Tunnel structure health monitoring method, system, equipment and medium

Abstract

The invention relates to a tunnel structure health monitoring method, system and device and a medium, and belongs to the technical field of tunnel structure monitoring, and the monitoring method comprises the steps: obtaining the structure monitoring data of a to-be-monitored tunnel, and carrying out the preprocessing; determining a tunnel structure type of the to-be-monitored tunnel, and performing feature extraction on the preprocessed structure monitoring data according to the tunnel structure type to obtain key features; performing structure health state analysis based on the key features to obtain a tunnel structure health state analysis result; identifying a potential damage part according to the health state analysis result of the tunnel structure, and evaluating the damage degree of the potential damage part; and generating a tunnel structure damage report according to the potential damage part and the damage degree, and sending the report to the management terminal. The tunnel health monitoring accuracy can be improved, and long-term safe and stable operation of the tunnel is ensured.

Description

Technical Field

[0001] The present application relates to the technical field of tunnel structure monitoring, and in particular to a tunnel structure health monitoring method, system, equipment and medium. Background Art

[0002] With the continuous development of transportation infrastructure, tunnel engineering plays an increasingly important role in the fields of railways, highways, and urban underground space development. As a special structure buried underground or on the seabed, tunnel structures are in a complex and changeable environment for a long time. They are affected by various factors such as soil, water pressure, waves, earthquakes, and vehicle dynamic loads. They are very likely to cause damage and deterioration problems, which in turn threaten the safety and durability of the overall structure of the tunnel. Therefore, how to effectively monitor the structural health status of tunnels, identify potential structural problems, and take timely remedial measures such as maintenance is a key research topic in the current engineering field.

[0003] At present, due to the diversity of tunnel structure types, different construction methods (such as shield method, mining method, immersed tube method, etc.) determine that the stress characteristics and damage modes of tunnel structures are different. Most common tunnel health monitoring methods are difficult to meet the differentiated needs of different tunnel types, resulting in low accuracy in tunnel structure damage assessment, which is easy to cause misjudgment or omission. Summary of the invention

[0004] In order to improve the accuracy of tunnel health monitoring and ensure the long-term safe and stable operation of the tunnel, the present application provides a tunnel structure health monitoring method, system, equipment and medium.

[0005] In a first aspect, the present application provides a tunnel structure health monitoring method, which adopts the following technical solution: A tunnel structure health monitoring method, the monitoring method comprising: Obtain the structural monitoring data of the tunnel to be monitored and perform preprocessing; Determine the tunnel structure type of the tunnel to be monitored, Performing feature extraction on the pre-processed structure monitoring data according to the tunnel structure type to obtain key features; Perform structural health status analysis based on the key features to obtain tunnel structural health status analysis results; Identifying potential damage locations according to the tunnel structure health status analysis results, and evaluating the damage extent of the potential damage locations; A tunnel structure damage report is generated according to the potential damage location and damage degree, and sent to the management terminal.

[0006] By adopting the above technical solutions, in the process of monitoring different types of tunnels, combined with the structural type of the tunnel, key features closely related to the health status of the tunnel are extracted from the huge amount of tunnel monitoring data, and possible structural damage is evaluated and located. This can not only effectively improve the analysis accuracy of the health status of the tunnel structure, but also quickly discover and locate potential damage that may exist in the tunnel. By timely evaluating the degree of damage, comprehensive tunnel health information is provided to managers, ensuring the long-term safe and stable operation of the tunnel.

[0007] Optionally, the step of extracting features from the pre-processed structure monitoring data according to the tunnel structure type to obtain key features includes: Determining corresponding feature extraction parameters according to the tunnel structure type; the feature extraction parameters include an extraction threshold of a structural vibration frequency feature, a calculation window size of a displacement change rate feature, and a sensitivity of a crack width change trend feature; The key features of the preprocessed structural monitoring data are extracted based on the feature extraction parameters; the key features include structural vibration frequency features, displacement change rate features and crack width change trend features.

[0008] By adopting the above technical solution, it is ensured that the feature extraction process can more accurately reflect the characteristics of different types of tunnel structures, so that the monitoring solution can better adapt to the monitoring needs of different types of tunnels and improve the universality and effectiveness of the monitoring solution.

[0009] Optionally, the step of performing structural health status analysis based on the key features to obtain tunnel structural health status analysis results includes: The key features are input into a pre-built structural health analysis model, and the tunnel structure health status analysis results are obtained according to the model output; wherein the structural health analysis model is pre-trained based on the historical structural monitoring data of the tunnel to be monitored and pre-labeled health analysis result labels.

[0010] By adopting the above technical solution, the extracted key features such as stress and displacement can be used to judge the current health status of the tunnel through a pre-trained structural health analysis model. By timely discovering potential structural problems in the tunnel, the risk of major accidents can be reduced, thereby ensuring the structural safety and operational stability of the tunnel.

[0011] Optionally, the method further includes a training step of the structural health analysis model, wherein the training step includes: Acquire a sample data set and perform preprocessing; wherein the sample data set includes historical structural monitoring data of the tunnel to be monitored and pre-labeled health analysis result labels; Performing feature extraction on the preprocessed sample data set according to the tunnel structure type of the tunnel to be monitored to obtain a key sample feature set; Dividing the key sample feature set into a training set and a test set; Inputting the training set into a pre-built machine learning model for training, optimizing model parameters, and obtaining the structural health analysis model; The structural health analysis model is tested and model parameters are corrected based on the test set until the loss function of the model meets a preset condition or the number of model iterations reaches a preset number, thereby obtaining the trained structural health analysis model.

[0012] By adopting the above technical solutions, key features were extracted from the historical data of the tunnel, and integrated into the machine learning model for training and correction, a model that can accurately analyze the health status of the tunnel structure was successfully constructed. The model can automatically learn the health pattern of the tunnel structure from complex data sets and provide more accurate predictions and diagnoses for future monitoring, which not only improves the efficiency of tunnel monitoring, but also provides a basis for tunnel maintenance and management, and has important engineering and economic value.

[0013] Optionally, the step of testing the structural health analysis model based on the test set includes: A test model is established according to the test set, and the performance of the structural health analysis model is tested based on the test model; wherein the test indicators of the test model include accuracy, precision, recall rate, F1 score and confusion matrix.

[0014] By adopting the above technical solutions, the model performance is tested to ultimately evaluate the applicability of the model in real-world environments and ensure that the model has high accuracy and reliability in actual applications.

[0015] Optionally, the step of identifying potential damage locations according to the tunnel structure health status analysis result includes: Acquiring structural information and load monitoring data of the tunnel to be monitored and performing data preprocessing; wherein the structural information includes tunnel design parameters and construction material characteristics; Based on the structural information of the tunnel to be monitored and the tunnel structure health status analysis result, a finite element model is constructed; wherein the tunnel structure health status analysis result includes at least one structural health status indicator and at least one abnormal characteristic statistical information, and the finite element model includes the geometry, material properties and boundary conditions of the tunnel structure; applying a load condition on the finite element model according to the load monitoring data; Calibrate the finite element model, and adjust the material properties and boundary conditions in the finite element model so that the simulation results of the finite element model match the abnormal characteristic statistical information in the tunnel structure health status analysis results; Perform static or dynamic analysis on the calibrated finite element model to obtain structural response data; Based on the structural response data, areas where the structural response is abnormally increased are identified and marked as potential damage sites.

[0016] By adopting the above technical solution, combined with the structural health status analysis results and load monitoring data, a more accurate finite element model can be constructed. By analyzing the structural response data, potential damage sites can be accurately identified, providing a scientific basis for maintenance decisions. This technical solution can effectively improve the accuracy and reliability of tunnel structure monitoring and provide strong support for the safe operation of tunnels.

[0017] Optionally, the step of assessing the degree of damage of the potential damage site comprises: Configure the damage degree quantitative index threshold according to the tunnel structure type; The damage degree quantification index threshold is compared with the structural response data of the potential damage part to obtain the damage degree corresponding to the potential damage part.

[0018] By adopting the above technical solution and setting reasonable thresholds and quantification methods according to different tunnel structure types, the health status of the structure can be identified more accurately, making maintenance and repair work more targeted, helping to reasonably arrange repair resources and avoid unnecessary or excessive repairs.

[0019] In a second aspect, the present application provides a tunnel structure health monitoring system, which adopts the following technical solution: A tunnel structure health monitoring system, the monitoring system comprising: An acquisition module is used to acquire the structural monitoring data of the tunnel to be monitored and perform preprocessing; A structure type determination module, used to determine the tunnel structure type of the tunnel to be monitored; A feature extraction module, used for extracting features from the pre-processed structure monitoring data according to the tunnel structure type to obtain key features; A health status analysis module, used to perform structural health status analysis based on the key features to obtain tunnel structure health status analysis results; A structural damage identification module, used to identify potential damage locations according to the tunnel structure health status analysis result, and evaluate the damage degree of the potential damage locations; The structural damage report push module is used to generate a tunnel structure damage report according to the potential damage location and damage degree, and send it to the management terminal.

[0020] In a third aspect, the present application provides a computer device, which adopts the following technical solution: A computer device comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect.

[0021] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the methods in the first aspect.

[0022] In summary, the present application includes at least one of the following beneficial technical effects: in the process of monitoring different types of tunnels, combined with the structural type of the tunnel, key features closely related to the health status of the tunnel are extracted from the huge amount of tunnel monitoring data, and possible structural damage is evaluated and located. This can not only effectively improve the analysis accuracy of the health status of the tunnel structure, but also quickly discover and locate potential damage that may exist in the tunnel. By timely evaluating the degree of damage, comprehensive tunnel health information is provided to managers, ensuring the long-term safe and stable operation of the tunnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a first flow chart of a tunnel structure health monitoring method according to one of the embodiments of the present application.

[0024] Figure 2 This is a second flow chart of a tunnel structure health monitoring method according to one of the embodiments of the present application.

[0025] Figure 3 It is a third flow chart of a tunnel structure health monitoring method according to one of the embodiments of the present application.

[0026] Figure 4 It is a fourth flow chart of a tunnel structure health monitoring method according to one of the embodiments of the present application.

[0027] Figure 5 It is a fifth flow chart of a tunnel structure health monitoring method according to one of the embodiments of the present application. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-5It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0029] The embodiment of the present application discloses a tunnel structure health monitoring method.

[0030] Reference Figure 1 , a tunnel structure health monitoring method, the monitoring method comprising: Step S101, obtaining the structural monitoring data of the tunnel to be monitored and performing preprocessing; Among them, structural monitoring data is an important basis for tunnel health monitoring, which usually includes various data types such as displacement, stress, vibration signal, crack width, etc. collected by sensors; data preprocessing, such as cleaning, normalization, and denoising of the original data, so as to extract key features from it more efficiently and accurately in the future.

[0031] Step S102, determining the tunnel structure type of the tunnel to be monitored, In some embodiments, the tunnel structure type can be divided into tunnels constructed by shield method, tunnels constructed by mining method, tunnels constructed by immersed tube method, tunnels constructed by jacking method, tunnels constructed by open cut and backfill method, etc. according to the construction method. The structure type is usually determined in advance during the tunnel design stage.

[0032] It is understandable that the tunnel structure type directly affects the structural behavior and stress characteristics of the tunnel. Tunnels of different structural types have different requirements and key indicators for health status monitoring. Therefore, the feature extraction scheme must be customized according to the specific structural type of the tunnel.

[0033] Step S103, extracting features from the pre-processed structural monitoring data according to the tunnel structure type to obtain key features; Among them, feature extraction is to extract the parts closely related to the health status of the tunnel from the huge data set, and through statistical analysis, physical modeling and other methods, extract indicators that can reflect the structural characteristics of the tunnel, such as structural vibration frequency characteristics, displacement change rate characteristics, crack width change trend characteristics, etc.

[0034] It can be understood that by extracting key features, the data size can be greatly reduced and the efficiency of subsequent analysis can be improved, while ensuring that attention is paid to the indicators most relevant to structural health and enhancing the accuracy of the tunnel structure health status analysis results.

[0035] Step S104, performing structural health status analysis based on key features to obtain tunnel structural health status analysis results; The tunnel structure health status analysis results include at least one structural health status indicator and at least one abnormal feature statistical information. In some embodiments, the extracted key features can be used to diagnose the tunnel health status through a machine learning model analysis method, assess whether the tunnel is in a normal, slightly damaged or severely damaged state, and timely discover potential structural problems in the tunnel to reduce the risk of major accidents, thereby ensuring tunnel structural safety and operational stability.

[0036] Step S105, identifying potential damage locations according to the tunnel structure health status analysis result, and evaluating the damage degree of the potential damage locations; Specifically, after obtaining the preliminary health analysis results, the specific potential damage sites are further located, and the severity of the damage is quantified by analyzing the key signals of each site, such as stress concentration area, crack width, displacement trend, etc.

[0037] It is understandable that accurately locating the damaged parts and assessing the damage extent will help to take timely repair and reinforcement measures, optimize the allocation of maintenance resources, and extend the service life of the tunnel.

[0038] Step S106, generating a tunnel structure damage report according to the potential damage location and damage degree, and sending it to the management terminal.

[0039] The comprehensive analysis results generate a detailed tunnel structure damage report, including damage location, severity, recommended countermeasures, etc., and the report is sent to the relevant management terminal to facilitate decision-making and maintenance operations.

[0040] In some embodiments, a comprehensive picture of the damaged area of ​​section B of a tunnel can be generated based on the comprehensive analysis results, marking the specific location of each damage point and its corresponding damage level, and the report can be synchronized to the tunnel operation management department to visualize the tunnel structure health monitoring results.

[0041] In the above implementation, when facing different types of tunnel monitoring processes, combined with the structural type of the tunnel, key features closely related to the health status of the tunnel are extracted from the huge amount of tunnel monitoring data, and possible structural damage is evaluated and located. This can not only effectively improve the analysis accuracy of the health status of the tunnel structure, but also quickly discover and locate potential damage that may exist in the tunnel. By timely evaluating the degree of damage, comprehensive tunnel health information is provided to managers, ensuring the long-term safe and stable operation of the tunnel.

[0042] Reference Figure 2 As an implementation method of step S103, the step of extracting features from the pre-processed structure monitoring data according to the tunnel structure type to obtain key features includes: Step S201, determining corresponding feature extraction parameters according to the tunnel structure type; Among them, the feature extraction parameters include the extraction threshold of the structural vibration frequency feature, the calculation window size of the displacement change rate feature, and the sensitivity of the crack width change trend feature; It is understandable that different types of tunnel structures may exhibit different physical properties and response patterns, so the parameters of feature extraction need to be adjusted to better accommodate these differences.

[0043] Among them, the extraction threshold of the structural vibration frequency feature refers to the threshold used to determine which frequency components are significant and worthy of further analysis when analyzing vibration data. By adjusting the extraction threshold, it can be ensured that the extracted features are more in line with the vibration characteristics of the specific tunnel structure, thereby improving the accuracy and reliability of monitoring; the calculation window size of the displacement change rate feature refers to the length of the data window used when calculating the displacement change rate over time. By adjusting the calculation window size and selecting different window lengths to calculate the displacement change rate, it can be ensured that the calculated displacement change rate more accurately reflects the displacement characteristics of the specific tunnel structure, thereby improving the accuracy and reliability of monitoring; the sensitivity of the crack width change trend feature refers to the sensitivity level used to determine which changes are significant and worthy of further analysis when analyzing crack width changes. By adjusting the sensitivity level to screen out meaningful crack width change trend features, it can be ensured that the extracted crack width change trend features are more in line with the crack characteristics of the specific tunnel structure, thereby improving the accuracy and reliability of monitoring.

[0044] Step S202, extracting key features of the preprocessed structural monitoring data based on feature extraction parameters; wherein the key features include structural vibration frequency features, displacement change rate features, and crack width change trend features.

[0045] In some specific embodiments, for tunnels constructed by the shield method, since this type of tunnel is usually located in softer geological conditions, its vibration frequency may be lower. Therefore, when extracting the structural vibration frequency characteristics, a lower extraction threshold can be set to identify low-frequency vibration modes, and a smaller displacement change rate calculation window size can be set; for tunnels constructed by the miner method, since this type of tunnel is often located in harder rocks, its vibration frequency may be higher and the displacement change rate may be smaller. Therefore, a higher vibration frequency threshold can be set, and a larger displacement change rate calculation window size can be set to capture high-frequency vibrations and small displacements occurring in the rock; for tunnels constructed by the immersed tube method, since this type of tunnel may be greatly affected by water flow, a higher sensitivity may need to be set when extracting the crack width change trend.

[0046] In the above implementation, by ensuring that the feature extraction process can more accurately reflect the characteristics of different types of tunnel structures, the monitoring scheme can better adapt to the monitoring needs of different types of tunnels, thereby improving the universality and effectiveness of the monitoring scheme.

[0047] As an implementation of step S104, the step of performing structural health status analysis based on key features to obtain tunnel structural health status analysis results includes: Input key features into the pre-built structural health analysis model and obtain the tunnel structural health status analysis results based on the model output; Among them, the structural health analysis model is pre-trained based on the historical structural monitoring data of the tunnel to be monitored and the pre-labeled health analysis result labels.

[0048] In the above implementation, the extracted key features such as stress and displacement can be used to judge the current health status of the tunnel through a pre-trained structural health analysis model, and the risk of major accidents can be reduced by timely discovering potential structural problems in the tunnel, thereby ensuring the structural safety and operational stability of the tunnel.

[0049] Reference Figure 3 As a further implementation of the tunnel structure health monitoring method, the method further includes a training step of a structure health analysis model, the training step including: Step S301, obtaining a sample data set and performing preprocessing; The sample dataset includes the historical structural monitoring data of the tunnel to be monitored and the pre-labeled health analysis result labels; Specifically, historical structural monitoring data may include time series data such as stress, displacement, vibration, etc. measured by sensors. In order to guide the training of machine learning models, health analysis result labels need to be introduced. These labels can be obtained through the experience of professional engineers or through simulation experiments, marking the potential damage locations and damage degrees of the tunnel, such as "healthy", "mildly damaged", "severely damaged", etc.

[0050] Among them, the preprocessing steps include cleaning up outliers, noise and other information in the original data set that are not conducive to model training, and standardizing or normalizing the data so that each feature has a similar scale, thereby improving the convergence speed and performance of the model.

[0051] Step S302, extracting features from the preprocessed sample data set according to the tunnel structure type of the tunnel to be monitored to obtain a key sample feature set; Among them, effective feature extraction based on the tunnel structure type of the tunnel to be monitored can more accurately reflect the characteristics of different types of tunnel structures, reduce the complexity of the model, and at the same time retain information that is crucial to structural health analysis, greatly improving the performance of the machine learning model.

[0052] It should be noted that the extraction process of the key sample feature set can refer to the detailed step description of step S201 to step S202 above.

[0053] Step S303, dividing the key sample feature set into a training set and a test set; In some embodiments, the data can be divided in a ratio of 8:2 or 7:3, with the larger part used for training the model and the smaller part used for testing the model, to ensure that the training set contains enough cases for the model to learn effective rules. At the same time, the test set is an independent data set of the model to verify the generalization ability of the model and avoid overfitting.

[0054] Step S304, inputting the training set into a pre-built machine learning model for training, optimizing model parameters, and obtaining a structural health analysis model; In one of the embodiments of the present application, the pre-built machine learning model may adopt a random forest model, which has strong robustness and generalization ability and good adaptability to high-dimensional data. In order to improve the robustness of the model, the random forest model of the embodiment of the present application is composed of multiple decision trees, each of which is generated by subsampling the sample data. The depth of each decision tree can be set in combination with the dimension and complexity of the tunnel data. When splitting nodes in each tree, some feature subsets are randomly selected for splitting to increase the diversity of the model and reduce the risk of overfitting.

[0055] During the model training process, random forest will automatically find the optimal combination of sample features and generate multiple decision trees. The depth of each tree and the number of trees can be adjusted through hyperparameters, and the loss function can be minimized by gradually adjusting its parameters.

[0056] Step S305 , testing the structural health analysis model and calibrating model parameters based on the test set until the loss function of the model meets a preset condition or the number of model iterations reaches a preset number, thereby obtaining a trained structural health analysis model.

[0057] Specifically, after the model is trained, it needs to be tested on a test set to test its prediction accuracy, stability, and generalization ability for unknown data. If the model does not perform well on the test set, it is necessary to further tune the model parameters, such as reducing the complexity of the model, increasing the size of the training data set, etc. Common methods include cross-validation and grid search.

[0058] In the above implementation, by extracting key features from the historical data of the tunnel and integrating them into the machine learning model for training and correction, a model that can accurately analyze the health status of the tunnel structure is successfully constructed. The model can automatically learn the health pattern of the tunnel structure from complex data sets and provide more accurate prediction and diagnosis for future monitoring, which not only improves the efficiency of tunnel monitoring, but also provides a basis for the maintenance and management of the tunnel, and has important engineering and economic value.

[0059] As an implementation of step S305, the step of testing the structural health analysis model based on the test set includes: A test model is established according to the test set, and the performance of the structural health analysis model is tested based on the test model; wherein the test indicators of the test model include accuracy, precision, recall rate, F1 score and confusion matrix.

[0060] Specifically, accuracy is the overall classification accuracy of the calculation model on the test set, that is, the ratio of the predicted label to the actual label; precision is to measure how many of the positive examples predicted by the model are truly positive examples for each category; recall measures the proportion of all samples that are actually positive that the model can correctly predict as positive; F1 score is the harmonic mean used to balance precision and recall; confusion matrix is ​​used to comprehensively evaluate the classification performance of the model in each category, including TP, TN, FP, and FN.

[0061] In the above implementation, the model performance is tested to ultimately evaluate the applicability of the model in a real environment and to ensure that the model has high accuracy and reliability in actual application.

[0062] Reference Figure 4 As an implementation of step S105, the step of identifying potential damage locations according to the tunnel structure health status analysis results includes: Step S401, obtaining structural information and load monitoring data of the tunnel to be monitored and performing data preprocessing; Among them, structural information includes tunnel design parameters and construction material properties; Specifically, the design parameters of the tunnel, such as geometric dimensions and cross-sectional shapes, and the properties of construction materials, such as concrete strength and reinforcement configuration, are the basis for accurately analyzing the health status of the tunnel structure; load monitoring data include real-time monitoring data of various loads that the tunnel bears in actual use, such as traffic loads, groundwater pressure, and seismic effects.

[0063] Among them, the data preprocessing step includes cleaning, filtering, normalizing and other processing of the acquired data to ensure the quality and availability of the data and improve the accuracy of subsequent analysis.

[0064] Step S402, constructing a finite element model based on the structural information of the tunnel to be monitored and the analysis result of the tunnel structure health status; The tunnel structure health status analysis result includes at least one structural health status indicator and at least one abnormal characteristic statistical information, and the finite element model includes the geometry, material properties and boundary conditions of the tunnel structure; Specifically, the structural health status indicators include but are not limited to health index, damage possibility prediction value, etc., and the abnormal characteristic statistical information includes but is not limited to abnormal value statistics of vibration, displacement, and crack width.

[0065] In one of the embodiments of the present application, a finite element model of the tunnel is constructed based on the structural information of the tunnel and the existing tunnel health status analysis results; the finite element method is a numerical calculation method based on discretization ideas, which can perform fine simulations on complex geometric shapes, material properties and boundary conditions. It is necessary to comprehensively utilize the design parameters and material properties of the tunnel, and combine the health status indicators (such as damage possibility prediction value, health index) and the statistical information of abnormal characteristics (such as vibration, displacement, crack width) to convert them into input conditions of the finite element model.

[0066] It can be understood that by constructing a finite element model based on the structural information and health status analysis results of the tunnel, and by considering the statistical information of abnormal characteristics in the health status analysis results, the actual behavior of the tunnel in a specific area can be simulated more accurately, making the model closer to the actual situation and helping to accurately identify potential damage sites.

[0067] Step S403, applying a load condition to the finite element model according to the load monitoring data; Specifically, actual load conditions are applied to the constructed finite element model. These conditions include but are not limited to traffic loads on the tunnel (such as loads when trains or vehicles pass through), groundwater pressure, earthquake effects, temperature changes and other incidental loads, so as to simulate the stress conditions of the tunnel in actual use. The real-time monitoring data of the tunnel is used as load input, which enhances the practical significance of the model and helps to accurately predict the structural response.

[0068] In some embodiments, the load conditions refer to various loads that the tunnel bears in actual use, including but not limited to traffic loads and groundwater pressure.

[0069] Step S404, calibrating the finite element model, adjusting the material properties and boundary conditions in the finite element model, so that the simulation results of the finite element model match the abnormal characteristic statistical information in the tunnel structure health status analysis results; Specifically, after the actual loading condition, the finite element model needs to be calibrated, that is, the material properties, boundary conditions and other parameters in the model need to be adjusted to make it more consistent with the actual structural health status. By comparing the simulation results obtained by the finite element model with the collected statistical information of abnormal characteristics (such as the actual observed crack width, vibration amplitude, etc.), the material stiffness, damping coefficient, boundary conditions and other parameters in the model are gradually adjusted until the error between the two is within the allowable range, so as to make the simulation results closer to the actual situation, ensure that the model can accurately reflect the actual state of the tunnel, and help to accurately identify potential damage sites.

[0070] Step S405, performing static or dynamic analysis in the calibrated finite element model to obtain structural response data; The structural response data include but are not limited to stress distribution, displacement change, crack width, etc.

[0071] Specifically, the calibrated finite element model can more accurately reflect the actual state of the tunnel structure. According to the actual use of the tunnel, static analysis (such as stress analysis) or dynamic analysis (such as vibration analysis) is selected. Static analysis is generally used to study long-term load effects, such as deadweight and continuous external loads; dynamic analysis is suitable for instantaneous loads such as earthquakes and impacts. Through calculation, structural response data such as stress, strain, displacement, and vibration can be obtained, which provides a basis for subsequent damage identification and helps to determine potential damage sites.

[0072] Step S406: Identify areas where the structural response is abnormally increased based on the structural response data, and mark them as potential damage areas.

[0073] Specifically, the obtained structural response data, especially key parameters such as stress distribution, displacement change and crack width, are used to analyze and identify areas where abnormal responses may exist. If the stress value and strain amplitude in some areas are significantly higher than those in other areas, or significantly exceed the safety threshold in the design specifications, these areas may be potential damage sites.

[0074] In some embodiments, for example, if the analysis shows that the stress concentration at a certain point on the left wall of the tunnel is significantly higher than that in other areas, and this stress value greatly exceeds the yield strength of the material, the area should be marked as a potential damage site, and it is recommended to strengthen monitoring or repair.

[0075] In the above implementation, a more accurate finite element model is constructed by combining the structural health status analysis results and load monitoring data. By analyzing the structural response data, potential damage sites can be accurately identified, providing a scientific basis for maintenance decisions. This technical solution can effectively improve the accuracy and reliability of tunnel structure monitoring and provide strong support for the safe operation of tunnels.

[0076] Reference Figure 5 As an implementation of step S105, the step of evaluating the damage degree of the potential damage site includes: Step S501, configuring a damage degree quantitative index threshold according to the tunnel structure type; In some embodiments, the damage degree quantitative index threshold may include a maximum stress value threshold, a maximum displacement threshold, and a maximum crack width threshold, which may be pre-set based on historical data, material characteristics, design standards and specifications, etc., and the quantitative index threshold may be adjusted according to the tunnel structure type (such as shield construction, mining construction, etc.); For example, tunnels constructed by the shield method may require lower stress thresholds and crack width thresholds because such tunnels are more sensitive to damage; tunnels constructed by the mining method may require higher stress thresholds and crack width thresholds because such tunnels generally have better bearing capacity.

[0077] Step S502: compare the damage degree quantification index threshold with the structural response data of the potential damage location to obtain the damage degree corresponding to the potential damage location.

[0078] Among them, the calculated structural response data index is compared with the configured damage degree quantitative index threshold. If the structural response data exceeds the threshold, it indicates that there is a certain degree of damage in the potential damage part. According to the difference between the structural response data and the threshold, the damage level (mild, moderate, severe, etc.) can be determined, such as mild (difference below 0.1), moderate (difference between 0.1 and 0.3), and severe (difference between 0.3 and 0.6).

[0079] In the above implementation, reasonable thresholds and quantification methods are set according to different tunnel structure types, which can more accurately identify the health status of the structure, make maintenance and repair work more targeted, and help to reasonably arrange repair resources and avoid unnecessary or excessive repairs.

[0080] It is understandable that, in actual application, configuring the damage degree quantification index threshold and calculating the damage degree are important steps to ensure the safety of tunnel structures and are also the core components of modern tunnel monitoring systems.

[0081] The embodiment of the present application also discloses a tunnel structure health monitoring system.

[0082] A tunnel structure health monitoring system, the monitoring system comprising: An acquisition module is used to acquire the structural monitoring data of the tunnel to be monitored and perform preprocessing; A structure type determination module, used to determine the tunnel structure type of the tunnel to be monitored; A feature extraction module is used to extract features from the pre-processed structural monitoring data according to the tunnel structure type to obtain key features; Health status analysis module, used to perform structural health status analysis based on key features and obtain tunnel structure health status analysis results; Structural damage identification module, used to identify potential damage sites based on tunnel structural health status analysis results and assess the damage extent of potential damage sites; The structural damage report push module is used to generate a tunnel structure damage report based on the potential damage location and damage degree, and send it to the management terminal.

[0083] In the above implementation, through the integrated multi-modular design, comprehensive monitoring and analysis of the health status of the tunnel structure is achieved. Feature extraction combined with the tunnel type can effectively identify key health status indicators. Based on these key features, the system can not only accurately assess the overall health status of the tunnel, but also identify possible damage locations and assess the extent of damage. Ultimately, by generating and pushing structural damage reports, the system can provide managers with timely and accurate decision support, significantly improving the safety and service life of the tunnel. In actual application, this technical solution helps to reduce dependence on manual inspections, reduce maintenance costs, and at the same time improve the operational safety and efficiency of the tunnel.

[0084] The tunnel structure health monitoring system of the embodiment of the present application can implement any of the above-mentioned tunnel structure health monitoring methods, and the specific working process of each module in the tunnel structure health monitoring system can refer to the corresponding process in the above-mentioned method embodiment.

[0085] In the several embodiments provided in this application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are only illustrative; for example, the division of a certain module is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0086] The embodiment of the present application also discloses a computer device.

[0087] The computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the tunnel structure health monitoring method as described above is implemented.

[0088] The embodiment of the present application also discloses a computer-readable storage medium.

[0089] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any of the tunnel structure health monitoring methods described above.

[0090] Among them, computer-readable storage media can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus or device; the program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0091] It should be noted that in the above embodiments, the description of each embodiment has different emphases, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0092] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A tunnel structure health monitoring method, characterized in that: The monitoring method comprises: Obtain the structural monitoring data of the tunnel to be monitored and perform preprocessing; Determine the tunnel structure type of the tunnel to be monitored, Performing feature extraction on the pre-processed structure monitoring data according to the tunnel structure type to obtain key features; Perform structural health status analysis based on the key features to obtain tunnel structural health status analysis results; Identifying potential damage locations according to the tunnel structure health status analysis results, and evaluating the damage extent of the potential damage locations; A tunnel structure damage report is generated according to the potential damage location and damage degree, and sent to the management terminal.

2. A tunnel structure health monitoring method according to claim 1, characterized in that: The step of extracting features from the pre-processed structure monitoring data according to the tunnel structure type to obtain key features comprises: Determining corresponding feature extraction parameters according to the tunnel structure type; the feature extraction parameters include an extraction threshold of a structural vibration frequency feature, a calculation window size of a displacement change rate feature, and a sensitivity of a crack width change trend feature; The key features of the preprocessed structural monitoring data are extracted based on the feature extraction parameters; the key features include structural vibration frequency features, displacement change rate features and crack width change trend features.

3. A tunnel structure health monitoring method according to claim 1, characterized in that: The steps of performing structural health status analysis based on the key features to obtain tunnel structural health status analysis results include: The key features are input into a pre-built structural health analysis model, and the tunnel structure health status analysis results are obtained according to the model output; wherein the structural health analysis model is pre-trained based on the historical structural monitoring data of the tunnel to be monitored and pre-labeled health analysis result labels.

4. A tunnel structure health monitoring method according to claim 3, characterized in that: The method further includes a training step of the structural health analysis model, wherein the training step includes: Acquire a sample data set and perform preprocessing; wherein the sample data set includes historical structural monitoring data of the tunnel to be monitored and pre-labeled health analysis result labels; Performing feature extraction on the preprocessed sample data set according to the tunnel structure type of the tunnel to be monitored to obtain a key sample feature set; Dividing the key sample feature set into a training set and a test set; Inputting the training set into a pre-built machine learning model for training, optimizing model parameters, and obtaining the structural health analysis model; The structural health analysis model is tested and model parameters are corrected based on the test set until the loss function of the model meets a preset condition or the number of model iterations reaches a preset number, thereby obtaining the trained structural health analysis model.

5. A tunnel structure health monitoring method according to claim 4, characterized in that: The step of testing the structural health analysis model based on the test set comprises: A test model is established according to the test set, and the performance of the structural health analysis model is tested based on the test model; wherein the test indicators of the test model include accuracy, precision, recall rate, F1 score and confusion matrix.

6. A tunnel structure health monitoring method according to claim 1, characterized in that: The steps of identifying potential damage locations according to the tunnel structure health status analysis results include: Acquiring structural information and load monitoring data of the tunnel to be monitored and performing data preprocessing; wherein the structural information includes tunnel design parameters and construction material characteristics; Based on the structural information of the tunnel to be monitored and the tunnel structure health status analysis result, a finite element model is constructed; wherein the tunnel structure health status analysis result includes at least one structural health status indicator and at least one abnormal characteristic statistical information, and the finite element model includes the geometry, material properties and boundary conditions of the tunnel structure; applying a load condition on the finite element model according to the load monitoring data; Calibrate the finite element model, and adjust the material properties and boundary conditions in the finite element model so that the simulation results of the finite element model match the abnormal characteristic statistical information in the tunnel structure health status analysis results; Perform static or dynamic analysis on the calibrated finite element model to obtain structural response data; Based on the structural response data, areas where the structural response is abnormally increased are identified and marked as potential damage sites.

7. A tunnel structure health monitoring method according to claim 6, characterized in that: The steps of assessing the extent of damage to the potential damage site include: Configure the damage degree quantitative index threshold according to the tunnel structure type; The damage degree quantification index threshold is compared with the structural response data of the potential damage part to obtain the damage degree corresponding to the potential damage part.

8. A tunnel structure health monitoring system, characterized in that: The monitoring system comprises: An acquisition module is used to acquire the structural monitoring data of the tunnel to be monitored and perform preprocessing; A structure type determination module, used to determine the tunnel structure type of the tunnel to be monitored; A feature extraction module, used for extracting features from the pre-processed structure monitoring data according to the tunnel structure type to obtain key features; A health status analysis module, used to perform structural health status analysis based on the key features to obtain tunnel structure health status analysis results; A structural damage identification module, used to identify potential damage locations according to the tunnel structure health status analysis result, and evaluate the damage degree of the potential damage locations; The structural damage report push module is used to generate a tunnel structure damage report according to the potential damage location and damage degree, and send it to the management terminal.

9. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the program.

10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 7.

Images