A multi-source data collaborative management and service method based on an internet of things

Abstract

The application relates to the technical field of data management, and discloses a multi-source data collaborative management and service method based on an Internet of Things, which comprises the following steps: dividing a tunnel into a plurality of monitoring intervals and collecting multi-source data of the monitoring intervals in real time; obtaining a three-dimensional interval model corresponding to each monitoring interval and dynamically mapping the three-dimensional interval model with the multi-source data to generate a digital interval model; simulating and predicting the multi-source data according to a pre-configured simulation period and a simulation mechanism; obtaining a risk parameter of the monitoring interval based on the predicted multi-source data, and judging whether a collaborative management and service operation needs to be performed based on the risk parameter; when needed, determining a collaborative management interval and generating management service information; the application decomposes a global data processing task into a plurality of local lightweight tasks, effectively reduces the data processing burden, and ensures the collection granularity and accuracy of local position data; meanwhile, based on an interval-level dynamic mapping mechanism, precise real-time synchronization of a physical tunnel and a virtual model in a local scale is realized.

Description

Technical Field

[0001] This invention relates to the field of data management technology, and more specifically, to a method for collaborative management and service of multi-source data based on the Internet of Things. Background Technology

[0002] As a key node on highways, tunnels are characterized by enclosed spaces, complex environments, and severe consequences of accidents. When accidents such as fires, rear-end collisions, traffic jams, or hazardous material leaks occur in tunnels, the casualty rate and economic losses far exceed those of ordinary road sections. In order to improve the real-time perception and proactive risk warning capabilities of tunnel operation status, it is necessary to combine Internet of Things (IoT) technology to conduct collaborative management and service analysis of multi-source data within the tunnel, thereby ensuring the safe and reliable operation of the tunnel.

[0003] Reference patent application CN120724355A discloses a tunnel environment monitoring method and system based on multi-source data fusion. This method involves multi-source monitoring of the target tunnel, data synchronization and preprocessing of the multi-source monitoring data; spatiotemporal feature extraction and unified feature mapping of multi-source standard data; multi-source fusion of multi-modal feature sequences; environmental classification and identification of the fused feature data; and the construction of an information visualization platform to monitor and display abnormal event data and multiple environmental identification data. This system enables multi-source monitoring, data synchronization and preprocessing, followed by spatiotemporal feature extraction, multi-source fusion, and environmental classification and identification, culminating in a monitoring and display platform. This allows for the coordinated analysis of various data sources, improving anomaly identification and response capabilities, and effectively reducing false alarm rates.

[0004] However, in existing multi-source data collaborative management methods for tunnels, including the aforementioned schemes, the underlying physical data acquisition architecture still generally adopts a method of collecting data on the tunnel as a whole, and then combining physical models and virtual models to identify potential risks within the tunnel. This holistic acquisition and modeling approach results in a large amount of data collected and analyzed in a single session, a heavy data processing burden, and difficulty in ensuring the granularity and accuracy of data at local locations. At the same time, it is difficult to achieve real-time synchronous mapping between the physical tunnel and the virtual model, making it impossible for operation and management personnel to accurately and comprehensively understand the potential risks at various locations within the tunnel, thus making it difficult to avoid potential risks in advance and reducing the collaborative management and service effectiveness of multi-source data.

[0005] Therefore, there is an urgent need for a new method for multi-source data collaborative management and service based on the Internet of Things to solve the above problems. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a multi-source data collaborative management and service method based on the Internet of Things, applied to a management service platform, comprising: S1: Based on the dual division criterion, the tunnel is divided into several monitoring sections arranged in sequence, and multi-source data of each monitoring section is collected in real time through IoT terminal devices. The multi-source data includes environmental monitoring data, traffic road data, equipment status data, and location structure data; the dual division criterion is used to limit the number and length of the monitoring sections. S2: Construct a three-dimensional tunnel model that matches the physical space of the tunnel. Divide the three-dimensional tunnel model into spatial segments according to the actual spatial boundaries of each monitoring segment to obtain multiple three-dimensional interval models that correspond one-to-one with the monitoring segments. Establish a dynamic mapping between each three-dimensional interval model and the multi-source data of the corresponding monitoring segment to generate a digital interval model. S3: Pre-configure the simulation period and simulation mechanism of the digital interval model. Based on the configured simulation period and simulation mechanism, simulate and obtain the predicted multi-source data of each digital interval model in the next simulation period. S4: Pre-construct a risk identification model for risk assessment, input the predicted multi-source data into the risk identification model, and obtain the risk parameters for the corresponding monitoring interval. The risk parameters include risk type and risk probability. S5: Based on the risk parameters of the monitoring interval, determine whether collaborative management and service operations need to be performed; if so, determine the collaborative management interval from all monitoring intervals and generate management service information adapted to the collaborative management interval. The management service information is used to guide the collaborative management and service operations of the collaborative management interval.

[0007] Furthermore, the dual division criterion is as follows: the number of monitoring intervals is not less than 3, and the length of the monitoring intervals at both ends is greater than the length of the monitoring interval in the middle.

[0008] Furthermore, based on the dual-division criterion, the tunnel is divided into several monitoring intervals arranged sequentially, including: S1.1: Query the maximum width, number of lanes, and length of the tunnel. Multiply the maximum width by the number of lanes and then multiply by the first coefficient to obtain the base length. Divide the tunnel length by the base length to calculate the division factor. S1.2: When the division multiple is less than the first set value, the base length is continuously reduced in increments of 5% of the base length until the division multiple is greater than or equal to the first set value, and then S1.3 or S1.4 is executed. S1.3: When the division factor equals the first set value, the tunnel is divided into two monitoring intervals with the reference length as the middle position and 110% of the reference length as the lengths of the two monitoring intervals at the two ends. A continuous monitoring interval; S1.4: When the division factor is greater than the first set value, subtract the second set value from the division factor to obtain the intermediate value. Use the integer part of the intermediate value as the number of monitoring intervals at the intermediate position. Use the reference length as the length of the monitoring interval at the intermediate position, and use half the remaining tunnel length as the length of the monitoring intervals at both ends, thus dividing the tunnel into... A continuous monitoring interval; S1.5: Based on the direction of vehicle travel within the tunnel, number the first section as 1. The monitoring intervals are numbered in ascending order.

[0009] Furthermore, a three-dimensional tunnel model matching the actual tunnel space is constructed. The three-dimensional tunnel model is spatially divided according to the actual spatial boundaries of each monitoring interval, resulting in multiple three-dimensional interval models, each corresponding one-to-one with the monitoring intervals, including: S2.1: Use laser scanning equipment to collect point cloud data inside the tunnel in real time. After noise reduction, registration and spatial alignment of the point cloud data, construct a three-dimensional tunnel model that matches the actual tunnel space through three-dimensional modeling technology. S2.2: Will The length of each monitoring interval is divided by the tunnel length to obtain... The percentage of the length of each monitoring interval; S2.3: According to the numbering sequence of the monitoring sections, mark them one by one in the three-dimensional tunnel model along the vehicle's direction of travel. The model boundary points corresponding to the length ratio of each monitoring interval are then connected sequentially to obtain the dividing surface; S2.4: Using each cutting plane as the cutting point, divide the 3D tunnel model into sections. A three-dimensional interval model.

[0010] Furthermore, a dynamic mapping is established between each three-dimensional interval model and the multi-source data of the corresponding monitoring interval to generate a digital interval model, including: S2.5: Establish There are four blank datasets, each containing four mapping units. Environmental monitoring data, traffic road data, equipment status data, and location structure data of each monitoring interval are imported into the corresponding four mapping units to convert the blank datasets into mapped datasets. S2.6: Using digital twin technology, a one-way transmission mapping channel is established between each three-dimensional interval model and its corresponding mapping dataset. The mapping channel is divided into four parallel sub-channels, and the input port of each sub-channel is connected to the corresponding mapping unit, and the output port is connected to the corresponding three-dimensional interval model. S2.7: Set a mapping clock on each of the four sub-channels belonging to the same mapping channel, and adjust the four mapping clocks to keep them synchronized in time; S2.8: Will Dynamically map multi-source data within a mapping dataset to corresponding... A digital interval model is generated by overlaying multi-source data on a three-dimensional interval model along the time axis.

[0011] Furthermore, the steps for configuring the simulation time period for the digital interval model include: S3.1: The smallest detectable change that occurs in each of the environmental monitoring data, traffic road data, equipment status data, and location structure data is obtained, and is denoted as the unit amplitude. S3.2: Record the data that changes by a unit amplitude as the target data. Divide the number of target data by the number of environmental monitoring data, traffic road data, equipment status data and location structure data respectively, and calculate the environmental proportion, traffic proportion, equipment proportion and location proportion. S3.3: Query the duration corresponding to the environmental percentage from 0 to 0.6, the traffic percentage from 0 to 0.5, the equipment percentage from 0 to 0.4, and the location percentage from 0 to 0.5, and record them as the first duration, the second duration, the third duration, and the fourth duration, respectively; S3.4: Based on the proportional relationship between the environmental proportion, traffic proportion, equipment proportion and location proportion, the proportional coefficients of the first duration, second duration, third duration and fourth duration are set in reverse, and the first duration, second duration, third duration and fourth duration are weighted and summed to calculate the comprehensive duration; S3.5: Starting from the current moment, and using a comprehensive duration extending backward from the starting point as the standard, set the simulation time period.

[0012] Furthermore, the simulation mechanism of the digital interval model is as follows: simulation is performed using multi-source data from four consecutive simulation periods in the past.

[0013] Furthermore, based on the risk parameters of the monitoring interval, it is determined whether collaborative management and service operations need to be performed, including: S5.1: The risk status of a risk type whose risk probability is greater than the baseline probability value and less than the first probability value is recorded as low risk; the risk status of a risk type whose risk probability is greater than or equal to the first probability value and less than the second probability value is recorded as medium risk; and the risk status of a risk type whose risk probability is greater than the second probability value is recorded as high risk. S5.2: When the number of low-risk risk types in each risk parameter of the monitoring interval is greater than or equal to 3, it is determined that the monitoring interval needs to perform collaborative management and service operations. S5.3: When the number of medium-risk risk types in each risk parameter of the monitoring interval is greater than or equal to 2, it is determined that the monitoring interval needs to perform collaborative management and service operations. S5.4: When the number of high-risk types in each risk parameter of the monitoring interval is greater than or equal to 1, it is determined that the monitoring interval needs to perform collaborative management and service operations.

[0014] Furthermore, collaborative management intervals are determined from all monitoring intervals, including: S5.6: The monitoring intervals that require collaborative management and service operations are designated as risk intervals, the interval numbers of the risk intervals are designated as risk numbers, and the arrangement of risk numbers for all risk intervals is identified; the arrangement includes continuous and intermittent states. S5.7: When the risk numbers are in a continuous state, the monitoring interval whose interval number is before the minimum value of all risk numbers is recorded as the preceding interval, and the monitoring interval whose interval number is after the maximum value of all risk numbers is recorded as the following interval. The preceding interval, all risk intervals and the following interval are summarized to form a collaborative management interval. S5.8: When the risk number is in an interval state, the monitoring interval that is before and after each risk number is recorded as the collaborative interval, and all risk intervals and collaborative intervals are summarized to form a collaborative management interval.

[0015] Furthermore, the management service information includes forecasts of multi-source data, risk types, risk probabilities, and interval numbers.

[0016] Beneficial effects: This invention proposes a multi-source data collaborative management and service method based on the Internet of Things (IoT). Based on a dual partitioning criterion, the tunnel is divided into multiple sequentially arranged monitoring intervals, and an independently mapped digital interval model is constructed for each monitoring interval. This decomposes the global data processing task into multiple local lightweight tasks, effectively reducing the data processing burden and ensuring the granularity and accuracy of local location data collection. Simultaneously, based on an interval-level dynamic mapping mechanism, precise real-time synchronization between the physical tunnel and the virtual model at a local scale is achieved, solving the latency and inaccuracy problems of traditional overall mapping. Furthermore, by combining simulation prediction and risk identification, proactive and quantitative prediction of potential risks in each monitoring interval is achieved, thereby determining the collaborative management interval and appropriate management service information. This allows operators to accurately and comprehensively understand the potential risks at various locations within the tunnel, facilitating early avoidance of potential risks and improving the collaborative management and service effect of multi-source data. This invention significantly enhances the proactive risk warning capability and collaborative management efficiency of tunnel operation. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a multi-source data collaborative management and service method based on the Internet of Things, provided in Embodiment 1 of the present invention. Figure 2 This is a schematic diagram of a multi-source data collaborative management and service system based on the Internet of Things, provided in Embodiment 2 of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Please see Figure 1 As shown in the figure, this embodiment provides a method for multi-source data collaborative management and service based on the Internet of Things, applied to a management service platform. The method includes: S1: Based on the dual division criterion, the tunnel is divided into several monitoring sections arranged in sequence, and multi-source data of each monitoring section is collected in real time through IoT terminal devices. The multi-source data includes environmental monitoring data, traffic road data, equipment status data and location structure data; the dual division criterion is used to limit the number and length of the monitoring sections.

[0020] Tunnels are a component of highways, and tunnels of varying lengths correspond to different scales of mountain structures. Larger mountain structures result in longer tunnels. To perform comprehensive IoT multi-source data collection and analysis on long tunnels presents several challenges. First, the number of IoT terminal devices and the amount of data collected are growing exponentially, placing immense pressure on data transmission, storage, and computing resources, significantly increasing the workload of data collection. Second, environmental parameters in different sections of long tunnels exhibit significant spatial heterogeneity. Globally uniform processing would smooth out local features, leading to decreased data granularity. Abnormal signals in local locations are easily averaged or submerged, reducing the accuracy of anomaly identification and increasing the likelihood of missed or false alarms. Therefore, it is necessary to break down long tunnels into smaller, manageable segments for processing.

[0021] In this embodiment, the monitoring interval is a portion of the tunnel that is continuously divided according to its length, so that multiple monitoring intervals can be spliced ​​together to form a complete tunnel.

[0022] When vehicles enter or exit tunnels, the light intensity in the entrance and exit areas undergoes drastic changes: entering from bright to dark can create a "black hole effect," while exiting from dark to bright can create a "white hole effect," both of which can cause drivers to have difficulty adapting visually for a moment, or even suffer temporary blindness. Considering the varying shapes and dimensions of different tunnels, to ensure the accuracy and rationality of the monitoring interval division, tunnel division operations need to be carried out under the constraints of a dual division criterion. Specifically, the dual division criterion is: the number of monitoring intervals should not be less than three, and the length of the monitoring intervals at both ends should be greater than the length of the monitoring interval at the middle. This dual division criterion ensures that a certain threshold for the number of monitoring intervals is maintained and also eliminates the negative impact on the driver's vision caused by changes in light intensity when vehicles enter and exit tunnels.

[0023] Based on a dual-division criterion, the tunnel is divided into several monitoring intervals arranged sequentially, including: S1.1: Query the maximum width, number of lanes, and length of the tunnel. Multiply the maximum width by the number of lanes and then multiply by the first coefficient to obtain the base length. Divide the tunnel length by the base length to calculate the division factor. Specifically, the maximum width of the tunnel is first determined by consulting the design drawings. Number of lanes and tunnel length The maximum width is multiplied by the number of lanes, then multiplied by the first coefficient to calculate the baseline length. Finally, the tunnel length is divided by the baseline length to calculate the division factor. The formula "maximum width × number of lanes" comprehensively reflects the tunnel's bandwidth and traffic flow density. Multiplying by a first coefficient determines a suitable baseline length, which serves as the initial length of the monitoring interval. As the number of lanes in the tunnel increases, the maximum width also increases, consequently increasing the number of vehicles passing through the tunnel per unit time and the traffic flow density. This necessitates a corresponding increase in the monitoring interval distance to cover a more comprehensive vehicle data collection window. Therefore, the baseline length increases with the number of lanes or the maximum width in the tunnel, meaning the division factor decreases as the number of lanes or the maximum width increases.

[0024] The value of the first coefficient should not be too large or too small. If it is too large, the reference length will be too long, resulting in multiple vehicles within a single monitoring interval. This reduces the data granularity and accuracy at local locations, making it impossible to accurately locate potential risks in the tunnel and achieve interval-level collaborative management. If it is too small, the reference length will be too short, resulting in insufficient perception time for vehicles within the interval, making it impossible to collect complete data. Additionally, the number of divided monitoring intervals will be too large, placing a heavy burden on the system. In this embodiment, the first coefficient is set to 2, and the formula for calculating the division multiple is as follows: ; In the formula, To divide into multiples, The length of the tunnel. For the maximum width, This refers to the number of lanes.

[0025] S1.2: When the division multiple is less than the first set value, the base length is continuously reduced in increments of 5% of the base length until the division multiple is greater than or equal to the first set value, and then S1.3 or S1.4 is executed. Under the constraint of the dual division criterion, the number of monitoring intervals must be at least three, and the length of the monitoring intervals at both ends must be greater than the length of the monitoring interval in the middle. When the number of monitoring intervals is the minimum of three, to ensure that the length of the monitoring intervals at both ends is not too large and to meet the subsequent data collection requirements, this scheme sets the length of the monitoring intervals at both ends to be 1.1 times the length of the middle monitoring interval (i.e., the baseline length). At this time, the total tunnel length is equivalent to the sum of 1.1 baseline lengths, 1 baseline length, and 1.1 baseline lengths, which is 3.2 baseline lengths, i.e., the first set value is 3.2. Therefore, the division multiple of 3.2 is the minimum requirement to meet the dual division criterion of monitoring intervals in this scheme; below this value, it is impossible to form a division of at least three monitoring intervals while ensuring a reasonable ratio between the end monitoring intervals and the middle monitoring intervals; above this value, it means that the tunnel has the capacity to accommodate more middle intervals.

[0026] When the division factor is less than 3.2, the baseline length needs to be adjusted by gradually reducing it to ensure that the division factor is greater than or equal to 3.2. Specifically, a 5% reduction increment is used as the reduction range. This ensures that the baseline length changes within a controllable and reasonable range during each reduction operation, preventing large fluctuations in the calculated division factor due to excessively large or small changes. Furthermore, the 5% reduction increment is set based on the average reduction increment observed in a single reduction operation across a large number of historical monitoring intervals. After reducing the baseline length, the final new division factor will always be greater than or equal to 3.2.

[0027] S1.3: When the division factor equals the first set value, the tunnel is divided into two monitoring intervals with the reference length as the middle position and 110% of the reference length as the lengths of the two monitoring intervals at the two ends. A continuous monitoring interval; When the division factor is exactly 3.2, there is no need to adjust the reference length, and the tunnel can be directly divided. The length of the monitoring intervals at both ends is equal to 3, which is 1.1 times the baseline length. The length of the monitoring interval in the middle is equal to the baseline length.

[0028] S1.4: When the division factor is greater than the first set value, subtract the second set value from the division factor to obtain the intermediate value. Use the integer part of the intermediate value as the number of monitoring intervals at the intermediate position. Use the reference length as the length of the monitoring interval at the intermediate position, and use half the remaining tunnel length as the length of the two monitoring intervals at the two ends, thus dividing the tunnel into... A continuous monitoring interval; The value of the second setpoint is related to the value of the first setpoint. In S1.2, to meet the requirements of the dual division criterion, the length of the monitoring intervals at both ends is set to 1.1 times the length of the monitoring interval in the middle. Regardless of the tunnel dimensions, the monitoring intervals at both ends must always occupy at least 1.1 times the reference length. Therefore, the total share of the reference length corresponding to the monitoring intervals at both ends of the tunnel is at least 2.2, that is, the second setpoint is 2.2. Subtracting this fixed value from the total division multiple yields the number of reference lengths included in the remaining part, which is the middle value. If the midpoint value is an integer multiple of the baseline length, it can be directly used as the number of central monitoring sections. The length of the central monitoring sections is the baseline length, and the length of the end monitoring sections is 1.1 times the baseline length. If the midpoint value is not an integer multiple of the baseline length, the integer part is used as the number of central monitoring sections. The length of the central monitoring sections is the baseline length. Then, the total tunnel length is subtracted from the length of all central monitoring sections to obtain the length of the two end monitoring sections. Then, this length is divided by 2 to obtain the length of a single end monitoring section. This is how the tunnel can be divided into monitoring sections.

[0029] S1.5: Based on the direction of vehicle travel within the tunnel, number the first section as 1. The monitoring intervals are numbered in ascending order.

[0030] By dividing the tunnel into continuously arranged monitoring intervals using a dual division criterion, the long tunnel can be broken down into smaller parts. This allows for the segmentation of a large amount of multi-source data within a long tunnel, reducing the data collection burden within each monitoring interval and ensuring the granularity and accuracy of the data. At the same time, dividing the monitoring intervals at both ends into longer intervals than the middle intervals also takes into account the impact of light changes on the driver's vision when vehicles enter and exit the tunnel. This provides more space for collecting dynamic data generated by the driver when entering and exiting the tunnel, improving the rationality and accuracy of the multi-source data.

[0031] To better understand the various tunnel classifications mentioned above, three specific examples are given below.

[0032] Example 1: Take a relatively short tunnel in a city as an example. The tunnel is 230 meters long, has 3 lanes, and the maximum width of the tunnel is 13.25 meters.

[0033] First, the base length is calculated to be 79.5 meters, with a division factor of approximately 2.89. Comparing this to the first set value of 3.2, 2.89 is less than 3.2. Therefore, step S1.2 needs to be executed to reduce the base length by 79.5 * 0.05. After the first reduction, the base length is 75.525 meters, with a division factor of approximately 3.05. Comparing this to the first set value, 3.05 is less than 3.2. Therefore, the base length needs to be further reduced by 79.5 * 0.05. After the second reduction, the base length is 71.55 meters, with a division factor of approximately 3.215. Since 3.215 is greater than 3.2, the reduction is stopped, and step S1.4 is executed.

[0034] When performing step S1.4, firstly, subtract the second set value 2.2 from the division multiple 3.215 to obtain the intermediate value 1.015. Take its integer part 1 as the number of monitoring intervals at the intermediate position. The length of this monitoring interval is the baseline length of 71.55 meters at this time. Then, subtract the length of the monitoring interval at the intermediate position 71.55 from the total tunnel length 230 to obtain the sum of the lengths of the two end position monitoring intervals, which is 158.45 meters. Then, take half of 158.45 meters, 79.225 meters, as the length of a single end position monitoring interval. In this way, a total of three monitoring intervals are obtained. According to the driving direction of vehicles in the tunnel, the three monitoring intervals are numbered. The length of the monitoring interval numbered 1 is 79.225 meters, the length of the monitoring interval numbered 2 is 71.55 meters, and the length of the monitoring interval numbered 3 is 79.225 meters.

[0035] Example 2: Taking a mountain highway tunnel as an example, the tunnel is 1507 meters long, has 3 lanes, and the maximum width of the tunnel is 14.25 meters.

[0036] First, the baseline length is calculated to be 85.5 meters, with a division factor of approximately 17.63. This is compared to the first set value of 3.2. Since 17.63 is greater than 3.2, step S1.4 needs to be executed. First, subtract the second set value of 2.2 from the division factor of 17.63 to obtain the intermediate value of 15.43. Take the integer part 15 as the number of monitoring intervals at the intermediate position. The length of this monitoring interval is the baseline length of 85.5 meters at this point. Therefore, the total length of the monitoring intervals at the intermediate position is 1282.5 meters. Then, subtract the number of monitoring intervals at the intermediate position from the total tunnel length of 1507 meters. The total length of the monitoring interval is 1282.5 meters, which is the sum of the lengths of the two end position monitoring intervals, 224.5 meters. Then, half of 224.5 meters, 112.25 meters, is taken as the length of a single end position monitoring interval. In this way, a total of 17 monitoring intervals are obtained. According to the driving direction of vehicles in the tunnel, the 17 monitoring intervals are numbered. The length of monitoring interval number 1 is 112.25 meters, the length of monitoring intervals numbered 2-16 is 71.55 meters, and the length of monitoring interval number 17 is 112.25 meters.

[0037] Example 3: Taking a cross-river underwater tunnel as an example, the tunnel is 244.8 meters long, has 3 lanes, and the maximum width of the tunnel is 12.75 meters.

[0038] First, the base length is calculated to be 76.5 meters, and the division factor is 3.2, which is equal to the first set value. Step S1.3 needs to be executed to divide the tunnel into 3 consecutive monitoring intervals. According to the driving direction of vehicles in the tunnel, the three monitoring intervals are numbered. The length of monitoring interval number 1 is 84.15 meters, the length of monitoring interval number 2 is 76.5 meters, and the length of monitoring interval number 3 is 84.15 meters.

[0039] In this embodiment, any two adjacent monitoring intervals are in a continuous geographical position, which enables the monitoring intervals to continuously divide the internal spatial position of the tunnel and avoid the phenomenon of local position loss or the occurrence of discontinuity at the critical position of the interval.

[0040] After the tunnel is divided into multiple continuous monitoring sections, real-time multi-source data from various IoT terminal devices in each monitoring section can be collected, and the real-time multi-source data can be used as the basis for subsequent judgment on whether any abnormalities have occurred in the monitoring section.

[0041] IoT terminal devices refer to IoT sensors installed and deployed inside the tunnel to detect multi-dimensional changes within the monitoring range. The type and number of IoT terminal devices installed and deployed in each monitoring range other than the two ends are the same. Only the installation spacing of IoT terminal devices in the monitoring ranges at the two ends is slightly larger than that in the monitoring range at the middle position. This ensures that the data monitoring length range of the IoT terminal devices is proportional to the specific length of the monitoring range, so that the IoT terminal devices can comprehensively collect data in the monitoring ranges at both ends and avoid the omission or loss of data in local positions.

[0042] Specifically, IoT terminal devices include, but are not limited to, temperature sensors, humidity sensors, smoke sensors, and cameras; and all of these IoT terminal devices communicate wirelessly with the management service platform.

[0043] Multi-source data is used to represent the dynamic changes that occur in a monitoring area in real time and in multiple dimensions; specifically, multi-source data includes environmental monitoring data, traffic and road data, equipment status data, and location and structure data.

[0044] Environmental monitoring data is used to represent the real-time external environment in the spatial area where the monitoring range is located in multiple dimensions. Environmental monitoring data includes, but is not limited to, visibility, carbon monoxide concentration, temperature, humidity, wind speed and direction, smoke concentration, etc. In this embodiment, environmental monitoring data is collected in real time through IoT terminal devices such as visibility meters, temperature sensors, and humidity sensors.

[0045] Traffic road data is used to represent the real-time traffic situation on the tunnel surface where the monitoring section is located in a multi-dimensional way. Traffic road data includes, but is not limited to, traffic flow, average vehicle speed, traffic density, vehicle spacing, and vehicle dwell time. In this embodiment, traffic road data is collected in real time through IoT terminal devices such as speed radar and video surveillance probes.

[0046] Equipment status data is used to represent the real-time status of auxiliary equipment (blowers, lights, audible and visual alarms, etc.) in the tunnel where the monitoring section is located in a multi-dimensional way. Equipment status data includes, but is not limited to, operating voltage, operating current, historical fault frequency, load energy consumption, start-stop frequency, equipment running time, etc. In this embodiment, equipment status data is collected in real time after the equipment self-tests.

[0047] Location structure data is used to represent the spatial location and real-time status of the wall structure at the location of the tunnel in the monitoring section in multiple dimensions. Location structure data includes, but is not limited to, lining deformation, crack width, settlement displacement, etc. In this embodiment, location structure data is collected in real time by IoT terminal devices such as multi-point displacement gauges, hydrostatic levels, and crack gauges.

[0048] S2: Construct a three-dimensional tunnel model that matches the physical space of the tunnel. Divide the three-dimensional tunnel model into spatial segments according to the actual spatial boundaries of each monitoring segment to obtain multiple three-dimensional interval models that correspond one-to-one with the monitoring segments. Establish a dynamic mapping between each three-dimensional interval model and the multi-source data of the corresponding monitoring segment to generate a digital interval model.

[0049] A 3D tunnel model is a 3D virtual model constructed using 3D modeling technology that maintains consistency with the overall shape, size, and structural position of the tunnel. It can convert the physical tunnel into a virtual simulation model to facilitate subsequent simulation and analysis of relevant data within the tunnel.

[0050] Since the multi-source data is collected for the monitoring interval, while the three-dimensional tunnel model is a virtual representation of the entire tunnel, the multi-source data and the three-dimensional tunnel model are incompatible. Therefore, it is necessary to segment the three-dimensional tunnel model into three-dimensional interval models that match the multi-source data.

[0051] This involves constructing a three-dimensional tunnel model that matches the physical space of the tunnel. The three-dimensional tunnel model is then spatially divided according to the actual spatial boundaries of each monitoring interval, resulting in multiple three-dimensional interval models that correspond one-to-one with the monitoring intervals. These include: S2.1: Use laser scanning equipment to collect point cloud data inside the tunnel in real time. After noise reduction, registration and spatial alignment of the point cloud data, construct a three-dimensional tunnel model that matches the actual tunnel space through three-dimensional modeling technology. It should be noted that the method used to construct the 3D tunnel model is existing technology. Specifically, the construction process involves first acquiring point cloud data of the tunnel's interior space in real time using a laser scanning device. Then, the point cloud data undergoes point cloud sampling and noise reduction to remove outliers caused by suspended particles, equipment vibrations, etc. The ICP algorithm is used to stitch together point cloud data from different locations. Ground control points are used to spatially align and unify the model coordinates with the actual geographical coordinates of the tunnel. Finally, Delaunay triangulation is used to convert the processed point cloud data into triangular mesh surfaces, and the triangular mesh surfaces are connected to generate the 3D tunnel model. This 3D tunnel model is generated based on real-time acquisition of point cloud data of the tunnel's interior space, therefore it matches the actual tunnel space.

[0052] S2.2: Will The length of each monitoring interval is divided by the tunnel length to obtain... The percentage of the length of each monitoring interval; S2.3: According to the numbering sequence of the monitoring sections, mark them one by one in the three-dimensional tunnel model along the vehicle's direction of travel. The model boundary points corresponding to the length ratio of each monitoring interval are then connected sequentially to obtain the dividing surface; Model boundary points are the points at the critical positions of two adjacent three-dimensional interval models. In this embodiment, the number of model boundary points on the same cutting surface is huge. Therefore, the shape of the cutting surface can be formed by connecting the model boundary points on the same cutting surface in sequence. S2.4: Using each cutting plane as the cutting point, divide the 3D tunnel model into sections. A three-dimensional interval model.

[0053] The three-dimensional interval model can only perform virtual simulation of the tunnel in terms of shape, size and location structure. It does not dynamically associate with the multi-source data in each monitoring interval of the tunnel. As a result, the three-dimensional tunnel model cannot be effectively associated and mapped with the actual situation of the tunnel. Therefore, it is necessary to dynamically map the three-dimensional interval model with the multi-source data to obtain a digital interval model that can perform real-time dynamic simulation of the tunnel part corresponding to the monitoring interval.

[0054] A dynamic mapping is established between each three-dimensional interval model and the multi-source data of the corresponding monitoring interval to generate a digital interval model, including: S2.5: Establish There are four blank datasets, each containing four mapping units. Environmental monitoring data, traffic road data, equipment status data, and location structure data of each monitoring interval are imported into the corresponding four mapping units to convert the blank datasets into mapped datasets. Specifically, the number of blank datasets is consistent with the number of monitoring intervals. Each mapping unit is used to store a blank data group of a specific type of multi-source data, ensuring that different types of multi-source data can maintain a relatively independent state. S2.6: Using digital twin technology, a one-way transmission mapping channel is established between each three-dimensional interval model and its corresponding mapping dataset. The mapping channel is divided into four parallel sub-channels, and the input port of each sub-channel is connected to the corresponding mapping unit, and the output port is connected to the corresponding three-dimensional interval model. Specifically, the mapping channel is used to provide dynamic data mapping between the three-dimensional interval model and the mapping dataset, enabling the three-dimensional interval model to maintain a real-time dynamic association with the corresponding monitoring interval. By dividing the internal space of the mapping channel into four sub-channels, a one-to-one mapping effect between a mapping unit and a sub-channel can be achieved, improving the independence of the dynamic mapping process and the reliability of the dynamic mapping results.

[0055] S2.7: Set a mapping clock on each of the four sub-channels belonging to the same mapping channel, and adjust the four mapping clocks to keep them synchronized in time; S2.8: Will Dynamically map multi-source data within a mapping dataset to corresponding... A digital interval model is generated by superimposing multi-source data on a three-dimensional interval model along the time axis of the mapped clock.

[0056] It should be noted that by setting a time-synchronized mapping clock, it can be ensured that the multi-source data in each dynamic mapping has the same timestamp, preventing time misalignment or asynchrony of multi-source data in the digital interval model; at the same time, the method of one type of multi-source data corresponding to one sub-channel can ensure the stability of dynamic mapping of different types of multi-source data and prevent cross-interference of different types of multi-source data during dynamic mapping.

[0057] By using a mapping channel with four equal molecular channels, a three-dimensional interval model can be dynamically mapped to construct a digital interval model corresponding to the monitoring interval. This achieves a one-to-one mapping effect between a type of mapping data and a sub-channel, improving the independence of the dynamic mapping process and the reliability of the dynamic mapping results. It also realizes the synchronous association effect between the physical entity tunnel and the virtual model.

[0058] S3: Pre-configure the simulation period and simulation mechanism of the digital interval model. Based on the configured simulation period and simulation mechanism, simulate the predicted multi-source data of each digital interval model in the next simulation period.

[0059] The digital interval model can only represent the real-time changes in the monitoring interval within the tunnel, and cannot predict future changes in the monitoring interval. In order to obtain the specific values ​​of the predicted multi-source data at the current moment, it is necessary to perform advanced virtual simulation processing on the digital interval model so that the digital interval model can make advanced predictions based on the existing multi-source data.

[0060] When performing advanced simulation and prediction on digital interval models, it is necessary to first set the simulation period and simulation mechanism of the digital interval models.

[0061] The simulation period refers to the time span during which the digital interval model performs advanced simulation and prediction, thus providing a time dimension limit for the simulation of the digital interval model, enabling the digital interval model to accurately simulate and predict multi-source data for a certain period of time in the future. Specifically, the steps for configuring the simulation time period for the digital interval model include: S3.1: Query the database to find the smallest detectable change in all data in the environmental monitoring data, traffic road data, equipment status data, and location structure data, and record it as the unit amplitude; Specifically, the minimum detectable change refers to the minimum change in the values ​​of environmental monitoring data, traffic road data, equipment status data, and location structure data that can be collected and monitored by the corresponding IoT terminal devices. The unit of change for each specific multi-source data may be different. For example, the unit of change for temperature in environmental monitoring data is 1 degree Celsius. S3.2: Record the data that changes by a unit amplitude as the target data. Divide the number of target data by the number of environmental monitoring data, traffic road data, equipment status data and location structure data respectively, and calculate the environmental proportion, traffic proportion, equipment proportion and location proportion. Once the unit amplitude is established, the system monitors the change in each data parameter in real time. If, within the current observation window, the change in a parameter reaches its unit amplitude, this data is marked as "target data." Then, the system calculates the number of parameters in each of the four main categories—environmental monitoring data, traffic and road data, equipment status data, and location and structure data—that become target data. This number is then divided by the total number of parameters in that category to obtain four percentages: environmental percentage, traffic percentage, equipment percentage, and location percentage. These percentages reflect the activity level of various data types within the tunnel—a higher percentage indicates that more parameters in that category are changing.

[0062] S3.3: Query the duration corresponding to the environmental percentage from 0 to 0.6, the traffic percentage from 0 to 0.5, the equipment percentage from 0 to 0.4, and the location percentage from 0 to 0.5, and record them as the first duration, the second duration, the third duration, and the fourth duration, respectively; The system internally establishes a mapping relationship based on historical data or experimental calibration: for each type of data (environment, traffic, equipment, location), how long does it typically take for its proportion to rise from 0 to a specified value (e.g., environment proportion 0.6, traffic proportion 0.5, equipment proportion 0.4, location proportion 0.5)? This step involves looking up the actual proportion value calculated in step S3.2 in a table or calculating the corresponding duration using a function: first duration (corresponding to environment proportion), second duration (corresponding to traffic proportion), third duration (corresponding to equipment proportion), and fourth duration (corresponding to location proportion). These durations essentially reflect the current rate of change for each type of data—the shorter the time taken to reach the current level of activity, the more drastic the change.

[0063] S3.4: Based on the proportional relationship between the environmental proportion, traffic proportion, equipment proportion and location proportion, the proportional coefficients of the first duration, second duration, third duration and fourth duration are set in reverse, and the first duration, second duration, third duration and fourth duration are weighted and summed to calculate the comprehensive duration; Because the four data categories have different weights in influencing the overall tunnel situation, and the more drastic the changes in a category, the shorter the simulation period should be (to capture subsequent changes more quickly), a reverse weighting strategy is adopted here. Specifically, based on the proportional relationship between the environmental, traffic, equipment, and location proportions, corresponding proportional coefficients are assigned to the four durations—the larger the proportion of a category, the smaller the coefficient of its corresponding duration in the weighted summation. Then, the four durations are multiplied by their respective coefficients and summed, followed by normalization to obtain a final composite duration. This composite duration represents the length of the simulation prediction time window that the system should set under the current level of data activity.

[0064] In this embodiment, the formula for calculating the overall duration is: ; In the formula, To account for the total duration, For the first duration, For the second duration, This is the third duration. This is the fourth duration.

[0065] It should be noted that the aforementioned reverse setting of the proportional coefficient means that the larger the value of the environmental proportion, traffic proportion, equipment proportion, or location proportion, the smaller the corresponding proportional coefficient. In this embodiment, it is known from common knowledge that the sum of the proportional coefficients is 1. At the same time, in the above content, the sum of the environmental proportion, traffic proportion, equipment proportion, and location proportion is: 0.6 + 0.5 + 0.4 + 0.5 = 2. And after dividing 0.6, 0.5, 0.4, and 0.5 by 2, they are 0.3, 0.25, 0.2, and 0.25, respectively. Therefore, when setting the proportional coefficient in reverse, the proportional coefficients of the environmental proportion, traffic proportion, equipment proportion, and location proportion can be reversed to 0.2, 0.25, 0.3, and 0.25.

[0066] S3.5: Starting from the current moment, and using a comprehensive duration extending backward from the starting point as the standard, set the simulation time period.

[0067] Specifically, starting from the current moment, the simulation period is extended forward by a comprehensive duration to obtain the end time of the simulation period. Thus, the simulation period is defined as [current moment, current moment + comprehensive duration]. Within this period, the system will use a pre-configured simulation mechanism to extrapolate each digital interval model and calculate their predicted multi-source data at the end of the next simulation period (i.e., current moment + comprehensive duration). In this way, the length of the simulation period can adaptively follow the changing rhythm of the multi-source data within the tunnel—a shorter period and more intensive predictions for faster data changes, and a longer period and more computational resources for slower data changes.

[0068] The simulation mechanism is the specific constraint condition used to perform advanced simulation and prediction of the digital interval model, thereby limiting the simulation of the digital interval model. Specifically, the simulation mechanism of the digital interval model is as follows: simulation is performed using multi-source data from four consecutive simulation periods in the past; this ensures that the amount of multi-source data participating in the prediction of the next simulation period remains sufficient, and also ensures that the multi-source data participating in the prediction changes as much as possible.

[0069] Predictive multi-source data refers to the multi-source data simulated by the digital interval model in the next simulation period at the current moment, so that the predicted multi-source data and the real-time multi-source data are consistent in terms of specific data type and quantity. When simulating and predicting multi-source data, environmental monitoring data, traffic road data, equipment status data, and location structure data within the digital interval model are arranged in chronological order. Environmental monitoring data, traffic road data, equipment status data, and location structure data from four consecutive simulation periods before the current time are selected as simulation data. Environmental monitoring data, traffic road data, equipment status data, and location structure data from the next simulation period after the current time are used as simulation targets to simulate the digital interval model and obtain the predicted multi-source data for the next simulation period.

[0070] By combining simulation mechanisms to simulate digital interval models, potential risks and hazards at different locations within the tunnel can be identified in advance, thus avoiding the lag inherent in real-time risk analysis methods. This provides sufficient response time for collaborative management services for risk warning, thereby mitigating potential risks and hazards that may arise within the tunnel in the future.

[0071] S4: Pre-construct a risk identification model for risk assessment, input the predicted multi-source data into the risk identification model, and obtain the risk parameters for the corresponding monitoring interval. The risk parameters include risk type and risk probability.

[0072] The risk identification model is an artificial intelligence model based on deep learning technology that comprehensively learns and identifies risk parameters in multi-source data. Its input is the predicted multi-source data of each monitoring interval in the next simulation period, specifically including environmental monitoring data, traffic road data, equipment status data and location structure data of the interval, and the output is the risk parameters of the corresponding monitoring interval.

[0073] Risk parameters refer to the specific content of abnormal risk situations that multi-source data may cause to the tunnel, thus serving as a direct basis for judging the specific risk situation of the tunnel. Specifically, risk parameters include risk type and risk probability. Among them, risk type is used to represent the specific category of risk in multi-source data, including environmental risk, traffic risk, equipment risk, and structural risk; risk probability is used to represent the specific probability that each type of risk may occur, and the risk probability ranges from 0 to 1.

[0074] The aforementioned risk identification model is based on existing deep learning models, using a large amount of historical multi-source data and corresponding risk parameters as training data, and is obtained through continuous iterative training. During training, a large amount of multi-source data and risk parameters are pre-collected, and a set of multi-source data and corresponding risk parameters are bound into data samples. These data samples are then divided into training samples and test samples in a 75%:25% ratio. The deep learning model parameters are set, including learning rate, batch size, and maximum number of epochs. Cross-validation is used to optimize the deep learning model and avoid overfitting. The deep learning model learns the risk types and probabilities from environmental monitoring data, traffic road data, equipment status data, and location structure data, and outputs the corresponding risk parameters. The deep learning model is trained using training samples and tested and validated using test samples until the risk parameter identification accuracy in the test samples reaches a preset accuracy threshold, thus obtaining the desired risk identification model. In this embodiment, the preset accuracy threshold can be customized; for example, the preset accuracy threshold is 0.92.

[0075] S5: Based on the risk parameters of the monitoring interval, determine whether collaborative management and service operations need to be performed; if so, determine the collaborative management interval from all monitoring intervals and generate management service information adapted to the collaborative management interval. The management service information is used to guide the collaborative management and service operations of the collaborative management interval.

[0076] Collaborative management and service operations are the collaborative management and service measures that need to be implemented when abnormal risk phenomena occur in the tunnel. That is, when abnormal risks occur in the tunnel, the risks and hidden dangers in the tunnel are handled collaboratively from multiple dimensions. Specifically, based on the risk parameters of the monitoring range, it is determined whether collaborative management and service operations need to be performed, including: S5.1: The risk status of a risk type whose risk probability is greater than the baseline probability value and less than the first probability value is recorded as low risk; the risk status of a risk type whose risk probability is greater than or equal to the first probability value and less than the second probability value is recorded as medium risk; and the risk status of a risk type whose risk probability is greater than the second probability value is recorded as high risk. The risk probability is compared with the baseline probability value, the first probability value, and the second probability value to determine the risk status of the risk type. The baseline probability value, the first probability value, and the second probability value are critical values ​​used to distinguish between low-risk, medium-risk, and high-risk risk types. The baseline probability value, the first probability value, and the second probability value can be customized according to tunnel operation safety requirements and historical experience. For example, the baseline probability value P0 is 0.1, the first probability value P1 is 0.3, and the second probability value P2 is 0.7.

[0077] Furthermore, to better manage tunnel operations, different values ​​can be configured based on the multi-source data categories (environmental monitoring data, traffic data, equipment status data, and location and structure data) and their severity levels corresponding to different risk types. For example, for risks with severe consequences such as excessive carbon monoxide concentration or lining deformation, lower baseline probability values, first probability values, and second probability values ​​(e.g., P0 = 0.05, P1 = 0.15, P2 = 0.45) should be used to achieve early warning; while for risks with relatively low severity, such as mild congestion, higher thresholds (e.g., P0 = 0.15, P1 = 0.40, P2 = 0.75) can be set to avoid frequent false alarms. Through this differentiated configuration, tunnel operation safety and collaborative management efficiency can be more accurately balanced, making risk classification more closely reflect the actual hazard level.

[0078] In practice, a globally unified threshold (such as P0 = 0.1, P1 = 0.3, and P2 = 0.7) can be used for initial operation. Then, based on historical data and accident records, the probability thresholds for different risk types can be optimized and adjusted through machine learning or expert experience, ultimately forming a set of differentiated threshold configuration tables.

[0079] When the risk probability of a monitoring interval is less than the baseline probability value, there is no potential risk in that monitoring interval, and there is no risk type in that monitoring interval. In this case, the risk type of the monitoring interval is set to the default value (such as 0-0). In specific implementation, when the risk type of the monitoring interval is the default value, no further determination is made on whether to perform collaborative management and service operations for that monitoring interval.

[0080] S5.2: When the number of low-risk risk types in each risk parameter of the monitoring interval is greater than or equal to 3, it is determined that the monitoring interval needs to perform collaborative management and service operations. If the number of risk types classified as low-risk within a certain monitoring interval is greater than or equal to 3, it means that collaborative management and service operations need to be performed on that monitoring interval. When a large number of potential risks, although low in probability, are diverse in type, the combined hidden dangers may accumulate to the point where proactive intervention is required to prevent multiple low-probability events from overlapping and causing accidents.

[0081] S5.3: When the number of medium-risk risk types in each risk parameter of the monitoring interval is greater than or equal to 2, it is determined that the monitoring interval needs to perform collaborative management and service operations. If the number of risk types classified as medium risk within a certain monitoring interval is not less than two, it means that collaborative management and service operations need to be performed on that monitoring interval. Medium risk means that the possibility of risk occurrence can no longer be ignored. When two or more medium risk types exist simultaneously, their interaction may escalate into a high-risk situation, requiring advance intervention.

[0082] S5.4: When the number of high-risk types in each risk parameter of the monitoring interval is greater than or equal to 1, it is determined that the monitoring interval needs to perform collaborative management and service operations. If the number of risk types identified as high-risk within a given monitoring interval is not less than one, it means that collaborative management and service operations need to be performed on that monitoring interval. High risk indicates a high probability of a serious incident, and collaborative management must be initiated immediately.

[0083] The above three judgment conditions are based on OR logic, meaning that as long as any one of them is met, the monitoring interval will trigger the collaborative management requirement.

[0084] S5.5: When the number of high-risk risk types is less than 1, the number of medium-risk risk types is less than 2, and the number of low-risk risk types is less than 3, it is determined that the monitoring interval does not need to perform collaborative management and service operations.

[0085] The collaborative management interval refers to the monitoring interval that requires collaborative management and service operations, and serves as the direct object of this collaborative management and service. Since multiple monitoring intervals within the tunnel are geographically continuous, and multi-source data within adjacent monitoring intervals can also experience spatial and physical diffusion, the collaborative management interval is not only for monitoring intervals with potential risks, but also needs to be diffused in location centered on the monitoring intervals with potential risks, in order to achieve a dynamic correlation effect between multiple collaborative management intervals within the tunnel.

[0086] When one or more monitoring intervals are determined to require collaborative management and service operations (i.e., designated as "risk intervals"), these risk intervals should not be handled in isolation. Instead, adjacent intervals should also be included in the management scope to form a continuous spatial segment with sufficient buffering and collaborative capabilities. Specifically, collaborative management intervals should be determined from all monitoring intervals, including: S5.6: The monitoring intervals that require collaborative management and service operations are designated as risk intervals, the interval numbers of the risk intervals are designated as risk numbers, and the arrangement of risk numbers for all risk intervals is identified; the arrangement includes continuous and intermittent states. First, all monitoring intervals deemed requiring collaborative management and service operations are marked as "risk intervals," and their monitoring interval numbers are extracted as risk numbers. Then, the arrangement of these risk numbers within the overall tunnel interval sequence is analyzed. The arrangement refers to the specific intervals and continuity of the risk numbers arranged in ascending order, thus providing a basis for subsequent identification of collaborative management intervals. A continuous state means all risk interval numbers are consecutive natural numbers, such as {3,4,5}, with no non-risk intervals in between. An intermittent state means the risk interval numbers are not consecutive, with at least one non-risk interval in between, such as {2,5,7}.

[0087] S5.7: When the risk numbers are in a continuous state, the monitoring interval whose interval number is before the minimum value of all risk numbers is recorded as the preceding interval, and the monitoring interval whose interval number is after the maximum value of all risk numbers is recorded as the following interval. The preceding interval, all risk intervals and the following interval are summarized to form a collaborative management interval. When risk numbers are consecutive, it means that risk areas are clustered together. To ensure that management and service operations can cover the risk area and its surrounding impact range, it is necessary to first identify the smallest risk number and take its preceding monitoring interval as the preceding interval (if it exists). Then, identify the largest risk number and take its following monitoring interval as the following interval (if it exists). Finally, merge the preceding interval, all risk intervals, and the following interval to form the final collaborative management interval. This step adds a buffer zone before and after the risk cluster area to prevent risk spread or reserve space for emergency response.

[0088] S5.8: When the risk number is in an interval state, the monitoring interval that is before and after each risk number is recorded as the collaborative interval, and all risk intervals and collaborative intervals are summarized to form a collaborative management interval.

[0089] When risk numbers are in an interleaved state, it indicates that the risk intervals are scattered and separated by non-risk areas. In this case, for each risk interval, it is necessary to first take the monitoring interval before and after it (if any) as a coordination interval. Then, all risk intervals and these coordination intervals are aggregated and merged to obtain the final coordinated management interval. It is important to note that coordination intervals corresponding to different risk intervals may overlap, resulting in several discontinuous coordinated management segments after merging, or a larger set containing all relevant intervals. This step also covers the adjacent intervals around each scattered risk point to address the spread or synergistic effects of local risks, but without forming a continuous large segment as in a continuous state (because the non-risk intervals in between may indeed not require management).

[0090] It should be noted that when the minimum value of the risk number is the first monitoring interval, there is no preceding interval, only the following interval; when the minimum value of the risk number is the last monitoring interval, there is no following interval, only the preceding interval.

[0091] The steps described above for determining the collaborative management area can ensure that the spatial scope of collaborative management can not only fully cover the risk area and its affected neighborhood, but also be expanded at different scales according to the clustering or dispersion of the risk area, thereby facilitating the subsequent accurate and efficient allocation of management resources.

[0092] To better understand the above-mentioned collaborative management interval confirmation process, the following example, a mountain highway tunnel, will be used as an example to explain the determination process of collaborative management intervals in both continuous and intermittent states. As described above, this tunnel is divided into 17 monitoring intervals, numbered sequentially as 1, 2...17.

[0093] Scenario A: After steps S5.2 to S5.5, the risk numbers of the monitoring intervals (i.e., risk intervals) that require collaborative management and service operations have been determined to be 3, 4, and 5. It can be seen that the risk numbers of all these risk intervals are arranged in a continuous state. At this time, the monitoring interval with interval number 2 is recorded as the preceding interval, the monitoring interval with interval number 6 is recorded as the following interval, and finally the monitoring intervals corresponding to interval numbers 2, 3, 4, 5, and 6 of "preceding interval + all risk intervals + following interval" are recorded as the collaborative management intervals.

[0094] Scenario B: After steps S5.2 to S5.5, the risk numbers of the monitoring intervals (i.e., risk intervals) that require collaborative management and service operations have been determined to be 3, 6, and 7. It can be seen that the risk numbers of all these risk intervals are arranged in an intermittent state. At this time, it is necessary to mark the collaborative intervals for each risk number. The interval numbers of the marked collaborative intervals are 2, 4, 5, and 8. Finally, the monitoring intervals corresponding to the interval numbers 2, 3, 4, 5, 6, 7, and 8 of "all risk intervals + all collaborative intervals" are recorded as collaborative management intervals.

[0095] Case C: After steps S5.2 to S5.5, the risk numbers of the monitoring intervals (i.e., risk intervals) that need to perform collaborative management and service operations have been determined to be 1, 2, and 3. It can be seen that the risk numbers of all these risk intervals are arranged in a continuous state. At this time, there is no preceding interval. The monitoring interval with interval number 4 is recorded as the following interval. Finally, the monitoring intervals corresponding to interval numbers 1, 2, 3, and 4 of "all risk intervals + following intervals" are recorded as collaborative management intervals.

[0096] Case D: After steps S5.2 to S5.5, the risk numbers of the monitoring intervals (i.e., risk intervals) that require collaborative management and service operations have been determined to be 15, 16, and 17. It can be seen that the risk numbers of all these risk intervals are arranged in a continuous state. At this time, there is no subsequent interval. The monitoring interval with interval number 14 is recorded as the preceding interval. Finally, the monitoring intervals corresponding to interval numbers 14, 15, 16, and 17 of "preceding interval + all risk intervals" are recorded as collaborative management intervals.

[0097] After identifying the collaborative management area, corresponding management service information can be generated within the management service platform based on the specific circumstances and data information of the collaborative management area. This management service information can serve as a theoretical basis for sending to the tunnel operation management department and help the tunnel operation management department to formulate targeted control measures in advance. This will reduce or avoid potential risks and hazards in the tunnel in the future, improve the safety of driving in the tunnel, and ultimately achieve the collaborative management and service effect of multi-source data.

[0098] In this embodiment, the management service information is generated within the management service platform and serves as the basis for the tunnel operation management department to formulate specific measures. Therefore, the management service information needs to include multi-dimensional data of the collaborative management area. Specifically, the management service information includes predicted multi-source data, risk type, risk probability, and area number.

[0099] It should be noted that since the number of collaborative management intervals is not fixed, the number of specific information in the corresponding management service information is also not fixed; among them, the number of interval numbers is equal to the number of collaborative management intervals, and the number of risk types is equal to the number of risk probabilities; thus providing multi-source and accurate data support for the subsequent specific measures of the tunnel operation management department, and realizing the collaborative management and service effect of multi-source data.

[0100] Example 2: Please see Figure 2 As shown, the parts not described in detail in this embodiment are described in Embodiment 1. A multi-source data collaborative management and service system based on the Internet of Things is provided and applied to a management service platform to implement the multi-source data collaborative management and service method based on the Internet of Things described in Embodiment 1. The system includes a data acquisition module, a model building module, a data simulation module, a parameter identification module, and an interval information module, wherein the modules are connected to each other through wired or wireless networks. The data acquisition module is configured to divide the tunnel into several monitoring sections arranged in a continuous sequence based on a dual division criterion, and to collect multi-source data of each monitoring section in real time through IoT terminal devices. The model building module is configured to build a three-dimensional tunnel model that matches the physical space of the tunnel. The three-dimensional tunnel model is spatially divided according to the actual spatial boundaries of each monitoring interval to obtain multiple three-dimensional interval models that correspond one-to-one with the monitoring intervals. The module also establishes a dynamic mapping between each three-dimensional interval model and the multi-source data of the corresponding monitoring interval to generate a digital interval model. The data simulation module configures the simulation period and simulation mechanism for pre-configuring the digital interval model. Based on the configured simulation period and simulation mechanism, it simulates and obtains the predicted multi-source data of each digital interval model in the next simulation period. The parameter identification module is configured to pre-build a risk identification model for risk assessment, and input the predicted multi-source data into the risk identification model to obtain the risk parameters for the corresponding monitoring interval. The interval information module is configured to determine whether collaborative management and service operations need to be performed on the monitored interval based on the risk parameters of the monitored interval; if so, it determines the collaborative management interval from all monitored intervals and generates management service information adapted to the collaborative management interval.

[0101] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for multi-source data collaborative management and service based on the Internet of Things, characterized in that, include: S1: Based on the dual division criterion, the tunnel is divided into several monitoring sections arranged in sequence, and multi-source data of each monitoring section is collected in real time through IoT terminal devices. The multi-source data includes environmental monitoring data, traffic road data, equipment status data, and location structure data; the dual division criterion is used to limit the number and length of the monitoring sections. S2: Construct a three-dimensional tunnel model that matches the physical space of the tunnel. Divide the three-dimensional tunnel model into spatial segments according to the actual spatial boundaries of each monitoring segment to obtain multiple three-dimensional interval models that correspond one-to-one with the monitoring segments. Establish a dynamic mapping between each three-dimensional interval model and the multi-source data of the corresponding monitoring segment to generate a digital interval model. S3: Pre-configure the simulation period and simulation mechanism of the digital interval model. Based on the configured simulation period and simulation mechanism, simulate and obtain the predicted multi-source data of each digital interval model in the next simulation period. S4: Pre-construct a risk identification model for risk assessment, input the predicted multi-source data into the risk identification model, and obtain the risk parameters for the corresponding monitoring interval. The risk parameters include risk type and risk probability. S5: Based on the risk parameters of the monitoring interval, determine whether collaborative management and service operations need to be performed; if so, determine the collaborative management interval from all monitoring intervals and generate management service information adapted to the collaborative management interval. The management service information is used to guide the collaborative management and service operations of the collaborative management interval.

2. The method for multi-source data collaborative management and service based on the Internet of Things according to claim 1, characterized in that, The dual division criterion is as follows: the number of monitoring intervals is not less than 3, and the length of the monitoring intervals at both ends is greater than the length of the monitoring interval in the middle.

3. The method for multi-source data collaborative management and service based on the Internet of Things according to claim 2, characterized in that, Based on the dual-division criterion, the tunnel is divided into several monitoring intervals arranged sequentially, including: S1.1: Query the maximum width, number of lanes, and length of the tunnel. Multiply the maximum width by the number of lanes and then multiply by the first coefficient to obtain the base length. Divide the tunnel length by the base length to calculate the division factor. S1.2: When the division multiple is less than the first set value, the base length is continuously reduced in increments of 5% of the base length until the division multiple is greater than or equal to the first set value, and then S1.3 or S1.4 is executed. S1.3: When the division factor equals the first set value, the tunnel is divided into two monitoring intervals with the reference length as the middle position and 110% of the reference length as the lengths of the two monitoring intervals at the two ends. A continuous monitoring interval; S1.4: When the division factor is greater than the first set value, subtract the second set value from the division factor to obtain the intermediate value. Use the integer part of the intermediate value as the number of monitoring intervals at the intermediate position. Use the reference length as the length of the monitoring interval at the intermediate position, and use half the remaining tunnel length as the length of the monitoring intervals at both ends, thus dividing the tunnel into... A continuous monitoring interval; S1.5: Based on the direction of vehicle travel within the tunnel, number the first section as 1. The monitoring intervals are numbered in ascending order.

4. The method for multi-source data collaborative management and service based on the Internet of Things according to claim 3, characterized in that, A three-dimensional tunnel model matching the physical space of the tunnel is constructed. The three-dimensional tunnel model is then spatially divided according to the actual spatial boundaries of each monitoring section, resulting in multiple three-dimensional interval models, each corresponding one-to-one with the monitoring intervals. These include: S2.1: Use laser scanning equipment to collect point cloud data inside the tunnel in real time. After noise reduction, registration and spatial alignment of the point cloud data, construct a three-dimensional tunnel model that matches the actual tunnel space through three-dimensional modeling technology. S2.2: Will The length of each monitoring interval is divided by the tunnel length to obtain... The percentage of the length of each monitoring interval; S2.3: According to the numbering sequence of the monitoring sections, mark them one by one in the three-dimensional tunnel model along the vehicle's direction of travel. The model boundary points corresponding to the length ratio of each monitoring interval are then connected sequentially to obtain the dividing surface; S2.4: Using each cutting plane as the cutting point, divide the 3D tunnel model into sections. A three-dimensional interval model.

5. The method for multi-source data collaborative management and service based on the Internet of Things according to claim 4, characterized in that, A dynamic mapping is established between each three-dimensional interval model and the multi-source data of the corresponding monitoring interval to generate a digital interval model, including: S2.5: Establish There are four blank datasets, each containing four mapping units. Environmental monitoring data, traffic road data, equipment status data, and location structure data of each monitoring interval are imported into the corresponding four mapping units to convert the blank datasets into mapped datasets. S2.6: Using digital twin technology, a one-way transmission mapping channel is established between each three-dimensional interval model and its corresponding mapping dataset. The mapping channel is divided into four parallel sub-channels, and the input port of each sub-channel is connected to the corresponding mapping unit, and the output port is connected to the corresponding three-dimensional interval model. S2.7: Set a mapping clock on each of the four sub-channels belonging to the same mapping channel, and adjust the four mapping clocks to keep them synchronized in time; S2.8: Will Dynamically map multi-source data within a mapping dataset to corresponding... A digital interval model is generated by overlaying multi-source data on a three-dimensional interval model along the time axis.

6. The method for multi-source data collaborative management and service based on the Internet of Things according to claim 5, characterized in that, The steps for configuring the simulation time period for the digital interval model include: S3.1: The smallest detectable change that occurs in each of the environmental monitoring data, traffic road data, equipment status data, and location structure data is obtained, and is denoted as the unit amplitude. S3.2: Record the data that changes by a unit amplitude as the target data. Divide the number of target data by the number of environmental monitoring data, traffic road data, equipment status data and location structure data respectively, and calculate the environmental proportion, traffic proportion, equipment proportion and location proportion. S3.3: Query the duration corresponding to the environmental percentage from 0 to 0.6, the traffic percentage from 0 to 0.5, the equipment percentage from 0 to 0.4, and the location percentage from 0 to 0.5, and record them as the first duration, the second duration, the third duration, and the fourth duration, respectively; S3.4: Based on the proportional relationship between the environmental proportion, traffic proportion, equipment proportion and location proportion, the proportional coefficients of the first duration, second duration, third duration and fourth duration are set in reverse, and the first duration, second duration, third duration and fourth duration are weighted and summed to calculate the comprehensive duration; S3.5: Starting from the current moment, and using a comprehensive duration extending backward from the starting point as the standard, set the simulation time period.

7. A method for multi-source data collaborative management and service based on the Internet of Things according to claim 6, characterized in that, The simulation mechanism of the digital interval model is as follows: simulation is performed using multi-source data from four consecutive simulation periods in the past.

8. The method for multi-source data collaborative management and service based on the Internet of Things according to claim 7, characterized in that, Based on the risk parameters of the monitoring range, determine whether collaborative management and service operations are required, including: S5.1: The risk status of a risk type whose risk probability is greater than the baseline probability value and less than the first probability value is recorded as low risk; the risk status of a risk type whose risk probability is greater than or equal to the first probability value and less than the second probability value is recorded as medium risk; and the risk status of a risk type whose risk probability is greater than the second probability value is recorded as high risk. S5.2: When the number of low-risk risk types in each risk parameter of the monitoring interval is greater than or equal to 3, it is determined that the monitoring interval needs to perform collaborative management and service operations. S5.3: When the number of medium-risk risk types in each risk parameter of the monitoring interval is greater than or equal to 2, it is determined that the monitoring interval needs to perform collaborative management and service operations. S5.4: When the number of high-risk types in each risk parameter of the monitoring interval is greater than or equal to 1, it is determined that the monitoring interval needs to perform collaborative management and service operations.

9. A method for multi-source data collaborative management and service based on the Internet of Things according to claim 8, characterized in that, The collaborative management intervals are determined from all monitoring intervals, including: S5.6: The monitoring intervals that require collaborative management and service operations are designated as risk intervals, the interval numbers of the risk intervals are designated as risk numbers, and the arrangement of risk numbers for all risk intervals is identified; the arrangement includes continuous and intermittent states. S5.7: When the risk numbers are in a continuous state, the monitoring interval whose interval number is before the minimum value of all risk numbers is recorded as the preceding interval, and the monitoring interval whose interval number is after the maximum value of all risk numbers is recorded as the following interval. The preceding interval, all risk intervals and the following interval are summarized to form a collaborative management interval. S5.8: When the risk number is in an interval state, the monitoring interval that is before and after each risk number is recorded as the collaborative interval, and all risk intervals and collaborative intervals are summarized to form a collaborative management interval.

10. A method for multi-source data collaborative management and service based on the Internet of Things according to claim 9, characterized in that, The management service information includes forecast multi-source data, risk type, risk probability, and interval number.

Images