Special risk grade assessment method and system for tunnel construction

Abstract

The invention relates to a special risk grade assessment method and system for tunnel construction, and relates to the technical field of tunnel construction risk assessment. The method mainly comprises the following steps: acquiring a similar tunnel of which the similarity with a target tunnel exceeds a preset value from a preset database based on attribute data; comparing the change trends of the similar tunnel and the target tunnel in the historical time, and determining weight values respectively corresponding to tunnel geological data, tunnel environment data, tunnel design data and tunnel construction data in the attribute data of the target tunnel; according to tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data and weight values corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data and the tunnel construction data in the attribute data of the target tunnel, the accident occurrence probability of the target tunnel is calculated; and determining the risk level of the target tunnel according to the accident probability of the target tunnel.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel construction risk assessment, and in particular to a special risk level assessment method and system for tunnel construction. Background Art

[0002] Tunnel excavation often faces risks arising from the complexity of construction technology and geological uncertainties. This can lead to frequent accidents such as large tunnel deformation and collapse, resulting in casualties, economic losses, and construction delays. Tunnel construction safety risks are highly hidden, sudden, and rapidly evolving. Previous static risk assessment methods have made it difficult to fully track and control construction safety risks, and they lack real-time feedback on dynamic tunnel construction safety risk information.

[0003] Tunnel construction safety management under complex geological conditions faces numerous challenges and shortcomings, most notably the lag and incompleteness of safety risk assessments. Current risk assessment methods, mostly based on existing geological exploration data and traditional empirical formulas, fail to fully reflect the complexity of geological conditions and the dynamic nature of construction. This static assessment model ignores the variability of geological conditions and the real-time conditions at the construction site. Consequently, assessment results often lag behind actual conditions, failing to provide timely warnings of potential risks. Summary of the Invention

[0004] The present invention aims to provide a special risk level assessment method and system for tunnel construction to address the deficiencies in the prior art. The technical problems to be solved by the present invention are achieved through the following technical solutions.

[0005] An embodiment of the present invention provides a method for assessing a specific risk level for tunnel construction, the method comprising:

[0006] Acquire attribute data of the target tunnel at the current time point, wherein the attribute data includes tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data;

[0007] Acquire similar tunnels having a similarity with the target tunnel exceeding a predetermined value from a preset database based on the attribute data;

[0008] Comparing the historical change trends of the attribute data of the similar tunnel and the target tunnel, determining the weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel;

[0009] Calculating the probability of an accident occurring in the target tunnel based on tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data, and their corresponding weight values ​​in the attribute data of the target tunnel;

[0010] The risk level of the target tunnel is determined according to the probability of an accident occurring in the target tunnel.

[0011] In an optional embodiment, obtaining a similar tunnel having a similarity with the target tunnel exceeding a predetermined value from a preset database based on the attribute data includes:

[0012] Calculating similarity values ​​between the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data of the target tunnel and the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data of each tunnel in the preset database;

[0013] Performing weighted calculation on the calculated similarity values ​​to obtain the similarity of each tunnel in the preset database;

[0014] Tunnels in the preset database whose similarity exceeds a predetermined value are determined as similar tunnels.

[0015] In an optional embodiment, comparing the historical change trends of the attribute data of the similar tunnel and the target tunnel, and determining the weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel, respectively, includes:

[0016] Obtaining, from the preset database, influencing factors corresponding to tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the similar tunnel;

[0017] Based on the influencing factors and the historical change trends of the attribute data of the similar tunnels and the target tunnel, the weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel are determined.

[0018] In an optional embodiment, determining the weight values ​​corresponding to the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel based on the influencing factors and the historical change trends of the attribute data of the similar tunnels and the target tunnel, respectively, includes:

[0019] Drawing tunnel geological data curves, tunnel environment data curves, tunnel design data curves, and tunnel construction data curves respectively according to the attribute data of the similar tunnels and the target tunnel;

[0020] Comparing the tunnel geological data curve graph, the tunnel environment data curve graph, the tunnel design data curve graph, and the tunnel construction data curve graph of the similar tunnel and the target tunnel respectively to obtain a first correction value, a second correction value, a third correction value, and a fourth correction value;

[0021] The influencing factors are corrected by using the first correction value, the second correction value, the third correction value, and the fourth correction value to obtain weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel.

[0022] In an optional embodiment, the comparing the tunnel geological data curve graphs, the tunnel environment data curve graphs, the tunnel design data curve graphs, and the tunnel construction data curve graphs of the similar tunnel and the target tunnel to obtain the first correction value, the second correction value, the third correction value, and the fourth correction value includes:

[0023] Comparing the tunnel geological data curve graphs, tunnel environment data curve graphs, tunnel design data curve graphs, and tunnel construction data curve graphs of the similar tunnel and the target tunnel, respectively, to obtain comparison curve results of the respective curve graphs, wherein the comparison curve results include a curve slope difference, a curve maximum difference, and a curve minimum difference;

[0024] The first correction value, the second correction value, the third correction value, and the fourth correction value are calculated according to the curve slope difference, the curve maximum difference, and the curve minimum difference.

[0025] In an optional embodiment, calculating the first correction value, the second correction value, the third correction value, and the fourth correction value by using the curve slope difference, the curve maximum difference, and the curve minimum difference includes:

[0026] The curve slope difference, curve maximum difference, and curve minimum difference corresponding to the tunnel geological data curve graph, the tunnel environment data curve graph, the tunnel design data curve graph, and the tunnel construction data curve graph are input into the correction value prediction model to obtain the first correction value, the second correction value, the third correction value, and the fourth correction value.

[0027] In an optional embodiment, the calculating the probability of an accident occurring in the target tunnel based on the tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data and their corresponding weight values ​​in the attribute data of the target tunnel includes:

[0028] Determining physical characteristics and spatiotemporal characteristics based on tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data and their corresponding weight values ​​in the attribute data of the target tunnel;

[0029] The probability of an accident occurring in the target tunnel is calculated based on the physical characteristics and the spatiotemporal characteristics.

[0030] In an optional embodiment, the calculating the probability of an accident occurring in the target tunnel by using the physical characteristics and the spatiotemporal characteristics includes:

[0031] Calculating a first accident probability and a second accident probability respectively according to the physical characteristics and the spatiotemporal characteristics;

[0032] The probability of an accident occurring in the target tunnel is calculated according to the first accident probability and the second accident probability.

[0033] In an optional embodiment, the calculating the probability of an accident occurring in the target tunnel by using the physical characteristics and the spatiotemporal characteristics includes:

[0034] Performing feature fusion on the physical feature and the spatiotemporal feature to obtain a fused feature;

[0035] The probability of an accident occurring in the target tunnel is calculated based on the fusion features.

[0036] An embodiment of the present invention provides a special risk level assessment system for tunnel construction, the system comprising:

[0037] An acquisition module is used to acquire attribute data of the target tunnel at the current time point, wherein the attribute data includes tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data;

[0038] The acquisition module is further configured to acquire, from a preset database based on the attribute data, similar tunnels having a similarity with the target tunnel exceeding a predetermined value;

[0039] a comparison module, configured to compare the historical change trends of the attribute data of the similar tunnel and the target tunnel, and determine the weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel;

[0040] a calculation module, configured to calculate the probability of an accident occurring in the target tunnel based on tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data, and their corresponding weight values ​​in the attribute data of the target tunnel;

[0041] The determination module is configured to determine the risk level of the target tunnel according to the probability of an accident occurring in the target tunnel.

[0042] The embodiments of the present invention include the following advantages:

[0043] An embodiment of the present invention provides a special risk level assessment method and system for tunnel construction. The method first obtains attribute data of a target tunnel at a current time point, wherein the attribute data includes tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data. Then, based on the attribute data, similar tunnels whose similarity to the target tunnel exceeds a predetermined value are obtained from a preset database. The attribute data of the similar tunnels and the target tunnel are compared in historical time trends to determine the weight values ​​corresponding to the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel. Then, the probability of an accident occurring in the target tunnel is calculated based on the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel and their corresponding weight values. Finally, the risk level of the target tunnel is determined based on the probability of an accident occurring in the target tunnel. Compared with the current tunnel risk assessment methods which are mostly based on existing geological exploration data and traditional empirical formulas, this application obtains the attribute data of the target tunnel and compares and matches it with the tunnels in the preset database to determine the weight values ​​corresponding to the attribute data, and then calculates the probability of an accident in the tunnel based on the attribute data and its corresponding weight values. Since the attribute data in this application can reflect the complexity of geological conditions and the dynamic change characteristics during the construction process, this application can improve the accuracy of tunnel construction risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of a special risk level assessment method for tunnel construction provided by an embodiment of the present invention;

[0045] Figure 2 This is a flow chart for determining the weight value of attribute data provided by an embodiment of the present invention;

[0046] Figure 3 It is a structural diagram of a special risk level assessment system for tunnel construction provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0048] See also Figure 1 , which is a special risk level assessment method for tunnel construction provided by an embodiment of the present invention, and specifically includes S101-S105:

[0049] S101, acquiring attribute data of a target tunnel at a current time point, wherein the attribute data includes tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data.

[0050] The target tunnel is the tunnel for which construction risk is to be assessed. In this embodiment, tunnel geological data may include data such as compressive strength, permeability coefficient, and joint density; tunnel environmental data may include data such as daily cumulative rainfall, surface settlement rate, and pore water pressure; tunnel design data may include data such as steel arch spacing, shotcrete thickness, and anchor density; and tunnel construction data may include data such as daily footage, support stress time series curves, and overexcavation rate, although this embodiment does not impose specific limitations on this.

[0051] S102: Acquire similar tunnels having a similarity with the target tunnel exceeding a predetermined value from a preset database based on the attribute data.

[0052] The preset database stores attribute data corresponding to a plurality of tunnels respectively; the preset values ​​can be set according to actual needs, such as 70%, 80%, etc.

[0053] In an optional embodiment provided in the application, the obtaining of similar tunnels having a similarity exceeding a predetermined value with the target tunnel from a preset database based on the attribute data includes: calculating similarity values ​​between the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data of the target tunnel and the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data of each tunnel in the preset database; performing a weighted calculation on the calculated similarity values ​​to obtain the similarity of each tunnel in the preset database; and determining the tunnels in the preset database having a similarity exceeding a predetermined value as similar tunnels.

[0054] It should be noted that the weight value of the similarity value in this embodiment is determined based on the data stored in the preset database, that is, in addition to storing the attribute data corresponding to each tunnel, the preset database also stores the impact factors corresponding to each item of data in the attribute data when an accident occurs in the tunnel. The impact factor is used to represent the weight ratio that affects the occurrence of the accident, and the impact factor is determined as the corresponding weight value. After calculating the similarity value, this embodiment performs a weighted calculation on the similarity value to obtain the similarity of each tunnel in the preset database, and then determines the tunnel whose similarity exceeds the predetermined value as a similar tunnel of the target tunnel, or determines the tunnel with the highest similarity as a similar tunnel of the target tunnel.

[0055] S103 , comparing the historical change trends of the attribute data of the similar tunnels and the target tunnel, and determining the weight values ​​corresponding to the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel.

[0056] like Figure 2In an optional embodiment provided by the application, comparing the historical change trends of the attribute data of the similar tunnel and the target tunnel, and determining the weight values ​​corresponding to the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel, respectively, includes:

[0057] S1031, obtaining influencing factors corresponding to tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in attribute data of similar tunnels according to a preset database.

[0058] S1032, based on the influencing factors and the historical change trends of the attribute data of the similar tunnels and the target tunnel, determine the weight values ​​corresponding to the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel.

[0059] Specifically, based on the influencing factors and the historical change trends of the attribute data of the similar tunnels and the target tunnel, weight values ​​corresponding to the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel are determined, including:

[0060] S10321: Draw a tunnel geological data curve graph, a tunnel environment data curve graph, a tunnel design data curve graph, and a tunnel construction data curve graph based on the attribute data of the similar tunnel and the target tunnel.

[0061] S10322, respectively compare the tunnel geological data curve graph, tunnel environment data curve graph, tunnel design data curve graph, and tunnel construction data curve graph of the similar tunnel and the target tunnel to obtain a first correction value, a second correction value, a third correction value, and a fourth correction value.

[0062] Specifically, the respectively comparing the tunnel geological data curve graphs, tunnel environment data curve graphs, tunnel design data curve graphs, and tunnel construction data curve graphs of the similar tunnel and the target tunnel to obtain the first correction value, the second correction value, the third correction value, and the fourth correction value includes: respectively comparing the tunnel geological data curve graphs, tunnel environment data curve graphs, tunnel design data curve graphs, and tunnel construction data curve graphs of the similar tunnel and the target tunnel to obtain comparison curve results of each curve graph, the comparison curve results including curve slope difference, curve maximum difference, and curve minimum difference; calculating the first correction value, the second correction value, the third correction value, and the fourth correction value through the curve slope difference, the curve maximum difference, and the curve minimum difference.

[0063] In this embodiment, the maximum curve difference is the value with the largest difference between the curves in each curve segment of the two curve graphs, and the minimum curve difference is the value with the smallest difference between the curves in each curve segment of the two curve graphs. It should be noted that the curve comparison results in this embodiment include the curve slope difference, maximum curve difference, and minimum curve difference for multiple time periods. That is, after obtaining the curve graph, the curve in the curve graph is divided into multiple curve segments according to the time length of a point, and then the curve slope difference, maximum curve difference, and minimum curve difference are determined for each curve graph.

[0064] Among them, the first correction value, the second correction value, the third correction value and the fourth correction value are calculated by the curve slope difference, the curve maximum difference and the curve minimum difference, including: inputting the curve slope difference, the curve maximum difference and the curve minimum difference corresponding to the tunnel geological data curve graph, the tunnel environment data curve graph, the tunnel design data curve graph and the tunnel construction data curve graph into the correction value prediction model to obtain the first correction value, the second correction value, the third correction value and the fourth correction value.

[0065] Specifically, in this embodiment, the curve slope difference, curve maximum difference, and curve minimum difference corresponding to each curve segment in the tunnel geological data curve graph are input into the correction value prediction model to obtain a first correction value; the curve slope difference, curve maximum difference, and curve minimum difference corresponding to each curve segment in the tunnel environment data curve graph are input into the correction value prediction model to obtain a second correction value; the curve slope difference, curve maximum difference, and curve minimum difference corresponding to each curve segment in the tunnel design data curve graph are input into the correction value prediction model to obtain a third correction value; the curve slope difference, curve maximum difference, and curve minimum difference corresponding to each curve segment in the tunnel construction data curve graph are input into the correction value prediction model to obtain a fourth correction value.

[0066] More specifically, this embodiment can convert the curve slope difference, curve maximum difference, and curve minimum difference corresponding to all time periods into a feature matrix, where each row of the feature matrix represents the curve slope difference, curve maximum difference, and curve minimum difference of a time period, and then input the feature matrix into the correction value prediction model to obtain the corresponding correction value.

[0067] It should be noted that the correction value prediction model in this embodiment is a pre-trained neural network model. The sample data used by this neural network model is the sample feature matrix, and the sample labels are the corresponding correction values. After the correction value prediction model is trained, the feature matrix corresponding to each curve graph can be input into the model to obtain the corresponding correction value.

[0068] S10323, correcting the influencing factors using the first correction value, the second correction value, the third correction value, and the fourth correction value to obtain weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel.

[0069] In this embodiment, after obtaining the first, second, third, and fourth correction values, the influencing factors obtained from the pre-set database are corrected to obtain weight values ​​corresponding to the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the target tunnel's attribute data. Specifically, in this embodiment, the corrected values ​​and the corresponding influencing factors are averaged to obtain the corresponding weight values.

[0070] S104, calculating the probability of an accident occurring in the target tunnel according to the tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data and their corresponding weight values ​​in the attribute data of the target tunnel.

[0071] In an optional embodiment provided in the application, the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel can be converted into eigenvectors respectively, and then each eigenvector is multiplied by the corresponding weight value to form an eigenvector matrix. Each row in the eigenvector matrix represents an attribute data (i.e., tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data) and the corresponding weight value. The eigenvector matrix is ​​then input into the tunnel accident prediction network model to obtain the probability of an accident occurring in the target tunnel.

[0072] Among them, the tunnel accident prediction network model is obtained by training based on the sample eigenvector matrix and the corresponding accident labels. The method for obtaining the sample eigenvector matrix is ​​the same as the method for obtaining the above-mentioned eigenvector matrix, and this embodiment will not be repeated here.

[0073] In another optional embodiment provided by the application, the calculating the probability of an accident occurring in the target tunnel based on the tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data and their corresponding weight values ​​in the attribute data of the target tunnel includes:

[0074] S1041 , determining physical characteristics and spatiotemporal characteristics based on tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data and their corresponding weight values ​​in the attribute data of the target tunnel.

[0075] This embodiment can calculate physical characteristics and spatiotemporal characteristics based on tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data, and then multiply the calculated physical characteristics and spatiotemporal characteristics by corresponding weight values ​​to obtain the final physical characteristics and spatiotemporal characteristics.

[0076] In this example, physical characteristics reflect the intrinsic causes of risk (e.g., insufficient rock mass strength), and include the surrounding rock stability coefficient, seepage-stress coupling factor, and construction disturbance index. Spatiotemporal characteristics reveal the propagation patterns of risk (e.g., deformation diffusion along faults). These characteristics are derived by integrating the spatial relationship and temporal correlation changes of construction segments. The spatial relationship is the connection relationship between nodes in the construction segments (adjacency matrix), and the temporal correlation is the temporal changes of construction parameters (e.g., support stress and deformation rate).

[0077] Specifically, the surrounding rock stability coefficient reflects the stability of the surrounding rock under the current ground stress state, which can be calculated using the following formula:

[0078]

[0079] in, is the uniaxial compressive strength of rock, the maximum stress at which the rock sample reaches failure in a uniaxial compression test, reflecting the rock's compressive capacity. is the von Mises equivalent stress, which is calculated based on the three-dimensional stress state of the surrounding rock ( , , ) calculated equivalent stress, s vm = 1 2 [ ( s 1 − s 2 ) 2 + ( s 2 − s 3 ) 2 + ( s 3 − s 1 ) 2 ] , G is the elastic modulus, which is the stress-strain proportional coefficient of the rock in the elastic deformation stage, characterizing the material stiffness, E is the shear modulus, the ability of the rock to resist shear deformation, which is related to the elastic modulus and Poisson's ratio (v), In this embodiment, the stiffness ratio is introduced , considering both compressive strength and rock brittleness, which is more consistent with the rock burst mechanism.

[0080] Specifically, the seepage-stress coupling factor is used to quantify the effect of groundwater seepage on surrounding rock stress, i.e., static water pressure calculation, which can predict the critical conditions of hydraulic fracturing and avoid water inrush accidents. Its governing equation is as follows:

[0081]

[0082] in, is the permeability, which is the ability of the rock mass to allow water to flow through it, and is related to porosity and fracture connectivity; is the hydraulic head, the total energy of the groundwater level; is the water storage rate, that is, the amount of water released by the rock mass when the unit head decreases, reflecting the elastic water storage capacity of the aquifer; is the source / sink term, which is the amount of water flowing in / out per unit volume per unit time (such as rainfall infiltration or drainage); is the stress tensor, and its physical meaning is the normal stress and shear stress distributed inside the rock mass. is the density of water, that is, the density of groundwater (usually taken as 1000 kg / m³); is the acceleration due to gravity (usually taken as 9.81 m / s²).

[0083] Specifically, the construction disturbance index represents the disturbance intensity of the excavation activity on the stratum, which can be calculated using the following formula:

[0084]

[0085] in, The daily advance is the length of tunnel excavated every day, reflecting the construction speed; is the cutterhead thrust, i.e. the total thrust of the tunnel boring machine (TBM) cutterhead acting on the tunnel face; is the elastic modulus of the surrounding rock, which characterizes the stiffness of the surrounding rock. The lower the elastic modulus (soft rock), the greater the deformation under the same thrust. The comprehensive construction speed and thrust reflect the instantaneous disturbance intensity of mechanical excavation on the surrounding rock. The greater the stiffness of the surrounding rock in the denominator D, the more it can resist deformation and the smaller the disturbance effect. When it is >0.5 (empirical threshold), it is considered that the construction disturbance exceeds the self-stabilizing capacity of the surrounding rock, and support reinforcement measures need to be initiated.

[0086] Through the above parameter definitions, the physical mechanism characteristics link the geological conditions and environmental effects during tunnel construction with the depth of construction activities, providing quantitative and interpretable input indicators for dynamic risk assessment. For example, in a deep tunnel project, when the surrounding rock stability coefficient S dropped from 1.2 to 0.9, the system automatically triggered a rockburst warning. The seepage-stress coupling model predicted that the pore water pressure in a fault zone would rise to 2.1 MPa after heavy rain, exceeding the support design limit (1.8 MPa), prompting the initiation of pre-grouting. After the construction disturbance index D reached 0.68, the system recommended reducing the daily advance from 4 meters to 2.5 meters. Actual monitoring showed a 42% reduction in the surrounding rock deformation rate.

[0087] Specifically, spatiotemporal features can be constructed through graph convolutional networks (GCNs). The construction section spatial relationship graph is a topological structure that represents the spatial correlation between different sections of the tunnel, and is used to model the mutual influence between construction sections (such as deformation propagation and risk diffusion). The tunnel is divided into multiple construction sections according to a certain length (such as 5m or 10m), and each section is a node; the edge weights of the nodes can be based on the following rules: physical proximity: adjacent construction sections are automatically connected; geological similarity: sections with the same surrounding rock grade or similar permeability coefficient are connected; construction impact range: connected according to the propagation distance of the construction disturbance (such as the blasting impact radius of 20m); the node feature is a multidimensional attribute vector of each node, such as geological, construction, and environmental data. Then construct the spatial adjacency matrix A, where A ij represents the connection weight between nodes i and j. The support stress time series and the spatial adjacency matrix A are then input into the graph convolutional network to obtain spatiotemporal features. The support stress time series can be obtained by installing fiber Bragg grating sensors or resistance strain gauges on tunnel support structures (such as steel arches and anchors) for real-time monitoring. The acquisition frequency is typically set to a high-frequency sampling of 1 to 10 Hz to capture dynamic changes.

[0088] For example, the 50-meter radius around the fault zone is divided into 20 5-meter segments (nodes 21-40). Physically adjacent edges have a weight of 1.0 for adjacent segments; geologically similar edges connect segments within the fault zone (nodes 25-35) with a weight of 0.8. Node characteristics are surrounding rock grade (IV→V), permeability (1e-5→3e-5 m / s), and daily advance (2.1→0.8 m / d). A three-day (72-hour) time series of support stresses within the fault zone segment, along with the spatial adjacency matrix A, is fed into a graph convolutional network. The network extracts the stress propagation pattern within the fault zone and discovers that stress spikes occur during the daily peak of tunneling. The resulting spatiotemporal features indicate that the stress anomaly propagates from node 28 to node 32 at a rate of 1.5 m / h.

[0089] S1042: Calculate the probability of an accident occurring in the target tunnel based on physical characteristics and spatiotemporal characteristics.

[0090] Specifically, the calculating the probability of an accident occurring in the target tunnel through the physical characteristics and the spatiotemporal characteristics includes: calculating a first accident probability and a second accident probability through the physical characteristics and the spatiotemporal characteristics respectively; and calculating the probability of an accident occurring in the target tunnel based on the first accident probability and the second accident probability.

[0091] More specifically, calculating the probability of an accident occurring in the target tunnel using the physical characteristics and the spatiotemporal characteristics includes: fusing the physical characteristics and the spatiotemporal characteristics to obtain a fused feature; and calculating the probability of an accident occurring in the target tunnel based on the fused feature. Specifically, the fused feature is input into a tunnel accident prediction model to obtain the probability of an accident occurring in the target tunnel. The tunnel accident prediction model is a network model trained based on the physical characteristics and spatiotemporal characteristics.

[0092] S105: Determine the risk level of the target tunnel according to the probability of an accident occurring in the target tunnel.

[0093] In this embodiment, the risk level of the target tunnel may be determined based on a predefined correspondence between an accident probability range and a risk level.

[0094] This embodiment provides a special risk level assessment method for tunnel construction. First, attribute data of a target tunnel at a current point in time is obtained, where the attribute data includes tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data. Then, based on the attribute data, similar tunnels whose similarity to the target tunnel exceeds a predetermined value are obtained from a preset database. The changing trends of the attribute data of the similar tunnels and the target tunnel over historical time are compared to determine the weight values ​​corresponding to the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel. Then, the probability of an accident occurring in the target tunnel is calculated based on the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel and their corresponding weight values. Finally, the risk level of the target tunnel is determined based on the probability of an accident occurring in the target tunnel. Compared with the current tunnel risk assessment methods which are mostly based on existing geological exploration data and traditional empirical formulas, this application obtains the attribute data of the target tunnel and compares and matches it with the tunnels in the preset database to determine the weight values ​​corresponding to the attribute data, and then calculates the probability of an accident in the tunnel based on the attribute data and its corresponding weight values. Since the attribute data in this application can reflect the complexity of geological conditions and the dynamic change characteristics during the construction process, this application can improve the accuracy of tunnel construction risk assessment.

[0095] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0096] In one embodiment, a special risk level assessment system for tunnel construction is provided. Figure 3 As shown, the functional modules of the special risk level assessment system for tunnel construction are described in detail as follows:

[0097] An acquisition module 31 is used to acquire attribute data of the target tunnel at a current time point, wherein the attribute data includes tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data;

[0098] The acquisition module 31 is further configured to acquire similar tunnels having a similarity with the target tunnel exceeding a predetermined value from a preset database based on the attribute data;

[0099] a comparison module 32 for comparing the historical change trends of the attribute data of the similar tunnel and the target tunnel, and determining the weight values ​​corresponding to the tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the target tunnel;

[0100] A calculation module 33 is configured to calculate the probability of an accident occurring in the target tunnel based on the tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data, and their corresponding weight values ​​in the attribute data of the target tunnel;

[0101] The determination module 34 is configured to determine the risk level of the target tunnel according to the probability of an accident occurring in the target tunnel.

[0102] In an optional embodiment, the acquisition module 31 is specifically configured to:

[0103] Calculating similarity values ​​between the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data of the target tunnel and the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data of each tunnel in the preset database;

[0104] Performing weighted calculation on the calculated similarity values ​​to obtain the similarity of each tunnel in the preset database;

[0105] Tunnels in the preset database whose similarity exceeds a predetermined value are determined as similar tunnels.

[0106] In an optional embodiment, the comparison module 32 is specifically configured to:

[0107] Obtaining, from the preset database, influencing factors corresponding to tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the similar tunnel;

[0108] Based on the influencing factors and the historical change trends of the attribute data of the similar tunnels and the target tunnel, the weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel are determined.

[0109] In an optional embodiment, the determination module 34 is further configured to:

[0110] Drawing tunnel geological data curves, tunnel environment data curves, tunnel design data curves, and tunnel construction data curves respectively according to the attribute data of the similar tunnels and the target tunnel;

[0111] Comparing the tunnel geological data curve graph, the tunnel environment data curve graph, the tunnel design data curve graph, and the tunnel construction data curve graph of the similar tunnel and the target tunnel respectively to obtain a first correction value, a second correction value, a third correction value, and a fourth correction value;

[0112] The influencing factors are corrected by using the first correction value, the second correction value, the third correction value, and the fourth correction value to obtain weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel.

[0113] In an optional embodiment, the determination module 34 is specifically configured to:

[0114] Comparing the tunnel geological data curve graphs, tunnel environment data curve graphs, tunnel design data curve graphs, and tunnel construction data curve graphs of the similar tunnel and the target tunnel, respectively, to obtain comparison curve results of the respective curve graphs, wherein the comparison curve results include a curve slope difference, a curve maximum difference, and a curve minimum difference;

[0115] The first correction value, the second correction value, the third correction value, and the fourth correction value are calculated according to the curve slope difference, the curve maximum difference, and the curve minimum difference.

[0116] In an optional embodiment, the calculation module 33 is configured to:

[0117] The curve slope difference, curve maximum difference, and curve minimum difference corresponding to the tunnel geological data curve graph, the tunnel environment data curve graph, the tunnel design data curve graph, and the tunnel construction data curve graph are input into the correction value prediction model to obtain the first correction value, the second correction value, the third correction value, and the fourth correction value.

[0118] In an optional embodiment, the calculation module 33 is configured to:

[0119] Determining physical characteristics and spatiotemporal characteristics based on tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data and their corresponding weight values ​​in the attribute data of the target tunnel;

[0120] The probability of an accident occurring in the target tunnel is calculated based on the physical characteristics and the spatiotemporal characteristics.

[0121] In an optional embodiment, the calculation module 33 is configured to:

[0122] Calculating a first accident probability and a second accident probability respectively according to the physical characteristics and the spatiotemporal characteristics;

[0123] The probability of an accident occurring in the target tunnel is calculated according to the first accident probability and the second accident probability.

[0124] In an optional embodiment, the calculation module 33 is configured to:

[0125] Performing feature fusion on the physical feature and the spatiotemporal feature to obtain a fused feature;

[0126] The probability of an accident occurring in the target tunnel is calculated based on the fusion features.

[0127] It should be noted that the above detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs.

[0128] The specific limitations of the specialized risk level assessment system for tunnel construction can be found in the limitations of the specialized risk level assessment method for tunnel construction described above and will not be further elaborated here. Each module in the aforementioned device may be implemented in whole or in part via software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in hardware form, or may be stored in a computer device memory in software form, allowing the processor to call and execute operations corresponding to each module.

[0129] Those skilled in the art will clearly understand that for the sake of convenience and brevity in description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0130] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A special risk level assessment method for tunnel construction, characterized in that: The method comprises: Acquire attribute data of the target tunnel at the current time point, wherein the attribute data includes tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data; Acquire similar tunnels having a similarity with the target tunnel exceeding a predetermined value from a preset database based on the attribute data; Comparing the historical change trends of the attribute data of the similar tunnel and the target tunnel, determining the weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel; Calculating the probability of an accident occurring in the target tunnel based on tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data, and their corresponding weight values ​​in the attribute data of the target tunnel; The risk level of the target tunnel is determined according to the probability of an accident occurring in the target tunnel.

2. The method according to claim 1, characterized in that The acquiring, from a preset database based on the attribute data, similar tunnels having a similarity with the target tunnel exceeding a predetermined value, includes: Calculating similarity values ​​between the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data of the target tunnel and the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data of each tunnel in the preset database; Performing weighted calculation on the calculated similarity values ​​to obtain the similarity of each tunnel in the preset database; Tunnels in the preset database whose similarity exceeds a predetermined value are determined as similar tunnels.

3. The method according to claim 2, characterized in that The comparing the historical change trends of the attribute data of the similar tunnel and the target tunnel, and determining the weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel, respectively, includes: Obtaining, from the preset database, influencing factors corresponding to tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data in the attribute data of the similar tunnel; Based on the influencing factors and the historical change trends of the attribute data of the similar tunnels and the target tunnel, the weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel are determined.

4. The method according to claim 3, characterized in that The determining, based on the influencing factors and the historical change trends of the attribute data of the similar tunnels and the target tunnel, weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel, respectively, includes: Drawing tunnel geological data curves, tunnel environment data curves, tunnel design data curves, and tunnel construction data curves respectively according to the attribute data of the similar tunnels and the target tunnel; Comparing the tunnel geological data curve graph, the tunnel environment data curve graph, the tunnel design data curve graph, and the tunnel construction data curve graph of the similar tunnel and the target tunnel respectively to obtain a first correction value, a second correction value, a third correction value, and a fourth correction value; The influencing factors are corrected by using the first correction value, the second correction value, the third correction value, and the fourth correction value to obtain weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel.

5. The method according to claim 4, characterized in that The step of respectively comparing the tunnel geological data graphs, the tunnel environment data graphs, the tunnel design data graphs, and the tunnel construction data graphs of the similar tunnel and the target tunnel to obtain a first correction value, a second correction value, a third correction value, and a fourth correction value includes: Comparing the tunnel geological data curve graphs, tunnel environment data curve graphs, tunnel design data curve graphs, and tunnel construction data curve graphs of the similar tunnel and the target tunnel, respectively, to obtain comparison curve results of the respective curve graphs, wherein the comparison curve results include a curve slope difference, a curve maximum difference, and a curve minimum difference; The first correction value, the second correction value, the third correction value, and the fourth correction value are calculated according to the curve slope difference, the curve maximum difference, and the curve minimum difference.

6. The method according to claim 5, characterized in that The calculating of the first correction value, the second correction value, the third correction value, and the fourth correction value by using the curve slope difference, the curve maximum difference, and the curve minimum difference includes: The curve slope difference, curve maximum difference, and curve minimum difference corresponding to the tunnel geological data curve graph, the tunnel environment data curve graph, the tunnel design data curve graph, and the tunnel construction data curve graph are input into the correction value prediction model to obtain the first correction value, the second correction value, the third correction value, and the fourth correction value.

7. The method according to claim 2, characterized in that The calculating the probability of an accident occurring in the target tunnel according to the tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data and their corresponding weight values ​​in the attribute data of the target tunnel includes: Determining physical characteristics and spatiotemporal characteristics based on tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data and their corresponding weight values ​​in the attribute data of the target tunnel; The probability of an accident occurring in the target tunnel is calculated based on the physical characteristics and the spatiotemporal characteristics.

8. The method according to claim 7, characterized in that Calculating the probability of an accident occurring in the target tunnel based on the physical characteristics and the spatiotemporal characteristics includes: Calculating a first accident probability and a second accident probability respectively according to the physical characteristics and the spatiotemporal characteristics; The probability of an accident occurring in the target tunnel is calculated according to the first accident probability and the second accident probability.

9. The method according to claim 7, characterized in that Calculating the probability of an accident occurring in the target tunnel based on the physical characteristics and the spatiotemporal characteristics includes: Performing feature fusion on the physical feature and the spatiotemporal feature to obtain a fused feature; The probability of an accident occurring in the target tunnel is calculated based on the fusion features.

10. A special risk level assessment system for tunnel construction, characterized in that: The system comprises: An acquisition module is used to acquire attribute data of the target tunnel at the current time point, wherein the attribute data includes tunnel geological data, tunnel environment data, tunnel design data, and tunnel construction data; The acquisition module is further configured to acquire, from a preset database based on the attribute data, similar tunnels having a similarity with the target tunnel exceeding a predetermined value; a comparison module, configured to compare the historical change trends of the attribute data of the similar tunnel and the target tunnel, and determine the weight values ​​corresponding to the tunnel geological data, the tunnel environment data, the tunnel design data, and the tunnel construction data in the attribute data of the target tunnel; a calculation module, configured to calculate the probability of an accident occurring in the target tunnel based on the tunnel geological data, tunnel environment data, tunnel design data, tunnel construction data, and their corresponding weight values ​​in the attribute data of the target tunnel; The determination module is configured to determine the risk level of the target tunnel according to the probability of an accident occurring in the target tunnel.

Images