Subway tunnel lining disease detection method and system, terminal and medium

Abstract

The invention relates to the field of subway tunnel lining disease detection technologies, in particular to a subway tunnel lining disease detection method and system, a terminal and a medium, and the method comprises the steps: obtaining a detection instruction; obtaining tunnel information based on the detection instruction; identifying disease information based on the tunnel information; acquiring the disease type based on the disease information; obtaining the number of the disease types based on the disease types, and calculating a difference value between the number of each disease type and a preset value; sorting the plurality of difference values; obtaining maintenance priorities of the plurality of disease types based on the sorting of the plurality of difference values; and obtaining a maintenance scheme of each disease type in sequence based on the maintenance priorities of the plurality of disease types. The detection efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of subway tunnel lining defect detection technology, and in particular to a subway tunnel lining defect detection method, system, terminal and medium. Background Art

[0002] As a permanent supporting structure for tunnel engineering, tunnel lining is an important guarantee for the normal operation and maintenance of traffic tunnels. As the tunnel service life increases, various defects such as cracks, cavities, water leakage, spalling, deformation and misalignment begin to appear one after another due to various internal and external factors. These defects not only affect the service life of the tunnel, but also endanger the driving safety of traffic vehicles. In order to ensure the safe operation of the tunnel, it is particularly important to conduct regular lining defect detection.

[0003] The commonly used inspection method for subway tunnels now is the traditional manual inspection method, that is, manual inspection with handheld inspection tools. This method is affected by the working status and experience level of the inspectors, and the quality of the inspection data cannot be guaranteed. At the same time, it consumes a lot of manpower and material resources and is inefficient. It is becoming increasingly difficult to adapt to the actual needs of tunnel inspection engineering sites. Summary of the invention

[0004] In a first aspect, in order to improve detection efficiency, the present application provides a subway tunnel lining defect detection method, system, terminal and medium.

[0005] This application provides a method for detecting subway tunnel lining defects, which adopts the following technical solutions: A method for detecting lining defects of a subway tunnel, comprising: Get the detection instruction; Obtain tunnel information based on detection instructions; Identify disease information based on tunnel information; Obtain the disease type based on the disease information; Obtain the number of disease types based on the disease type and calculate the difference between the number of each disease type and a preset value; Sort multiple differences; Obtain maintenance priorities for multiple types of damage based on the sorting of multiple differences; Based on the maintenance priorities of multiple types of defects, a maintenance plan for each type of defect is obtained in turn.

[0006] By adopting the above technical solution, the disease information is automatically identified based on tunnel information (such as images, point cloud data, sensor data, etc.), and classified according to different types of diseases. This automated process greatly reduces the errors and time consumption of manual inspection, and can detect the diseases of the tunnel structure in real time. The classification and identification of diseases are carried out in a systematic way, which can avoid misjudgments caused by differences in experience or fatigue of manual inspectors; by calculating the difference between the number of each disease type and the preset value, the severity of each disease can be quantitatively evaluated, which makes the disease assessment more scientific and objective, rather than relying solely on the intuitive feeling of manual assessment. This method can clearly identify which disease types are too numerous and exceed the preset standards, and then focus on those serious disease types; based on the maintenance priority of different types of diseases, more targeted maintenance plans can be formulated. For example, for different types of diseases such as cracks, corrosion, and water seepage, personalized maintenance plans can be formulated according to the specific characteristics of the disease and the difficulty of maintenance, avoiding a one-size-fits-all approach.

[0007] Preferably, the step of identifying disease information based on tunnel information includes: Perform preprocessing based on tunnel information and obtain preprocessing information and a disease identification model; The preprocessed information is input into the disease recognition model for recognition and disease information is obtained.

[0008] By adopting the above technical solutions, for large-scale subway tunnel systems, automated diagnosis can greatly reduce the frequency and manpower requirements of manual inspections, thereby reducing the cost of long-term operation and maintenance. By timely identifying defects and conducting data analysis, the system can provide a forecast of the development trend of defects and provide predictive support for tunnel maintenance. This approach is more cost-effective than traditional reactive maintenance and helps avoid the high cost of emergency repairs or large-scale repairs.

[0009] Preferably, the step of obtaining the disease type based on the disease information includes: Obtain disease information; The disease information is identified and matched with the corresponding disease type in the disease tree knowledge base.

[0010] By adopting the above technical solution, the determination of the disease type no longer relies on manual judgment by matching the disease information with the disease tree knowledge base, but is completed through automated knowledge base retrieval. This not only greatly improves the detection efficiency, but also reduces the need for manual intervention, freeing up the workload of maintenance personnel. The automatic matching process greatly shortens the time from obtaining disease information to the final confirmation of the disease type. Especially when facing a large number of tunnel disease detections, automated processing significantly improves the overall work efficiency, allowing detection and repair work to be carried out in a timely manner.

[0011] Preferably, the tunnel information is collected by ground detection equipment and multiple types of sensors, which includes tunnel images, point cloud data, temperature and humidity information, strain information, displacement information and crack information.

[0012] By adopting the above technical solutions, the data collected by different sensors and detection equipment can comprehensively monitor the health status of the tunnel from multiple dimensions. For example, tunnel images can provide clear visual information, point cloud data can accurately reflect the spatial changes of the tunnel structure, temperature and humidity information can reveal the impact of environmental conditions on the tunnel, displacement and strain information can directly reflect the structural deformation and stress state of the tunnel, and crack information can help to find tiny cracks on the surface of the tunnel. By integrating multi-dimensional data, it is possible to accurately identify potential tunnel diseases (such as cracks, settlement, deformation, etc.) and conduct a full range of structural health diagnosis. The collaborative work of multiple sensors can make up for the limitations of single sensor data and improve the accuracy of tunnel disease monitoring.

[0013] Preferably, the method comprises: Collect high-definition images of subway tunnels; Obtaining image samples of the subway tunnel, where the image samples are images taken when the subway tunnel was just built; Compare high-definition images with image samples to obtain detection information on tunnel changes and changed locations; Obtain disease information based on the detection information of the changed parts.

[0014] By adopting the above technical solution, the location and area of ​​the tunnel structure that has changed can be clearly identified by comparing the high-definition image with the sample image. This method can help engineers locate the specific diseased part in the shortest time without the need to conduct a comprehensive inspection of the entire tunnel, reducing the scope and time of the inspection; by regularly comparing high-definition images at different time points, the evolution trend of the disease can be understood. For example, whether a crack is expanding, whether the settlement is increasing, etc. In this way, the severity of the disease can be judged, and the maintenance plan can be adjusted in time to prevent the problem from expanding.

[0015] Preferably, the step of sequentially acquiring a maintenance plan for each type of damage based on the maintenance priorities of multiple types of damage comprises: Get the disease type; Obtain the corresponding disease cause based on the disease type; Obtain corresponding maintenance solutions based on the cause of the disease; Get the repair priority of multiple disease types; Based on the maintenance priority of each disease type, the maintenance plan for each disease type is output in turn.

[0016] By adopting the above technical solutions, the maintenance plan corresponding to each disease type is based on an in-depth analysis of the cause of the disease. By understanding the root causes of the disease (such as structural cracks, material aging, water seepage, etc.), more effective and targeted maintenance plans can be developed. For example, structural cracks may require grouting reinforcement, while material aging may require the replacement of damaged parts. This can improve the accuracy and effectiveness of maintenance work and ensure that the implementation of the plan can solve the fundamental problem. The maintenance plan based on the cause of the disease can be more personalized and customized, ensuring that different types of diseases receive the most appropriate repair method. It avoids simple "one-size-fits-all" repair methods and reduces secondary damage or waste of resources caused by improper maintenance.

[0017] Preferably, the method further comprises: Obtain the historical maintenance personnel corresponding to the maintenance plan; If there are multiple historical maintenance personnel, the historical maintenance personnel with the highest repair success rate corresponding to the disease type is obtained as the target maintenance personnel; The target maintenance personnel performs maintenance work according to the maintenance plan.

[0018] By adopting the above technical solution, by selecting the maintenance personnel with the highest repair success rate based on historical maintenance data, it can be ensured that the maintenance work is completed by the personnel with the most technical capabilities and experience. Such maintenance personnel not only have more practical experience, but can also effectively avoid repair failures or rework caused by operating errors or unfamiliarity with the type of disease, thereby improving the quality and success rate of maintenance. Selecting the maintenance personnel with the highest success rate as the target maintenance personnel can reduce the risk of failure caused by differences in maintenance personnel's skills. Especially when facing more complex or serious diseases, maintenance personnel with a high success rate can better deal with complex situations and reduce the occurrence of failures and accidents.

[0019] Metro tunnel lining defect detection In the second aspect, in order to improve the detection efficiency, the present application provides a subway tunnel lining disease detection system, which adopts the following technical solutions: A subway tunnel lining disease detection system, comprising: An instruction acquisition module is used to acquire detection instructions; A detection module, used for obtaining tunnel information based on a detection instruction; A disease identification module is used to identify disease information based on tunnel information; A disease type determination module is used to obtain the disease type based on the disease information; A difference calculation module, used for obtaining the number of disease types based on the disease type and calculating the difference between the number of each disease type and a preset value; A sorting module, used for sorting the multiple differences and obtaining the maintenance priorities of multiple disease types based on the sorting of the multiple differences; A maintenance plan generation module is used to obtain a maintenance plan for each type of damage in turn based on the maintenance priorities of multiple types of damage.

[0020] In a third aspect, the present application provides a smart terminal, which adopts the following technical solution: An intelligent terminal comprises a memory and a processor, wherein the memory stores a computer program which can be loaded by the processor and execute the above-mentioned subway tunnel lining disease detection method.

[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 the above-mentioned subway tunnel lining disease detection method.

[0022] In summary, the present application includes at least one of the following beneficial technical effects: 1. Automatically identify disease information based on tunnel information (such as images, point cloud data, sensor data, etc.) and classify them according to different types of diseases. This automated process greatly reduces the errors and time consumption of manual inspections, and can detect tunnel structure diseases in real time. The classification and identification of diseases are carried out in a systematic way, which can avoid misjudgments due to differences in experience or fatigue of manual inspectors; by calculating the difference between the number of each disease type and the preset value, the severity of each disease can be quantitatively evaluated, which makes the disease assessment more scientific and objective, rather than relying solely on the intuitive feeling of manual assessment. This method can clearly identify which types of diseases are too numerous and exceed the preset standards, and then focus on those serious types of diseases; based on the maintenance priorities of different types of diseases, more targeted maintenance plans can be formulated. For example, for different types of diseases such as cracks, corrosion, and water seepage, personalized maintenance plans can be formulated according to the specific characteristics of the disease and the difficulty of maintenance to avoid a one-size-fits-all approach; 2. Data collected by different sensors and detection equipment comprehensively monitor the health status of the tunnel from multiple dimensions. For example, tunnel images can provide clear visual information, point cloud data can accurately reflect the spatial changes of the tunnel structure, temperature and humidity information reveals the impact of environmental conditions on the tunnel, displacement and strain information directly reflects the structural deformation and stress state of the tunnel, and crack information helps to detect tiny cracks on the surface of the tunnel. By integrating multi-dimensional data, potential tunnel diseases (such as cracks, settlement, deformation, etc.) can be accurately identified, and a full range of structural health diagnosis can be performed. The collaborative work of multiple sensors can make up for the limitations of single sensor data and improve the accuracy of tunnel disease monitoring; 3. By comparing high-definition images with sample images, the locations and areas where changes have occurred in the tunnel structure can be clearly identified. This method can help engineers locate specific diseased parts in the shortest possible time without the need for a comprehensive inspection of the entire tunnel, reducing the scope and time of the inspection; by regularly comparing high-definition images at different time points, the evolution trend of the disease can be understood. For example, whether a crack is expanding, whether settlement is increasing, etc. This can determine the severity of the disease, adjust the maintenance plan in a timely manner, and prevent the problem from expanding. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a method flow chart of the subway tunnel lining disease detection method according to an embodiment of the present application, mainly showing steps S100-S800.

[0024] Figure 2 It is a method flow chart of the subway tunnel lining disease detection method according to an embodiment of the present application, mainly showing steps SC1-SC4.

[0025] Figure 3 It is a method flow chart of the subway tunnel lining disease detection method according to an embodiment of the present application, mainly showing steps SD1-SD5. DETAILED DESCRIPTION

[0026] The present application is further described in detail below in conjunction with all the accompanying drawings.

[0027] The command can be obtained by mechanical button triggering or virtual button triggering. In the mechanical button triggering method, the command can be automatically obtained after turning on the power by pressing the power button, or the current behavior information can be obtained by pressing the corresponding trigger button again after turning on the power. In the virtual button triggering method, the command can be obtained by pressing the relevant virtual trigger button in the interface of the corresponding software.

[0028] The present application embodiment discloses a method for detecting subway tunnel lining defects. Figure 1 , Figure 2 , Figure 3 , a subway tunnel lining disease detection method includes: Step S100: obtaining a detection instruction; Specifically, the tester can trigger the detection instruction by clicking or touching a button on the human-computer interaction interface of the terminal.

[0029] Step S200: obtaining tunnel information based on the detection instruction; Specifically, tunnel information is collected by ground detection equipment and various sensors, including tunnel images, point cloud data, temperature and humidity information, strain information, displacement information and crack information. Real-time video images are obtained through cameras installed in the tunnel to help detect whether there are cracks, peeling, corrosion, water seepage and other defects on the lining surface. These images can be used for visual inspection and can automatically identify defects through image processing algorithms. The temperature distribution of the tunnel lining is detected by infrared imaging technology to identify potential heat sources or temperature abnormalities (such as moisture infiltration, pipeline leakage, etc.). This data helps to analyze hidden problems in the tunnel structure. Laser scanners (LiDAR) or ground laser scanning (TLS) technology are used to obtain three-dimensional point cloud data of the tunnel to help build a three-dimensional model of the tunnel. Point cloud data can accurately capture the geometry of the tunnel and its deformation, such as cracking and displacement of the lining. By comparing point cloud data collected at different time points, it is possible to identify small deformations or displacements of the tunnel structure and discover potential structural problems or defects. Point cloud data can also be used to monitor whether the tunnel wall has subsidence or uneven deformation. Temperature sensors are used to monitor temperature changes in the tunnel, especially in areas with large seasonal changes. Temperature differences may cause expansion and contraction of lining materials, leading to cracks or other defects. Humidity sensors can monitor humidity levels in the tunnel. Excessive humidity may cause groundwater infiltration, corrosion of concrete, and mold growth. Long-term high humidity environments may also accelerate the corrosion process of the lining. Combining temperature and humidity data can provide a more comprehensive understanding of the environmental conditions of the tunnel and predict the types of defects that may be caused. For example, in an environment with high temperature and humidity, steel reinforcement corrosion of reinforced concrete is prone to occur. Strain sensors can detect small deformations on the surface of the tunnel lining, especially in areas where cracks or subsidence may exist. By monitoring strain in real time, abnormal changes in the tunnel structure can be discovered in advance; displacement sensors are used to monitor the displacement of the tunnel inner wall and detect whether the lining has uneven settlement, slippage or deformation; doors are used to monitor the width of cracks on the tunnel lining surface to help determine the development trend of the cracks and whether reinforcement measures are needed. The above ground detection settings and multiple types of sensors are set up in advance by the staff.

[0030] Step S300: identifying damage information based on tunnel information; The specific steps of identifying disease information include: Step SA1: preprocessing based on tunnel information and obtaining preprocessing information and a disease identification model; Step SA2: Input the preprocessed information into the disease recognition model for recognition and obtain disease information. Specifically, the raw data collected by ground detection equipment or multiple sensors are de-noised, filtered and completed to ensure the accuracy and integrity of the data for subsequent analysis. The surface image of the tunnel is processed using image processing techniques (such as edge detection, morphological transformation, texture analysis, etc.) to extract the potential characteristics of the disease. Common diseases such as cracks, spalling, corrosion, etc. may be characterized by color changes, irregular shapes, surface depressions, etc. The data obtained from different sensors (such as infrared, laser scanning, video images) are fused to further improve the accuracy of identifying disease characteristics. For example, infrared sensors can help identify surface temperature differences and reveal cracks or water leakage problems in tunnel linings; laser scanning can accurately obtain the geometric characteristics of cracks (such as width and depth). The features extracted from the tunnel information are compared with the pre-established disease pattern library. The pattern library contains feature information of common disease types, such as crack width, depth, shape, and area of ​​spalling areas. The system identifies diseases by establishing a disease recognition model for pre-processed information. Convolutional neural network (CNN) in deep learning is one of the most commonly used models and can be used as a disease recognition model. CNN can automatically extract features from the original image and identify the morphological characteristics of defects such as cracks, spalling, and corrosion. Defect recognition models also include multimodal data fusion models. These models combine data from different sensors and comprehensively analyze defect information through weighted average, decision tree, or neural network methods to improve the accuracy of defect recognition. In addition, the specific location of the defect can be determined through geometric analysis, and the spatial coordinates of the defect can be generated to facilitate subsequent positioning and maintenance operations. The spatial location of the defect is usually calibrated by the coordinate system of the tunnel or the detailed drawings of the tunnel.

[0031] Step S400: obtaining the disease type based on the disease information; The above steps specifically include: Step SB1: Obtain disease information; Step SB2: Identify the disease information and match the corresponding disease type in the disease tree knowledge base.

[0032] Specifically, after obtaining the disease information, it is necessary to identify and match the disease type based on this information. This process usually relies on a disease tree knowledge base, which contains the definition, characteristics, impact and other information of different types of diseases. The disease tree knowledge base can be a structured data set containing multiple disease categories. Each disease type usually includes the following information: disease name: such as "crack", "corrosion", etc.; feature description: including the morphology of the disease, the conditions of occurrence, appearance characteristics, etc.; diagnostic criteria: how to judge the occurrence of the disease, including sensor data thresholds, image features, etc. According to the disease information obtained from the disease recognition model (such as disease type, location information, confidence, etc.), find the corresponding disease type in the knowledge base and compare it with the existing disease information. For the matching of disease types, directly compare the disease type identified by the model with the disease name in the knowledge base; for the matching of features such as location and morphology, the image features, sensor data and diagnostic criteria in the knowledge base can be calculated to match. For example, morphological analysis can be used to analyze the morphology of cracks to determine whether they conform to the definition of cracks in the knowledge base; for confidence matching, a certain confidence threshold can be set, and only when the identified confidence is greater than the threshold, the match is considered successful. Once a matching disease type is found, the corresponding disease type and related information can be returned. The types of defects can be divided into crack defects: types of cracks (longitudinal, transverse, oblique cracks, micro cracks) and their causes; peeling / stripping defects: causes of peeling or shedding of the lining surface; corrosion defects: causes and consequences of corrosion problems such as steel corrosion and concrete carbonization; water seepage defects: causes of tunnel water seepage, such as cracks, poor joints or failure of the drainage system; deformation defects: settlement, deformation, tilting, etc. of the tunnel lining, and their possible causes; void defects: causes of the formation of voids in the lining and their effects; expansion defects: lining rupture caused by expansion reactions; defects caused by geological changes: lining defects caused by changes in geological conditions (such as earthquakes, groundwater flow, etc.); surface coating aging: aging, peeling, etc. of the tunnel lining surface coating; construction defects: defects caused by insufficient quality control during construction, such as loose joints and improper concrete mix.

[0033] Step S500: obtaining the number of disease types based on the disease type and calculating the difference between the number of each disease type and a preset value; Step S600: sorting multiple difference values; Step S700: Obtaining maintenance priorities of multiple types of defects based on the sorting of multiple differences; Step S800: Obtaining a maintenance plan for each type of fault in turn based on the maintenance priorities of multiple types of faults.

[0034] Specifically, after the disease information is identified and classified, the number of each disease type is counted. Then it is compared with the preset threshold and the difference is calculated. The preset value can be a threshold based on historical data, engineering standards or industry standards, such as "the number of cracks per kilometer does not exceed 10". Calculation of the difference: Difference = Current number of disease types - preset threshold. Sort the differences of all disease types and start processing from the disease type with the largest difference, because the larger the difference, the more serious the disease problem of this type, and it needs priority repair. According to the sorting results, the repair priority of each disease type is determined. The repair priority is determined based on the difference size and the severity of the disease type. Generally, the disease type with a large difference is repaired first. A priority list can be set according to the actual situation. According to the repair priority of each disease type, the maintenance plan related to the disease type is obtained in turn. Each disease type may correspond to a different repair plan, for example, cracks may require grouting repair, corrosion may require surface coating or replacement of part of the structure, and water seepage problems may require replacement of the waterproof layer.

[0035] In addition to the above solutions, other methods are also included. The specific steps include: Step SC1: Collect high-definition images of the subway tunnel; Step SC2: obtaining image samples of the subway tunnel, where the image samples are images taken when the subway tunnel is just built; Step SC3: Compare the high-definition image with the image sample to obtain the tunnel change status and detection information of the changed parts; Step SC4: Obtain disease information based on the detection information of the changed part. Specifically, high-definition images of subway tunnels usually require the use of high-resolution cameras or laser scanners. Modern subway tunnel inspections often use a combination of laser radar (LiDAR) and high-resolution cameras to obtain a three-dimensional model and high-definition images of the tunnel. Obtaining image samples when the tunnel is built is the key to comparing the current tunnel changes. Image samples need to be taken at the beginning of tunnel construction, and these images should contain detailed information about each part of the tunnel. After obtaining the current high-definition image and the initial image samples of the tunnel, the next step is to compare the two sets of images to identify the changes in the tunnel and the parts that have changed. This process can be achieved through image difference detection and image registration technology. Image Registration: Ensure that the current high-definition image is aligned with the image sample. Classic image registration methods can be used, such as feature point-based registration (such as SIFT, ORB) or deep learning-based image alignment methods (such as convolutional neural networks). Compare the two images and identify changes in the tunnel by calculating the image difference. Common methods include: directly calculating the difference between pixels (such as through image difference, image background subtraction, etc.). Detect changes in the surface structure of the tunnel, such as cracks and shedding. After completing the change detection, the next step is to extract specific disease information based on the detected changed parts. The disease information is obtained in the same way as the above method.

[0036] In addition, the specific steps for obtaining maintenance solutions include: Step SD1: Obtain disease type; Step SD2: Obtain the corresponding disease cause based on the disease type; Step SD3: Obtaining a corresponding maintenance plan based on the cause of the disease; Step SD4: Obtain maintenance priorities of multiple disease types; Step SD5: Output the maintenance plan for each disease type in turn based on the maintenance priority of each disease type.

[0037] Specifically, different types of diseases have different inducing causes. Based on historical data, literature research and expert knowledge, a rule base for the causes of diseases is established. When a certain type of disease is detected, the possible cause of the disease can be deduced according to the rules. If supported by historical data, the cause of the disease can be predicted by a machine learning model. For each type of disease and its cause, a corresponding maintenance plan can be formulated. The maintenance plan is usually determined based on the nature and cause of the disease, and the feasibility and cost of the maintenance need to be considered. Crack maintenance plan: Cracks can be treated by injection repair, reinforcement, coating repair, etc. Water seepage maintenance plan: For water seepage problems, waterproof layer repair, sealing and other work can be carried out. Corrosion maintenance plan: Corrosion can be repaired by replacing the corroded part, applying anti-corrosion layer, improving ventilation and other measures. Shedding and deformation maintenance plan: It needs to be repaired by reinforcement, support, re-construction and other methods. Finally, it is necessary to sort and output the maintenance plan for each type of disease according to the maintenance priority. This can be done by sorting the disease types and then outputting the corresponding maintenance plans one by one.

[0038] After the maintenance plan is in place, the maintenance personnel need to be determined. The specific steps include: Step SE1: Obtain the historical maintenance personnel corresponding to the maintenance plan; Step SE2: If there are multiple historical maintenance personnel, the historical maintenance personnel with the highest repair success rate corresponding to the disease type are obtained as the target maintenance personnel; Step SE3: The target maintenance personnel performs maintenance work according to the maintenance plan.

[0039] Specifically, each maintenance plan may involve multiple historical maintenance personnel, who are usually recorded in a maintenance personnel database or record system. The database contains basic information of each maintenance personnel (such as name, work number), the type of maintenance plan implemented, and their repair success rate when implementing the plan. Establish a historical maintenance personnel record table, which contains: maintenance personnel ID; maintenance plan type (such as crack repair, water seepage repair, etc.); repair success rate (which can be calculated based on the results of historical maintenance); time and location of maintenance tasks and other information. Sort historical maintenance personnel by repair success rate. Select the maintenance personnel with the highest success rate. Assign specific tasks to maintenance personnel, which can be through a notification system, a work management system, or by directly generating work orders. Maintenance personnel perform repair work according to the given maintenance plan. This step may include providing maintenance tools, materials, arranging maintenance time, etc.

[0040] The implementation principle of a subway tunnel lining disease detection method in the embodiment of the present application is: automatically identifying disease information based on tunnel information (such as images, point cloud data, sensor data, etc.), and classifying according to different types of diseases. This automated process greatly reduces the error and time consumption of manual inspection, and can detect the diseases of the tunnel structure in real time. The classification and identification of diseases are carried out in a systematic way, which can avoid misjudgment caused by experience differences or fatigue of manual inspectors; by calculating the difference between the number of each disease type and the preset value, the severity of each disease can be quantitatively evaluated, which makes the disease assessment more scientific and objective, rather than relying solely on the intuitive feeling of manual assessment. This method can clearly identify which disease types are too numerous and exceed the preset standards, and then focus on those serious disease types; based on the maintenance priorities of different types of diseases, more targeted maintenance plans can be formulated. For example, for different types of diseases such as cracks, corrosion, and water seepage, personalized maintenance plans can be formulated according to the specific characteristics of the disease and the difficulty of maintenance to avoid a one-size-fits-all approach.

[0041] The embodiment of the present application discloses a subway tunnel lining defect detection system.

[0042] A subway tunnel lining defect detection system, comprising: An instruction acquisition module is used to acquire detection instructions; A detection module, used for obtaining tunnel information based on a detection instruction; A disease identification module is used to identify disease information based on tunnel information; A disease type determination module is used to obtain the disease type based on the disease information; A difference calculation module, used for obtaining the number of disease types based on the disease type and calculating the difference between the number of each disease type and a preset value; A sorting module, used for sorting the multiple differences and obtaining the maintenance priorities of multiple disease types based on the sorting of the multiple differences; A maintenance plan generation module is used to obtain a maintenance plan for each type of damage in turn based on the maintenance priorities of multiple types of damage.

[0043] The present application provides a smart terminal.

[0044] An intelligent terminal comprises a memory and a processor, wherein the memory stores a computer program which can be loaded by the processor and execute the above-mentioned subway tunnel lining disease detection method.

[0045] The present application provides a computer-readable storage medium.

[0046] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute the above-mentioned subway tunnel lining disease detection method.

[0047] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for detecting lining defects in a subway tunnel, characterized in that: include: Get the detection instruction; Obtain tunnel information based on detection instructions; Identify disease information based on tunnel information; Obtain the disease type based on the disease information; Obtain the number of disease types based on the disease type and calculate the difference between the number of each disease type and a preset value; Sort multiple differences; Obtain maintenance priorities for multiple types of damage based on the sorting of multiple differences; Based on the maintenance priorities of multiple types of defects, a maintenance plan for each type of defect is obtained in turn.

2. A subway tunnel lining defect detection method according to claim 1, characterized in that: The step of identifying the damage information based on the tunnel information includes: Perform preprocessing based on tunnel information and obtain preprocessing information and a disease identification model; The preprocessed information is input into the disease recognition model for recognition and disease information is obtained.

3. A subway tunnel lining defect detection method according to claim 1, characterized in that: The step of obtaining the disease type based on the disease information includes: Obtain disease information; The disease information is identified and matched with the corresponding disease type in the disease tree knowledge base.

4. A subway tunnel lining defect detection method according to claim 1, characterized in that: The tunnel information is collected by ground detection equipment and multiple types of sensors, and includes tunnel images, point cloud data, temperature and humidity information, strain information, displacement information and crack information.

5. The method for detecting subway tunnel lining defects according to claim 1, characterized in that: The method comprises: Collect high-definition images of subway tunnels; Obtaining image samples of the subway tunnel, where the image samples are images taken when the subway tunnel was just built; Compare high-definition images with image samples to obtain detection information on tunnel changes and changed locations; Obtain disease information based on the detection information of the changed parts.

6. A subway tunnel lining defect detection method according to claim 1, characterized in that: The step of sequentially acquiring a maintenance plan for each type of damage based on the maintenance priorities of the multiple types of damage comprises: Get the disease type; Obtain the corresponding disease cause based on the disease type; Obtain corresponding maintenance solutions based on the cause of the disease; Get the repair priority of multiple disease types; Based on the maintenance priority of each disease type, the maintenance plan for each disease type is output in turn.

7. A subway tunnel lining defect detection method according to claim 6, characterized in that: The method further comprises: Obtain the historical maintenance personnel corresponding to the maintenance plan; If there are multiple historical maintenance personnel, the historical maintenance personnel with the highest repair success rate corresponding to the disease type is obtained as the target maintenance personnel; The target maintenance personnel performs maintenance work according to the maintenance plan.

8. A subway tunnel lining disease detection system, characterized by: include, An instruction acquisition module is used to acquire detection instructions; A detection module, used for obtaining tunnel information based on a detection instruction; A disease identification module is used to identify disease information based on tunnel information; A disease type determination module is used to obtain the disease type based on the disease information; A difference calculation module, used for obtaining the number of disease types based on the disease type and calculating the difference between the number of each disease type and a preset value; A sorting module, used for sorting the multiple differences and obtaining the maintenance priorities of multiple disease types based on the sorting of the multiple differences; A maintenance plan generation module is used to obtain a maintenance plan for each type of damage in turn based on the maintenance priorities of multiple types of damage.

9. An intelligent terminal, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the subway tunnel lining defect detection method according to any one of claims 1 to 7.

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

Images