Highway tunnel construction safety intelligent system

Abstract

The invention provides an intelligent safety system for highway tunnel construction, which relates to the technical field of construction safety warning and comprises the following structures: a data sensing layer consisting of various sensors, such as a temperature and humidity sensor, a gas sensor, a displacement sensor, a vibration sensor and a pressure sensor, the data acquisition module is responsible for acquiring environmental parameters, structural states and equipment operation data in a tunnel in real time; the high-definition camera and the intelligent identification equipment are used for acquiring video images, personnel and vehicles in the tunnel; a data transmission layer; a data processing and analyzing layer; and an application layer. By setting the adaptive calibration technology in various sensors, the sensors can be ensured to calibrate monitoring results for special environments, errors of the monitoring results can be reduced, the probability of making wrong decisions can be reduced, the accuracy of collecting surrounding environment parameters by the sensors can be improved, and the monitoring accuracy of the sensors can be improved. And potential safety hazards of tunnel construction can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of construction safety warning, and particularly to an intelligent system for highway tunnel construction safety. Background Technique

[0002] With the rapid development of highway construction in China, tunnel engineering, as a key component thereof, the problem of construction safety has become increasingly prominent. During the construction process of highway tunnels, there are many complex and extremely challenging safety risk factors. On the one hand, the construction environment of the tunnel is extremely complex and changeable, with large fluctuations in internal temperature and humidity. The high-temperature and high-humidity environment is likely to cause construction equipment failures and affect the performance of construction materials. At the same time, various harmful gases, such as carbon monoxide, hydrogen sulfide, etc., will be generated during the construction process. If not monitored and controlled in time, it will pose a fatal threat to the health of personnel. In addition, the tunnel structure bears huge stresses during construction, and local stratum changes and blasting operations may cause displacements and cracks in the tunnel lining, thereby triggering collapse accidents. On the other hand, there are many personnel and equipment in the tunnel and their activities are frequent. Construction workers shuttle between various operation areas, and large construction machinery and transport vehicles come and go continuously. If there is a lack of effective management and monitoring, safety hazards are likely to occur.

[0003] To solve the above problems, sensors and monitoring devices are used to monitor the environment inside the tunnel. However, due to the complex and changeable environment inside the tunnel, monitoring errors are likely to occur, leading to wrong judgments, and further increasing the potential hazards of tunnel construction. Summary of the Invention

[0004] (I) Technical Problems to be Solved

[0005] In view of the deficiencies of the prior art, the present invention provides an intelligent system for highway tunnel construction safety, which solves the problem that in order to solve the above problems, sensors and monitoring devices are used to monitor the environment inside the tunnel. However, due to the complex and changeable environment inside the tunnel, monitoring errors are likely to occur, leading to wrong judgments, and further increasing the potential hazards of tunnel construction.

[0006] (II) Technical Solutions

[0007] To achieve the above object, the present invention is realized through the following technical solutions: An intelligent system for highway tunnel construction safety includes the following architecture:

[0008] A data perception layer, which consists of various sensors, such as temperature and humidity sensors, gas sensors, displacement sensors, vibration sensors and pressure sensors, and is responsible for real-time collection of environmental parameters, structural states and equipment operation data inside the tunnel; there are also high-definition cameras and intelligent recognition devices for obtaining video images, personnel and vehicles inside the tunnel;

[0009] The data transmission layer transmits the data collected by the sensing layer to the data processing center or the cloud through wired or wireless communication networks, such as optical fibers, 4G / 5G, and LoRa;

[0010] The data processing and analysis layer cleans, denoises, and fuses the transmitted data, and uses big data analysis, machine learning, and artificial intelligence to mine the potential laws and features in the data, evaluate the safety status of tunnel construction, and predict possible safety risks;

[0011] The application layer includes various application systems and system function components for tunnel construction safety management;

[0012] The system function components include a video monitoring module, a sensor monitoring module, a personnel and equipment positioning module, an intelligent early warning module, a safety management module, and an emergency response module.

[0013] Furthermore, an adaptive calibration technology is set in various sensors in the data sensing layer, and the adaptive calibration technology allows the sensor to automatically adjust the calibration parameters according to changes in environmental conditions.

[0014] Furthermore, the adaptive calibration technology first has the ability to sense environmental parameters, then constructs an algorithm model, calculates through the constructed algorithm model, and then performs real-time calibration execution through the calculated error.

[0015] Furthermore, constructing the algorithm model includes the following steps:

[0016] Step 1: First, it is necessary to determine the factors affecting the parameters in the tunnel environment;

[0017] Step 2: Collect experimental data;

[0018] Step 3: Select and construct an appropriate model according to the actual situation;

[0019] Step 4: Verify and optimize the constructed model.

[0020] Furthermore, the environmental parameters affecting the sensor measurement error in Step 1 are mainly temperature, humidity, air pressure, electromagnetic interference, and dust. First, it is necessary to analyze these factors separately to understand how they affect the sensor performance.

[0021] Furthermore, in Step 2 of collecting experimental data, it is first necessary to design the experimental scheme, then analyze the relationship between each environmental parameter and the measurement error by controlling variables, and finally conduct multiple groups of repeated experiments.

[0022] Further, in the third step, when encountering simple situations, only a linear regression model needs to be constructed; when encountering complex situations, a polynomial regression model needs to be constructed; and when considering multiple factors comprehensively, a multiple regression model needs to be constructed.

[0023] Further, in the fourth step, independent validation data is required to verify the accuracy of the constructed mathematical model. If the verification result of the model is not satisfactory, optimization measures can be taken.

[0024] Further, the video monitoring module uses high-definition cameras and intelligent video analysis technology to monitor and record the construction conditions, personnel and vehicle activities, equipment operation, etc. in the tunnel in real time. The sensor monitoring module monitors the structural safety status of the tunnel in real time through various sensors arranged in the tunnel. The personnel and equipment positioning module uses UWB and Zigbee positioning technologies to be able to track the position information of construction personnel and equipment in real time, and master the distribution and activity trajectories of personnel and equipment.

[0025] Further, the intelligent early warning module gives real-time early warnings of potential safety risks during the tunnel construction process based on the results of the data processing and analysis layer. The safety management module realizes the informatization management of construction resources by establishing a database of information such as construction personnel, equipment, and materials. The emergency response module automatically activates the emergency response plan and coordinates various resources for rescue and disposal in case of a safety accident or emergency.

[0026] (III) Beneficial Effects

[0027] The present invention provides an intelligent system for the safety of highway tunnel construction, which has the following beneficial effects:

[0028] In this solution, by setting an adaptive calibration technology in various sensors, it can ensure that the sensors calibrate the monitoring results for special environments, which is beneficial to reducing the error of the monitoring results, thereby reducing the probability of making wrong decisions, improving the accuracy of the sensors in collecting surrounding environmental parameters, and being beneficial to reducing the safety hazards of tunnel construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is an architecture diagram of an intelligent system for the safety of highway tunnel construction proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0031] Embodiment:

[0032] As Figure 1 shown, the embodiment of the present invention provides an intelligent system for highway tunnel construction safety, including the following architecture:

[0033] The data perception layer is composed of various sensors, such as temperature and humidity sensors, gas sensors, displacement sensors, vibration sensors and pressure sensors, which are responsible for real-time collection of environmental parameters, structural states and equipment operation data in the tunnel; there are also high-definition cameras and intelligent recognition devices for obtaining video images, personnel and vehicles in the tunnel;

[0034] The data transmission layer transmits the data collected by the perception layer to the data processing center or the cloud through wired or wireless communication networks, such as optical fibers, 4G / 5G and LoRa;

[0035] The data processing and analysis layer cleans, denoises and fuses the transmitted data, and uses big data analysis, machine learning and artificial intelligence to mine potential laws and features in the data, evaluate the safety status of tunnel construction and predict possible safety risks;

[0036] The application layer includes various application systems and system function components for tunnel construction safety management;

[0037] The system function components include a video monitoring module. The video monitoring module uses high-definition cameras and intelligent video analysis technology to monitor and record the construction conditions, personnel and vehicle activities, equipment operation, etc. in the tunnel in real time, and can also automatically identify and analyze abnormal behaviors in the video, such as personnel not wearing safety helmets, illegal operations, vehicle speeding, etc., and issue alarms in a timely manner;

[0038] The sensor monitoring module, through various sensors arranged in the tunnel, monitors the structural safety status of the tunnel in real time, such as the deformation, settlement, and crack development of the tunnel; monitors the air quality in the tunnel, including oxygen content, concentrations of harmful gases (such as carbon monoxide, hydrogen sulfide, etc.); monitors environmental parameters such as temperature, humidity, wind speed, and light in the tunnel; monitors the operation status and working parameters of construction equipment, such as ventilation equipment, drainage equipment, blasting equipment, etc.;

[0039] Personnel and Equipment Location Module. The Personnel and Equipment Location Module adopts UWB and Zigbee positioning technologies, which can track the location information of construction personnel and equipment in real time, master the distribution and movement trajectories of personnel and equipment, and can also set up electronic fences to issue early warnings in a timely manner when personnel or equipment enter dangerous areas or exceed the specified range. In case of an accident, it can quickly determine the locations of trapped personnel and equipment, providing accurate information for rescue work;

[0040] Intelligent Early Warning Module. The Intelligent Early Warning Module is based on the results of the data processing and analysis layer to give real-time early warnings of potential safety risks during tunnel construction. Different early warning levels and thresholds can be set. When the monitored data exceeds the threshold, alarms are sent to construction management personnel and on-site operators in a timely manner through methods such as sound and light alarms, text messages, and APP push. Machine learning and artificial intelligence algorithms are used to learn and analyze historical data and real-time data to predict possible safety accidents and take preventive measures in advance;

[0041] Safety Management Module. The Safety Management Module realizes the informatization management of construction resources by establishing a database of information such as construction personnel, equipment, and materials. Subsequently, it formulates and manages construction safety rules and regulations, operating procedures, and emergency plans, and conducts safety training and education for construction personnel. Records and tracks the safety inspections, hidden danger investigations, and rectification situations at the construction site to achieve closed-loop operation of safety management;

[0042] Emergency Response Module. In case of a safety accident or emergency, the Emergency Response Module automatically activates the emergency response plan, coordinates various resources for rescue and disposal, and can also provide emergency command and dispatch functions, including personnel dispatch, equipment allocation, material distribution, etc. Furthermore, it can display information such as videos and data of the accident site in real time to support emergency decision-making, record and evaluate the process and effect of emergency disposal, and provide a basis for subsequent summary and improvement.

[0043] Adaptive calibration technology is set in various sensors in the data perception layer. The adaptive calibration technology allows the sensor to automatically adjust calibration parameters according to changes in environmental conditions. The adaptive calibration technology first has the ability to sense environmental parameters. For example, temperature sensors, humidity sensors, and other auxiliary sensors are integrated inside the sensor to monitor environmental conditions. These auxiliary sensors convert changes in environmental parameters into electrical signals, which are used as inputs to the adaptive calibration algorithm;

[0044] Then, an algorithm model is constructed and calculations are performed through the constructed algorithm model. For example, for a strain sensor, temperature changes will cause thermal expansion of the sensor material, resulting in measurement errors. Through experiments, the functional relationship between temperature and strain measurement errors can be obtained, and this functional relationship constitutes the core part of the adaptive calibration algorithm;

[0045] Subsequently, real-time calibration is performed based on the calculated error. When the sensor is working, the adaptive calibration algorithm calculates the measurement error in real time according to the perceived environmental parameters and the pre-constructed mathematical model, and calibrates the output of the sensor. For example, if the temperature sensor detects a 10°C increase in temperature, according to the temperature-strain error model, the algorithm calculates that the strain measurement value may be 2% larger due to thermal expansion, and then the output of the strain sensor will be corrected accordingly.

[0046] The construction of the algorithm model includes the following steps:

[0047] Step 1: First, it is necessary to determine the factors affecting the parameters in the tunnel environment. The environmental parameters affecting the measurement error of the sensor mainly include temperature, humidity, air pressure, electromagnetic interference, and dust. First, these factors need to be analyzed separately to understand how they affect the performance of the sensor;

[0048] Temperature: Temperature changes will cause the thermal expansion and contraction of the sensor material. For example, for a strain sensor, when the temperature rises, the material of the sensor itself will expand, resulting in a deviation in the strain measurement value. At the same time, the increase in temperature will also affect the performance of the internal electronic components of the sensor, such as changing the values of resistance and capacitance, thereby affecting the signal processing and transmission;

[0049] Humidity: A high-humidity environment may cause the sensor to get damp. Especially for some humidity-sensitive materials, such as certain gas sensors based on electrical principles, the increase in humidity may change the conductivity of the sensor surface, interfere with the reaction between gas molecules and the sensor surface, and thus affect the measurement accuracy;

[0050] Air pressure: During the tunnel construction process, the change in air pressure will affect some pressure sensors or sensors related to air pressure. For example, when the air pressure changes, the air density will also change accordingly, which may affect the propagation speed of the acoustic sensor and thus cause measurement errors;

[0051] Electromagnetic interference: There are various electrical devices in the tunnel, such as ventilators and lighting devices. The electromagnetic interference generated by these devices will affect the signal transmission of the sensor. For example, electromagnetic interference may superimpose noise on the signal line of the sensor, change the amplitude and frequency of the signal, and cause deviations in the measurement results;

[0052] Dust: The accumulation of dust may block the air inlet of the sensor or cover the sensitive components. For optical sensors, since dust will scatter or absorb light, it will reduce the intensity and quality of the optical signal, thereby affecting the measurement accuracy;

[0053] Step 2: Collect experimental data. To collect experimental data, first, an experimental plan needs to be designed. A series of experiments are required to build a mathematical model. For example, to study the influence of temperature on the measurement error of a sensor, the sensor can be placed in an experimental chamber with controllable temperature. At different temperature settings (such as -20°C, -10°C, 0°C, 10°C, 20°C, etc.), measure a known standard quantity and record the measured values of the sensor and the corresponding standard true values at each temperature;

[0054] Then, analyze the relationship between each environmental parameter and the measurement error by controlling variables. When studying the influence of a certain environmental parameter on the sensor, other environmental parameters should be kept as constant as possible. For example, when studying the influence of humidity, factors such as temperature, air pressure, and electromagnetic interference need to be kept stable, and only the humidity is changed. In this way, the relationship between each environmental parameter and the measurement error can be analyzed separately;

[0055] Finally, conduct multiple sets of repeated experiments. To ensure the accuracy and reliability of the experimental data, multiple repeated experiments are required under each experimental condition. For example, when measuring a standard quantity at a certain temperature, repeat the experiment more than 10 times, and then calculate the average value and standard deviation of the measured values to reduce the experimental error;

[0056] Step 3: Select and build an appropriate model according to the actual situation. When dealing with simple situations, only a linear regression model needs to be built. When dealing with complex situations, a polynomial regression model needs to be built. When considering multiple factors comprehensively, a multiple regression model needs to be built;

[0057] Linear regression model: If the experimental data shows an approximately linear relationship between the environmental parameter and the measurement error, a linear regression model can be used. For example, for the relationship between temperature and the measurement error of a strain sensor, assume the relationship between the measurement error E and temperature T is E = aT + b, where a and b are coefficients to be determined. By fitting the experimental data using the least squares method, the values of a and b can be obtained, thus getting the linear mathematical model between temperature and the measurement error;

[0058] Polynomial regression model: When the relationship between the environmental parameter and the measurement error is not a simple linear relationship, a polynomial regression model can be used. For example, for the relationship between humidity and the measurement accuracy of a gas sensor, a quadratic polynomial model E = aH 2 + bH + c is needed, where H is humidity, and a, b, and c are coefficients obtained by fitting the experimental data. This model can better describe complex non-linear relationships;

[0059] Multiple regression model: The multiple regression model is used when the effects of multiple environmental parameters on the sensor measurement error need to be considered simultaneously. For example, the relationship between the measurement error E and temperature T, humidity H, and air pressure P may be E = aT + bH + cP + d, where a, b, c, and d are coefficients obtained by multiple linear regression analysis of experimental data. In practical applications, non-linear terms such as quadratic terms and interaction terms can also be added to the multiple regression model to more accurately describe complex actual situations;

[0060] Step 4: Verify and optimize the constructed model. Independent validation data is needed to verify the accuracy of the constructed mathematical model. If the verification result of the model is not satisfactory, optimization measures can be taken;

[0061] Verification method: Use an independent validation data set to verify the accuracy of the constructed mathematical model. Take a part of the experimental data not used for model construction as the validation set. Substitute the environmental parameters into the mathematical model to calculate the predicted value of the measurement error, and then compare it with the actual measurement error. Metrics such as root mean square error (RMSE) and mean absolute error (MAE) can be used to measure the prediction accuracy of the model;

[0062] Optimization measures: If the verification result of the model is not satisfactory, the following optimization measures can be taken. One is to increase the scope and diversity of experimental data to more comprehensively reflect the relationship between environmental parameters and measurement error. The other is to try other more complex model structures such as neural network models and support vector regression models. These models may have better performance in dealing with complex non-linear relationships, but they also require more computing resources and training data;

[0063] The steps of verification are as follows:

[0064] S1: Divide and prepare the data;

[0065] Divide the data set: Divide the collected experimental data into a training set, a validation set, and a test set. Generally, the training set is used to construct the mathematical model and accounts for 60%-80% of the total data set; the validation set is used to adjust the hyperparameters of the model during model training and accounts for 10%-20%; the test set is used to finally evaluate the accuracy and reliability of the model and accounts for 10%-20%. For example, if there are 100 groups of experimental data, 70 groups can be used as the training set, 15 groups as the validation set, and 15 groups as the test set;

[0066] Data standardization: Standardize the data before using it to make different environmental parameter and measurement error data have the same scale. A common standardization method is to convert the data into a standard normal distribution with a mean of 0 and a standard deviation of 1. This can avoid model bias caused by different data dimensions and help improve the convergence speed and accuracy of the model;

[0067] S2: Verification method based on statistical metrics;

[0068] Root Mean Square Error (RMSE): This is a commonly used metric to measure the prediction accuracy of a model. It is calculated by first calculating the average of the squares of the prediction errors (the differences between the predicted values and the true values) and then taking the square root.

[0069] Its formula is:

[0070] where n is the number of samples in the test set, y i is the true value of the i-th sample, is the prediction of the i-th sample. The smaller the value of RMSE, the higher the prediction accuracy of the model. For example, a model with an RMSE of 0.1 is usually more accurate in prediction than a model with an RMSE of 0.5;

[0071] Mean Absolute Error (MAE): It measures the average of the absolute values of the prediction errors.

[0072] Its formula is:

[0073] Compared with RMSE, MAE is less sensitive to outliers. If the MAE of a model is small, it means that the prediction error of the model is small on average and the reliability is high;

[0074] Coefficient of Determination (R 2 ): It represents the proportion of the data variance that the model can explain.

[0075] Its formula is:

[0076] where is the average of the true values, and the value range of R 3 is from 0 to 1. The closer it is to 1, the better the model fits the data and the higher the accuracy. For example, a model with R 2 = 0.9 can explain the relationship between environmental parameters and measurement errors better than a model with R 2 = 0.6;

[0077] S3: Use the cross-validation method;

[0078] K-fold Cross-Validation: The dataset is divided into K subsets of equal size (usually K is taken as 5 or 10). For example, when K = 5, the dataset is first divided into 5 subsets. Each time, one of the subsets is selected as the test set, and the remaining 4 subsets are used as the training set. In this way, 5 different training-test combinations can be obtained. Model training and testing are performed on each combination, and corresponding evaluation metrics (such as RMSE, MAE, etc.) are calculated. Finally, the average value of these K groups of evaluation metrics is used as the final evaluation result of the model. This method can make full use of limited data and can more comprehensively evaluate the performance of the model on different data subsets, reducing the risk of overfitting;

[0079] Leave-One-Out Cross-Validation: This is a special case of K-fold cross-validation. When K is equal to the number of samples n in the dataset, each time only one sample is left as the test set, and the remaining n - 1 samples are used as the training set. The advantage of this method is that it can make the most of the data, but the computational cost is relatively high, and it is suitable for cases with a small amount of data;

[0080] S4: Comparative validation method;

[0081] Comparison with existing models: If there is an existing and widely recognized model of the same type used to describe the relationship between environmental parameters and sensor measurement errors, the newly constructed model can be compared with the existing model, and their evaluation metrics (such as RMSE, MAE, R 2 etc.) on the same test set can be compared. If these metrics of the new model are better than those of the existing model, it indicates that the new model has better accuracy and reliability;

[0082] Comparison with theoretical models: In some cases, a theoretical model between environmental parameters and measurement errors can be derived based on physical principles or mathematical theories. The empirically constructed model is compared with the theoretical model to check their consistency in prediction trends and values. If the empirical model is consistent with the theoretical model and can better fit the actual data, this also proves the accuracy and reliability of the empirical model;

[0083] S5: Verification through actual application scenarios;

[0084] Verification by simulating the actual environment: In the laboratory, try to simulate the actual complex environment of the tunnel as much as possible, including changes in environmental parameters such as temperature, humidity, and air pressure, as well as factors such as electromagnetic interference and dust. Place the sensor in the simulated environment, use the constructed mathematical model to predict the measurement error, and compare it with the actually measured error. If the model can accurately predict the measurement error in the simulated actual environment, then it is more likely to be reliable in actual tunnel applications;

[0085] Field test verification: Install the sensor at the tunnel construction site, collect the measurement data and corresponding environmental parameters in the actual environment, substitute these data into the mathematical model to calculate the prediction error, and compare it with the error measured on site. The field test can most truly reflect the accuracy and reliability of the model because it takes into account various actual complex factors during the tunnel construction process, such as the interference of construction equipment, personnel activities, etc.

[0086] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent safety system for highway tunnel construction, characterized in that: It includes the following architecture: The data perception layer, which consists of various sensors such as temperature and humidity sensors, gas sensors, displacement sensors, vibration sensors, and pressure sensors, is responsible for collecting environmental parameters, structural status, and equipment operation data in the tunnel in real time; there are also high-definition cameras and intelligent recognition devices for obtaining video images, personnel, and vehicles in the tunnel; The data transmission layer transmits the data collected by the perception layer to the data processing center or the cloud through wired or wireless communication networks such as optical fibers, 4G / 5G, and LoRa; The data processing and analysis layer cleans, denoises, and fuses the transmitted data, and uses big data analysis, machine learning, and artificial intelligence to mine potential laws and characteristics in the data, evaluate the safety status of tunnel construction, and predict possible safety risks; The application layer includes various application systems and system function components for tunnel construction safety management; The system function components include a video monitoring module, a sensor monitoring module, a personnel and equipment positioning module, an intelligent early warning module, a safety management module, and an emergency response module.

2. The intelligent system for highway tunnel construction safety according to claim 1, wherein: An adaptive calibration technology is set in various sensors in the data perception layer, and the adaptive calibration technology allows the sensors to automatically adjust calibration parameters according to changes in environmental conditions.

3. The intelligent safety system for highway tunnel construction according to claim 2, wherein: The adaptive calibration technology first has the ability to sense environmental parameters, then constructs an algorithm model, calculates through the constructed algorithm model, and then performs real-time calibration execution through the calculated error.

4. An intelligent system for highway tunnel construction safety according to claim 3, characterized in that: Constructing the algorithm model includes the following steps: Step 1: First, it is necessary to determine the factors affecting the parameters in the tunnel environment; Step 2: Collect experimental data; Step 3: Select and construct an appropriate model according to the actual situation; Step 4: Verify and optimize the constructed model.

5. The intelligent safety system for highway tunnel construction according to claim 4, wherein: The environmental parameters affecting the measurement error of the sensor in Step 1 are mainly temperature, humidity, air pressure, electromagnetic interference, and dust. First, these factors need to be analyzed separately to understand how they affect the sensor performance.

6. The intelligent safety system for highway tunnel construction according to claim 4, characterized in that: In Step 2, when collecting experimental data, first design the experimental scheme, then analyze the relationship between each environmental parameter and the measurement error through controlling variables, and finally conduct multiple groups of repeated experiments.

7. The intelligent system for highway tunnel construction safety according to claim 4, characterized in that: In Step 3, when encountering a simple situation, only a linear regression model needs to be constructed; when encountering a complex situation, a polynomial regression model needs to be constructed; when considering multiple factors comprehensively, a multiple regression model needs to be constructed.

8. An intelligent safety system for highway tunnel construction according to claim 1, wherein: In Step 4, independent verification data is needed to verify the accuracy of the constructed mathematical model. If the verification result of the model is not ideal, optimization measures can be taken.

9. The intelligent system for highway tunnel construction safety according to claim 1, wherein: The video monitoring module uses high-definition cameras and intelligent video analysis technology to monitor and record the construction situation, personnel and vehicle activities, equipment operation, etc. in the tunnel in real time. The sensor monitoring module monitors the structural safety status of the tunnel in real time through various sensors arranged in the tunnel. The personnel and equipment positioning module uses UWB and Zigbee positioning technologies to be able to track the position information of construction personnel and equipment in real time, and master the distribution and movement trajectories of personnel and equipment.

10. The intelligent safety system for highway tunnel construction according to claim 1, characterized in that: The intelligent early warning module conducts real-time early warning of potential safety risks during tunnel construction based on the results of the data processing and analysis layer. The safety management module realizes the informatization management of construction resources by establishing a database of information such as construction personnel, equipment, and materials. The emergency response module automatically activates the emergency response plan and coordinates various resources for rescue and disposal in case of a safety accident or emergency situation.

Images