Road subsidence real-time detection and repair method and system

Abstract

The invention relates to the technical field of road detection and maintenance, in particular to a road subsidence real-time detection and repair method and system. The method comprises the following steps: acquiring historical road surface data of a road, screening out abnormal subsidence data, establishing a road surface subsidence identification model, acquiring real-time road surface data along a target road based on a sensor pre-embedded along the target road, determining abnormal nodes on the target road, extracting road surface data of the abnormal nodes, and identifying the road surface subsidence. And inputting the degree of pavement subsidence in the pavement subsidence identification model, establishing repair measures corresponding to different degrees of pavement subsidence, and in response to the output degree of pavement subsidence, triggering the corresponding repair measures to perform adaptive repair on the subsidence position of the abnormal node. According to the subsidence degree output by the subsidence identification model, a differential repair strategy can be conveniently triggered, and through self-adaptive repair, the workload of traditional manual detection is greatly reduced, pavement excavation is reduced, road defect detection is responded in time, and road passing is rapidly recovered.

Description

Technical Field

[0001] The present invention relates to the technical field of road detection, and in particular to a method and system for real-time detection and repair of road subsidence. Background Art

[0002] Current road subsidence detection technology primarily utilizes a collaborative approach of manual inspections and specialized equipment. During the inspection phase, manual inspections collect basic data through visual observation and simple measurement tools, while offline equipment such as geological radar and laser scanners can identify structural defects in key road sections with millimeter-level accuracy. Once structural defects are identified, appropriate measures must be taken to repair the road. For pavement repair, standardized construction processes have been established using engineering technologies such as grouting reinforcement and asphalt resurfacing. Grouting repairs can locally stabilize the foundation by filling underground cavities, while resurfacing can quickly restore the smoothness of damaged pavement.

[0003] However, the existing technology system still has significant limitations: at the detection level, manual inspections are limited by differences in personnel experience and are prone to missed inspections. Equipment such as geological radars require pre-set detection routes, making it difficult to achieve all-weather coverage. In the repair phase, grouting operations rely on heavy machinery to enter the construction site, which often leads to long traffic interruptions due to the need for construction fencing. Traditional resurfacing processes require closing the work surface for layered compaction, making it difficult to respond to road defect detection in a timely manner and quickly restore road traffic. Summary of the Invention

[0004] The technical problem to be solved by the embodiments of the present invention is to provide a method and system for real-time detection and repair of road subsidence, so as to solve the problem in the prior art that road subsidence detection cannot identify road defects in real time and make rapid repairs to the road surface.

[0005] The present invention discloses a method for real-time detection and repair of road subsidence, comprising: Obtain historical road surface data of the road and filter out abnormal subsidence data from the acquired historical road surface data; Extracting time domain features from the subsidence anomaly data, and establishing a road subsidence identification model based on the extracted time domain features; Based on the sensors embedded along the target road, real-time road surface data along the target road is obtained, and abnormal nodes on the target road are determined based on the obtained real-time road surface data; Extracting road surface data of abnormal nodes from the real-time road surface data and inputting the data into the road surface subsidence identification model to output the degree of road surface subsidence; Establish repair measures corresponding to different degrees of road subsidence, and in response to the road subsidence degree output by the road subsidence identification model, trigger the corresponding repair measures to adaptively repair the subsidence position of the abnormal node.

[0006] Optionally, establishing a road subsidence recognition model based on the extracted time domain features includes: According to a preset road subsidence classification standard, the time domain features of slight subsidence and severe subsidence are respectively extracted from the subsidence abnormal data; Dividing all the extracted time domain features into a training set and a test set, selecting a network architecture to build a basic classification model, and inputting all the time domain features in the training set into the basic classification model in batches for training to obtain the road subsidence recognition model; All the time domain features in the test set are batch-inputted into the trained road surface subsidence recognition model for testing, and the model parameters of the road surface subsidence recognition model are updated through an iterative optimization algorithm.

[0007] Optionally, determining abnormal nodes on the target road based on the acquired real-time road surface data includes: By pre-burying multiple first-class sensors along the target road, real-time road surface data of each section along the target road can be obtained. Compare the real-time road surface data of the road section area with the preset standard threshold to determine whether the road section area is an abnormal road section; Determine the prone-to-sinking locations in the road section area based on the historical road surface data; if the road section area is determined to be an abnormal road section, obtain real-time road surface data corresponding to the prone-to-sinking locations through a second type of sensor pre-buried at each prone-to-sinking location in the abnormal road section; The real-time road surface data of all subsidence-prone locations are compared with the preset standard thresholds, and abnormal nodes where road subsidence occurs are searched for.

[0008] Optionally, establishing repair measures corresponding to different degrees of road subsidence includes: If the road surface subsidence is mild, microcapsules containing a repair agent are pre-installed and embedded in each subsidence-prone location according to the determined subsidence-prone location; According to the embedded microcapsules, a trigger element for destroying the capsule is preset near the microcapsules, and data interaction is established between the server and the trigger element through a communication module; If the road subsidence is severe, an unmanned repair vehicle equipped with rapid grouting equipment will be pre-set, and a vehicle dispatching network architecture will be established to automatically dispatch the unmanned repair vehicle to the abnormal node.

[0009] Optionally, the road subsidence real-time detection and repair method further includes a method for triggering repair measures when the road subsidence identification model outputs a slight subsidence, including: In response to the road subsidence identification model outputting a slight subsidence, sending a trigger signal and controlling a heating element arranged near the microcapsule to heat and destroy the microcapsule; or, In response to the output of the mild subsidence by the road surface subsidence recognition model, send a trigger signal and control the piezoelectric element arranged near the microcapsule to crush and damage the microcapsule; or, In response to the output of the mild subsidence by the road surface subsidence recognition model, send a trigger signal and control the sensitive material arranged near the microcapsule to release and dissolve the microcapsule.

[0010] Optionally, the method for establishing a vehicle scheduling network architecture to self-schedule an unmanned repair vehicle to an abnormal node includes: Based on the determined subsidence-prone positions, obtain the coordinate information of each subsidence-prone position based on GPS positioning; Number each subsidence-prone position in sequence along the target road, and bind and store the number with the coordinate information of the corresponding subsidence-prone position one by one; In response to the output of the severe subsidence recognition result by the road surface subsidence recognition model, obtain the number of the abnormal node, and extract the corresponding coordinate information according to the number of the abnormal node; Establish a vehicle scheduling network architecture, send the extracted coordinate information to the vehicle scheduling network architecture, and the vehicle scheduling network architecture schedules an unmanned repair vehicle to the abnormal node according to the extracted coordinate information.

[0011] Optionally, the method for the vehicle scheduling network architecture to schedule an unmanned repair vehicle to the abnormal node according to the extracted coordinate information includes: Obtain the coordinate positions of each unmanned repair vehicle based on GPS positioning; Match the coordinate information of the abnormal node with the coordinate positions of all unmanned repair vehicles one by one, and sort all unmanned repair vehicles in the order of distance from near to far; Schedule to obtain the traffic flow information around the nearest unmanned repair vehicle, and calculate and determine the driving time of the nearest unmanned repair vehicle to the abnormal node according to the obtained traffic flow information; Compare the calculated driving time with a preset time. If the driving time is greater than the preset time, schedule to obtain the traffic flow information around the second nearest unmanned repair vehicle, and calculate and determine the driving time of the second nearest unmanned repair vehicle to the abnormal node until the calculated driving time is less than the preset time; If the calculated driving time is less than the preset time, schedule the corresponding unmanned repair vehicle to the abnormal node.

[0012] Optionally, after determining the abnormal nodes on the target road, the method for real-time detection and repair of road subsidence further includes a method for determining the repair priorities of multiple abnormal nodes, including: Calculate the first deviation rate between the real-time road surface data of each abnormal section and the preset standard threshold respectively; Divide each abnormal section into different repair priorities in the order of the first deviation rate from large to small; According to the number of abnormal nodes in each abnormal road section and the degree of subsidence of each abnormal node, the comprehensive abnormality score of each abnormal node is calculated, and the repair priority of each abnormal road section is dynamically adjusted according to the comprehensive abnormality score. The calculation formula of the comprehensive abnormality score is:

[0013]

[0014] Where, represents the abnormal node density coefficient, Indicates the number of abnormal nodes in the abnormal road section, Indicates the number of subsidence-prone locations in abnormal road sections, represents the comprehensive abnormality score, Indicates the sinking degree of abnormal nodes, Indicates the first deviation rate between the real-time road surface data of the abnormal road section and the preset standard threshold, and represents a fixed weight.

[0015] Optionally, the road subsidence real-time detection and repair method further includes a method for issuing an early warning after the road subsidence recognition model outputs the road subsidence degree, including: In response to the road subsidence degree output by the road subsidence identification model, the coordinate information, subsidence degree and comprehensive abnormality score of the abnormal node are extracted to generate a structured report; Based on the generated structured report, early warning information including repair suggestions and different repair priorities is pushed through the client or mobile terminal.‌

[0016] The present invention also discloses a system, which adopts the above-mentioned road subsidence real-time detection and repair method, and the system includes: The data screening module is used to obtain historical road surface data and filter out abnormal subsidence data from the acquired historical road surface data; A model building module, used to extract time domain features from the subsidence abnormal data and establish a road subsidence recognition model based on the extracted time domain features; An abnormal node determination module is used to obtain real-time road surface data along the target road based on sensors pre-buried along the target road, and to determine abnormal nodes on the target road based on the obtained real-time road surface data; a subsidence degree identification module, configured to extract road surface data of abnormal nodes from the real-time road surface data, input the data into the road surface subsidence identification model, and output the degree of road surface subsidence; The repair trigger module is used to establish repair measures corresponding to different degrees of road subsidence, and in response to the road subsidence degree output by the road subsidence identification model, trigger the corresponding repair measures to adaptively repair the subsidence position of the abnormal node.

[0017] Compared with the prior art, the real-time road subsidence detection and repair method and system provided by the embodiments of the present invention have the following beneficial effects: By analyzing historical road surface data and combining it with AI algorithms to extract time-domain features, a high-precision subsidence identification model is established. Real-time road surface data is collected along the target road to identify abnormal nodes on the target road. Based on the subsidence level output by the subsidence identification model, a differentiated repair strategy is triggered. By adaptively repairing the subsidence locations at abnormal nodes, the workload of traditional manual inspections is significantly reduced, and timely responses to road defect detection and rapid restoration of traffic are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments, in which: Figure 1 A schematic block diagram of the overall steps of the method for real-time detection and repair of road subsidence provided by an embodiment of the present invention; Figure 2 A schematic block diagram of the execution process of the method for real-time detection and repair of road subsidence provided by an embodiment of the present invention; Figure 3 A schematic block diagram of the execution process of searching for abnormal nodes provided by an embodiment of the present invention; Figure 4 A schematic diagram of the architecture for scheduling unmanned repair vehicles according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. Now, in conjunction with the accompanying drawings, the preferred embodiments of the present invention will be described in detail.

[0020] The present invention discloses a method for real-time detection and repair of road subsidence, such as Figure 1 Shown, including: S1. Obtain historical road surface data of the road, and filter out abnormal subsidence data from the acquired historical road surface data; S2. extracting time domain features from the subsidence anomaly data and establishing a road subsidence recognition model based on the extracted time domain features; S3. Based on sensors embedded along the target road, real-time road surface data along the target road is acquired, and abnormal nodes on the target road are determined based on the acquired real-time road surface data; S4. Extract the road surface data of the abnormal nodes from the real-time road surface data and input it into the road surface subsidence recognition model to output the degree of road surface subsidence; S5. Establish repair measures corresponding to different degrees of road surface subsidence, and in response to the degree of road surface subsidence output by the road surface subsidence recognition model, trigger the corresponding repair measures to adaptively repair the subsidence position of the abnormal nodes.

[0021] Through the implementation of the above-mentioned embodiments of the real-time road subsidence detection and repair method, analyze the historical road surface data, combine with the AI algorithm to extract time-domain features, which helps to establish a high-precision subsidence recognition model. And by collecting the real-time road surface data along the target road, determine the abnormal nodes on the target road, and according to the subsidence degree output by the subsidence recognition model, it is convenient to trigger a differentiated repair strategy, and by adaptively repairing the subsidence position of the abnormal nodes, the workload of traditional manual detection is greatly reduced, and the road defect detection can be responded to in time and the road traffic can be quickly restored. Among them, the historical road surface data can be obtained from the database long-term monitored and recorded by sensors, or provided by the exclusive municipal department. The road surface data required to be applied in the embodiments of the present invention mainly includes numerical values such as road surface subsidence depth, stress, and strain, and the extracted time-domain features include subsidence rate, etc. Through the recognition and judgment of the numerical values such as subsidence depth, stress, and strain by the subsidence recognition model, the subsidence degree of the abnormal road surface is determined, and then the corresponding repair measures are matched, and a decision on priority repair is made for the road surface with different subsidence degrees.

[0022] Further, in combination with Figure 2 As shown, establish a road surface subsidence recognition model according to the extracted time-domain features, including: According to the preset road surface subsidence division standard, respectively extract the time-domain features of slight subsidence and severe subsidence from the subsidence abnormal data; Divide all the extracted time-domain features into a training set and a test set, select a network architecture to construct a basic classification model, and batch input all the time-domain features in the training set into the basic classification model for training to obtain a road surface subsidence recognition model; Batch input all the time-domain features in the test set into the trained road surface subsidence recognition model for testing, and update the model parameters of the road surface subsidence recognition model through an iterative optimization algorithm.

[0023] Through the implementation of the above-mentioned real-time road subsidence detection and repair method embodiment, the road subsidence classification standard is to set different subsidence thresholds based on road grade and traffic volume. For example, the subsidence threshold for highways, urban expressways, and urban arterial roads is 5mm. Subsidence on highways, urban expressways, and urban arterial roads is considered minor, while subsidence exceeding 5mm is considered severe. The subsidence threshold for ordinary roads is 10mm. Subsidence on ordinary roads is considered minor, while subsidence exceeding 10mm is considered severe. Preferably, the training set and test set can be divided in a 7:3 ratio to ensure a balanced distribution of the two types of subsidence samples. Classification machine learning algorithms such as XGBoost (Extreme Gradient Boosting), Bayes (Bayes Classifier), CNN (Convolutional Neural Network), and SVM (Support Vector Machine) can be selected as the basic classification model to train the training set. The algorithm's hyperparameters and model structure can be adjusted to optimize model performance. A convolutional neural network algorithm model is preferred. Its input layer is designed based on the shape of the motion data after filtering and other processing. If the data is two-dimensional, the input layer should accept two-dimensional input. Convolutional layers extract local features from the input data by adding one or more convolutional layers, with appropriate kernel size and number. Pooling layers (such as max pooling) are added after each convolutional layer to reduce the spatial dimensionality of the features and reduce computational effort. Typically, multiple convolutional and pooling layers are stacked to form multiple convolutional blocks. Following the convolutional layers, one or more fully connected layers are added to learn high-level features and perform classification. The number of neurons in the fully connected layer depends on the needs of the output layer. For example, for multi-classification problems, the number of neurons in the final layer should equal the number of classes. After the final fully connected layer, a Softmax function (normalized exponential function) is used to convert the output into a probability distribution. This allows for high-precision road subsidence identification using a classification algorithm.

[0024] Further, combined with Figure 3 As shown, the abnormal nodes on the target road are determined based on the acquired real-time road surface data, including: By pre-burying multiple first-class sensors along the target road, real-time road surface data of each section along the target road can be obtained. Compare the real-time road surface data of the road section area with the preset standard threshold to determine whether the road section area is an abnormal road section; Determine the prone-to-sinking locations in the road section area based on the historical road surface data; if the road section area is determined to be an abnormal road section, obtain real-time road surface data corresponding to the prone-to-sinking locations through a second type of sensor pre-buried at each prone-to-sinking location in the abnormal road section; The real-time road surface data of all subsidence-prone locations are compared with the preset standard thresholds, and abnormal nodes where road subsidence occurs are searched for.

[0025] In the implementation of the above-mentioned real-time road subsidence detection and repair method, the first type of sensor is preferably a fiber optic sensor. Fiber optic sensors offer long-distance continuous monitoring capabilities (covering several kilometers), high sensitivity (capable of detecting micro-strains), and resistance to electromagnetic interference. Deployed continuously at 200-meter intervals, they are suitable for preliminary screening of large-scale road deformation. Their wide coverage and low cost allow them to effectively monitor global data across the entire road, capturing trends in uneven settlement across the entire road section and quickly identifying suspected abnormal sections. This avoids the waste of resources associated with high-precision monitoring of all sections and reduces data processing. Preset thresholds are used to initially exclude normal areas. Specifically, the stress and strain values in the real-time road data are compared with preset standard thresholds. If the stress or strain value in a section exceeds the preset threshold, the section is considered abnormal; otherwise, it is considered normal, effectively improving system response speed. The second type of sensor is preferably a piezoelectric sensor. Piezoelectric sensors are sensitive to dynamic pressure changes (such as instantaneous roadbed collapse) and have millimeter-level resolution. They are used for high-precision local monitoring. Deployed at intervals of 10 meters, they can capture microscopic deformation characteristics of subsidence-prone locations, such as crack expansion and pore water pressure changes, and provide node-level monitoring of each subsidence-prone location. When an abnormal road section triggers an alert, the second type of sensor at each subsidence-prone location within the abnormal section is activated for millimeter-level monitoring. Specifically, only after the abnormal section is triggered, the second type of sensor at each subsidence-prone location within the abnormal section is dynamically activated. High-frequency sampling captures the core location of the subsidence. The stress and strain values in the real-time road surface data at the subsidence-prone location are then compared with preset standard thresholds to locate the specific subsidence anomaly node. This improves subsidence identification response time, reduces false alarms, and reduces energy consumption and losses during long-term high-load operation. Consequently, the coordinated dual-sensor monitoring mode significantly increases the monitoring frequency of the entire target road, effectively shortens the time required to detect minor subsidence, and reduces the density of sensor deployment on the target road, saving costs. Among them, both the first and second types of sensors transmit data to the cloud platform through low-power wide area network (LPWAN) technology.

[0026] Furthermore, corresponding repair measures for different degrees of road subsidence are established, including: If the road surface subsidence is mild, pre-install microcapsules containing a repair agent, and embed microcapsules at each subsidence location according to the determined subsidence locations; According to the embedded microcapsules, a trigger element for destroying the capsule is preset near the microcapsules, and data interaction is established between the server and the trigger element through a communication module; If the road surface subsidence is severe, a driverless repair vehicle equipped with rapid grouting equipment is preset, and a vehicle dispatching network architecture is established to automatically dispatch the driverless repair vehicle to the abnormal node.

[0027] Furthermore, the real-time road subsidence detection and repair method further includes a method for triggering repair measures when the road surface subsidence recognition model outputs mild subsidence, including In response to the road surface subsidence recognition model outputting mild subsidence, sending a trigger signal and controlling a heating element arranged near the microcapsule to heat and damage the microcapsule; or, In response to the road surface subsidence recognition model outputting mild subsidence, sending a trigger signal and controlling a piezoelectric element arranged near the microcapsule to squeeze and damage the microcapsule; or, In response to the road surface subsidence recognition model outputting mild subsidence, sending a trigger signal and controlling a sensitive material arranged near the microcapsule to release and dissolve the microcapsule.

[0028] Through the implementation of the above embodiments of the real-time road subsidence detection and repair method, the microcapsule shell is made of polyurethane material, and the inside is filled with epoxy resin repair agent. When the repair agent flows out of the ruptured microcapsule, it can fill the cracks in the slightly sunken road surface. At this time, the repair agent reacts chemically with the road surface material and can quickly restore the road surface flatness after curing, that is, the epoxy resin repair agent reacts crosslinkingly with the components in the road surface material to form a strong polymer network. The repair agent usually initially cures within 10 - 30 minutes and completely cures within 24 hours. The strength of the repaired road surface reaches more than 95% of the original road surface, ensuring traffic safety.

[0029] Preferably, the diameter of the microcapsule is 0.1mm - 0.5mm, and the filling amount of the repair agent is 100g - 200g per square meter. Among them, when the triggering element is a heating element, it is preferably a resistance wire. When the trigger signal arrives, the heating element is powered on to soften and rupture the microcapsule shell; when the triggering element is a piezoelectric element, when the trigger signal arrives, the piezoelectric element generates mechanical pressure to damage the microcapsule shell; when the triggering element is a chemically sensitive material, it is preferably a pH-sensitive material. When the trigger signal arrives, it can release chemical substances, such as acids or alkalis, to facilitate dissolving the shell of the microcapsule. Among them, the trigger signal can be transmitted to the triggering element through a wireless network. For example: using an electrical signal, it can be transmitted through a wire to the triggering element near the microcapsule; or using a wireless signal, activating the triggering element near the microcapsule through radio frequency or infrared signals.

[0030] Furthermore, as shown in Figure 4 establishing a vehicle dispatching network architecture to automatically dispatch the driverless repair vehicle to the abnormal node includes: According to the determined subsidence-prone positions, obtaining the coordinate information of each subsidence-prone position based on GPS positioning; Number each subsidence - prone location in sequence along the target road, and bind and store the number with the coordinate information of the corresponding subsidence - prone location one by one; In response to the severe subsidence recognition result output by the road surface subsidence recognition model, obtain the number of the abnormal node, and extract the corresponding coordinate information according to the number of the abnormal node; Establish a vehicle scheduling network architecture, send the extracted coordinate information to the vehicle scheduling network architecture, and let the vehicle scheduling network architecture dispatch the unmanned repair vehicle to the abnormal node according to the extracted coordinate information.

[0031] Furthermore, the vehicle scheduling network architecture dispatches the unmanned repair vehicle to the abnormal node according to the extracted coordinate information, including: Obtain the coordinate positions of each unmanned repair vehicle based on GPS positioning; Match the coordinate information of the abnormal node with the coordinate positions of all unmanned repair vehicles one by one, and sort all unmanned repair vehicles in the order of distance from near to far; Dispatch to obtain the traffic flow information around the nearest unmanned repair vehicle, and calculate and determine the driving time of the nearest unmanned repair vehicle to the abnormal node according to the obtained traffic flow information; Compare the calculated driving time with the preset time. If the driving time is greater than the preset time, dispatch to obtain the traffic flow information around the second - nearest unmanned repair vehicle, and calculate and determine the driving time of the second - nearest unmanned repair vehicle to the abnormal node until the calculated driving time is less than the preset time; If the calculated driving time is less than the preset time, dispatch the corresponding unmanned repair vehicle to the abnormal node.

[0032] Through the implementation of the above - mentioned embodiments of the real - time road subsidence detection and repair method, obtaining the coordinates of the subsidence - prone location based on GPS positioning can ensure that the repair vehicle is accurately navigated to the abnormal node, avoiding the positioning deviation caused by the error of traditional manual marking. Sorting according to the double - constraint conditions of "distance priority + time tolerance" can screen the unmanned repair vehicles near the abnormal node, and dynamically adjust in combination with the traffic flow, and preferentially select the unmanned repair vehicle with the fastest expected arrival time. Thus, through the dynamic programming of the dispatching of unmanned repair vehicles, the total process time from the abnormal recognition of road surface subsidence to the dispatch of unmanned repair vehicles is greatly shortened, and the efficiency is significantly improved compared with manual dispatching. In the multi - vehicle cooperation mode, the dispatching tasks of multiple abnormal nodes can be processed in parallel, improving the resource utilization rate. Among them, the preset time threshold can be set differently according to the road grade. For example, the threshold for the main road is 20 minutes, and the threshold for the branch road is 30 minutes, avoiding the dispatching failure caused by a single standard.

[0033] Furthermore, the real - time road subsidence detection and repair method also includes a method for determining the repair priorities of multiple abnormal nodes after determining the abnormal nodes on the target road, including: Calculate the first deviation rate between the real-time road surface data of each abnormal road section and the preset standard threshold respectively; Divide each abnormal road section into different repair priorities in descending order of the first deviation rate; Based on the number of abnormal nodes in each abnormal road section and the subsidence degree of each abnormal node, calculate the comprehensive abnormal score of each abnormal node, and dynamically adjust the repair priority of each abnormal road section according to the comprehensive abnormal score. The calculation formula of the comprehensive abnormal score is:

[0034]

[0035] In the formula, represents the abnormal node density coefficient, represents the number of abnormal nodes in the abnormal road section, represents the number of easily subsiding positions in the abnormal road section, represents the comprehensive abnormal score, represents the subsidence degree of the abnormal node, represents the first deviation rate between the real-time road surface data of the abnormal road section and the preset standard threshold, and represents the fixed weight.

[0036] Through the implementation of the above embodiments of the real-time road subsidence detection and repair method, first calculate the deviation rate between the real-time stress and deformation data of each road section area and the preset threshold. For example, if the strain value exceeds the threshold by 10%, it is determined as a first-level abnormality, and if it exceeds 5%, it is a second-level abnormality. Then, within the abnormal road section, perform weighted calculation on the deformation data of the abnormal nodes to generate a comprehensive abnormal score, thereby dynamically adjusting the repair priority of each abnormal road section according to the comprehensive abnormal score. For example: when the comprehensive abnormal score is greater than the threshold, directly trigger a first-level response. Furthermore, by setting different thresholds, the abnormal situations are divided into different levels, which helps to classify and manage the abnormal situations, prioritize the handling of more serious abnormalities, ensure that resources are preferentially allocated to the road sections most in need of repair, improve the efficiency of road maintenance, reduce the interference of human factors, and improve the accuracy and reliability of decision-making.

[0037] Furthermore, the real-time road subsidence detection and repair method further includes a method for giving an early warning after the road surface subsidence recognition model outputs the road surface subsidence degree, including: In response to the road surface subsidence recognition model outputting the road surface subsidence degree, extract the coordinate information, subsidence degree, and comprehensive abnormal score of the abnormal nodes to generate a structured report; According to the generated structured report, push warning information including repair suggestions and different repair priorities through the client or mobile terminal.

[0038] Through the implementation of the above-mentioned embodiment of the method for real-time detection and repair of road subsidence, the generated structured report can be used to present the road subsidence situation in an intuitive and easy-to-understand manner, so that relevant personnel can quickly obtain key information. And by providing the coordinate information of the abnormal nodes, the road subsidence area that needs attention can be accurately located, thereby improving work efficiency. By timely pushing early warning information through the client or mobile terminal, relevant personnel can receive early warnings at any time and any place, which improves the convenience and flexibility of information transmission, avoids preventing further expansion of road subsidence, and helps to quickly respond to and handle problems. The pushed early warning information includes repair suggestions and different repair priorities, and can also provide decision-making support for road management departments, help formulate reasonable repair plans, ensure that resources are first allocated to the most urgent and critical repair tasks, and improve work efficiency.‌

[0039] The present invention also discloses a system, which adopts the above-mentioned road subsidence real-time detection and repair method, and the system includes: The data screening module is used to obtain historical road surface data and filter out abnormal subsidence data from the acquired historical road surface data; A model building module is used to extract time domain features from subsidence anomaly data and establish a road subsidence identification model based on the extracted time domain features; An abnormal node determination module is used to obtain real-time road surface data along the target road based on sensors pre-buried along the target road, and to determine abnormal nodes on the target road based on the obtained real-time road surface data; The subsidence degree identification module is used to extract the road surface data of abnormal nodes from the real-time road surface data, input it into the road surface subsidence identification model, and output the degree of road surface subsidence; The repair trigger module is used to establish repair measures corresponding to different degrees of road subsidence. In response to the road subsidence degree output by the road subsidence identification model, the corresponding repair measures are triggered to adaptively repair the subsidence position of the abnormal node.

[0040] The present invention also discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned road subsidence real-time detection and repair method are implemented.

[0041] The present invention also discloses a computer device, including 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 steps of the above-mentioned method for real-time detection and repair of road subsidence are implemented.

[0042] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to specific embodiments. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce a means for implementing the functions specified in one or more flows and / or blocks Figure 1 one or more flows and / or blocks Figure 1 or in one or more blocks.

[0043] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including an instruction means that implements the functions specified in one or more flows and / or blocks Figure 1 one or more flows and / or blocks Figure 1 or in one or more blocks.

[0044] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more flows and / or blocks Figure 1 one or more flows and / or blocks Figure 1 or in one or more blocks.

[0045] It should be understood that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Those skilled in the art can modify the technical solutions described in the above embodiments or perform equivalent replacements for some of the technical features; and all such modifications and replacements should fall within the protection scope of the present invention.

Claims

1. A real-time detection and repair method for road subsidence, characterized in that, The road subsidence real-time detection and repair method comprises: Obtain historical road surface data of the road and filter out abnormal subsidence data from the acquired historical road surface data; Extracting time domain features from the subsidence anomaly data, and establishing a road subsidence identification model based on the extracted time domain features; Based on the sensors embedded along the target road, real-time road surface data along the target road is obtained, and abnormal nodes on the target road are determined based on the obtained real-time road surface data; Extracting road surface data of abnormal nodes from the real-time road surface data and inputting the data into the road surface subsidence identification model to output the degree of road surface subsidence; Repair measures corresponding to different degrees of road subsidence are established, and in response to the road subsidence degree output by the road subsidence identification model, corresponding repair measures are triggered to adaptively repair the subsidence position of the abnormal node.

2. The real-time detection and repair method for road subsidence according to claim 1, characterized in that, The road subsidence identification model is established based on the extracted time domain features, including: According to a preset road subsidence classification standard, the time domain features of slight subsidence and severe subsidence are respectively extracted from the subsidence abnormal data; Dividing all the extracted time domain features into a training set and a test set, selecting a network architecture to build a basic classification model, and inputting all the time domain features in the training set into the basic classification model in batches for training to obtain the road subsidence recognition model; All the time domain features in the test set are batch-inputted into the trained road surface subsidence recognition model for testing, and the model parameters of the road surface subsidence recognition model are updated through an iterative optimization algorithm.

3. The real-time detection and repair method for road subsidence according to claim 1, characterized in that Determining abnormal nodes on the target road based on the acquired real-time road surface data includes: By pre-burying multiple first-class sensors along the target road, real-time road surface data of each section along the target road can be obtained. Compare the real-time road surface data of the road section area with the preset standard threshold to determine whether the road section area is an abnormal road section; Determine the prone-to-sinking locations in the road section area based on the historical road surface data; if the road section area is determined to be an abnormal road section, obtain real-time road surface data corresponding to the prone-to-sinking locations through a second type of sensor pre-buried at each prone-to-sinking location in the abnormal road section; The real-time road surface data of all subsidence-prone locations in abnormal road sections are compared with the preset standard thresholds, and abnormal nodes where road subsidence occurs are searched.

4. The real-time detection and repair method for road subsidence according to claim 3, wherein The establishment of repair measures corresponding to different degrees of road subsidence includes: If the road surface subsidence is mild, microcapsules containing a repair agent are pre-installed and embedded in each subsidence-prone location according to the determined subsidence-prone location; According to the embedded microcapsules, a trigger element for destroying the capsule is preset near the microcapsules, and data interaction is established between the server and the trigger element through a communication module; If the road subsidence is severe, an unmanned repair vehicle equipped with rapid grouting equipment will be pre-set, and a vehicle dispatching network architecture will be established to automatically dispatch the unmanned repair vehicle to the abnormal node.

5. The real-time detection and repair method for road subsidence according to claim 4, characterized in that The road subsidence real-time detection and repair method also includes a method for triggering repair measures when the road subsidence identification model outputs a slight subsidence, including: In response to the output of a mild subsidence by the road surface subsidence recognition model, send a trigger signal and control a heating element arranged near the microcapsule to heat and damage the microcapsule; or, In response to the output of a mild subsidence by the road surface subsidence recognition model, send a trigger signal and control a piezoelectric element arranged near the microcapsule to squeeze and damage the microcapsule; or, In response to the output of a mild subsidence by the road surface subsidence recognition model, send a trigger signal and control a sensitive material arranged near the microcapsule to release and dissolve the microcapsule.

6. The real-time detection and repair method for road subsidence according to claim 4, characterized in that The establishment of a vehicle scheduling network architecture to self-schedule an unmanned repair vehicle to an abnormal node includes: Based on the determined subsidence-prone positions, obtain the coordinate information of each subsidence-prone position based on GPS positioning; Number each subsidence-prone position in sequence along the target road, and bind and store the number with the coordinate information of the corresponding subsidence-prone position one by one; In response to the output of a severe subsidence recognition result by the road surface subsidence recognition model, obtain the number of the abnormal node, and extract the corresponding coordinate information according to the number of the abnormal node; Establish a vehicle scheduling network architecture, send the extracted coordinate information to the vehicle scheduling network architecture, and have the vehicle scheduling network architecture schedule an unmanned repair vehicle to the abnormal node according to the extracted coordinate information.

7. The real-time detection and repair method for road subsidence according to claim 6, characterized in that The scheduling of an unmanned repair vehicle to the abnormal node by the vehicle scheduling network architecture according to the extracted coordinate information includes: Obtain the coordinate positions of each unmanned repair vehicle based on GPS positioning; Match the coordinate information of the abnormal node with the coordinate positions of all unmanned repair vehicles one by one, and sort all unmanned repair vehicles in ascending order of distance from near to far; Schedule to obtain the traffic flow information around the nearest unmanned repair vehicle, and calculate and determine the driving time of the nearest unmanned repair vehicle to the abnormal node according to the obtained traffic flow information; Compare the calculated driving time with a preset time. If the driving time is greater than the preset time, then schedule to obtain the traffic flow information around the second nearest unmanned repair vehicle, and calculate and determine the driving time of the second nearest unmanned repair vehicle to the abnormal node until the calculated driving time is less than the preset time; If the calculated driving time is less than the preset time, then schedule the corresponding unmanned repair vehicle to the abnormal node.

8. The real-time detection and repair method for road subsidence according to any one of claims 6 or 7, characterized in that, The road surface subsidence real-time detection and repair method further includes a method for determining the repair priorities of multiple abnormal nodes after determining the abnormal nodes on the target road, including: Calculate the first deviation rate between the real-time road surface data of each abnormal road section and a preset standard threshold respectively; Divide each abnormal road section into different repair priorities in descending order of the first deviation rate; According to the number of abnormal nodes in each abnormal road section and the subsidence degree of each abnormal node, calculate the comprehensive abnormal score of each abnormal node, and dynamically adjust the weight of the repair priority of each abnormal road section according to the comprehensive abnormal score. The calculation formula of the comprehensive abnormal score is: Wherein, represents the abnormal node density coefficient, represents the number of abnormal nodes in the abnormal section, represents the number of easily subsiding positions in the abnormal section, represents the comprehensive abnormal score, represents the subsidence degree of the abnormal node, represents the first deviation rate between the real-time road surface data of the abnormal section and the preset standard threshold, and represents the fixed weight.

9. The real-time detection and repair method for road subsidence according to claim 8, characterized in that The road surface subsidence real-time detection and repair method further includes a method for giving an early warning after the road surface subsidence recognition model outputs the degree of road surface subsidence, including: In response to the road subsidence degree output by the road subsidence identification model, the coordinate information, subsidence degree and comprehensive abnormality score of the abnormal node are extracted to generate a structured report; Based on the generated structured report, early warning information containing repair suggestions and different repair priorities is pushed through the client or mobile terminal.

10. A system, characterized in that, The method for real-time detection and repair of road subsidence according to any one of claims 1 to 9 is adopted, wherein the system comprises: The data screening module is used to obtain historical road surface data and filter out abnormal subsidence data from the acquired historical road surface data; A model building module, used to extract time domain features from the subsidence abnormal data and establish a road subsidence recognition model based on the extracted time domain features; An abnormal node determination module is used to obtain real-time road surface data along the target road based on sensors pre-buried along the target road, and to determine abnormal nodes on the target road based on the obtained real-time road surface data; a subsidence degree identification module, configured to extract road surface data of abnormal nodes from the real-time road surface data, input the data into the road surface subsidence identification model, and output the degree of road surface subsidence; The repair trigger module is used to establish repair measures corresponding to different degrees of road subsidence, and in response to the road subsidence degree output by the road subsidence identification model, trigger the corresponding repair measures to adaptively repair the subsidence position of the abnormal node.

Images