Smart city road collapse emergency supervision internet of things large model system and method

Abstract

This invention relates to the field of road surface monitoring, and provides a large-scale IoT model system and method for emergency monitoring of road surface collapses in smart cities. The system includes an emergency monitoring management platform configured to: acquire multi-source monitoring data; determine the collapse risk distribution of a target road segment based on the multi-source monitoring data; determine exploration parameters based on the collapse risk distribution; control the vehicle-mounted ground-penetrating radar of the emergency monitoring platform to explore target exploration points based on the exploration parameters and collect exploration data; determine whether the target exploration point has geological hazard characteristics based on the exploration data; and if the target exploration point has geological hazard characteristics, generate a traffic control instruction and send the traffic control instruction to the emergency monitoring platform to control the raising of the gates on the associated roads of the target exploration point. This invention can improve the accuracy of determining the location, size, and development trend of potential road surface collapses.

Description

Technical Field

[0001] This manual relates to the field of road surface monitoring, and in particular to a large-scale IoT model system and method for emergency monitoring of road surface collapse in smart cities. Background Technology

[0002] Urban underground pipe networks are highly susceptible to subsidence or collapse of urban roads under traffic loads or geological disturbances, causing property damage and casualties. Currently, urban subsidence risk monitoring typically relies on a single type of data (such as surface subsidence or pipe network flow) for anomaly identification, making it difficult to accurately determine the spatial location, scale, and development trend of underground cavities or damaged pipes without excavation or the installation of underground sensors.

[0003] Therefore, it is hoped that a large-scale IoT model system and method for emergency monitoring of road collapses in smart cities can be provided to improve the accuracy of determining the location, size and development trend of potential road collapses. Summary of the Invention

[0004] The invention includes a large-scale IoT model system for emergency monitoring of road collapses in smart cities. The system comprises an emergency monitoring user platform, an emergency monitoring service platform, an emergency monitoring management platform, an emergency monitoring sensor network platform, and an emergency monitoring object platform. The emergency monitoring management platform is configured to: acquire multi-source monitoring data through the emergency monitoring object platform; determine the collapse risk distribution of a target road segment based on the multi-source monitoring data; determine exploration parameters based on the collapse risk distribution, the exploration parameters including target exploration points and operating frequency groups; control the vehicle-mounted ground-penetrating radar of the emergency monitoring object platform to explore the target exploration points based on the exploration parameters and collect exploration data; determine whether the target exploration points have geological hazard characteristics based on the exploration data; and if the target exploration points have geological hazard characteristics, generate a traffic control command and send the traffic control command to the emergency monitoring object platform to control the raising of the gates on the associated roads of the target exploration points.

[0005] The invention includes a method for emergency monitoring of road collapses in smart cities. The method is executed by an emergency monitoring management platform within a large-scale IoT model system for emergency monitoring of road collapses in smart cities. The method includes: acquiring multi-source monitoring data through an emergency monitoring object platform; determining the collapse risk distribution of a target road segment based on the multi-source monitoring data; determining exploration parameters based on the collapse risk distribution, the exploration parameters including target exploration points and operating frequency groups; controlling the vehicle-mounted ground-penetrating radar of the emergency monitoring object platform to explore the target exploration points based on the exploration parameters and collect exploration data; determining whether the target exploration points have geological hazard characteristics based on the exploration data; and if the target exploration points have geological hazard characteristics, generating a traffic control command and sending the traffic control command to the emergency monitoring object platform to control the raising of the gates on the associated roads of the target exploration points.

[0006] The beneficial effects of the present invention include, but are not limited to: (1) By integrating multi-source monitoring data to assess the risk of collapse, and dynamically optimizing the exploration parameters of the vehicle-mounted ground-penetrating radar according to the assessment results, accurate and efficient geological hazard detection can be achieved, potential risks can be detected in a timely manner, traffic control instructions can be automatically generated and executed, road collapse accidents can be effectively prevented, and the safety and emergency response capabilities of urban roads can be significantly improved; (2) By constructing a pipe segment map containing standard pipe segment information, connection information and multi-source monitoring data, the connection relationship and characteristics of the pipeline network can be effectively extracted, and the collapse risk of each pipe segment can be accurately assessed through machine learning models, significantly improving the accuracy of road collapse prediction; (3) By comprehensively analyzing the spatial aggregation and temporal trend of the collapse risk distribution, the core exploration area with the highest risk can be accurately locked from the dynamic data, effectively filtering static or isolated abnormal signals, making the positioning of exploration targets more accurate. Attached Figure Description

[0007] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0008] Figure 1 This is a schematic diagram of the platform structure of a smart city road collapse emergency monitoring IoT large model system according to some embodiments of this specification; Figure 2 This is an exemplary flowchart of an emergency monitoring method for road collapse in smart cities, as shown in some embodiments of this specification. Figure 3 This is an exemplary schematic diagram of a collapse risk model according to some embodiments of this specification; Figure 4 This is an exemplary schematic diagram illustrating the determination of exploration parameters according to some embodiments of this specification. Detailed Implementation

[0009] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. The accompanying drawings do not represent all implementation methods.

[0010] Unless the context clearly indicates an exception, words such as "a," "an," "a kind," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0011] In the embodiments of this specification, the order of the steps described in the step-by-step instructions is interchangeable unless otherwise specified, and steps may be omitted. Other steps may also be included in the operation process.

[0012] Figure 1 This is a schematic diagram of the platform structure of a smart city road collapse emergency monitoring IoT big data model system according to some embodiments of this specification.

[0013] The IoT big data model in the smart city road collapse emergency monitoring IoT big data model system can be a model architecture built on the Internet of Things (IoT) to enable the efficient operation of large amounts of data within the system. Simultaneously, various big data models (such as general big data models, industry-specific big data models, lightweight big data models, and multimodal big data models) can be applied to the IoT big data model to assist it in data perception and processing.

[0014] In some embodiments, such as Figure 1 As shown, the smart city road collapse emergency monitoring IoT big data model system 100 includes an emergency monitoring user platform 110, an emergency monitoring service platform 120, an emergency monitoring management platform 130, an emergency monitoring sensor network platform 140, and an emergency monitoring object platform 150.

[0015] Emergency monitoring user platform 110 refers to a platform for interaction with users (such as regulatory personnel). In some embodiments, emergency monitoring user platform 110 includes terminal devices. For example, terminal devices include mobile devices, tablet computers, and consoles.

[0016] Emergency monitoring service platform 120 refers to a platform used for receiving and transmitting data and / or information. In some embodiments, emergency monitoring service platform 120 is configured as a server or processor, etc. Emergency monitoring service platform 120 interacts bidirectionally with emergency monitoring user platform 110 and emergency monitoring management platform 130.

[0017] The emergency monitoring and management platform 130 refers to a comprehensive management platform that manages and coordinates the connections and collaboration between multiple platforms. In some embodiments, the emergency monitoring and management platform 130 is configured as a server or processor, etc. The emergency monitoring and management platform 130 interacts bidirectionally with the emergency monitoring service platform 120 and the emergency monitoring sensor network platform 140.

[0018] In some embodiments, the emergency monitoring and management platform 130 includes a data center 131. The data center 131 is configured to store data, information, or instructions generated or received by the emergency monitoring and management platform 130. The data center 131 includes a model library. The emergency monitoring and management platform 130 can access various models or large models through the model library.

[0019] Emergency monitoring sensor network platform 140 refers to a platform used for the comprehensive management of sensor information. In some embodiments, emergency monitoring sensor network platform 140 is configured as a communication network or gateway.

[0020] The emergency monitoring object platform 150 refers to a platform that generates sensor information and executes control commands. In some embodiments, the emergency monitoring object platform 150 includes drones, vehicle-mounted ground-penetrating radar, and road facilities (such as barriers, warning devices, and height restriction barriers). The emergency monitoring object platform 150 interacts bidirectionally with the emergency monitoring sensor network platform 140. The drone is configured to collect surface images of the target road section.

[0021] Vehicle-mounted ground-penetrating radar can be a radar device used to explore the internal structure and defects of non-visible areas beneath the road surface. In some embodiments, vehicle-mounted ground-penetrating radar can emit high-frequency electromagnetic waves (such as 900MHz), medium-frequency electromagnetic waves (such as 400MHz), and low-frequency electromagnetic waves (such as 100MHz).

[0022] Road barriers are used to restrict vehicle traffic on roads. Height restriction barriers are used to limit the height of vehicles that can travel on roads. Warning devices are used to provide warning information to drivers of vehicles on roads, including displays and broadcasting equipment.

[0023] In some embodiments, road facilities may be deployed along roads or at intersections.

[0024] For further explanation of the above content, please refer to [link / reference]. Figures 2 to 4 And its related descriptions.

[0025] In some embodiments of this specification, the smart city road collapse emergency monitoring IoT big data model system 100 can form an information operation closed loop between various functional platforms and operate in a coordinated and regular manner under the unified management of the emergency monitoring and management platform, thereby realizing the informatization and intelligentization of urban road collapse emergency monitoring.

[0026] Figure 2This is an exemplary flowchart of a smart city road collapse emergency monitoring method according to some embodiments of this specification. In some embodiments, process 200 of the smart city road collapse emergency monitoring method is executed by the emergency monitoring management platform (hereinafter referred to as the management platform) in the smart city road collapse emergency monitoring IoT big data model system. Figure 2 As shown, the process 200 of the smart city road collapse emergency monitoring method includes the following steps: Step 210: Obtain multi-source monitoring data through the emergency monitoring object platform.

[0027] Multi-source monitoring data refers to various data used to monitor road-related conditions. In some embodiments, multi-source monitoring data may include pipeline operation data, soil data, and traffic load data, etc. Once a target road segment is determined, the management platform may acquire only the multi-source monitoring data corresponding to that target road segment. For an explanation of the target road segment, see step 220 and its related description.

[0028] Pipeline operation data refers to data related to underground pipelines such as underground water supply, drainage, and gas supply, such as pressure and flow rate within the pipelines.

[0029] Soil data refers to data related to the soil beneath the road surface, such as soil moisture content, pore water pressure, and soil stress.

[0030] Traffic load data refers to data related to traffic conditions above the road surface, such as the volume of vehicles on the road, their speed, vehicle type, and vehicle load.

[0031] In some embodiments, the management platform can acquire multi-source monitoring data through the emergency monitoring object platform. For example, the emergency monitoring object platform acquires pipeline network operation data and soil data through sensors (such as humidity sensors, pressure sensors, pore water pressure gauges, settlement displacement sensors, and flow sensors) installed in the pipeline network and buried in the soil, and uploads the pipeline network operation data and soil data to the management platform.

[0032] In some embodiments, the management platform may also obtain traffic load data from third-party platforms such as map service providers or public transport operation platforms.

[0033] In some embodiments, multi-source monitoring data may also include surface deformation data.

[0034] Surface deformation data refers to data characterizing minute deformations occurring on a road surface. In some embodiments, surface deformation data includes settlement displacement values ​​and settlement rates at multiple monitoring points within a monitoring period. Multiple monitoring points and monitoring periods (e.g., one year) are pre-set based on historical experience. Minor deformations can be settlement displacement values ​​and settlement rates both on the order of millimeters.

[0035] Settlement displacement refers to the amount of vertical displacement of a monitoring point relative to a reference point. The reference point can be a fixed point that does not undergo vertical displacement.

[0036] Settlement rate refers to the speed at which a monitoring point settles.

[0037] In some embodiments, the management platform can acquire surface deformation data through methods such as satellite synthetic aperture radar interferometry. For example, the management platform performs synthetic aperture radar imaging of the monitoring point by the satellite at the beginning and end of the monitoring period, and performs interferometry processing on the radar wave phase information of the two synthetic aperture radar images to calculate the distance from the monitoring point to the satellite during the two imaging periods. The difference between the distance corresponding to the beginning time and the distance corresponding to the end time is taken as the settlement displacement value.

[0038] In some embodiments, the management platform can calculate the ratio of the settlement displacement value of the monitoring point to the duration of the monitoring period within the monitoring period, thereby obtaining the settlement rate of the monitoring point.

[0039] In some embodiments of this specification, millimeter-level settlement displacement values ​​and settlement rates are incorporated into multi-source monitoring, enabling early and large-scale surveys of minor surface deformations and providing a basis for identifying areas at risk of subsidence.

[0040] Step 220: Based on multi-source monitoring data, determine the collapse risk distribution of the target road section.

[0041] The target road segment refers to the road segment that requires a road surface subsidence assessment. In some embodiments, the target road segment can be preset.

[0042] In some embodiments, the management platform can also identify historical collapse cases based on historical disaster data, road monitoring data, and road construction logs, and identify target road sections based on these historical collapse cases.

[0043] Historical disaster data refers to information related to road subsidence that occurred in the past. In some embodiments, historical disaster data includes the location, time, morphological characteristics, and causes of road subsidence that occurred in the past. Morphological characteristics of road subsidence include the depth, diameter, area, and shape of the sinkhole. Causes of road subsidence include burst water pipes, illegal construction, or heavy rain. Historical disaster data is stored in the historical disaster database of the data center.

[0044] A historical disaster database is a structured database used to store and manage detailed information about road collapse events that have occurred.

[0045] Road monitoring data refers to image data related to road surface collapses. For example, road monitoring data includes video recordings before and after a road surface collapse. In some embodiments, the emergency monitoring platform acquires road monitoring data through image acquisition devices on both sides of the road and uploads it to the processor.

[0046] Road construction logs are used to record information related to road and underground pipeline construction, such as the location, time, and excavation depth. These logs are stored in a management platform.

[0047] A historical collapse case refers to a complete set of information about a historical road surface collapse event. For example, a historical collapse case includes the location, time, morphological characteristics, cause, corresponding road monitoring data, and road construction logs for the corresponding time period of the historical road surface collapse event.

[0048] In some embodiments, the management platform can retrieve road collapse events that occurred in historical time periods based on historical disaster data, treat each road collapse event as a historical collapse case, and include the historical disaster data, road monitoring data, and road construction logs corresponding to the road collapse event into the corresponding historical collapse case.

[0049] In some embodiments, the management platform can identify the location with the most collapses within a preset historical period (e.g., the past two years) from multiple historical collapse cases, designate the area containing that location as the target area, and determine all roads within the target area as target road segments. The target area can be a circular area formed by taking the location with the most collapses as the center and a preset radius (e.g., 1 kilometer).

[0050] Some embodiments in this specification identify high-risk target road sections by integrating historical collapse cases, which can more effectively identify target road sections based on actual collapse experience and data.

[0051] The collapse risk distribution can characterize the distribution of collapse risk among multiple road segment units within a target road segment. Collapse risk refers to the probability of road surface collapse.

[0052] A road segment unit can be a pre-divided road segment within a target road segment. In some embodiments, the management platform can define a road surface corresponding to a standard pipeline segment as a road segment unit.

[0053] A standard pipe segment can be a segment of the underground pipe network corresponding to the target road segment. In some embodiments, the management platform uses the underground pipe network corresponding to the target road segment as a dividing point, with the manholes in the underground pipe network as the dividing point, and the pipe segment between two manholes as a standard pipe segment.

[0054] Since the target road segment contains multiple road segment units, the management platform can obtain multi-source monitoring data corresponding to each road segment unit.

[0055] In some embodiments, the management platform can determine the collapse risk distribution of a target road segment based on multi-source monitoring data. For example, for each road segment unit, the management platform can construct a target feature vector based on the multi-source monitoring data corresponding to the road segment unit, calculate the vector similarity between multiple reference feature vectors in the vector database and the target feature vector, and sort them from high to low. Then, based on the labels corresponding to the reference feature vectors that meet the matching conditions, the average value is calculated to determine the collapse risk of the road segment unit. The matching conditions include being among the top k in the vector similarity ranking, where k is a preset value (e.g., 3). Vector similarity is negatively correlated with vector distance, which includes Euclidean distance, etc.

[0056] In some embodiments, the vector database can be pre-configured based on historical data. For example, the management platform constructs multiple reference feature vectors based on multi-source monitoring data corresponding to multiple historical road segment units, and determines the labels of the reference feature vectors based on whether the historical road segment units have collapsed. The management platform records the label of historical road segment units that have collapsed as 1, and the label of historical road segment units that have not collapsed as 0.

[0057] In some embodiments, the management platform can average the labels corresponding to the reference feature vectors that meet the matching conditions to obtain the collapse risk of a single road segment unit. The management platform determines and combines the collapse risks of multiple road segment units using the above method to obtain the collapse risk distribution.

[0058] Step 230: Determine exploration parameters based on the collapse risk distribution.

[0059] Exploration parameters are parameters that guide vehicle-mounted ground-penetrating radar in conducting exploration. In some embodiments, exploration parameters may include target exploration points and operating frequency groups, etc. There may be multiple target exploration points within the target road segment.

[0060] The target exploration point refers to the road section unit where underground exploration is required. The operating frequency group refers to the combination of electromagnetic wave frequencies used by the vehicle-mounted ground-penetrating radar during exploration.

[0061] In some embodiments, the management platform can determine exploration parameters based on the distribution of collapse risk and through preset rules.

[0062] In some embodiments, preset rules can be pre-set. For example, based on the distribution of subsidence risk, the management platform identifies road segment units with a subsidence risk greater than a risk threshold as target exploration points, and determines working frequency groups based on the relationship between the number of target exploration points and a preset number. Where the number of target exploration points within a target road segment is greater than the preset number, the working frequency group can use a combination of high-frequency, medium-frequency, and low-frequency electromagnetic waves. Where the number of target exploration points is not greater than the preset number, the working frequency group can use a combination of high-frequency and medium-frequency electromagnetic waves. The preset number is pre-set based on historical experience.

[0063] In some embodiments, the management platform can also determine target exploration points based on the distribution of subsidence risks, the surface subsidence pattern of the target road section, and abnormal surface features, and generate an exploration task list based on the target exploration points. The management platform can then generate multiple candidate exploration paths based on the exploration task list, and determine the target exploration path from these candidate paths based on traffic load data and the business priority of the target exploration points.

[0064] Surface subsidence models can be used to characterize the temporal and spatial morphology or trend of surface deformation data for a target road segment. In some embodiments, surface subsidence models include temporal models and spatial models. Temporal models characterize subsidence trends, including stable subsidence rates, accelerating subsidence rates, and decelerating subsidence rates. Spatial models characterize the morphology of subsidence, including funnel-shaped or sheet-like patterns.

[0065] In some embodiments, the management platform can calculate the difference between the settlement rate at the second time point and the settlement rate at the first time point based on surface deformation data of the target road segment obtained from the two closest time points (denoted as the first time point and the second time point, where the first time point is later than the second time point). If the absolute value of the difference is not greater than a rate threshold, the settlement rate is stable. If the difference is positive and greater than the rate threshold, the settlement rate decelerates. If the difference is negative and its absolute value is greater than the rate threshold, the settlement rate accelerates. The rate threshold is preset based on historical experience.

[0066] In some embodiments, the management platform can draw a settlement plane along the direction of the target road segment based on surface deformation data acquired multiple times, determine the settlement pattern of the target road segment based on the settlement plane, and thus determine the spatial pattern. The settlement plane can characterize the changes in settlement displacement and settlement rate along the direction of the target road segment.

[0067] For example, if there is a clear settlement center in the settlement plane, the settlement displacement value gradually decreases from the settlement center to the surrounding area, and the spatial pattern can be funnel-shaped.

[0068] Abnormal surface features refer to unusual conditions on the road surface, including road surface cracks, unusual wet stains, and changes in surface texture.

[0069] In some embodiments, the management platform can control drones to collect surface images of target road sections through the emergency monitoring object platform, and determine abnormal surface features of the target road sections based on the surface images. The surface images can be images of the target road section from multiple different angles.

[0070] In some embodiments, the management platform can identify anomalous surface features based on surface images using image recognition models (such as YOLO).

[0071] In some embodiments, in addition to identifying road segment units with a collapse risk greater than a risk threshold as target exploration points, the management platform can also identify target exploration points in various other ways. For example, the management platform can identify road segment units with an accelerating subsidence rate as target exploration points, as well as road segment units with abnormal surface features.

[0072] An exploration task list is a list of information related to target exploration points. For example, it includes the identifier, coordinates, and working frequency group of the target exploration points. The identifier and coordinates of the target exploration points are stored in the management platform.

[0073] In some embodiments, the management platform can also determine target exploration points with surface subsidence patterns within a preset future time period based on surface deformation data and a subsidence characteristic model, and update the exploration task list. The preset future time period is pre-set based on historical experience, such as the next 3 months.

[0074] In some embodiments, the settlement feature model can be a machine learning model. For example, the settlement feature model can include any one or a combination of a Recurrent Neural Network (RNN) model, a Convolutional Long Short-Term Memory Network (ConvLSTM) model, or other custom model structures.

[0075] In some embodiments, the input to the settlement feature model includes surface deformation data corresponding to multiple road segment units, and the output includes the probability that each road segment unit has a surface settlement pattern (e.g., the probability is represented by a value from 0 to 1).

[0076] In some embodiments, the management platform can train a settlement feature model based on a large number of first training samples with first labels, using methods such as gradient descent. The first training samples may include sample surface deformation data of sample road segment units, and the first label may be whether the sample road segment unit has experienced road collapse.

[0077] In some embodiments, the first training sample and the first label can be obtained based on historical data. For example, the management platform uses historical surface deformation data of historical road segment units as the first training sample, and determines the first label based on whether road collapse occurs in a subsequent period (e.g., 3 months) for historical road segment units corresponding to a set of similar historical surface deformation data in the first training sample. For example, the ratio of the number of road collapses to the number of road collapses that did not occur in historical road segment units corresponding to each set of similar historical surface deformation data is used as the label for that set.

[0078] In some embodiments, the settlement feature model can be trained as follows: multiple first training samples with a first label are input into the initial settlement feature model; a loss function is constructed using the first label and the prediction results of the initial settlement feature model; the initial settlement feature model is iteratively updated based on the loss function; and the training of the settlement feature model is completed when a preset condition is met. The preset condition may be that the loss function converges, the number of iterations reaches a set value, etc.

[0079] In some embodiments, the management platform can identify road segment units with a probability higher than a probability threshold (e.g., 0.7) as target exploration points with a surface subsidence pattern in a preset future time period based on the probability of multiple road segment units output by the subsidence characteristic model, and add them to the exploration task list. The probability threshold is determined based on empirical prediction.

[0080] Some embodiments in this specification actively screen surface deformation data through settlement characteristic models, thereby achieving large-scale, low-cost risk survey capabilities and enhancing the comprehensiveness and foresight of risk monitoring.

[0081] Candidate exploration paths refer to exploration paths to be determined. Exploration paths refer to the routes taken by the vehicle-mounted ground-penetrating radar to explore target exploration points.

[0082] In some embodiments, the management platform can determine multiple candidate exploration paths by enumeration based on the exploration task list.

[0083] Operational priority refers to an indicator that characterizes the importance and urgency of a target exploration point. In some embodiments, the operational priority of a target exploration point may be the same as its subsidence risk, or it may be determined based on its location. For example, a target exploration point located above a main urban road or a gas pipeline has a much higher operational priority than a target exploration point located above a remote side road.

[0084] The target exploration path refers to the final determined exploration path.

[0085] In some embodiments, the management platform can calculate the time cost and priority cost corresponding to multiple candidate exploration paths based on traffic load data and business priorities, and determine the candidate exploration path with the lowest weighted sum of time cost and priority cost as the target exploration path. The weights of time cost and priority cost are pre-set based on historical experience.

[0086] In some embodiments, the management platform can use the average vehicle speed in traffic load data as the moving speed of the vehicle-mounted ground penetrating radar, and calculate the distance required for the vehicle-mounted ground penetrating radar to traverse all target exploration points in the candidate exploration path, and use the ratio of distance to moving speed as the time cost of the candidate exploration path.

[0087] In some embodiments, for each target exploration point in the candidate exploration path, the management platform can calculate the priority cost of that target exploration point by multiplying its operational priority by the time taken for the vehicle-mounted ground-penetrating radar to travel from the starting point to the target exploration point. The management platform then sums the priority costs of each target exploration point to obtain the priority cost of the candidate exploration path.

[0088] Some embodiments in this specification, after identifying multiple target exploration points, dynamically plan target exploration routes by comprehensively evaluating traffic load data and business priorities, ensuring that limited exploration resources are prioritized for allocation to important and urgent areas.

[0089] Step 240: Control the vehicle-mounted ground-penetrating radar of the emergency monitoring platform to conduct exploration of the target exploration point based on the exploration parameters and collect exploration data.

[0090] In some embodiments, the management platform controls the vehicle-mounted ground-penetrating radar to travel to the target exploration point in the exploration parameters through the emergency monitoring object platform, and controls the vehicle-mounted ground-penetrating radar to explore the target exploration point at the working frequency group in the exploration parameters and return the exploration data to the management platform.

[0091] The exploration data can be two-dimensional or three-dimensional radar images collected by vehicle-mounted ground-penetrating radar.

[0092] Step 250: Based on the exploration data, determine whether there are any geological hazards at the target exploration point.

[0093] Geological hazard characteristics refer to geological features associated with the potential for road surface collapse. In some embodiments, geological hazard characteristics may include underground cavities, loose soil, and abnormally water-bearing areas (where a large amount of water is present under the road surface).

[0094] In some embodiments, the management platform can preprocess and analyze exploration data using computer vision algorithms to identify abnormal areas within the data, thereby determining whether the target exploration point exhibits geological hazard characteristics. Specifically, areas with strong and clearly defined reflection signals in the exploration data can be identified as underground cavities; areas with discontinuous reflection signals can be identified as loose soil; and areas with abnormally strong reflection signals and the presence of water pipes can be identified as anomalous water-bearing areas. Strong reflection signals can be characterized by a clear waveform with complete peaks and troughs. Abnormally strong reflection signals can be characterized by the peaks and troughs of the waveform being flattened, forming a plateau.

[0095] Step 260: If the target exploration point has geological hazards, generate a traffic control instruction and send the traffic control instruction to the emergency monitoring platform to control the raising of the gates on the roads associated with the target exploration point.

[0096] In some embodiments, traffic control instructions may include roads associated with the target exploration point. Upon receiving the traffic control instructions, the emergency monitoring platform can control the raising of the gates on the associated roads.

[0097] Associated roads refer to roads that are related to the target exploration point and need to be temporarily closed. In some embodiments, the management platform can identify roads connected to the target exploration point as associated roads.

[0098] Some embodiments in this specification utilize multi-source monitoring data to assess road collapse risks and dynamically optimize the exploration parameters of vehicle-mounted ground-penetrating radar based on the assessment results, achieving precise and efficient detection of geological hazards. This method can promptly identify potential risks, automatically generate and execute traffic control instructions, effectively prevent road collapse accidents, and significantly improve the safety and emergency response capabilities of urban roads.

[0099] Figure 3 This is an exemplary schematic diagram of a collapse risk model according to some embodiments of this specification.

[0100] In some embodiments, such as Figure 3 As shown, the management platform can construct a pipe segment map 340 based on standard pipe segment information 310, standard pipe segment connection information 320, and multi-source monitoring data 330, and determine the collapse risk distribution 360 based on the pipe segment map 340 through the collapse risk model 350.

[0101] For explanations regarding standard pipe sections, multi-source monitoring data, and the distribution of collapse risk, please refer to [link / reference needed]. Figure 2 And its related descriptions.

[0102] Standard pipe segment information refers to information related to standard pipe segments, such as the material, diameter, and corresponding spatial pattern of the standard pipe segment. For an explanation of the spatial pattern, please refer to step 230 and its related description.

[0103] The connection information for standard pipe segments refers to information related to the connection of standard pipe segments. In some embodiments, the connection information includes the physical connection relationship between standard pipe segments, upstream and downstream relationships, and shared inspection well information. The physical connection relationship includes whether the standard pipe segments are connected via a shared inspection well. A shared inspection well can be an inspection well jointly connected to two standard pipe segments. Shared inspection well information includes the shared inspection well number, coordinates, and last inspection time.

[0104] Standard pipe section information and connection information are pre-stored in the management platform.

[0105] A pipe segment map is a graph structure consisting of at least one node and edges between nodes, used to characterize the features of multiple standard pipe segments and their connection relationships.

[0106] In some embodiments, the management platform can use a standard pipe segment as a node in the pipe segment map (e.g., node 341). If two standard pipe segments are connected by a shared inspection well, then there is an edge between the two nodes corresponding to the two standard pipe segments (e.g., edge 342). Node characteristics include standard pipe segment information and multi-source monitoring data. Edge characteristics include upstream and downstream relationships and shared inspection well information.

[0107] In some embodiments, the collapse risk model can be a machine learning model. For example, the settlement feature model can include any one or a combination of graph neural network (GNN) models or other custom model structures.

[0108] In some embodiments, the input to the collapse risk model includes a pipe segment map, and the output includes the collapse risk of each node in the pipe segment map. The management platform generates a collapse risk distribution based on the collapse risk of each node in the pipe segment map.

[0109] In some embodiments, the management platform can train a collapse risk model based on a large number of second training samples with second labels, using methods such as gradient descent. The second training samples may include a sample pipe segment map, and the second label may be the collapse risk of multiple nodes in the sample pipe segment map.

[0110] In some embodiments, the second training sample and the second label can be obtained based on historical data. For example, the management platform can construct a sample pipe segment map based on historical standard pipe segment information, historical connection information of historical standard pipe segments, and historical multi-source monitoring data. In the sample pipe segment map, the collapse risk of nodes where road collapse has occurred is 1, and the collapse risk of nodes where road collapse has not occurred is 0.

[0111] In some embodiments, the training process of the collapse risk model is similar to that of the settlement characteristic model, and will not be described in detail here.

[0112] Some embodiments in this specification construct pipe segment maps that include standard pipe segment information, connection information, and multi-source monitoring data, effectively extracting the connection relationships and characteristics of the pipeline network, and accurately assessing the collapse risk of each pipe segment through machine learning models, significantly improving the accuracy of road collapse prediction.

[0113] Figure 4 This is an exemplary schematic diagram illustrating the determination of exploration parameters according to some embodiments of this specification. In some embodiments, such as Figure 4 As shown, the process 400 for determining exploration parameters includes the following steps: Step 410: Obtain the collapse risk distribution of the target road section at multiple time points.

[0114] For details regarding the target road section and the distribution of collapse risks, please refer to [link / reference]. Figure 2 And its related descriptions. Multiple time points, including the time points for obtaining the current collapse risk distribution and the historical collapse risk distribution.

[0115] Step 420: Based on the collapse risk distribution at each time point in the collapse risk distribution at multiple time points, determine the first abnormal point of the target road segment.

[0116] The first abnormal location refers to a road segment where the risk of collapse is spatially concentrated.

[0117] In some embodiments, the management platform identifies road segment units (i.e., target exploration points) with a collapse risk greater than a risk threshold based on the collapse risk distribution corresponding to each time point in the collapse risk distribution across multiple time points. If target exploration points exhibit clustering (e.g., target exploration points exceeding a preset clustering number are adjacent road segment units), the management platform determines the clustered target exploration points as the first abnormal point of the target road segment. The preset clustering number is pre-set based on historical experience.

[0118] Step 430: Based on the collapse risk distribution corresponding to each spatial point in the collapse risk distribution at multiple time points, determine the second abnormal point of the target road segment.

[0119] One spatial point corresponds to one road segment unit.

[0120] The second abnormal location refers to a road segment where the risk of collapse gradually increases over time.

[0121] In some embodiments, the management platform is based on the collapse risk distribution across multiple time points, specifically the collapse risk at each spatial point corresponding to multiple time points. The management platform identifies spatial points where the collapse risk continuously increases across multiple time points, and the increase in collapse risk at adjacent time points exceeds an amplitude threshold, as second abnormal points. The amplitude threshold is pre-set based on historical experience.

[0122] Step 440: Determine exploration parameters based on the first and second anomaly points.

[0123] For an explanation of exploration parameters, please refer to [link / reference]. Figure 2 And its related descriptions.

[0124] In some embodiments, the management platform can include road segment units identified as a first anomaly and a second anomaly in the target exploration points, and determine different working frequency groups for different target exploration points. For example, if a target exploration point has both a first and a second anomaly, the working frequency group includes a combination of high-frequency, mid-frequency, and low-frequency electromagnetic waves. If a target exploration point has only a first anomaly, the working frequency group includes a combination of mid-frequency and low-frequency electromagnetic waves. If a target exploration point has only a second anomaly, the working frequency group includes a combination of high-frequency and mid-frequency electromagnetic waves.

[0125] Some embodiments in this specification, through comprehensive analysis of the spatial clustering and temporal trends of collapse risk distribution, can accurately pinpoint the core exploration area with the highest risk from dynamic data, effectively filtering out static or isolated abnormal signals, thus making the location of exploration targets more precise.

[0126] It should be noted that the above descriptions of process 200 and process 400 for determining exploration parameters in the emergency monitoring method for road collapses in smart cities are merely illustrative and do not limit the scope of this specification. Those skilled in the art can make various modifications and changes to the above processes under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0127] In some embodiments, the management platform can determine the risk level of the target road segment based on the first and second anomaly points, and determine the exploration parameters based on the risk level.

[0128] Risk level refers to the classification of the risk level of a target road section in assessing its potential for collapse. For example, risk levels include Level 1, Level 2, and Level 3. Level 1 risk corresponds to the presence of both a first and second anomalous points on the target road section. Level 2 risk corresponds to the presence of only the first or second anomalous point on the target road section. Level 3 risk corresponds to the absence of both the first and second anomalous points on the target road section.

[0129] In some embodiments, the management platform can determine exploration parameters based on the level of risk. For example, if the risk level is a first or second risk level, the working frequency group includes a combination of high-frequency, mid-frequency, and low-frequency electromagnetic waves. If the risk level is a third risk level, the working frequency group includes high-frequency electromagnetic waves to explore shallow subsurface layers.

[0130] In some embodiments, when the risk level is at the first risk level, the management platform generates a traffic warning instruction and sends it to the emergency monitoring platform, controlling the warning devices on the associated roads to issue warning information. For example, based on the traffic warning instruction, the emergency monitoring platform controls a display screen to show warning information or controls a broadcasting device to play warning information. See the description of the warning devices for details. Figure 1 And its related description. For an explanation of associated roads, see step 260 and its related description.

[0131] The warning message is used to alert drivers that a road section may be prone to collapse.

[0132] In some embodiments, when the risk level is classified as Level 2, the management platform determines the load limit and height limit of the target road segment based on historical load data and the current time point's collapse risk distribution, and generates load limit and height limit instructions. The management platform sends these instructions to the emergency monitoring platform, controlling the prompting device to issue load limit information and controlling the height of the height restriction barriers on associated roads. For example, based on the load limit and height limit instructions, the emergency monitoring platform controls the display screen to show the load limit information in the instruction or controls the broadcasting equipment to play the load limit information, and adjusts the height of the height restriction barriers on associated roads to match the height limit value in the instruction.

[0133] Historical load data refers to the traffic load data of the target road segment at historical times. For an explanation of traffic load data, please refer to step 210 and its related description.

[0134] In some embodiments, the management platform can calculate the sum of the collapse risks of multiple road segment units in the target road segment based on the collapse risk distribution at the current time point, and determine the load limit and height limit of the target road segment based on the obtained sum and historical carrying capacity data. The load limit and height limit are negatively correlated with the obtained sum and positively correlated with the historical carrying capacity data.

[0135] Load limit information is used to remind drivers of the load limit for the target road section, such as "The load limit for the road section ahead is 30 tons".

[0136] In some embodiments, when the risk level is the second risk level, the management platform controls a drone to collect surface images of the target road segment through the emergency monitoring object platform, and determines abnormal surface features of the target road segment based on the surface images. If abnormal surface features exist in the target road segment, the management platform corrects the risk level corresponding to the target road segment from the second risk level to the first risk level. For an explanation of surface images and abnormal surface features, please refer to step 230 and its related description.

[0137] Some embodiments in this specification can adjust the risk level of target road sections in a timely manner based on intuitive surface features, significantly improving the accuracy of risk assessment.

[0138] In some embodiments, when the risk level is level three, the management platform determines the exploration period and generates a periodic exploration plan based on the collapse risk of the target exploration point. The management platform sends the periodic exploration plan to the emergency monitoring platform, which then controls the vehicle-mounted ground-penetrating radar to conduct exploration of the target exploration point based on the periodic exploration plan.

[0139] An exploration cycle refers to the time interval between repeated explorations of a target exploration point. A cycled exploration plan includes the coordinates of each target exploration point and the corresponding exploration cycle.

[0140] In some embodiments, the management platform determines the exploration period based on the subsidence risk of the target exploration point. For example, the greater the subsidence risk, the shorter the exploration period.

[0141] Some embodiments in this specification can automatically implement differentiated management measures, ranging from early warning and traffic diversion to periodic exploration, based on the severity of the risk of collapse in the target road section. This tiered response mechanism enables the refined use of public resources, allowing for both decisive handling of emergencies and long-term management of potential hazards, ensuring the rationality of emergency supervision and reducing economic costs.

[0142] Furthermore, certain features, structures, or characteristics in one or more embodiments of this specification may be appropriately combined.

[0143] Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, the numerical parameters should take into account specified significant digits and employ a general method of digit preservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0144] If there is any inconsistency or conflict between the descriptions, definitions, and / or terms used in the materials referenced in this specification and the content described in this specification, the descriptions, definitions, and / or terms used in this specification shall prevail.

Claims

1. A smart city road collapse emergency supervision Internet of Things large model system, characterized in that, The system includes an emergency monitoring user platform, an emergency monitoring service platform, an emergency monitoring management platform, an emergency monitoring sensor network platform, and an emergency monitoring object platform. The emergency monitoring management platform is configured as follows: Multi-source monitoring data is obtained through the aforementioned emergency monitoring platform; Based on the multi-source monitoring data, the collapse risk distribution of the target road section is determined; Based on the collapse risk distribution, exploration parameters are determined, including target exploration points and working frequency groups; The vehicle-mounted ground-penetrating radar controlling the emergency monitoring platform explores the target exploration point based on the exploration parameters and collects exploration data; Based on the exploration data, determine whether the target exploration point has any geological hazard characteristics; as well as, If the target exploration point has the geological hazard characteristics, a traffic control instruction is generated and sent to the emergency monitoring platform to control the raising of the gates on the road associated with the target exploration point.

2. The system of claim 1, wherein, The emergency monitoring and management platform is also configured as follows: Based on standard pipe segment information, standard pipe segment connection information, and the multi-source monitoring data, a pipe segment map is constructed; the pipe segment map includes at least one node and edges between nodes, the at least one node has node features, and the at least one node and the node features are determined based on the multi-source monitoring data, the standard pipe segment information, and the connection information; the edges have edge features, and the edges and the edge features are determined based on the standard pipe segment information and the connection information; and... Based on the pipe segment map, the collapse risk distribution is determined by a collapse risk model, which is a machine learning model.

3. The system of claim 1, wherein, The emergency monitoring and management platform is also configured as follows: Obtain the collapse risk distribution of the target road segment at multiple time points; Based on the collapse risk distribution corresponding to each time point in the multiple time point collapse risk distributions, the first abnormal point of the target road segment is determined; Based on the collapse risk distribution corresponding to each spatial point in the collapse risk distribution of the multiple time points, the second abnormal point of the target road segment is determined; as well as, The exploration parameters are determined based on the first and second anomaly points.

4. The system of claim 3, wherein, The emergency monitoring and management platform is also configured as follows: Based on the first and second anomaly points, the risk level of the target road segment is determined, and the risk level includes a first risk level, a second risk level, and a third risk level. as well as, The exploration parameters are determined based on the risk level.

5. The system of claim 4, wherein, The emergency monitoring and management platform is also configured as follows: When the risk level is the first risk level, a traffic warning instruction is generated and sent to the emergency monitoring object platform to control the prompting device of the associated road to issue a warning message; When the risk level is the second risk level Based on historical load data and the current time point's collapse risk distribution, the load limit and height limit of the target road section are determined, and load limit instructions and height limit instructions are generated. The load limit command and the height limit command are sent to the emergency monitoring object platform to control the prompting device to issue load limit information and control the height of the height limit frame on the associated road; When the risk level is the third risk level. Based on the collapse risk of the target exploration point, the exploration period is determined and a periodic exploration plan is generated. as well as, The periodic exploration plan is sent to the emergency monitoring platform, and the vehicle-mounted ground-penetrating radar is controlled to explore the target exploration point based on the periodic exploration plan.

6. A smart city road collapse emergency supervision method, characterized in that, The method is executed by the emergency monitoring and management platform in the smart city road collapse emergency monitoring IoT big data model system, including: Obtain multi-source monitoring data through the emergency monitoring target platform; Based on the multi-source monitoring data, the collapse risk distribution of the target road section is determined; Based on the collapse risk distribution, exploration parameters are determined, including target exploration points and working frequency groups; The vehicle-mounted ground-penetrating radar controlling the emergency monitoring platform explores the target exploration point based on the exploration parameters and collects exploration data; Based on the exploration data, determine whether the target exploration point exhibits any geological hazard characteristics; and, If the target exploration point has the geological hazard characteristics, a traffic control instruction is generated and sent to the emergency monitoring platform to control the raising of the gates on the road associated with the target exploration point.

7. The method of claim 6, wherein, The determination of the collapse risk distribution of the target road section based on the multi-source monitoring data includes: Based on standard pipe segment information, standard pipe segment connection information, and the multi-source monitoring data, a pipe segment map is constructed; the pipe segment map includes at least one node and edges between nodes, the at least one node has node features, and the at least one node and the node features are determined based on the multi-source monitoring data, the standard pipe segment information, and the connection information; the edges have edge features, and the edges and the edge features are determined based on the standard pipe segment information and the connection information; and... Based on the pipe segment map, the collapse risk distribution is determined by a collapse risk model, which is a machine learning model.

8. The method of claim 7, wherein, The determination of exploration parameters based on the collapse risk distribution includes: Obtain the collapse risk distribution of the target road segment at multiple time points; Based on the collapse risk distribution corresponding to each time point in the multiple time point collapse risk distributions, the first abnormal point of the target road segment is determined; Based on the collapse risk distribution corresponding to each spatial point in the collapse risk distribution at the multiple time points, the second abnormal point location of the target road segment is determined; and... The exploration parameters are determined based on the first and second anomaly points.

9. The method of claim 8, wherein, The method further includes: Based on the first and second anomaly locations, the risk level of the target road segment is determined, including a first risk level, a second risk level, and a third risk level; and, The exploration parameters are determined based on the risk level.

10. The method as described in claim 9, characterized in that, The method further includes: When the risk level is the first risk level, a traffic warning instruction is generated and sent to the emergency monitoring object platform to control the prompting device of the associated road to issue a warning message; When the risk level is the second risk level Based on historical load data and the current time point's collapse risk distribution, the load limit and height limit of the target road section are determined, and load limit instructions and height limit instructions are generated. The load limit command and the height limit command are sent to the emergency monitoring object platform to control the prompting device to issue load limit information and control the height of the height limit frame on the associated road; When the risk level is the third risk level. Based on the collapse risk of the target exploration point, the exploration period is determined and a periodic exploration plan is generated; and... The periodic exploration plan is sent to the emergency monitoring platform, and the vehicle-mounted ground-penetrating radar is controlled to explore the target exploration point based on the periodic exploration plan.

Images