltzmann constant, and T is the flame temperature.

6. The method for disaster prevention and emergency rescue during the construction period of a submarine tunnel according to claim 1, characterized in that: The central processor uses the weighted average method to integrate the sensor parameters into a risk index. When the risk index exceeds the threshold, a fire alarm is triggered. Fire risk determination: , Among them, R fire is the fire risk index, I smoke is the real-time smoke concentration, T is the real-time temperature, I flame is the flame detection signal strength, CO is the real-time carbon monoxide concentration, RH is the real-time humidity, I smoke，max 、T max ,I flame，max , CO max RH max is the maximum allowed value of each parameter, w1, ω2, w3, w4, w5 are the weights of each parameter; If R fire >R fire，threshold , the fire alarm is triggered to warn the staff to evacuate in time.

7. The disaster prevention and emergency rescue method during the construction period of a submarine tunnel according to claim 1, characterized in that: Landslide prevention sensors include displacement sensors, vibration sensors, and inclination sensors; Displacement Sensor: Where C is the capacitance, ∈ is the dielectric constant, ug is the electrode area, and d is the distance between electrodes; Vibration sensor: V = k × a, where V is the output voltage; k is the sensor sensitivity coefficient, and a is the vibration acceleration; Tilt sensor: Where θ is the inclination angle, a x and a y are the gravitational acceleration components along the x and y axes.

8. The method for disaster prevention and emergency rescue during the construction period of a submarine tunnel according to claim 1, characterized in that: The central processor uses the weighted average method to integrate the sensor parameters into a risk index. When the risk index exceeds the threshold, a landslide alarm is triggered and the landslide risk is determined: , Among them, R collapse is the landslide risk index, d is the real-time displacement, a is the real-time vibration acceleration, θ is the real-time inclination angle, ΔI is the real-time sound wave intensity change, ε is the real-time strain, and d max 、a max ,θ max , ΔI max , ε max is the maximum allowed value of each parameter, w1, w2, w3, w4, w5 are the weights of each parameter; If R collapse >R collapse , threshold, a landslide alarm is triggered to warn staff to evacuate in time.

Images