A tunnel risk prediction method, device and computer equipment based on multi-sensing approach

Abstract

The present application relates to a tunnel risk prediction method, device and computer equipment with a multi-sensory approach. The method comprises: collecting physical feature data of each segmented area of ​​the tunnel, generating risk prediction data of each area using a corresponding risk prediction model, and splicing these data according to time and space dimensions to form a global time and global space risk view of the tunnel. The global spliced ​​time and space data are respectively input into the global time and space risk prediction model to obtain the overall time and space risk prediction data. By combining the global time and space risk prediction data, the overall risk of the tunnel is comprehensively predicted to generate risk prediction data for the target tunnel. The use of this method can effectively improve the accuracy and timeliness of risk prediction for the tunnel.

Description

Technical Field

[0001] The present application relates to the technical field of risk prediction, and in particular to a tunnel risk prediction method, device and computer equipment based on a multi-sensory approach. Background Art

[0002] In the process of tunnel construction and operation, risk factors are complex and diverse, and traditional tunnel risk prediction methods often rely on a single monitoring method, such as geological exploration, on-site testing, historical experience, etc. However, these single-channel data are easily limited by factors such as insufficient data collection, complex environmental conditions, and delayed warning in practical applications, resulting in low accuracy and timeliness of tunnel risk prediction. Summary of the invention

[0003] Based on this, it is necessary to provide a tunnel risk prediction method, device, computer equipment, computer-readable storage medium and computer program product with a multi-sensory approach that can effectively improve the accuracy and timeliness of tunnel risk prediction in response to the above-mentioned technical problems.

[0004] In a first aspect, the present application provides a tunnel risk prediction method based on a multi-sensory approach, comprising:

[0005] Obtaining tunnel physical characteristic data corresponding to each segmented area of ​​the target tunnel, and obtaining a tunnel risk prediction model corresponding to each tunnel physical characteristic data;

[0006] Inputting the physical characteristic data of each tunnel into the corresponding tunnel risk prediction model respectively to obtain the tunnel risk prediction data corresponding to each segmented area;

[0007] According to the tunnel time dimension and the tunnel space dimension, the tunnel risk prediction data are spliced ​​to obtain time global tunnel risk splicing data and space global tunnel risk splicing data;

[0008] Inputting the temporal global tunnel risk splicing data into the global temporal risk prediction model of the target tunnel to obtain global temporal risk prediction data;

[0009] Inputting the spatial global tunnel risk splicing data into the global spatial risk prediction model of the target tunnel to obtain global spatial risk prediction data;

[0010] The tunnel risk of the target tunnel is predicted by combining the global time risk prediction data and the global space risk prediction data to obtain target tunnel risk prediction data.

[0011] In a second aspect, the present application also provides a tunnel risk prediction device with multiple sensing paths, comprising:

[0012] A tunnel data acquisition module is used to acquire tunnel physical characteristic data corresponding to each segmented area of ​​the target tunnel, and to acquire a tunnel risk prediction model corresponding to each tunnel physical characteristic data;

[0013] A tunnel risk prediction module, used to input the physical characteristic data of each tunnel into the corresponding tunnel risk prediction model to obtain the tunnel risk prediction data corresponding to each segmented area;

[0014] A risk data splicing module is used to splice the risk prediction data of each tunnel according to the tunnel time dimension and the tunnel space dimension to obtain time global tunnel risk splicing data and space global tunnel risk splicing data;

[0015] The tunnel risk prediction module is further used to input the temporal global tunnel risk splicing data into the global temporal risk prediction model of the target tunnel to obtain the global temporal risk prediction data;

[0016] The tunnel risk prediction module is further used to input the spatial global tunnel risk splicing data into the global spatial risk prediction model of the target tunnel to obtain the global spatial risk prediction data;

[0017] The tunnel risk prediction module is also used to predict the tunnel risk of the target tunnel in combination with the global time risk prediction data and the global space risk prediction data to obtain target tunnel risk prediction data.

[0018] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0019] Obtaining tunnel physical characteristic data corresponding to each segmented area of ​​the target tunnel, and obtaining a tunnel risk prediction model corresponding to each tunnel physical characteristic data;

[0020] Inputting the physical characteristic data of each tunnel into the corresponding tunnel risk prediction model respectively to obtain the tunnel risk prediction data corresponding to each segmented area;

[0021] According to the tunnel time dimension and the tunnel space dimension, the tunnel risk prediction data are spliced ​​to obtain time global tunnel risk splicing data and space global tunnel risk splicing data;

[0022] Inputting the temporal global tunnel risk splicing data into the global temporal risk prediction model of the target tunnel to obtain global temporal risk prediction data;

[0023] Inputting the spatial global tunnel risk splicing data into the global spatial risk prediction model of the target tunnel to obtain global spatial risk prediction data;

[0024] The tunnel risk of the target tunnel is predicted by combining the global time risk prediction data and the global space risk prediction data to obtain target tunnel risk prediction data.

[0025] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0026] Obtaining tunnel physical characteristic data corresponding to each segmented area of ​​the target tunnel, and obtaining a tunnel risk prediction model corresponding to each tunnel physical characteristic data;

[0027] Inputting the physical characteristic data of each tunnel into the corresponding tunnel risk prediction model respectively to obtain the tunnel risk prediction data corresponding to each segmented area;

[0028] According to the tunnel time dimension and the tunnel space dimension, the tunnel risk prediction data are spliced ​​to obtain time global tunnel risk splicing data and space global tunnel risk splicing data;

[0029] Inputting the temporal global tunnel risk splicing data into the global temporal risk prediction model of the target tunnel to obtain global temporal risk prediction data;

[0030] Inputting the spatial global tunnel risk splicing data into the global spatial risk prediction model of the target tunnel to obtain global spatial risk prediction data;

[0031] The tunnel risk of the target tunnel is predicted by combining the global time risk prediction data and the global space risk prediction data to obtain target tunnel risk prediction data.

[0032] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0033] Obtaining tunnel physical characteristic data corresponding to each segmented area of ​​the target tunnel, and obtaining a tunnel risk prediction model corresponding to each tunnel physical characteristic data;

[0034] Inputting the physical characteristic data of each tunnel into the corresponding tunnel risk prediction model respectively to obtain the tunnel risk prediction data corresponding to each segmented area;

[0035] According to the tunnel time dimension and the tunnel space dimension, the tunnel risk prediction data are spliced ​​to obtain time global tunnel risk splicing data and space global tunnel risk splicing data;

[0036] Inputting the temporal global tunnel risk splicing data into the global temporal risk prediction model of the target tunnel to obtain global temporal risk prediction data;

[0037] Inputting the spatial global tunnel risk splicing data into the global spatial risk prediction model of the target tunnel to obtain global spatial risk prediction data;

[0038] The tunnel risk of the target tunnel is predicted by combining the global time risk prediction data and the global space risk prediction data to obtain target tunnel risk prediction data.

[0039] The above-mentioned multi-sensory tunnel risk prediction method, device, computer equipment, storage medium and computer program product obtain the tunnel physical characteristic data corresponding to each segmented area of ​​the target tunnel, and obtain the tunnel risk prediction model corresponding to each tunnel physical characteristic data; input the physical characteristic data of each tunnel into the corresponding tunnel risk prediction model respectively, and obtain the tunnel risk prediction data corresponding to each segmented area; according to the tunnel time dimension and the tunnel space dimension, the data of each tunnel risk prediction data is spliced ​​to obtain the time global tunnel risk splicing data and the space global tunnel risk splicing data; the time global tunnel risk splicing data is input into the global time risk prediction model of the target tunnel to obtain the global time risk prediction data; the space global tunnel risk splicing data is input into the global space risk prediction model of the target tunnel to obtain the global space risk prediction data; combined with the global time risk prediction data and the global space risk prediction data, the tunnel risk of the target tunnel is predicted to obtain the target tunnel risk prediction data.

[0040] By dividing the tunnel into multiple areas, carefully collecting the physical characteristic data of each area and inputting it into the corresponding risk prediction model, the risk prediction data of each area can be obtained. Then, according to the global splicing processing of time and space dimensions, the risk prediction data of each area are integrated into the overall time risk and space risk data, so as to make a multi-dimensional and multi-scale accurate prediction of the overall risk of the tunnel. It can not only effectively improve the accuracy and timeliness of the risk prediction of the tunnel, but also timely capture the potential safety hazards under specific time and space conditions. In addition, through the joint analysis of the global time and space model, the risk evolution trend of the tunnel in the long-term operation can be fully grasped, which provides strong support for the forward-looking management and preventive maintenance of the tunnel, thereby further improving the safety and reliability of the tunnel operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 A diagram showing an application environment of a tunnel risk prediction method using multiple sensing approaches in one embodiment;

[0043] Figure 2 A schematic diagram of a flow chart of a tunnel risk prediction method using multiple sensing approaches in one embodiment;

[0044] Figure 3 A schematic flow chart of a method for obtaining target tunnel risk prediction data in one embodiment;

[0045] Figure 4 A schematic diagram of a flow chart of a method for obtaining information on interactive impact of change data in one embodiment;

[0046] Figure 5 It is a flowchart of a method for calculating risk value interaction information and risk change rate interaction information in one embodiment;

[0047] Figure 6 It is a flowchart of a method for obtaining temporal global tunnel risk splicing data and spatial global tunnel risk splicing data in one embodiment;

[0048] Figure 7 It is a flowchart of a method for obtaining temporal global tunnel risk splicing data and spatial global tunnel risk splicing data in another embodiment;

[0049] Figure 8 A flowchart of a method for obtaining abnormal repair information and comprehensive traffic dispatch information in a high-risk area in one embodiment;

[0050] Fig. 9 It is a structural block diagram of a tunnel risk prediction device with multiple sensing paths in one embodiment;

[0051] Fig.10 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0053] The tunnel risk prediction method of a multi-sensing approach provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The server 104 obtains the tunnel physical feature data corresponding to each segmented area of ​​the target tunnel through the terminal 102, and obtains the tunnel risk prediction model corresponding to each tunnel physical feature data; inputs the physical feature data of each tunnel into the corresponding tunnel risk prediction model respectively, and obtains the tunnel risk prediction data corresponding to each segmented area; according to the tunnel time dimension and the tunnel space dimension, the data of each tunnel risk prediction data is spliced ​​to obtain the time global tunnel risk splicing data and the space global tunnel risk splicing data; the time global tunnel risk splicing data is input into the global time risk prediction model of the target tunnel to obtain the global time risk prediction data; the space global tunnel risk splicing data is input into the global space risk prediction model of the target tunnel to obtain the global space risk prediction data; combined with the global time risk prediction data and the global space risk prediction data, the tunnel risk of the target tunnel is predicted to obtain the target tunnel risk prediction data. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0054] In an exemplary embodiment, Figure 2 As shown in the figure, a tunnel risk prediction method with multiple sensing approaches is provided. Figure 1 The server in the example is used to illustrate, including the following steps 202 to 212. Among them:

[0055] Step 202, obtaining tunnel physical characteristic data corresponding to each segmented area of ​​the target tunnel, and obtaining a tunnel risk prediction model corresponding to each tunnel physical characteristic data.

[0056] The segmented area can be to divide the entire tunnel into several areas, each area covering a certain length of the tunnel section. These area divisions can be based on factors such as the physical characteristics of the tunnel, construction structure, geological environment, etc., so as to monitor physical data and conduct risk assessment in each independent area.

[0057] Among them, tunnel physical characteristic data can be collected by sensors or monitoring equipment, which can reflect various key data information of tunnel environment and structural status. These data include structural vibration, humidity, temperature, stress, displacement, crack width and other parameters, which can provide basic input data for tunnel risk prediction.

[0058] The tunnel risk prediction model can be a trained or customized mathematical model or algorithm based on tunnel characteristics and historical data. It can perform calculations and analysis based on the input tunnel physical characteristic data and output the risk status of the tunnel.

[0059] Specifically, monitoring equipment is set up in the target tunnel, which is divided into several segments (such as the entrance section, the middle section and the exit section) to obtain the tunnel physical characteristic data of different segments, including structural vibration, temperature, humidity, displacement, stress, etc., which can be collected in real time through the sensor network. Based on the environmental characteristics and data types of each segmented area, the corresponding tunnel risk prediction model is selected from the preset model library to ensure that the selected model can accurately reflect the characteristics of the area.

[0060] Step 204, input the physical characteristic data of each tunnel into the corresponding tunnel risk prediction model to obtain the tunnel risk prediction data corresponding to each segmented area.

[0061] Among them, the tunnel risk prediction data can be the result data processed by the risk prediction model, showing the risk status of each area in the tunnel, including the risk level (calculated based on the risk prediction data of each segmented area, and a score between 0 and 100 is obtained, or expressed in a level of 1 to 5. For a certain tunnel, the risk level is 67, indicating that the tunnel is of medium to high risk under the current conditions and may face certain safety hazards), possible damage types (based on the physical characteristics and historical data of the tunnel, predict the possible damage types, and estimate the probability of occurrence of each type and its specific impact. The possible damage types of the tunnel Damage types include cracks (40% probability, expected crack length of 10 meters), water seepage (30% probability, expected water seepage area of ​​15 square meters) and structural failure (20% probability, failure degree of 10%). These data help determine potential risk factors and damage modes), occurrence time (the risk prediction model can predict the occurrence time of specific damage types. For example, cracks may occur on June 15, 2025, while water seepage is expected to occur within 2025 and structural failure may occur within the next year. Accurate prediction of time provides tunnel managers with a basis for taking repair and preventive measures in advance).

[0062] Specifically, the collected physical feature data of each segmented area are respectively input into the corresponding tunnel risk prediction model. After receiving the input data, the tunnel risk prediction model will process the data in each area based on its training or configured parameters, and output the tunnel risk prediction data of the segmented area. Tunnel risk prediction data usually includes the structural integrity score in the area (the structural health of the tunnel is quantified by a score, usually a value between 0 and 1, indicating the integrity of the tunnel. The structural integrity score of this tunnel is 0.80, indicating that its structure is generally intact, but there is some damage and it needs regular inspection and maintenance), potential damage risk level (the potential damage risk level is obtained through a comprehensive analysis of the tunnel's physical characteristics, historical data and environmental impacts. For example, for this tunnel, the potential damage risk level is 75, indicating that the tunnel faces a high risk of damage and requires high-priority prevention and repair measures.), environmental impact factors (the impact of environmental factors on tunnel risk is quantified through a series of data, such as groundwater level, temperature and humidity, and seismic activity. The environmental impact factor score of this tunnel is 7, indicating that the environmental conditions in the area have a great impact on tunnel risk. The groundwater level is 1.5 meters, the temperature and humidity are high, and there are several seismic activities every year. These factors may cause accelerated aging or damage to the structure), etc., to identify the degree of safety risk in the area.

[0063] For example, suppose the risk assessment results of a tunnel are as follows. Risk level: 67 (medium to high risk). Possible damage types: cracks, 40% probability, expected crack length 10 meters; water seepage, 30% probability, water seepage area 15 square meters; structural failure, 20% probability, failure degree 10%. Occurrence time: cracks, expected on June 15, 2025; water seepage, expected within 2025, structural failure, expected within the next year. Structural integrity score, 0.80 (relatively healthy); potential damage risk level: 75; environmental impact factor, groundwater level, 1.5 meters; temperature and humidity: 10°C to 30°C, humidity 80%; seismic activity, 3 small earthquakes per year; soil type, clay; environmental impact factor score, 7.

[0064] Step 206, splicing the risk prediction data of each tunnel according to the tunnel time dimension and the tunnel space dimension to obtain time global tunnel risk splicing data and space global tunnel risk splicing data.

[0065] Among them, the tunnel time dimension can be a time-varying factor considered in risk prediction, which is used to analyze and track the evolution of risks over time. Through the time dimension, the state changes of the tunnel in different time periods can be observed, thereby identifying potential time-sensitive risks.

[0066] Among them, the tunnel spatial dimension can be a spatial distribution factor considered in risk analysis, which is used to show the geographical distribution of risks in the tunnel structure. Through the analysis of spatial dimensions, we can intuitively understand which areas in the tunnel are more likely to generate risks and help identify specific spatial risk hotspots.

[0067] Among them, the time-global tunnel risk splicing data can be a set of predicted data that integrates the risks of various areas of the tunnel based on the time dimension, showing the continuous change trend of tunnel risk over time.

[0068] Among them, the spatial global tunnel risk splicing data can be a collection of risk prediction data from various regions integrated in the spatial dimension, showing the risk distribution of the tunnel at each spatial location.

[0069] Specifically, the tunnel risk prediction data of each segmented area is spliced ​​in the time and space dimensions. In the time dimension, the tunnel risk prediction data of each segmented area is arranged in chronological order to generate a set of time-continuous time-global tunnel risk splicing data, which reflects the evolution of the risk in the tunnel over time; in the spatial dimension, the data of each segmented area are combined according to the spatial distribution to form a set of spatial global tunnel risk splicing data to visualize the spatial risk distribution of the tunnel from the entrance to the exit. In this way, the global risk data obtained through the spatiotemporal splicing process not only retains the details of the regional data, but also provides support for the temporal and spatial change trend of the overall risk.

[0070] Step 208, input the time global tunnel risk splicing data into the global time risk prediction model of the target tunnel to obtain the global time risk prediction data.

[0071] Among them, the global time risk prediction model can be a prediction model that analyzes the changes in tunnel risk in the time dimension. It receives the time global tunnel risk splicing data as input, and predicts the risk status in the future period by analyzing the risk evolution trend in the time series.

[0072] The core task of the global time risk prediction model is to analyze the evolution of tunnel risk from the time dimension. The model receives the time global risk splicing data from each segmented area of ​​the tunnel, which contains the quantitative data of the tunnel risk status at different time points. By analyzing these time series data, the model can identify potential time trends, such as periodic risks, periods of sudden events, and risk peaks during construction or use. Specifically, the model uses statistical learning methods (such as time series analysis, regression analysis, or deep learning models) to capture temporal dependencies, evaluate risk patterns in past time periods, and predict future risks based on these historical data.

[0073] Among them, the global time risk prediction data can be the result calculated by the global time risk prediction model, which describes the risk trend of the tunnel in the future time dimension.

[0074] Specifically, the time-global tunnel risk splicing data is input into the global time risk prediction model, and the risk trend analysis in the time dimension is performed through the global time risk prediction model. The global time risk prediction model predicts the time-sensitive risks that may occur in the future according to the risk change trend in different time periods in the tunnel, and obtains global time risk prediction data, such as the risk peak caused by changes in environmental factors in certain time periods. This global time risk prediction model can effectively identify the time points when the risks will intensify in the future, so as to facilitate early warning and formulate corresponding prevention and control measures to ensure the safety of the tunnel in these critical periods. For the global time risk prediction data, the output of the global time risk prediction model is usually a risk index or risk curve that changes with time, indicating the possibility of accidents or other safety problems in the tunnel at different time points. These outputs can be displayed in the form of time series, such as "the risk of a major accident in the tunnel is 5% in the next 12 months", and this value may change over time.

[0075] Step 210, input the spatial global tunnel risk splicing data into the global spatial risk prediction model of the target tunnel to obtain the global spatial risk prediction data.

[0076] Among them, the global spatial risk prediction model can be a risk prediction model focusing on the spatial dimension, which is mainly used to analyze the overall risk distribution of different spatial areas of the tunnel. It receives spatial global tunnel risk splicing data and evaluates the risk level and changes at each location of the tunnel.

[0077] The goal of the global spatial risk prediction model is to comprehensively evaluate the risk characteristics of each area of ​​the tunnel from a spatial dimension and consider the impact of spatial factors on the overall tunnel risk. The model receives spatial global risk splicing data from different tunnel areas, which covers different physical characteristics of the tunnel in spatial distribution (such as structure, geology, traffic load, environmental conditions and other quantitative data). The model uses spatial analysis methods (such as spatial regression, geographic information system (GIS) analysis, machine learning algorithms, etc.) to model the risks of different areas and identify the risk correlation between areas. For example, some areas may have a higher risk of landslides due to poor geological conditions, while other areas may have a higher risk of collisions due to high traffic density. The global spatial risk prediction model generates a global risk assessment report by integrating this spatial information.

[0078] Among them, the global spatial risk prediction data can be the analysis results generated by the global spatial risk prediction model, reflecting the risk status of the tunnel in the spatial dimension.

[0079] Specifically, the spatial global tunnel risk splicing data is input into the global spatial risk prediction model, and the overall risk analysis in the spatial dimension is performed through the global spatial risk prediction model. The global spatial risk prediction model generates a risk distribution map of the entire tunnel space based on the risk data of different spatial locations of the tunnel, helps identify the spatial risk hotspots in the tunnel, and generates global spatial risk prediction data. This analysis can reveal potential structural weaknesses or specific areas that are susceptible to environmental influences in the tunnel, providing a reference for the reinforcement and maintenance of the tunnel's physical structure.

[0080] For global spatial risk prediction data, the output of the global spatial risk prediction model is usually expressed as a spatial risk distribution map or heat map, showing the risk level of each location in the tunnel. Each spatial area will have a risk score or index, which can be assigned based on factors such as the structural characteristics, geological conditions, and equipment layout of the tunnel. The output data may be a two-dimensional or three-dimensional map, and the color or shade of each area in the map represents the risk level of the area. For example, the middle section of the tunnel may have a higher risk due to complex geological conditions, while the entrance location may face higher safety hazards due to heavy traffic.

[0081] Step 212, combining the global time risk prediction data and the global space risk prediction data, predicting the tunnel risk of the target tunnel, and obtaining target tunnel risk prediction data.

[0082] Among them, the target tunnel risk prediction data can be the risk prediction result for the entire tunnel, which combines the global time and space risk prediction data to provide a comprehensive risk overview.

[0083] Specifically, on the basis of completing the time and space risk prediction, the global time risk prediction data is combined with the global space risk prediction data, and the risk trend in the time dimension and the risk hotspot in the space dimension are integrated for comprehensive analysis to generate the target tunnel risk prediction data of the target tunnel, forming a more comprehensive tunnel risk prediction result. This step ensures that the risk prediction results have continuity and accuracy in both time and space, thereby providing strong data support for the long-term safety management of the tunnel, the formulation of maintenance strategies, and emergency plans for sudden risks.

[0084] In the above-mentioned tunnel risk prediction method of multi-sensing approach, tunnel physical feature data corresponding to each segmented area of ​​the target tunnel is obtained, and a tunnel risk prediction model corresponding to each tunnel physical feature data is obtained; the physical feature data of each tunnel is respectively input into the corresponding tunnel risk prediction model to obtain the tunnel risk prediction data corresponding to each segmented area; the risk prediction data of each tunnel is spliced ​​according to the tunnel time dimension and the tunnel space dimension to obtain time global tunnel risk splicing data and space global tunnel risk splicing data; the time global tunnel risk splicing data is input into the global time risk prediction model of the target tunnel to obtain global time risk prediction data; the space global tunnel risk splicing data is input into the global space risk prediction model of the target tunnel to obtain global space risk prediction data; the tunnel risk of the target tunnel is predicted by combining the global time risk prediction data and the global space risk prediction data to obtain the target tunnel risk prediction data.

[0085] By dividing the tunnel into multiple areas, carefully collecting the physical characteristic data of each area and inputting it into the corresponding risk prediction model, the risk prediction data of each area can be obtained. Then, according to the global splicing processing of time and space dimensions, the risk prediction data of each area are integrated into the overall time risk and space risk data, so as to make a multi-dimensional and multi-scale accurate prediction of the overall risk of the tunnel. It can not only effectively improve the accuracy and timeliness of the risk prediction of the tunnel, but also timely capture the potential safety hazards under specific time and space conditions. In addition, through the joint analysis of the global time and space model, the risk evolution trend of the tunnel in the long-term operation can be fully grasped, which provides strong support for the forward-looking management and preventive maintenance of the tunnel, thereby further improving the safety and reliability of the tunnel operation.

[0086] In an exemplary embodiment, Figure 3 As shown, combining the global time risk prediction data and the global space risk prediction data, predicting the tunnel risk of the target tunnel, and obtaining the target tunnel risk prediction data, includes steps 302 to 306. Among them:

[0087] Step 302 , respectively performing risk characteristic analysis on the global time risk prediction data and the global space risk prediction data to obtain tunnel risk time variation data and tunnel risk space variation data.

[0088] Among them, the tunnel risk time variation data can be information reflecting the changes in the risk level of the tunnel in different time periods, revealing the trend of risk growth or fluctuation over time.

[0089] Among them, the spatial variation data of tunnel risk can be information reflecting the risk distribution in different spatial locations within the tunnel and the risk differences between regions.

[0090] Specifically, the global time risk prediction data is analyzed to extract the characteristics that reflect the changes in risk over time, including the trend of risk increase and decrease, the time periods of peak and mutation, the periodic fluctuation pattern and other information, so as to generate tunnel risk time change data. This step helps to identify the changes in the risk level of the tunnel in different time periods. Next, the global spatial risk prediction data is analyzed to extract the risk distribution characteristics of the tunnel in the spatial dimension, such as high-risk areas, differences in risk levels at different locations, and the distribution of spatial hotspots, to generate tunnel risk spatial change data. These temporal and spatial characteristic data reveal in detail the temporal and spatial changes in tunnel risk, providing a sufficient basis for more in-depth correlation analysis.

[0091] Step 304: Analyze the correlation between the tunnel risk time variation data and the tunnel risk space variation data to obtain the variation data interaction impact information.

[0092] The interactive impact information of the changing data can be the correlation and interaction between the time-varying data and the space-varying data, which reveals the comprehensive impact of time and space factors on tunnel risk. For example, through analysis, it can be found that the risk of certain high-risk areas in a specific time period is particularly significant, or the time trend of the risk is closely related to the characteristics of a specific spatial area.

[0093] Specifically, the tunnel risk time variation data and the tunnel risk space variation data are interactively analyzed to identify the correlation between time and space factors. Specifically, it is analyzed whether the time variation of risk will have an impact on a specific spatial area, or whether the risk at a specific spatial location increases significantly during certain time periods. For example, through data correlation analysis, trend comparison, pattern recognition and other technologies, it is possible to find out whether the time peak of the risk is concentrated in a specific spatial area, or whether the high-incidence points of spatial risk have temporal periodicity. Through these analyses, the interactive influence information between the two types of change data can be obtained, laying the foundation for comprehensive risk prediction.

[0094] Step 306, determining target tunnel risk prediction data based on the tunnel risk time variation data, the tunnel risk space variation data, and the variation data interaction impact information.

[0095] Specifically, the overall risk prediction model of the target tunnel is constructed based on the tunnel risk time variation data, tunnel risk space variation data and the interactive influence information between the variation data of the target tunnel. This overall risk prediction model can integrate the risk variation trends in the time and space dimensions, and integrate the analysis results of the interactive influence to generate the target tunnel risk prediction data. The target tunnel risk prediction data contains the risk trends and spatial distribution characteristics that may appear in the tunnel as a whole in the future, and can fully reflect the potential risk status of the tunnel in the time and space dimensions.

[0096] Among them, the target tunnel risk prediction data calculation formula corresponding to the target tunnel risk prediction data is determined according to the tunnel risk time change data, the tunnel risk space change data and the change data interaction impact information as follows:

[0097]

[0098] in, is the tunnel risk time variation data, is the spatial variation data of tunnel risk, To change the data interactively, is the tunnel risk time data change rate, is the change rate of tunnel risk spatial data, is the overall adjustment factor, as well as are nonlinear exponential parameters, and are all weight coefficients.

[0099] In this embodiment, by analyzing the characteristics of the global time and space risk prediction data, the time change data and space change data of the tunnel risk are generated respectively, and the correlation between the two is further analyzed in depth to generate the interactive impact information of the change data. This process can reveal the complex interactive characteristics of the risk changes of the tunnel in the time and space dimensions, so that the prediction data not only contains the independent risk trends of time and space, but also reflects the linkage effect between the two. The final target tunnel risk prediction data can provide more accurate risk distribution and change trends, which will help tunnel managers to fully grasp the dynamic risk situation in the tunnel, provide reliable data support for accurate risk prevention, safety inspections in key areas, and integrated management of time and space, and improve the overall safety and management efficiency of the tunnel.

[0100] In an exemplary embodiment, Figure 4 As shown, the correlation between the tunnel risk time variation data and the tunnel risk space variation data is analyzed to obtain the variation data interaction impact information, including steps 402 to 404. Among them:

[0101] Step 402, calculating risk value interaction information and risk change rate interaction information between tunnel risk time variation data and tunnel risk space variation data.

[0102] Among them, the risk value interaction information can be the correlation between the risk values ​​of various areas of the tunnel in time and space dimensions, revealing whether the risk levels of various areas at different time points show synchronous or correlated changes.

[0103] The risk change rate interaction information can be the speed of risk change in each area of ​​the tunnel in time and space dimensions and their mutual relationship, reflecting whether there is a correlation between the risk increase and decrease rates in different time periods and spatial locations. For example, in a certain time period, whether the risk change rate in different areas is accelerated or slowed down synchronously, or whether there is an area with abnormally rapid risk change in a specific time period.

[0104] Specifically, the risk value interaction information is calculated by performing point-to-point multiplication or other nonlinear combination of time risk data and space risk data to quantify the direct interaction between time and space risk values. Next, the change rate of time risk and space risk is calculated, and then these two change rates are multiplied or combined by other functions to calculate the risk change rate interaction information to reflect the interactive impact between the time and space risk change trends.

[0105] Step 404: Determine change data interaction impact information based on the risk value interaction information and the risk change rate interaction information.

[0106] Specifically, by using weighted aggregation or other fusion methods, the risk value interaction information (i.e., the direct association between temporal and spatial risk values) is combined with the risk change rate interaction information (i.e., the association between temporal and spatial risk change speeds). This comprehensive analysis allows the assessment of the joint impact of temporal and spatial risk factors, quantifies the degree of impact of their interaction on the overall risk of the tunnel, and obtains information on the interaction of change data. By integrating these two types of interactive information, the spatiotemporal dynamic characteristics of tunnel risk can be more fully understood, providing a solid basis for the final determination of target tunnel risk prediction data.

[0107] In this embodiment, by calculating the risk value interaction information and risk change rate interaction information between the tunnel risk time-varying data and the space-varying data, the complex interaction of tunnel risks in time and space dimensions can be deeply revealed. These interactive information not only reflects the synchronization of risk levels and increase and decrease rates in each time period and spatial location, but also helps identify key areas and high-risk periods of risk changes in time and space. The final determined interactive impact information of the changing data can more comprehensively reflect the dynamic correlation of risks in the time and space dimensions, and provide more insightful guidance for the safety management of the tunnel. Managers can formulate more accurate risk control measures based on this, identify and prioritize time and space risk hotspots, and effectively improve the safety and operational stability of the tunnel.

[0108] In an exemplary embodiment, Figure 5 As shown, calculating the risk value interaction information and risk change rate interaction information between the tunnel risk time change data and the tunnel risk space change data includes steps 502 to 510. Among them:

[0109] Step 502, multiplying the tunnel risk time variation data and the tunnel risk space variation data to obtain the product of the time and space variation data.

[0110] Among them, the product of time-space variation data can be a matrix obtained by multiplying the time-varying data and the space-varying data of tunnel risk at each corresponding time and space point, which reflects the degree of mutual influence of risks in time and space dimensions.

[0111] Specifically, the tunnel risk time variation data and space variation data are multiplied one by one at each time and space point to generate a matrix of the time and space variation data product. Each element of the product matrix represents the degree of mutual influence between the time variation and space variation of the risk value at a specific time and space position. For example, the simultaneous existence of risk growth at a certain point in time and high risk in a certain area may result in a large product value, indicating that there is a potential high-risk area at this time and space point.

[0112] Step 504: Perform nonlinear transformation on the product of time-space variation data to obtain risk value interaction information.

[0113] Specifically, nonlinear transformations (such as logarithmic, exponential or sigmoid functions) are applied to process the product matrix of the product of temporal and spatial variation data to further reveal more complex interactions between risk values. Nonlinear transformations can amplify high-risk interaction areas or suppress lower-risk areas, so that risk value interaction information can more clearly reflect significant risk associations in the temporal and spatial dimensions. This transformation can highlight the nonlinear characteristics of risk interactions, help identify potential high-risk points in temporal and spatial dimensions, and obtain risk value interaction information.

[0114] Step 506, respectively calculating the change rates of the tunnel risk time change data and the tunnel risk space change data to obtain the tunnel risk time data change rate and the tunnel risk space data change rate.

[0115] Among them, the tunnel risk time data change rate can refer to the rate of change of tunnel risk in the time dimension, reflecting the speed at which the risk increases or decreases between different time points.

[0116] Among them, the tunnel risk spatial data change rate can refer to the risk change rate between different spatial areas of the tunnel, reflecting the speed at which the risk increases or decreases in the spatial dimension.

[0117] Specifically, the time change rate of the tunnel risk time change data and the spatial change rate of the tunnel risk space change data are calculated respectively, where the time change rate represents the rate of increase or decrease of risk at different time points, such as the degree of change in risk value per hour; the spatial change rate represents the rate of change of risk at different spatial locations, such as the change in risk differences between regions. By calculating these change rates, the acceleration of risk changes in time and space can be obtained, and the change rate of tunnel risk time data and the change rate of tunnel risk space data can be obtained.

[0118] Step 508, multiplying the tunnel risk time data change rate and the tunnel risk space data change rate to obtain the product of the time and space change rates.

[0119] Among them, the product of time and space change rate can be a matrix obtained by multiplying the tunnel risk time data change rate and the space data change rate one by one at each time and space point, reflecting the combined effect of time and space change rate.

[0120] Specifically, the matrix of the tunnel risk time data change rate and the tunnel risk space data change rate is multiplied at each corresponding time and space point to obtain the matrix of the time and space change rate product. This product reflects the joint effect between the risk change rate in the time and space dimensions. For example, if the change rate product value of a certain area in a specific time period is large, it may indicate that the risk of the area has increased significantly during this time period, suggesting a trend of synchronous increase in the time and space change rates.

[0121] Step 510, taking the absolute value of the product of the time-space change rate and performing nonlinear transformation to obtain risk change rate interaction information.

[0122] Specifically, the absolute value of each value in the matrix of the product of the time-space change rate is taken to eliminate the influence of the risk change direction (increase or decrease), and then the matrix is ​​transformed nonlinearly, such as using a logarithmic or exponential function, to generate the final risk change rate interaction information. This processing step further enhances the significant interaction in the change rate, highlights the significant correlation between the time and space change rates, and more intuitively identifies the high-risk areas that accelerate or decelerate in the time and space dimensions. This interactive information provides an important basis for predicting the rapidly growing risk areas in the tunnel.

[0123] In this embodiment, by calculating the product of the tunnel risk time-varying data and the space-varying data, and performing a nonlinear transformation on the result, the risk value interaction information is extracted, revealing the risk synchronization between different time periods and spatial regions. At the same time, by calculating the change rate of the time and space-varying data respectively, multiplying the two and then performing a nonlinear transformation, the risk change rate interaction information is extracted, further revealing the interactive characteristics of the risk increase and decrease speed in the time and space dimensions. This process provides a refined spatiotemporal analysis of the risk dynamics of the tunnel, can identify key high-risk areas and time periods, and reveal the acceleration or deceleration trend of risk changes. The interactive information finally obtained provides tunnel managers with a more comprehensive risk warning basis, enabling them to make targeted safety protection and maintenance decisions, thereby effectively improving the safety and management efficiency of the tunnel.

[0124] In an exemplary embodiment, Figure 6 As shown, according to the tunnel time dimension and the tunnel space dimension, the risk prediction data of each tunnel is spliced ​​to obtain the time global tunnel risk splicing data and the space global tunnel risk splicing data, including steps 602 to 606. Among them:

[0125] Step 602, decompose each tunnel risk prediction data according to the tunnel time dimension and the tunnel space dimension to obtain each time dimension decomposition data and each space dimension decomposition data.

[0126] Among them, the time dimension decomposition data can be subset data obtained by decomposing the tunnel risk prediction data according to time points, indicating the risk status of the tunnel at each specific moment.

[0127] Among them, the spatial dimension decomposition data can be subset data obtained by disassembling the tunnel risk prediction data according to spatial position, indicating the risk situation of the tunnel in different spatial areas.

[0128] Specifically, the risk prediction data of each tunnel is processed and decomposed according to the time dimension and space dimension of the tunnel, and the risk data is disassembled into independent data fragments at each time point and each spatial area. The time dimension decomposition data represents the risk situation of each area in different time periods, while the space dimension decomposition data represents the risk distribution at each location in the tunnel.

[0129] Step 604 , feature fusion is performed on each time dimension decomposition data and each space dimension decomposition data to obtain time dimension fusion data and space dimension fusion data.

[0130] Among them, the time dimension fusion data can be comprehensive data generated by feature fusion of all time dimension decomposition data, reflecting the overall changing trend of tunnel risk over time.

[0131] Among them, the spatial dimension fusion data can be comprehensive data generated by feature fusion of all spatial dimension decomposition data, showing the overall risk distribution of the tunnel in the spatial dimension.

[0132] Specifically, all the time dimension decomposition data are fused on the time axis, that is, the data at each time point are integrated into a time dimension fusion data with overall characteristics, reflecting the overall trend of risk changes over time; similarly, the spatial dimension decomposition data are fused on the spatial axis to obtain spatial dimension fusion data, showing the comprehensive distribution of risks at different locations in the tunnel. This fusion step generates more holistic information by extracting features in time and space, making the risk pattern in the time and space dimensions more significant.

[0133] Step 606, reconstruct the time dimension fusion data and the space dimension fusion data respectively to obtain time global tunnel risk splicing data and space global tunnel risk splicing data.

[0134] Specifically, the time dimension fusion data and the space dimension fusion data are reconstructed respectively. In the time dimension, the time fusion data are arranged into a continuous time series to show the global changes in the risk of the entire tunnel in different time periods; in the spatial dimension, the spatial fusion data are reconstructed by position to generate global splicing data reflecting the risk distribution of different spatial areas of the tunnel, and finally the time global tunnel risk splicing data and the space global tunnel risk splicing data are obtained.

[0135] In this embodiment, the tunnel risk prediction data is decomposed according to the time and space dimensions, and the features are fused and reconstructed to generate time-global and space-global tunnel risk splicing data. This process enables the risk data to show the overall risk evolution trend in the time dimension, while clearly presenting the risk distribution characteristics of different areas in the tunnel in the spatial dimension. The time-global risk splicing data provides managers with a comprehensive perspective on the future development of risks, which helps to predict and respond to risks in specific periods in advance; the space-global risk splicing data provides a clear reference for the risk positioning of the tunnel structure, which facilitates the identification and management of high-risk areas. Ultimately, the global view of the spliced ​​data significantly enhances the visualization of tunnel risks, provides data support for refined risk prevention and control and resource scheduling, and helps to improve the safety management efficiency of tunnels.

[0136] In an exemplary embodiment, Figure 7 As shown, the time dimension fusion data and the space dimension fusion data are reconstructed respectively to obtain the time global tunnel risk splicing data and the space global tunnel risk splicing data, including steps 702 to 706. Among them:

[0137] Step 702, performing reverse tunnel time dimension decomposition on the time dimension fusion data to obtain each reverse time dimension decomposition data, and performing reverse tunnel space dimension decomposition on the space dimension fusion data to obtain each reverse space dimension decomposition data.

[0138] Among them, the reverse tunnel time dimension decomposition can be to re-disassemble the time dimension fusion data of tunnel risk into risk data subsets at each time point, with the aim of restoring the specific risk details in different time periods so as to analyze the risk characteristics at each time point.

[0139] Among them, the reverse tunnel spatial dimension decomposition can be to decompose the spatial dimension fusion data of tunnel risk into risk data subsets of multiple regions according to different spatial positions, and re-display the independent risk information of each divided area in the tunnel, which is convenient for further spatial risk analysis.

[0140] Among them, the reverse time dimension decomposition data can be a set of risk data at each time point obtained through reverse time dimension decomposition, which shows the risk status of the tunnel in different time periods and provides details for further analysis of risk trends over time.

[0141] Among them, the reverse spatial dimension decomposition data can be a set of risk data of various spatial areas obtained through reverse spatial dimension decomposition, showing the independent risk situations of different spatial positions of the tunnel and providing support for risk analysis in the spatial dimension.

[0142] Specifically, the time dimension fusion data is decomposed in reverse tunnel time dimension, and the reverse time dimension decomposition data at each time point is disassembled to reproduce the risk situation in different time periods; at the same time, the space dimension fusion data is decomposed in reverse tunnel space dimension, and the space area of ​​the tunnel is divided into reverse space dimension decomposition data of each area to reproduce the independent risk situation of each spatial position. Through this decomposition step, the risk details of the two dimensions of time and space are restored.

[0143] Step 704 , interactively influence nonlinear fusion of each inverse time dimension decomposition data and each space dimension decomposition data to obtain inverse time dimension fusion data and inverse space dimension fusion data.

[0144] Among them, nonlinear fusion of interactive impacts can be the process of processing the data decomposed in the reverse time dimension and space dimension, enhancing the spatiotemporal interaction characteristics through nonlinear operations (such as logarithmic or exponential functions), and revealing the complex correlation between time and space dimensions in risk changes.

[0145] Among them, the reverse time dimension fusion data can be the result data of the risk interaction between different time points after nonlinear fusion of interactive effects, showing the global risk characteristics in the time dimension.

[0146] Among them, the reverse spatial dimension fusion data can be the result data of the risk interaction between different spatial positions after nonlinear fusion of interactive influence, presenting the global risk distribution characteristics in the spatial dimension.

[0147] Specifically, the decomposed data of each reverse time dimension and each spatial dimension are nonlinearly fused to integrate the interaction of risk characteristics in time and space. Specifically, the data of each time point and spatial position are interacted with each other, and nonlinear functions (such as logarithmic or exponential functions) are applied to transform the data to amplify or suppress certain risk interaction characteristics. For example, if the risk value of a certain time period rises rapidly in a specific spatial area, this interaction can be more significantly presented through nonlinear fusion. The reverse time dimension fusion data and reverse space dimension fusion data obtained after fusion contain the characteristics of each dimension and highlight the risk characteristics brought about by the interaction of time and space.

[0148] Step 706 , reconstruct feature data of the reverse time dimension fusion data and the reverse space dimension fusion data respectively to obtain time global tunnel risk splicing data and space global tunnel risk splicing data.

[0149] Specifically, the reverse time dimension fusion data and the reverse space dimension fusion data are reconstructed for feature data. The reconstruction process combines the fusion data of all time periods and spatial regions to combine the overall risk change trend of the tunnel in different periods into a complete time global view (i.e., time global tunnel risk splicing data), and splices the risk distribution of each location of the tunnel into an overall space global view (i.e., space global tunnel risk splicing data), forming time global tunnel risk splicing data and space global tunnel risk splicing data. This reconstruction process provides a macro-risk overview of tunnel risk in a time-space global manner, supports the overall assessment of the safety status of the tunnel in time and space dimensions, and facilitates the formulation of targeted protection and management strategies.

[0150] In this embodiment, the risk characteristics of each time and space are refined into independent data by reverse decomposing the fused data of time and space dimensions, and the nonlinear fusion technology is applied to integrate the interaction of time and space, so that the interaction of risk characteristics in each time period and spatial area is more prominent. The subsequent feature data reconstruction step generates time and space global tunnel risk splicing data, showing the risk dynamics of the entire tunnel in time and space. Through this reverse decomposition and reconstruction, managers can obtain more detailed risk information and a more accurate global risk view, and identify key high-risk periods and high-risk areas. The global splicing data finally generated provides a comprehensive visualization of the tunnel's time and space risks, providing strong data support for the implementation of targeted safety management measures, optimization of maintenance plans, and rational allocation of resources, effectively improving the safety management efficiency of the tunnel.

[0151] In an exemplary embodiment, Figure 8 As shown, after predicting the tunnel risk of the target tunnel by combining the global time risk prediction data and the global space risk prediction data to obtain the target tunnel risk prediction data, the method further includes steps 802 to 806. Among them:

[0152] Step 802: Perform spatiotemporal analysis on the target tunnel risk prediction data to determine the spatiotemporal risk information of the target tunnel.

[0153] Among them, spatiotemporal analysis can be the process of analyzing tunnel risk prediction data in time and space, identifying the distribution patterns of risks in different time periods and spatial locations, and helping to understand the overall risk dynamics of the tunnel.

[0154] Among them, the risk spatiotemporal information can be the result data obtained through spatiotemporal analysis, which indicates the risk level of the tunnel at each time point and spatial area, and provides a detailed description of the spatiotemporal risk changes of the tunnel.

[0155] Specifically, a comprehensive spatiotemporal analysis is performed on the target tunnel risk prediction data to analyze its distribution characteristics in different time periods and spatial regions. Specifically, by analyzing the target tunnel risk prediction data time by time and region by region, the risk change curves of each segmented area in each time period are constructed to generate risk spatiotemporal information. This risk spatiotemporal information shows the risk level of each segmented area in the tunnel at different time points, helps to identify the high-incidence time periods and concentrated areas of risk, lays the foundation for further analysis of the overall dynamic risk distribution of the tunnel, and provides data support for predicting the risk change trend.

[0156] Step 804, based on the risk spatiotemporal information, determine the abnormality investigation information of the high-risk area corresponding to the target tunnel and the comprehensive risk management and control information of the tunnel.

[0157] Among them, abnormal inspection information in high-risk areas can be refined data generated based on risk spatiotemporal information, marking areas and time periods with higher risk levels in the tunnel, and guiding managers to conduct priority inspections and tests in these areas.

[0158] Among them, the comprehensive risk management information of the tunnel can be management data based on the overall risk situation of the tunnel, including comprehensive analysis results of the tunnel risk situation and corresponding safety management recommendations, providing guidance for the comprehensive risk control of the tunnel.

[0159] Specifically, the generated risk spatiotemporal information is used to identify the segmented areas with higher risk levels and mark them as high-risk areas. Based on these high-risk areas, detailed abnormal investigation information is generated to help the tunnel management team identify the areas that need to be investigated. At the same time, the risk status of each segmented area and time point is integrated to analyze the overall risk distribution and trend of the tunnel, and generate comprehensive risk management information for the tunnel.

[0160] Step 806, generating high-risk area anomaly repair information corresponding to the target tunnel based on the high-risk area anomaly investigation information, and generating comprehensive traffic dispatch information of the target tunnel based on the tunnel comprehensive risk management and control information.

[0161] Among them, the abnormal repair information of high-risk areas can be specific repair suggestions generated for high-risk areas, including the operations, technical means, resources and time arrangements required for repair, to help the maintenance team quickly reduce hidden dangers in high-risk areas.

[0162] Among them, the comprehensive traffic dispatch information can be a traffic management plan generated according to the overall risk situation of the tunnel, including traffic flow control, speed limit, diversion and other measures to ensure the safe passage of the tunnel under high-risk conditions.

[0163] Specifically, based on the abnormal investigation information of high-risk areas, the abnormal repair information of high-risk areas is further refined and generated to specifically guide the maintenance team to carry out necessary maintenance and repair actions in these areas to reduce the risk level. For example, it can include details such as specific repair steps, technologies used, and estimated completion time. At the same time, based on the comprehensive risk management information of the tunnel, comprehensive traffic dispatch information is generated to optimize traffic management in the tunnel by reasonably arranging traffic flow, adjusting traffic strategies, setting speed limits or diversions, etc., to ensure the safe operation of the tunnel during high-risk periods.

[0164] In this embodiment, detailed spatiotemporal risk information is generated through spatiotemporal analysis of the target tunnel risk prediction data, showing the risk changes of each segmented area at different times, providing an accurate data basis for dynamic risk management inside the tunnel. Based on this spatiotemporal information, high-risk areas are further identified and abnormal investigation information is generated, while comprehensive risk management and control information of the tunnel is constructed, thereby helping managers to quickly lock in key risk areas and systematically evaluate the overall safety status of the tunnel. Finally, specific repair plans are generated through abnormal investigation information in high-risk areas to ensure that maintenance measures are taken quickly in key areas; at the same time, traffic scheduling strategies are generated based on comprehensive management and control information to optimize tunnel traffic management at different risk levels. This process greatly improves the accuracy and response speed of tunnel risk management, supports targeted maintenance and traffic management, and comprehensively improves the safe operation level and management efficiency of tunnels.

[0165] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0166] Based on the same inventive concept, the embodiment of the present application also provides a multi-sensory tunnel risk prediction device for implementing the multi-sensory tunnel risk prediction method involved in the above-mentioned method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in the embodiments of one or more multi-sensory tunnel risk prediction devices provided below can be referred to the above limitations on a multi-sensory tunnel risk prediction method, which will not be repeated here.

[0167] In an exemplary embodiment, Fig. 9 As shown, a tunnel risk prediction device with multiple sensing paths is provided, including: a tunnel data acquisition module 902, a tunnel risk prediction module 904 and a risk data splicing module 906, wherein:

[0168] The tunnel data acquisition module 902 is used to acquire tunnel physical characteristic data corresponding to each segmented area of ​​the target tunnel, and to acquire a tunnel risk prediction model corresponding to each tunnel physical characteristic data;

[0169] The tunnel risk prediction module 904 is used to input the physical characteristic data of each tunnel into the corresponding tunnel risk prediction model to obtain the tunnel risk prediction data corresponding to each segmented area;

[0170] The risk data splicing module 906 is used to splice the risk prediction data of each tunnel according to the tunnel time dimension and the tunnel space dimension to obtain the time global tunnel risk splicing data and the space global tunnel risk splicing data;

[0171] The tunnel risk prediction module 904 is further used to input the time global tunnel risk splicing data into the global time risk prediction model of the target tunnel to obtain the global time risk prediction data;

[0172] The tunnel risk prediction module 904 is further used to input the spatial global tunnel risk splicing data into the global spatial risk prediction model of the target tunnel to obtain the global spatial risk prediction data;

[0173] The tunnel risk prediction module 904 is also used to predict the tunnel risk of the target tunnel by combining the global time risk prediction data and the global space risk prediction data to obtain the target tunnel risk prediction data.

[0174] In one embodiment, the tunnel risk prediction module 904 is also used to perform risk characteristic analysis on the global time risk prediction data and the global space risk prediction data respectively to obtain tunnel risk time variation data and tunnel risk space variation data; analyze the correlation between the tunnel risk time variation data and the tunnel risk space variation data to obtain variation data interaction impact information; determine the target tunnel risk prediction data based on the tunnel risk time variation data, the tunnel risk space variation data and the variation data interaction impact information.

[0175] In one embodiment, the tunnel risk prediction module 904 is also used to calculate the risk value interaction information and the risk change rate interaction information between the tunnel risk time change data and the tunnel risk space change data; and determine the change data interaction impact information based on the risk value interaction information and the risk change rate interaction information.

[0176] In one embodiment, the tunnel risk prediction module 904 is also used to multiply the tunnel risk time-varying data and the tunnel risk space-varying data to obtain the product of the time-space varying data; perform nonlinear transformation on the product of the time-space varying data to obtain risk value interaction information; respectively calculate the change rates of the tunnel risk time-varying data and the tunnel risk space-varying data to obtain the tunnel risk time data change rate and the tunnel risk space data change rate; multiply the tunnel risk time data change rate and the tunnel risk space data change rate to obtain the time-space change rate product; take the absolute value of the time-space change rate product and perform nonlinear transformation to obtain risk change rate interaction information.

[0177] In one embodiment, the risk data splicing module 906 is also used to decompose each tunnel risk prediction data according to the tunnel time dimension and the tunnel space dimension to obtain each time dimension decomposition data and each space dimension decomposition data; perform feature fusion on each time dimension decomposition data and each space dimension decomposition data respectively to obtain time dimension fusion data and space dimension fusion data; perform data reconstruction on the time dimension fusion data and the space dimension fusion data respectively to obtain time global tunnel risk splicing data and space global tunnel risk splicing data.

[0178] In one embodiment, the risk data splicing module 906 is also used to perform reverse tunnel time dimension decomposition on the time dimension fusion data to obtain each reverse time dimension decomposition data, and to perform reverse tunnel space dimension decomposition on the space dimension fusion data to obtain each reverse space dimension decomposition data; perform interactive nonlinear fusion on each reverse time dimension decomposition data and each space dimension decomposition data to obtain reverse time dimension fusion data and reverse space dimension fusion data; perform feature data reconstruction on the reverse time dimension fusion data and the reverse space dimension fusion data to obtain time global tunnel risk splicing data and space global tunnel risk splicing data.

[0179] In one embodiment, the tunnel risk prediction module 904 is also used to perform spatiotemporal analysis on the target tunnel risk prediction data to determine the risk spatiotemporal information of the target tunnel; the risk spatiotemporal information characterizes the risk level of each segmented area of ​​the target tunnel at different times; based on the risk spatiotemporal information, the high-risk area abnormality investigation information and the tunnel comprehensive risk management information corresponding to the target tunnel are determined; the high-risk area abnormality repair information corresponding to the target tunnel is generated based on the high-risk area abnormality investigation information, and the comprehensive traffic scheduling information of the target tunnel is generated based on the tunnel comprehensive risk management information.

[0180] Each module in the above-mentioned tunnel risk prediction device with multiple sensing paths can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each of the above modules.

[0181] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Fig.10 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store server data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a tunnel risk prediction method with a multi-sensory approach is implemented.

[0182] Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0183] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0184] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0185] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the above-mentioned method embodiments.

[0186] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0187] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0188] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0189] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A tunnel risk prediction method based on multiple sensing approaches, characterized in that: The method comprises: Obtaining tunnel physical characteristic data corresponding to each segmented area of ​​the target tunnel, and obtaining a tunnel risk prediction model corresponding to each tunnel physical characteristic data; Inputting the physical characteristic data of each tunnel into the corresponding tunnel risk prediction model respectively to obtain the tunnel risk prediction data corresponding to each segmented area; According to the tunnel time dimension and the tunnel space dimension, the tunnel risk prediction data are spliced ​​to obtain time global tunnel risk splicing data and space global tunnel risk splicing data; Inputting the temporal global tunnel risk splicing data into the global temporal risk prediction model of the target tunnel to obtain global temporal risk prediction data; Inputting the spatial global tunnel risk splicing data into the global spatial risk prediction model of the target tunnel to obtain global spatial risk prediction data; Predicting the tunnel risk of the target tunnel by combining the global time risk prediction data and the global space risk prediction data to obtain target tunnel risk prediction data; The tunnel risk of the target tunnel is predicted by combining the global time risk prediction data and the global space risk prediction data to obtain the target tunnel risk prediction data, including: Perform risk characteristic analysis on the global time risk prediction data and the global space risk prediction data respectively to obtain tunnel risk time variation data and tunnel risk space variation data; Analyze the correlation between the tunnel risk time variation data and the tunnel risk space variation data to obtain the variation data interaction impact information; Determining the target tunnel risk prediction data according to the tunnel risk time variation data, the tunnel risk space variation data and the variation data interaction impact information; Wherein, the target tunnel risk prediction data calculation formula corresponding to the target tunnel risk prediction data is determined according to the tunnel risk time variation data, the tunnel risk space variation data and the variation data interaction influence information as follows: in, is the tunnel risk time variation data, is the spatial variation data of tunnel risk, To change the data interactively, is the tunnel risk time data change rate, is the change rate of tunnel risk spatial data, is the overall adjustment factor, as well as are nonlinear exponential parameters, and are all weight coefficients.

2. The method according to claim 1, characterized in that The analyzing the correlation between the tunnel risk time variation data and the tunnel risk space variation data to obtain the variation data interaction impact information includes: Calculating risk value interaction information and risk change rate interaction information between the tunnel risk time change data and the tunnel risk space change data; The change data interaction impact information is determined according to the risk value interaction information and the risk change rate interaction information.

3. The method according to claim 2, characterized in that The calculating of the risk value interaction information and the risk change rate interaction information between the tunnel risk time change data and the tunnel risk space change data includes: Multiplying the tunnel risk time variation data and the tunnel risk space variation data to obtain a product of the time and space variation data; Performing nonlinear transformation on the product of the temporal and spatial variation data to obtain the risk value interaction information; Respectively calculating the change rates of the tunnel risk time change data and the tunnel risk space change data to obtain the tunnel risk time data change rate and the tunnel risk space data change rate; Multiplying the tunnel risk time data change rate and the tunnel risk space data change rate to obtain a time-space change rate product; The absolute value of the product of the time-space change rate is taken and a nonlinear transformation is performed to obtain the risk change rate interaction information.

4. The method according to claim 1, characterized in that: The tunnel risk prediction data are spliced ​​according to the tunnel time dimension and the tunnel space dimension to obtain the time global tunnel risk splicing data and the space global tunnel risk splicing data, including: Decomposing each of the tunnel risk prediction data according to the tunnel time dimension and the tunnel space dimension to obtain decomposition data of each time dimension and decomposition data of each space dimension; Perform feature fusion on each of the time dimension decomposition data and each of the space dimension decomposition data to obtain time dimension fusion data and space dimension fusion data; The time dimension fusion data and the space dimension fusion data are reconstructed respectively to obtain the time global tunnel risk splicing data and the space global tunnel risk splicing data.

5. The method according to claim 4, characterized in that The reconstructing the time dimension fusion data and the space dimension fusion data respectively to obtain the time global tunnel risk splicing data and the space global tunnel risk splicing data includes: Performing reverse tunnel time dimension decomposition on the time dimension fusion data to obtain each reverse time dimension decomposition data, and performing reverse tunnel space dimension decomposition on the space dimension fusion data to obtain each reverse space dimension decomposition data; Perform interactive nonlinear fusion on each of the reverse time dimension decomposition data and each of the space dimension decomposition data to obtain reverse time dimension fusion data and reverse space dimension fusion data; Feature data are reconstructed for the reverse time dimension fusion data and the reverse space dimension fusion data respectively to obtain the time global tunnel risk splicing data and the space global tunnel risk splicing data.

6. The method according to claim 1, characterized in that After the step of predicting the tunnel risk of the target tunnel by combining the global time risk prediction data and the global space risk prediction data to obtain target tunnel risk prediction data, the method further includes: Performing spatiotemporal analysis on the target tunnel risk prediction data to determine the risk spatiotemporal information of the target tunnel; the risk spatiotemporal information represents the risk level of each segmented area of ​​the target tunnel at different times; Determine, based on the risk spatiotemporal information, abnormality investigation information of high-risk areas corresponding to the target tunnel and comprehensive risk management and control information of the tunnel; The high-risk area abnormality repair information corresponding to the target tunnel is generated based on the high-risk area abnormality investigation information, and the comprehensive traffic dispatching information of the target tunnel is generated based on the tunnel comprehensive risk management and control information.

7. A tunnel risk prediction device with multiple sensing paths, comprising a tunnel data acquisition module, a tunnel risk prediction module and a risk data splicing module, characterized in that: The device implements the steps of the method according to any one of claims 1 to 6.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

Images