A method for evaluating the safety of an operating railway tunnel structure based on double-drive fusion

Abstract

This invention discloses a method for evaluating the structural safety of operating railway tunnels based on dual-drive fusion. The method includes: acquiring tunnel structural state parameters; filtering these parameters based on the correlation between data to obtain typical tunnel structural state parameters; dividing these typical parameters into two parts: the first part serves as input to a numerical simulation model, and the second part serves as input to a multi-source data model; using a portion of the numerical simulation model's output as input to the multi-source data model; evaluating the target tunnel based on the numerical simulation model and the multi-source data model; outputting evaluation indicators for tunnel structural stability, structural functionality, and material durability; and rating the target tunnel based on these evaluation indicators. This method achieves a hybrid fusion of the multi-source data-driven model and the numerical simulation model, while combining series, parallel, and embedded modes to overcome the defects or shortcomings of series fusion, parallel fusion, and embedded fusion.

Description

Technical Field

[0001] This invention relates to the field of tunnel engineering structural safety technology, and more specifically to a method for evaluating the structural safety of operating railway tunnels based on dual-drive fusion. Background Technology

[0002] With the increasing scale of railway tunnel construction and operation, the safety performance of tunnel structures will gradually decrease due to external environmental factors and dynamic loads (vibration loads, aerodynamic loads) if they are not properly maintained. When the safety performance of a railway tunnel falls below the critical state, safety accidents may occur at any time during tunnel operation, leading to a series of adverse consequences. Accurately identifying the current risk factors of operating railway tunnel structures, judging their current state, and evaluating their safety performance are of significant guiding value for the formulation of maintenance plans and the reporting of maintenance projects during tunnel operation and maintenance.

[0003] Therefore, accurately identifying the current risk factors of operating railway tunnel structures, judging the current state of tunnel structures, and evaluating the safety performance of tunnel structures are problems that urgently need to be solved by those skilled in the art. Summary of the Invention

[0004] In view of the above problems, the present invention provides a method for safety evaluation of operating railway tunnel structures based on dual-drive fusion to overcome or at least partially solve the above problems.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] This invention provides a method for structural safety evaluation of operating railway tunnels based on dual-drive fusion, comprising the following steps: S1. Obtain tunnel structural state parameters, which include basic data and monitoring data. The basic data includes geological conditions, structural design parameters, and construction process data. The monitoring data includes structural apparent defects data, dynamic stress data, and environmental impact data. S2. Based on the correlation between the data, the tunnel structure state parameters are filtered to obtain typical tunnel structure state parameters; S3. Divide the typical parameters of the tunnel structure state into two parts. The first part is used as the input of the numerical simulation model, and the second part is used as a part of the input of the pre-constructed multi-source data model. Use part of the output of the numerical simulation model as the other part of the input of the multi-source data model, and perform iterative training on the multi-source data model. S4. Test the optimal multi-source data model obtained by iterative training using the state parameters of tunnels that were not involved in training; S5. Based on the numerical simulation model and the trained optimal multi-source data model, evaluate the target tunnel, output evaluation indicators for tunnel structural stability, structural functionality and material durability, and rate the target tunnel according to the evaluation indicators.

[0007] Furthermore, the specific method for filtering the tunnel structure state parameters based on the correlation between data in step S2 is as follows: Calculate the Pearson correlation coefficient between any two tunnel structural state parameters. If the absolute value of the Pearson correlation coefficient is greater than 0.8, the two parameters are determined to be strongly correlated, and one of the parameters is retained as a typical parameter of the tunnel structural state.

[0008] Furthermore, the typical parameters of the first part of the tunnel structure status in step S3 include tunnel burial depth, special geology, structural geometry, structural parameters, structural material parameters, dynamic load, and the water level difference between groundwater and the arch. The typical parameters of the tunnel structure in the second part include structural deformation, structural cracks, dynamic loads, and the difference in water level between groundwater and the arch.

[0009] Furthermore, in step S3, the numerical simulation model is constructed using finite element software. The model construction and analysis process is as follows: Construct a calculation model, wherein the width of the calculation model is 3-5 times the tunnel clearance, and the tunnel arch is 2-3 times the tunnel clearance. Assign corresponding mechanical parameters to the surrounding rock, including cohesion, elastic modulus, and internal friction angle; Set the boundary conditions for the model: the top is a free boundary, and the left, right, front, back, and bottom are fixed constraint boundaries. Calculate the initial geostress of the model; The construction of the tunnel structure assigns corresponding mechanical parameters to the structure. Simulate tunnel structural defects or problems; To achieve a state of relative equilibrium between the surrounding rock and the structure; Obtain the internal forces of the tunnel structure and establish the mapping relationship between the internal forces of the tunnel structure and the input parameter dynamic load.

[0010] Furthermore, in step S3, during the iterative training of the multi-source data model, the mapping relationship between structural internal forces and dynamic loads obtained from the numerical simulation model analysis is used as the boundary constraint condition for model training.

[0011] Furthermore, the multi-source data model is a BP neural network model, and the output parameters of the multi-source data model include structural water leakage, concrete carbonation depth, and steel reinforcement corrosion depth.

[0012] Furthermore, the rating criteria for the evaluation indicators in step S5 are as follows: Structural stability rating is based on the ratio of internal forces in the structure under operating conditions to internal forces in the structure under design conditions: 0.75-1.0 is Level I, 0.50-0.75 is Level II, 0.25-0.50 is Level III, and 0-0.25 is Level IV. The structural functionality rating is based on the daily cumulative leakage of a single leak in the tunnel structure: 0-1.0 is Level I, 1.0-3.0 is Level II, 3.0-5.0 is Level III, and 5.0-+∞ is Level IV. Material durability rating is based on the ratio of concrete carbonation depth and steel corrosion depth to structural thickness and steel diameter, respectively: 0-0.25 is Grade I, 0.25-0.50 is Grade II, 0.50-0.75 is Grade III, and 0.75-1.0 is Grade IV.

[0013] Furthermore, the internal forces of the structure include axial force and bending moment, which are used as input parameters of the multi-source data model for iterative training.

[0014] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a method for evaluating the structural safety of operating railway tunnels based on dual-drive fusion, which has the following beneficial effects: 1. This invention integrates a multi-source data-driven model with a numerical simulation model, which combines three modes: serial, parallel and embedded. (1) The model input parameters are divided into two parts: one part is used as the input of the multi-source data model and the other part is used as the input of the numerical simulation model. The two parts have common parameters and the two models are analyzed in parallel, corresponding to the parallel fusion mode. (2) Part of the output results of the numerical simulation model are used as the input parameters of the multi-source data model, corresponding to the serial fusion mode. (3) The relationship between the input and output parameters obtained from the numerical simulation model is used as the boundary condition of the multi-source data model, thereby restricting the relationship between some parameters within the model, corresponding to the embedded fusion mode, thus overcoming the defects or deficiencies of serial fusion, parallel fusion and embedded fusion.

[0015] 2. This invention prioritizes and hierarchically analyzes three indicators of railway tunnels: structural stability, structural functionality, and material durability. It clarifies that tunnel structural stability is the prerequisite or foundation for evaluating and analyzing tunnel structural functionality and material durability. Based on the evaluation results, it provides corresponding handling strategies to ensure the structural safety of railway tunnels and ultimately achieve full-life and long-life operation and maintenance of railway tunnels. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0017] Figure 1 This is a flowchart of the method for evaluating the structural safety of operating railway tunnels provided in this embodiment of the invention; Figure 2 This is a schematic diagram of the structural safety evaluation method for operating railway tunnels provided in this embodiment of the invention. Figure 3 This is a schematic diagram of BP neural network prediction provided in an embodiment of the present invention. Detailed Implementation

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

[0019] This invention discloses a method for structural safety evaluation of operating railway tunnels based on dual-drive fusion, such as... Figure 1 As shown, it includes the following steps: S1. Obtain tunnel structural state parameters, which include basic data and monitoring data. The basic data includes geological conditions, structural design parameters, and construction process data. The monitoring data includes structural apparent defects data, dynamic stress data, and environmental impact data. S2. Based on the correlation between the data, the tunnel structure state parameters are filtered to obtain typical tunnel structure state parameters; S3. Divide the typical parameters of the tunnel structure state into two parts. The first part is used as the input of the numerical simulation model, and the second part is used as a part of the input of the pre-constructed multi-source data model. Use part of the output of the numerical simulation model as the other part of the input of the multi-source data model, and perform iterative training on the multi-source data model. S4. Test the optimal multi-source data model obtained by iterative training using the state parameters of tunnels that were not involved in training; S5. Based on the numerical simulation model and the trained optimal multi-source data model, evaluate the target tunnel, output evaluation indicators for tunnel structural stability, structural functionality and material durability, and rate the target tunnel according to the evaluation indicators.

[0020] This method prioritizes and hierarchically considers the structural stability, structural functionality, and material durability of railway tunnels. It clarifies that the structural stability of a tunnel is the prerequisite or foundation for evaluating and analyzing its structural functionality and material durability. Based on the evaluation results, it provides corresponding handling strategies to ensure the structural safety of railway tunnels and ultimately achieve full-life and long-life operation and maintenance of railway tunnels.

[0021] The following is a detailed description of each of the above steps, such as... Figure 2 The schematic diagram is shown below: S1. Obtaining tunnel structural state parameters; The main ways to obtain parameters of tunnel structural status include basic data and monitoring data. Monitoring data includes: regular inspections, special monitoring, operational monitoring, emergency inspections, and special inspections.

[0022] The basic data mainly includes geological conditions, structural design parameters, and construction process data; Geological conditions include the elastic modulus, Poisson's ratio, and cohesion of the surrounding rock, while construction process data includes tunnel depth, tunnel length, structural geometric dimensions, structural parameters, and structural material parameters. Regular inspections mainly include structural surface defects, such as structural cracks, structural water leakage, structural surface peeling, and structural crushing. Specialized monitoring mainly includes structural defects or problems, such as insufficient structural thickness, voids behind the structure, concrete carbonization, and steel corrosion. Operational monitoring mainly includes structural deformation or stress state, including structural deformation or displacement, clearance convergence, and contact pressure between surrounding rock and structure; Emergency and special inspections mainly include special structural conditions or sudden changes in the external environment, such as structural collapse, sudden changes in dynamic loads, and surges in groundwater.

[0023] S2, Typical Data Parameter Filtering; Because the tunnel structure state parameters obtained in S1 are numerous and large in volume, and some data are correlated, directly inputting them into the model would inevitably increase the model's analytical workload, reduce its generalization ability, and increase the probability of overfitting.

[0024] If a numerical simulation model is directly input, it is difficult to find software that can simultaneously simulate all defects or flaws. For example, finite element software has a fast analysis speed and high mesh generation quality, but its ability to simulate defects such as cracks is insufficient; discrete element software can simulate defects such as cracks well, but the analysis accuracy is significantly affected by the number of meshes, and its ability to simulate contact is poor; block analysis software can simulate the contact problems of surrounding rock or structure well, but it cannot simulate small cracks.

[0025] In summary, it is necessary to select typical data from the tunnel structural state parameters. The specific selection process is as follows: The correlation between two parameters is determined by calculating the Pearson correlation coefficient between each pair of parameters. If the absolute value of the Pearson correlation coefficient is greater than 0.8, the two parameters are considered to be strongly correlated, and only one parameter can be selected.

[0026] The calculation process for the Pearson correlation coefficient is as follows: 1. Calculate the covariance of two variables. Covariance indicates whether the overall trends of change between two variables are consistent, and can be calculated using the following formula: cov(X,Y)=E[(XE(X))(YE(Y))].

[0027] 2. Calculate the standard deviation of the two variables. The standard deviation represents the dispersion of a variable and can be calculated using the following formulas: std(X)=sqrt(E[(XE(X))^2]), std(Y)=sqrt(E[(YE(Y))^2]).

[0028] 3. Dividing the covariance by the product of the standard deviations of the two variables yields the Pearson correlation coefficient: r = cov(X,Y) / (std(X)). std(Y)).

[0029] 4. If the value of r is close to 1, it indicates a strong positive correlation between the two variables. If the value of r is close to -1, it indicates a strong negative correlation between the two variables. If the value of r is close to 0, it indicates that there is no linear correlation between the two variables.

[0030] Through the above process, typical parameters of the tunnel structure are obtained, including tunnel depth, special geological conditions, structural geometry, structural parameters, structural material parameters, structural deformation, structural cracks, structural water leakage, dynamic load, and the difference in water level between groundwater and the arch.

[0031] S3. Data analysis based on dual-drive fusion; 1. The typical parameters of the tunnel structure are divided into two parts. One part consists of structural deformation, structural cracks, dynamic load, and the water level difference between groundwater and the arch, which are used as the model input for multi-source data. The other part consists of tunnel depth, special geological conditions, structural geometry, structural parameters, structural material parameters, dynamic loads, and the water level difference between groundwater and the arch, which serve as inputs for the numerical simulation model.

[0032] 2. Data analysis based on numerical simulation models: Finite difference software is used for simulation analysis, which can effectively simulate minor defects or damage such as structural cracks. The analysis steps are as follows: (1) A three-dimensional numerical software was used to construct a calculation model. The model size width was 3-5 times the tunnel clearance to eliminate the boundary effect of the model size. The tunnel arch was taken as the actual burial depth of the tunnel and the tunnel arch was taken as 2-3 times the tunnel clearance. (2) Assign corresponding mechanical parameters to the surrounding rock, such as the cohesion, elastic modulus and internal friction angle of the surrounding rock; (3) Set the boundary conditions of the model. Set the top of the model as a free boundary, and set the left, right, front, back and bottom as fixed constraint boundaries. (4) Initial model calculations to obtain initial ground stress; model operation until unbalanced force is less than 10. -3 ; (5) Construction of tunnel structure and assignment of corresponding mechanical parameters to the structure, such as lining structure, invert arch structure, etc.; (6) Simulation of tunnel structural defects or diseases, such as structural cracks, insufficient structural thickness, voids behind the structure, etc. (7) After the tunnel structure and its defects or defects are repaired, the surrounding rock and the structure interact and reach a relatively balanced state. The model runs until the unbalanced force is less than 10. -3 ; (8) Obtain the internal forces of the tunnel structure, such as axial force and bending moment, and establish the mapping relationship between the internal forces of the structure and the input parameter dynamic load.

[0033] 3. For example Figure 3 As shown, the data analysis based on the multi-source data model uses the output results of the numerical simulation model, including the structural axial force and structural bending moment, as the input of the multi-source data model. In this embodiment, a typical BP neural network is selected as the multi-source data model. The model output is the leakage of water in the structure, the carbonation depth of the concrete material, and the degree of corrosion of the steel bars.

[0034] letter x Representing numerical values, such as x 1 represents the deformation value of the tunnel structure; x 2 represents the crack value of the tunnel structure; x 3 represents external dynamic load; and so on.

[0035] w represents the weight value, such as w 1 The weight representing the deformation value of the tunnel structure, w 2 The weight representing the crack value of the tunnel structure, w 3 This represents the weight of the external dynamic load; and so on.

[0036] b Represents bias. i This represents a variable, and its values ​​can be 1, 2, 3, 4, 5, 6, ...

[0037] Meanwhile, during the training process of the multi-source data model, the mapping relationship between the internal forces of the structure and the dynamic loads obtained from the numerical simulation model analysis is used as the boundary condition.

[0038] S4. Test the optimal multi-source data model obtained by iterative training using the state parameters of tunnels that were not involved in training; S5. Evaluation of the safety performance of tunnel structures; 1. Based on the classification results of dual-drive fusion data analysis, tunnels are divided according to three classification indicators: structural stability, structural functionality, and material durability. Among them, structural stability indicators include axial force and bending moment; structural functionality indicators include structural water leakage; and material durability indicators include concrete carbonation depth and steel reinforcement corrosion degree.

[0039] 2. The evaluation index is determined by classifying the internal forces (axial force and bending moment) of the tunnel under design conditions as the benchmark, and the ratio of the internal forces calculated under operational conditions to those under design conditions is used as the evaluation index for structural stability, as shown in the table below:

[0040] Table 1. Evaluation criteria for structural stability The daily cumulative leakage volume of a single leak in a tunnel structure is used as the functional evaluation standard for the tunnel structure, as shown in the table below:

[0041] Table 2. Evaluation Criteria for Structural Functionality The ratios of concrete carbonation depth and steel reinforcement corrosion depth under operational conditions to structural thickness and steel reinforcement diameter, respectively, are used as evaluation indicators of material durability, as shown in the table below:

[0042] Table 3. Evaluation Criteria for Material Durability 3. Evaluation of tunnel structure safety performance: The stability of the tunnel structure is evaluated. If the risk to the structural stability is high and maintenance work is required, the structural functionality and material durability are not evaluated.

[0043] Secondly, structural functionality is evaluated only after structural stability meets the requirements; finally, material durability is evaluated only after structural functionality meets the requirements.

[0044] S6. Selection of tunnel structure treatment strategies; Corresponding maintenance strategies were developed for different condition levels. For example, when the evaluation result indicates a condition level of II, the safety risk is high, requiring enhanced monitoring and routine maintenance; similar strategies apply to other condition levels. Details are shown in the table below:

[0045] Table 4. Maintenance strategies corresponding to different condition levels Example: Taking a tunnel in a certain location as an example; Step 1: Obtaining tunnel structural state parameters; Contact the project's tunnel construction unit, surveying unit, design and construction unit, supervision unit, government departments, quality supervision station, and third-party service providers to collect non-confidential materials, such as project approval documents, survey reports, design drawings, construction plans, supervision plans, research completion reports, and third-party service materials. Compile parameters characterizing the tunnel's structural condition from these materials. These include: elastic modulus of the surrounding rock, Poisson's ratio, cohesion, tunnel depth, tunnel length, structural geometric dimensions, structural parameters, structural material parameters, structural cracks, structural water leakage, structural surface spalling, structural crushing, insufficient structural thickness, cavities behind the structure, concrete carbonization, steel corrosion, structural deformation or displacement, clearance convergence, contact pressure between the surrounding rock and the structure, structural spalling, sudden changes in dynamic loads, and surges in groundwater levels.

[0046] Step 2: Screening typical data parameters; Typical parameters of the tunnel structure include tunnel depth, special geological conditions, structural geometry, structural parameters, structural material parameters, structural deformation, structural cracks, structural leakage, dynamic load, and the difference in water level between groundwater and the arch. Specific values ​​for each typical parameter are as follows: tunnel depth 400 meters, no special geological conditions, single-circle cross-section with a net width of 8.8m and a net height of 7.25m, composite lining structure, initial support of 10cm thick anchor-sprayed support, secondary lining of 0.25m thick C30 reinforced concrete with 50mm diameter steel bars, surrounding rock grade IV mudstone, elastic modulus of 20GPa, cohesion of 3.58kPa, internal friction angle of 35 degrees, tunnel has been in operation for 50 years, maximum crack width is 0.8m, maximum dynamic load is 100-230kN, and the difference in water level between groundwater and the arch is 15m.

[0047] Step 3: Data analysis based on dual-drive fusion; 1. The typical parameters of the tunnel structure are divided into two parts. One part consists of structural deformation, structural cracks, dynamic load, and the water level difference between groundwater and the arch, which serve as the model input for multi-source data. The other part consists of tunnel depth, special geological conditions, structural geometry, structural parameters, structural material parameters, dynamic load, and the water level difference between groundwater and the arch, which serve as the input for the numerical simulation model.

[0048] 2. Data analysis based on numerical simulation models; Numerical analysis of an operating railway tunnel was conducted using the finite difference software ABAQUS. The analysis results show that when the dynamic load is 200.0 kN, the axial force and bending moment of the structure are 350 kN and 520 kN·m, respectively. Furthermore, a significant correlation exists between the axial force, bending moment, and external dynamic load of the tunnel structure. This relationship can be fitted using a power-law function with a coefficient of determination of approximately 0.95, expressed by the following formula:

[0049] In the formula: x represents the dynamic load, y represents the axial force of the tunnel structure, and z represents the bending moment of the tunnel structure.

[0050] 3. Data analysis based on multi-source data models; The axial force and bending moment of the tunnel structure, respectively, of 350 kN and 520 kN·m, were added to the multi-source data set. Simultaneously, the axial force, bending moment, and external dynamic load of the tunnel structure were constrained to satisfy the aforementioned power-law relationship. Through BP neural network prediction and analysis, the maximum daily seepage volume at a single point in the tunnel structure was determined to be 2.5 m³. 3 On average, the carbonation depth of the structural concrete is 0.125m, and the corrosion depth of the reinforcing steel is 30.0mm. Step 4: Safety performance evaluation of the tunnel structure; If the secondary lining is made of 0.25m thick C30 reinforced concrete, then the axial force of the tunnel structure under design conditions is 3500kN and the bending moment is 5500kN·m.

[0051] The stability evaluation index for the tunnel structure is: the axial force evaluation index is 350 / 3500=0.1; The bending moment evaluation index is 520 / 5500=0.095.

[0052] The functional evaluation index for the tunnel structure is 2.5m. 3 / day; The material durability indices for the tunnel structure are as follows: the carbonation depth of the structural concrete is 0.125 / 0.25=0.5; The corrosion depth of the reinforcing steel is 30.0 / 50=0.6.

[0053] Step 5: Selection of tunnel structure treatment strategy.

[0054] According to the analysis results in step four, the ratios of axial force and bending moment of the tunnel structure under operating conditions to those under design conditions are 0.1 and 0.095, respectively, indicating that the stability level of the tunnel structure is Level I, the safety risk is negligible, the maintenance strategy is daily operation, and the maintenance measures are daily inspections.

[0055] Further analysis revealed that the functional evaluation index for the tunnel structure was a water leakage rate of 2.5m³. 3 / day indicates that the functional status of the tunnel structure is Level II, the safety risk is moderate, the maintenance strategy is to strengthen monitoring, and the maintenance measures are routine maintenance.

[0056] Finally, the material durability evaluation index of the tunnel structure is that the ratio of concrete carbonation depth and steel corrosion depth to structural thickness and steel diameter is 0.5 and 0.6, respectively. This indicates that the material durability status of the tunnel structure is Level III, with a high safety risk. The maintenance strategy is real-time monitoring, and the maintenance measure is to report the maintenance plan.

[0057] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0058] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for structural safety evaluation of operating railway tunnels based on dual-drive fusion, characterized in that, Includes the following steps: S1. Obtain tunnel structural state parameters, which include basic data and monitoring data. The basic data includes geological conditions, structural design parameters, and construction process data. The monitoring data includes structural apparent defects data, dynamic stress data, and environmental impact data. S2. Based on the correlation between the data, the tunnel structure state parameters are filtered to obtain typical tunnel structure state parameters; S3. Divide the typical parameters of the tunnel structure state into two parts. The first part is used as the input of the numerical simulation model, and the second part is used as a part of the input of the pre-constructed multi-source data model. Use part of the output of the numerical simulation model as the other part of the input of the multi-source data model, and perform iterative training on the multi-source data model. S4. Test the optimal multi-source data model obtained by iterative training using the state parameters of tunnels that were not involved in training; S5. Based on the numerical simulation model and the trained optimal multi-source data model, evaluate the target tunnel, output evaluation indicators for tunnel structural stability, structural functionality and material durability, and rate the target tunnel according to the evaluation indicators.

2. The method for structural safety evaluation of operating railway tunnels based on dual-drive fusion as described in claim 1, characterized in that, The specific method for filtering the tunnel structure state parameters based on the correlation between data in step S2 is as follows: Calculate the Pearson correlation coefficient between any two tunnel structural state parameters. If the absolute value of the Pearson correlation coefficient is greater than 0.8, the two parameters are determined to be strongly correlated, and one of the parameters is retained as a typical parameter of the tunnel structural state.

3. The method for structural safety evaluation of operating railway tunnels based on dual-drive fusion as described in claim 1, characterized in that, The typical parameters of the first part of the tunnel structure in step S3 include tunnel burial depth, special geology, structural geometry, structural parameters, structural material parameters, dynamic load, and the difference in water level between groundwater and the arch. The typical parameters of the tunnel structure in the second part include structural deformation, structural cracks, dynamic loads, and the difference in water level between groundwater and the arch.

4. The method for structural safety evaluation of operating railway tunnels based on dual-drive fusion as described in claim 1, characterized in that, In step S3, the numerical simulation model is constructed using finite element software. The model construction and analysis process is as follows: Construct a calculation model, wherein the width of the calculation model is 3-5 times the tunnel clearance, and the tunnel arch is 2-3 times the tunnel clearance. Assign corresponding mechanical parameters to the surrounding rock, including cohesion, elastic modulus, and internal friction angle; Set the boundary conditions for the model: the top is a free boundary, and the left, right, front, back, and bottom are fixed constraint boundaries. Calculate the initial geostress of the model; The construction of the tunnel structure assigns corresponding mechanical parameters to the structure. Simulate tunnel structural defects or problems; To achieve a state of relative equilibrium between the surrounding rock and the structure; Obtain the internal forces of the tunnel structure and establish the mapping relationship between the internal forces of the tunnel structure and the input parameter dynamic load.

5. The method for structural safety evaluation of operating railway tunnels based on dual-drive fusion as described in claim 1, characterized in that, In step S3, during the iterative training of the multi-source data model, the mapping relationship between structural internal forces and dynamic loads obtained from the numerical simulation model analysis is used as the boundary constraint condition for model training.

6. The method for structural safety evaluation of operating railway tunnels based on dual-drive fusion as described in claim 1, characterized in that, The multi-source data model is a BP neural network model, and the output parameters of the multi-source data model include structural water leakage, concrete carbonation depth, and steel reinforcement corrosion depth.

7. The method for structural safety evaluation of operating railway tunnels based on dual-drive fusion as described in claim 1, characterized in that, The rating criteria for the evaluation indicators in step S5 are as follows: Structural stability rating is based on the ratio of structural internal forces under operating conditions to structural internal forces under design conditions: 0.75-1.0 is Level I, 0.50-0.75 is Level II, 0.25-0.50 is Level III, and 0-0.25 is Level IV; The structural functionality rating is based on the daily cumulative leakage of a single leak in the tunnel structure: 0-1.0 is Level I, 1.0-3.0 is Level II, 3.0-5.0 is Level III, and 5.0-+∞ is Level IV. Material durability rating is based on the ratio of concrete carbonation depth and steel corrosion depth to structural thickness and steel diameter, respectively: 0-0.25 is Grade I, 0.25-0.50 is Grade II, 0.50-0.75 is Grade III, and 0.75-1.0 is Grade IV.

8. The method for structural safety evaluation of operating railway tunnels based on dual-drive fusion as described in claim 7, characterized in that, The internal forces of the structure include axial force and bending moment, which are used as input parameters of the multi-source data model for iterative training.

Images