Hollow slab beam bearing capacity evaluation method based on bridge technical condition index

Abstract

The application discloses a hollow slab beam bearing capacity evaluation method based on bridge technical condition indexes, which parameterizes disease index parameters influencing the bearing capacity of the bridge in the bridge technical condition evaluation, establishes finite element models under various disease working conditions, calculates deflection check coefficients under the corresponding working conditions, and takes the deflection check coefficients as bearing capacity evaluation indexes; the bridge disease indexes are taken as input parameters, the deflection check coefficients are taken as output parameters, and corresponding neural network models are established; the disease parameters in the technical condition evaluation data are inputted to realize the prediction of the deflection check coefficients of the bridge, so that the purpose of the bearing capacity evaluation of the hollow slab beam is achieved. The application can realize the connection between the bridge technical condition evaluation and the bearing capacity evaluation, makes up for the defect that the bridge technical condition evaluation cannot directly evaluate the bearing capacity of the bridge, and solves the problems of long test time, large workload, high cost, high risk, the need of interrupting traffic and the like existing in the traditional bridge static load test.

Description

Technical Field

[0001] This invention relates to the field of bridge safety inspection, specifically a method for evaluating the load-bearing capacity of hollow slab beams based on bridge technical condition indicators. The evaluation results can be used to determine the structural safety status of bridges. Background Technology

[0002] Bridges are crucial national infrastructure investment projects and key hubs in highway networks. During their service life, many bridges are subjected to erosion from various complex environments, the repeated effects of increasing vehicle loads and traffic flow, leading to material aging, fatigue, accumulated damage, and gradual performance degradation, thus raising concerns about the structural safety of bridges. Therefore, assessing the load-bearing capacity of bridges has become one of the important tasks in the field of bridge engineering structural testing.

[0003] According to the "Specifications for Testing and Evaluation of Load-Bearing Capacity of Highway Bridges" (JTG / TJ21-2011), bridge technical condition assessment and static load testing are considered the most effective and reliable methods for assessing bridge condition. Static load testing assesses bridge load-bearing capacity by applying static loads to the bridge and comparing the elastic displacement or strain values ​​at observation points with theoretically calculated values. However, it suffers from drawbacks such as requiring traffic disruption, high testing costs, and long testing times, and carries certain risks for bridge structures with defects and damage. Bridge technical condition assessment methods offer advantages such as specific bridge type classification, detailed description of defects, and objective calculation methods. Through meticulous observation and testing of all parts of the bridge, qualitative and quantitative descriptions of defects and localized damage are provided, facilitating a comprehensive understanding of the bridge's condition. However, it only determines the bridge's grade and does not directly assess its actual load-bearing capacity, resulting in a disconnect between bridge technical condition assessment and load-bearing capacity assessment. In fact, there is a close relationship between bridge load-bearing capacity and defects; the qualitative and quantitative descriptions of bridge defects obtained from bridge technical condition assessment can serve as a basis for assessing bridge load-bearing capacity. Summary of the Invention

[0004] This invention aims to overcome the shortcomings of existing bridge load-bearing capacity assessment technologies by proposing a method for assessing the load-bearing capacity of hollow slab beams based on bridge technical condition indicators. The goal is to establish a connection between bridge technical condition indicators and load-bearing capacity, thereby enabling rapid and efficient assessment of bridge load-bearing capacity. This solves the problems of long testing time, large workload, high cost, high risk, and traffic interruption required by traditional bridge static load tests.

[0005] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: The present invention provides a method for evaluating the load-bearing capacity of hollow slab beams based on bridge technical condition indicators, characterized by the following steps: Step 1: Obtain the measured parameters of bridge defects: Step 1.1: Determine the number of hollow slabs in the span of the static load test of the hollow slab beam in the year of testing. The calculation parameters for the corrosion rate of the longitudinal reinforcement in each hollow slab spanning the hole include: concrete compressive strength. Concrete protective layer thickness Concrete carbonation depth Longitudinal reinforcement diameter Service life of bridge structures and the bridge's average annual ambient humidity ;in, For the first The compressive strength of concrete in hollow core slab No. For the first Thickness of concrete protective layer for hollow core slab No. 1; For the first Carbonization depth of the hollow core plate For the first The diameter of the longitudinal reinforcement bars in the hollow core slab; Step 1.2: Calculate the longitudinal steel corrosion rate for each hollow slab based on the longitudinal steel corrosion rate calculation parameters. ,in, For the first Corrosion rate of longitudinal steel bars in hollow core slab; Step 1.3: Determine the measured bridge defect parameters for the year of inspection. ,in, Indicates the first The measured defect index parameters of the hollow core slab, and ,in, For the first The concrete compressive strength of the hollow core slab; For the first The elastic modulus of concrete for hollow core slab No. 1; For the first Corrosion rate of longitudinal steel bars in hollow core slab; For the first Average spacing of transverse cracks in hollow core slab No. 1; For the first Average height of transverse cracks in hollow core slab No. 1; For the first The length of hinge joint detachment in the hollow core slab; Step 2: Construct a training dataset for bridge defect index parameters; Step 2.1: Determine the orthogonal experiment parameters, including: determining the number of factors for the orthogonal experiment based on the total number of disease index types. The number of levels in the orthogonal experiment is ; Step 2.2: Based on the measured bridge defect index parameters Set the level values ​​for each factor in the orthogonal experiment, and use the number of orthogonal experiments as the basis for the result. The number of rows represents the number of factors in an orthogonal experiment. For each hollow slab, construct an orthogonal experimental table to determine the number of columns. Step 2.3: Define the current number of static load tests as... From the orthogonal experimental table of each hollow slab, the first... The first set of orthogonal experimental data is extracted and combined to form the second set of data. matrices As the first Bridge distress parameters from the second static load test, including Indicates from the first A set of orthogonal test data extracted from the orthogonal test table for hollow slab No. For the first The compressive strength of concrete in a set of orthogonal tests of hollow slab No. 1; For the first The elastic modulus of concrete in a set of orthogonal tests of hollow slab No. 1; For the first Corrosion rate of longitudinal steel bars in a set of orthogonal tests for hollow core slab No. 1; For the first Average spacing of transverse cracks in a set of orthogonal tests of hollow slab No. 1; For the first The average height of transverse cracks in a set of orthogonal tests of hollow slab No. 1; For the first The hinge breakage length in a set of orthogonal tests of the hollow slab was obtained, thus yielding Bridge distress parameters from static load test And used as a training dataset for bridge defect index parameters; Step 3: Construct a training dataset for bridge deflection verification coefficients: Step 3.1: Determine the bridge's geometric material parameters, including: bridge span. Bridge width Concrete density Reinforcing steel density Elastic modulus of steel bars Thus through Using parametric design language to build finite element models of bridges; Step 3.2: Determine the static load test conditions and control sections for the bridge finite element model, and arrange the test conditions on the control sections. After applying load to the finite element model of the bridge at several deflection measurement points, the static load test under defect-free conditions is calculated. The deflection value at each deflection measuring point is , The first static load test under defect-free working conditions Deflection values ​​at each measuring point; Step 3.3: According to the... Bridge distress parameters from static load test Simulation No. Bridge defects from the second static load test, calculation of the first In the secondary static load test The deflection value at each deflection measuring point is , For the first In the second static load test, the first The deflection value at the nth measuring point is calculated, and the deflection value at the th measuring point is calculated. In the second static load test, the first Deflection value verification coefficient at each measuring point Thus, the first Deflection verification coefficient of secondary static load test ; and thus obtain Deflection verification coefficients from the static load test were assembled into a matrix. This serves as a training dataset for bridge deflection verification coefficients. Step 4: Establish the Elman neural network model: Step 4.1: Establish the Elman neural network model, including: an input layer and an output layer, wherein the input layer has... One input node, The output layer contains the number of bridge defect indicators. One output node; Step 4.2: Training dataset for bridge defect index parameters Normalization was performed to obtain the normalized training dataset of bridge defect index parameters. ;in, The normalized parameters represent the bridge distress index parameters for the s-th static load test. Step 4.3: Train the Elman neural network model: Training dataset using normalized bridge defect index parameters Training dataset for bridge deflection verification coefficients with inverse normalization The Elman neural network model is trained to obtain the optimal connection weights and the optimal Elman neural network model with the optimal connection weights. Step 5: Assessment of bridge load-bearing capacity: The measured bridge defect parameters for the year of inspection The data is input into the optimal Elman neural network model, and the predicted value of the bridge deflection verification coefficient for the detection year is obtained. , For the year of detection Predicted values ​​of deflection verification coefficients for each measuring point; when This indicates that the actual load-bearing capacity of the bridge in the year of testing meets the design requirements; when This indicates that the actual load-bearing capacity of the bridge in the year of testing does not meet the design requirements. This represents the maximum value within the constant range of the deflection verification coefficient for hollow slab beams.

[0006] The present invention provides an electronic device, comprising a memory and a processor, wherein the memory is used to store a program that supports the processor in executing the hollow slab beam load-bearing capacity assessment method, and the processor is configured to execute the program stored in the memory.

[0007] The present invention discloses a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to perform the steps of the hollow slab beam bearing capacity assessment method.

[0008] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention establishes a correlation model between bridge technical condition assessment indicators and bearing capacity by parameterizing the defect indicators in the technical condition assessment data as input and deflection verification coefficient as the bearing capacity indicator output. This realizes the connection between bridge technical condition assessment and bridge bearing capacity assessment, making full use of and exploring the bridge technical condition assessment data. It makes up for the shortcomings of directly assessing the bearing capacity of the bridge after determining the bridge grade through technical condition assessment, and reduces the workload and test cost of assessing bearing capacity through static load tests.

[0009] 2. The Elman neural network of this invention belongs to the feedback type neural network. Compared with the traditional BP neural network, the Elman neural network adds feedback to the hidden layer and output layer nodes, thereby continuously optimizing the computation process. In addition, the Elman neural network has good application prospects in fields with nonlinear time series characteristics. It has strong robustness, good generalization ability, strong versatility and objectivity. The bridge distress index parameters also change with time, and the distress parameters have time-varying characteristics. Therefore, the Elman neural network model has good matching ability in solving the problem of establishing a correlation model between bridge distress index parameters and deflection verification coefficient, and realizing the connection between bridge technical condition indicators and bearing capacity assessment. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of a simply supported beam bridge in numerical simulation according to the present invention; Figure 2This is a cross-sectional view of the bridge according to the present invention; Figure 3 This is a diagram showing the longitudinal reinforcement arrangement of the present invention; Figure 4 This is a layout diagram of the static load test vehicle for the present invention; Figure 5 This is a diagram showing the arrangement of deflection measurement points according to the present invention; Figure 6 This is a comparison chart of the prediction results of deflection measurement point 1 in this invention; Figure 7 This is a comparison chart of the prediction results of deflection measurement point 2 in this invention; Figure 8 This is a comparison chart of the prediction results of deflection measurement point 3 in this invention. Detailed Implementation

[0011] In this embodiment, a method for assessing the load-bearing capacity of hollow slab beams based on bridge technical condition indicators consists of three parts. The first part involves acquiring measured bridge defect index parameters. By using technical condition assessment data from the testing year, the defect index parameters affecting the load-bearing capacity of the hollow slab beams are parameterized and input into the Elman neural network model established in the second part for bridge load-bearing capacity prediction. The second part involves establishing the Elman neural network model. Static load tests under different defect conditions are designed based on the measured bridge defect index parameters to obtain a training dataset of corresponding bridge defect index parameters. Finite element simulation is performed using Ansys to calculate the deflection verification coefficient of the bridge under different defect conditions, thus obtaining a training dataset of bridge deflection verification coefficients. The Elman neural network model is then established using the training datasets of bridge defect index parameters and bridge deflection verification coefficients. The third part involves inputting the measured bridge defect index parameters from the first part into the Elman neural network model established in the second part, outputting the predicted value of the deflection verification coefficient for assessing the load-bearing capacity of the hollow slab beam. Specifically, as shown in the example... Figure 1 Taking the hollow slab beam shown as an example, the bridge span is... Each span has 3 hollow slabs, with a cross-section as follows: Figure 2 As shown, the longitudinal reinforcement is arranged as follows: Figure 3 As shown, the method for evaluating the load-bearing capacity of hollow slab beams specifically includes the following steps: Step 1: Obtain the measured parameters of bridge defects: Step 1.1: Determine the number of hollow slabs in the span of the static load test of the hollow slab beam in the year of testing. The calculation parameters for the corrosion rate of longitudinal reinforcement in each hollow slab span include: concrete compressive strength. Concrete protective layer thickness Concrete carbonation depth Longitudinal reinforcement diameter Service life of bridge structures and the bridge's average annual ambient humidity ;in, For the first The compressive strength of concrete in hollow core slab No. For the first Thickness of concrete protective layer for hollow core slab No. 1; For the first Carbonization depth of the hollow core plate For the first The diameter of the longitudinal reinforcing bars in the hollow core slab; in this embodiment, the year of detection. Year, nianhe Number of hollow slabs in the load span of the static load test of hollow slab beams in 2018 Concrete compressive strength Concrete protective layer thickness Concrete carbonation depth Longitudinal reinforcement diameter Service life of bridge structures Annual average environmental humidity of bridges As shown in Table 1; Table 1

[0012] Step 1.2: Calculate the longitudinal steel corrosion rate for each hollow slab based on the longitudinal steel corrosion rate calculation parameters. ,in, For the first The corrosion rate of the longitudinal reinforcement of the hollow core slab; in this embodiment, the corrosion rate of the reinforcement is calculated using the formulas in Niu Ditao's "Durability and Life Prediction of Concrete Structures", the calculation formulas are (1)-(9), and the corrosion rate of the longitudinal reinforcement of each hollow core slab is As shown in Table 2; (1) (2) (3) (4) (5) (6) (7) (8) (9) In equations (1)-(9), Indicates the carbonation rate coefficient of concrete. Indicates the residual carbonization amount (mm). Indicates the time (a) when the steel reinforcement begins to corrode. This indicates the rate of steel corrosion (mm / a) before the steel bar cracks. This indicates the correction factor for the position of the reinforcing bars; This represents the correction factor for microenvironmental conditions. This indicates the time from the start of steel bar corrosion to rust expansion and cracking (a). This indicates the depth of steel reinforcement corrosion (mm) when the concrete cover cracks due to rust expansion. Indicates the influence coefficient of the position of the reinforcing bar. This indicates the depth of steel corrosion (mm) before rust expansion and cracking. This indicates the depth of steel corrosion (mm) after rust expansion and cracking, and before rust expansion and cracking. Pick After rust expansion and cracking Pick , Indicates the rate of steel reinforcement corrosion; Table 2

[0013] Step 1.3: Determine the measured bridge defect parameters for the year of inspection. ,in, Indicates the first The measured defect index parameters of the hollow core slab, and ,in, For the first The concrete compressive strength of the hollow core slab; For the first The elastic modulus of concrete for hollow core slab No. 1; For the first Corrosion rate of longitudinal steel bars in hollow core slab; For the first Average spacing of transverse cracks in hollow core slab No. 1; For the first Average height of transverse cracks in hollow core slab No. 1; For the first The length of the hinge joint detachment in the hollow slab; in this embodiment, the year is detected. Year, nianhe Bridge defect index parameters measured in 2018 As shown in Table 3; Table 3

[0014] Step 2: Construct a training dataset for bridge defect index parameters; Step 2.1: Determine the orthogonal experiment parameters, including: determining the number of factors for the orthogonal experiment based on the total number of disease index types. The number of levels in the orthogonal experiment is In this embodiment, the types of defects include concrete compressive strength, concrete elastic modulus, longitudinal steel corrosion rate, average spacing of transverse cracks, average height of longitudinal cracks, and hinge joint detachment length. Therefore, the number of factors in the orthogonal experiment is: To enable the trained neural network model to predict the load-bearing capacity under most fault conditions, the number of levels in the orthogonal experiment and the range of values ​​for each factor can be increased. In this embodiment, the number of levels in the orthogonal experiment is set to... ; Step 2.2: Based on the measured bridge defect index parameters Set the level values ​​for each factor in the orthogonal experiment, and use the number of orthogonal experiments as the basis for the result. The number of rows represents the number of factors in an orthogonal experiment. To determine the number of columns, an orthogonal experimental table is constructed for each hollow slab; in this embodiment, the table is based on the measured bridge defect index parameters. The level values ​​for each factor are set as shown in Table 4, and the orthogonal experimental design of L25.5.6 is used, with the number of orthogonal experiments as the parameter. The number of rows represents the number of factors in an orthogonal experiment. To determine the number of columns, an orthogonal experimental table is constructed for each hollow slab. In this embodiment, the same orthogonal experimental table is used for each hollow slab, as shown in Table 5. Table 4

[0015] Table 5

[0016] Step 2.3: Define the current number of static load tests as... From the orthogonal experimental table of each hollow slab, the first... The first set of orthogonal experimental data is extracted and combined to form the second set of data. matrices As the first Bridge distress parameters from the second static load test, including Indicates from the first A set of orthogonal experimental data extracted from the orthogonal experimental table of hollow slabs For the first The compressive strength of concrete in a set of orthogonal tests on a hollow slab; For the first The elastic modulus of concrete in a set of orthogonal tests on a hollow slab; For the first Corrosion rate of longitudinal steel reinforcement in a set of orthogonal tests for a hollow slab; For the first The average spacing of transverse cracks in a set of orthogonal tests on a hollow slab; For the first The average height of transverse cracks in a set of orthogonal tests on a hollow slab; For the first The hinge breakage length in a set of orthogonal tests of hollow slabs is obtained to obtain Bridge distress parameters from static load test This serves as a training dataset for bridge defect index parameters. In this embodiment, to enhance the diversity of the training dataset, all possible combinations of orthogonal test data for each hollow slab are considered, including the number of static load tests. Take as , No. matrices For the first Bridge distress parameters from the second static load test. Bridge distress parameters from static load test This is a training dataset for bridge defect index parameters; Step 3: Construct a training dataset for bridge deflection verification coefficients: Step 3.1: Determine the bridge's geometric material parameters, including: bridge span. Bridge width Concrete density Reinforcing steel density Elastic modulus of steel bars Thus through A parametric design language is used to establish a finite element model of the bridge; in this embodiment, the bridge span... Bridge width Concrete density Reinforcing steel density Elastic modulus of steel bars ; Step 3.2: Determine the static load test conditions and control sections for the bridge finite element model, and arrange the test sections on the control sections. After applying loads to the finite element model of the bridge at several deflection measurement points, the static load test under defect-free conditions was calculated. The deflection value at each deflection measuring point is , The first static load test under defect-free working conditions The deflection values ​​at each measuring point; in this embodiment, referring to the "Specifications for Load Testing of Highway Bridges" (JTG-TJ21-01-2015), the static load test condition is arranged at the most unfavorable position of bending moment and deflection at the mid-span section in the longitudinal direction. Due to the small bridge width, only the medium load is selected in the transverse direction, and the load distribution scheme is as follows. Figure 4 As shown, vehicles are arranged Vehicle width wheelbase The control section is the mid-span section, and each control section of the hollow slab is arranged at the bottom. Deflection measurement points common One deflection measurement point, such as Figure 5 As shown, after applying loads to the finite element model of the bridge, the static load test under defect-free conditions is calculated. The deflection value at each deflection measuring point is ; Step 3.3: According to the... Bridge distress parameters from static load test Simulation No. Bridge defects from the second static load test, calculation of the first In the secondary static load test The deflection value at each deflection measuring point is , For the first In the second static load test, the first The deflection value at the nth measuring point is calculated, and the deflection value at the th measuring point is calculated. In the second static load test, the first Deflection value verification coefficient at each measuring point Thus, the first Deflection verification coefficient of secondary static load test ; and thus obtain Deflection verification coefficients from the static load test were assembled into a matrix. This serves as the training dataset for bridge deflection verification coefficients; in this embodiment, the first... In the secondary static load test The deflection value at each deflection measuring point is , No. The deflection check coefficient for the second static load test is: ; by The matrix assembled from the deflection verification coefficients of the secondary static load test This serves as a training dataset for bridge deflection verification coefficients. Step 4: Establish the Elman neural network model: Step 4.1: Establish the Elman neural network model, including: an input layer and an output layer, wherein the input layer has... One input node, The number of defects in the bridge is defined in this embodiment. For each hollow slab, the defects are: concrete compressive strength, concrete elastic modulus, longitudinal steel corrosion rate, average spacing of transverse cracks, average height of transverse cracks, and hinge joint detachment length. The number of hollow slabs is also specified. Therefore , The output layer has There are one output node; for each static load test. One deflection verification coefficient; Step 4.2: Training dataset for bridge defect index parameters Normalization was performed to obtain the normalized training dataset of bridge defect index parameters. ;in, The normalized parameters represent the bridge distress index parameters for the s-th static load test. Step 4.3: Train the Elman neural network model: Training dataset using normalized bridge defect index parameters Training dataset for bridge deflection verification coefficients with inverse normalization The Elman neural network model is trained to obtain the optimal connection weights and the optimal Elman neural network model with the optimal connection weights; Step 5: Assessment of bridge load-bearing capacity: Year of detection Bridge defect index parameters measured in 2018 The input is fed into the optimal Elman neural network model, and the corresponding predicted values ​​of the bridge deflection verification coefficients are obtained. In this embodiment, The predicted bridge deflection verification coefficient for the year is [value missing]. , The predicted bridge deflection verification coefficient for the year is [value missing]. , The predicted bridge deflection verification coefficient for the year is [value missing]. ; Predicted results Figure 6 , Figure 7 and Figure 8 As shown, the maximum value of the constant value range of the hollow slab beam deflection verification coefficient in this invention is... Take as The prediction results show that The actual load-bearing capacity of the bridge in 2018 met the design requirements. nianhe The actual load-bearing capacity of the bridge in that year did not meet the design requirements, and The actual value of the annual bridge deflection verification coefficient is: , The actual value of the annual bridge deflection verification coefficient is: , The actual value of the annual bridge deflection verification coefficient is: , Consistent with the prediction results, and The error is very small, so a method for evaluating the bearing capacity of hollow slab beams based on bridge technical condition indicators can well replace the static load test for bridge bearing capacity evaluation.

[0017] In this embodiment, an electronic device includes a memory and a processor. The memory stores a program that supports the processor in executing the above-described method, and the processor is configured to execute the program stored in the memory.

[0018] In this embodiment, a computer-readable storage medium stores a computer program, which is executed by a processor to perform the steps of the above method.

Claims

1. A hollow slab beam carrying capacity evaluation method based on bridge technical condition index, characterized in that, Includes the following steps: Step 1: Obtain the measured parameters of bridge defects: Step 1.1: Determine the number N of hollow slabs in the span of the static load test of the hollow slab beam in the test year, and the calculation parameters for the corrosion rate of the longitudinal reinforcement in each hollow slab of the span, including: concrete compressive strength [fcu1,…,fcu n ,…,fcu N ], Concrete cover thickness [c1,…,c n ,…,c N ], Concrete carbonation depth [x1,…,x n ,…,x N ], longitudinal reinforcement diameter [d1,…,d n ,…,d N The service life (t) of the bridge structure and the average annual ambient humidity (RH) of the bridge; among which, fcu n Let c be the concrete compressive strength of the nth hollow slab. n x is the thickness of the concrete cover for the nth hollow slab; n Let d be the carbonization depth of the nth hollow slab. n The diameter of the longitudinal reinforcement in the nth hollow slab; Step 1.2: Calculate the longitudinal steel corrosion rate [η1,…,η] for each hollow slab based on the longitudinal steel corrosion rate calculation parameters. n ,…,η N ], where η n Let be the corrosion rate of the longitudinal reinforcement in the nth hollow slab; Step 1.3: Determine the measured bridge defect parameters for the year of inspection. in, Let represent the measured defect index parameters of the nth hollow slab, and Among them, fcu n Ec represents the concrete compressive strength of the nth hollow slab. n η is the elastic modulus of the concrete for the nth hollow slab; n Let be the corrosion rate of the longitudinal reinforcement in the nth hollow slab; The average spacing of transverse cracks in the nth hollow slab; The average height of the transverse crack in the nth hollow slab; The hinge joint detachment length of the nth hollow slab; Step 2: Construct a training dataset for bridge defect index parameters; Step 2.1: Determine the orthogonal experiment parameters, including: determining the number of factors in the orthogonal experiment as l based on the total number of disease index types, and the number of levels in the orthogonal experiment as m; Step 2.2: Based on the measured bridge defect index parameter B, set the level values ​​of each factor in the orthogonal experiment, and construct an orthogonal experiment table for each hollow slab with the orthogonal experiment number K as the number of rows and the orthogonal experiment number l as the number of columns; Step 2.3: Define the current static load test number as s, extract a set of orthogonal test data from the orthogonal test table of each hollow slab for the sth test, and combine them into the sth matrix B. s =[A s,1 ,…,A s,n ,…,A s,N As the bridge defect index parameter for the s-th static load test, among which fcu represents a set of orthogonal experimental data extracted from the nth hollow slab orthogonal experimental table. s,n Ec represents the compressive strength of concrete in a set of orthogonal tests for the nth hollow slab. s,n η is the elastic modulus of concrete in a set of orthogonal tests for the nth hollow slab; s,n The corrosion rate of longitudinal steel bars in a set of orthogonal tests for the nth hollow slab; The average spacing of transverse cracks in a set of orthogonal tests on the nth hollow slab; The average height of transverse cracks in a set of orthogonal tests on the nth hollow slab; Let be the hinge joint detachment length in a set of orthogonal tests for the nth hollow slab, thus obtaining the bridge distress index parameters [B1,…,B] from the S static load tests. s ,…,B S ] T And used as a training dataset for bridge defect index parameters; Step 3: Construct a training dataset for bridge deflection verification coefficients: Step 3.1: Determine the bridge's geometric material parameters, including: bridge span L0, bridge width B0, and concrete density ρ. c Reinforcing steel density ρ b Elastic modulus E of steel bars b Thus, a finite element model of the bridge is established using the Ansys parametric design language; Step 3.2: Determine the static load test conditions and control sections of the bridge finite element model, and arrange R deflection measuring points on the control sections. After applying the load to the bridge finite element model, calculate the deflection value C0 = [y] at the R deflection measuring points in the static load test under the defect-free condition. 0，1 ,…,y 0,r ,…,y 0,R ], y 0,r This represents the deflection value at the r-th measuring point during a static load test under no-damage conditions. Step 3.3: According to the bridge distress index parameter B from the s-th static load test. s Simulate the bridge defects in the s-th static load test, and calculate the deflection values ​​at R deflection measurement points in the s-th static load test [y]. s，1 ,…,y s,r ,…,y s,R ], y s,r Let be the deflection value at the r-th measuring point in the s-th static load test, and calculate the verification coefficient λ for the deflection value at the r-th measuring point in the s-th static load test. s,r =y s,r / y 0,r Thus, the deflection verification coefficient C of the s-th static load test is obtained. s =[λ s,1 ,…,λ s,r ,…,λ s,R ]; then the deflection verification coefficients of the S static load tests are obtained and assembled into a matrix [C1,…,C s ,…,C S ] T This serves as a training dataset for bridge deflection verification coefficients. Step 4: Establish the Elman neural network model: Step 4.1: Establish an Elman neural network model, including an input layer and an output layer. The input layer has M×N input nodes, where M is the number of bridge defect indicators, and the output layer has R output nodes. Step 4.2: The training data set of bridge disease index parameters [B1,..., B s ,…,B S ] T is normalized to obtain the normalized training data set of bridge disease index parameters wherein, represents the normalized bridge disease index parameter of the s-th static load test. Step 4.3: Train the Elman neural network model: Training dataset using normalized bridge defect index parameters The training dataset for the inverse normalized bridge deflection verification coefficients [C1,…,C] s ,…,C S ] T The Elman neural network model is trained to obtain the optimal connection weights and the optimal Elman neural network model with the optimal connection weights. Step 5: Assessment of bridge load-bearing capacity: The measured disease index parameters of the bridge in the detection year are obtained The measured disease index parameters of the bridge in the detection year are obtained The measured disease index parameters of the bridge in the detection year are obtained When the actual load-carrying capacity of the bridge in the detection year meets the design requirements; When , it indicates that the actual bearing capacity of the bridge in the detection year does not meet the design requirements, wherein Δ represents the maximum value of the constant value range of the hollow slab beam deflection checking coefficient.

2. An electronic device comprising a memory and a processor, characterized in that The memory is used to store a program that supports the processor in executing the method for evaluating the load-bearing capacity of the hollow slab beam according to claim 1, and the processor is configured to execute the program stored in the memory.

3. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program, when run by the processor, executes the steps of the method for evaluating the load-bearing capacity of hollow slab beams as described in claim 1.

Images