A method and device for detecting reinforced concrete bridge deck

Abstract

The embodiment of the present invention provides a method and device for detecting reinforced concrete bridge decks, by constructing a finite element calculation model for the bridge deck; determining the deflection information of the bridge deck under different load grades and corresponding action times, the respective average crack widths of several units near the bottom center in the transverse and longitudinal directions, and the total proportion of the cracked unit area for the bottom unit through nonlinear finite elements; constructing a neural network, and determining the actual average crack width corresponding to each unit after the bridge deck is meshed according to a specified size, the area percentage of the actual cracked grid for the bridge deck, and the measured deflection data of the bridge deck; outputting the target load grade and the target action times for the target load grade based on the back propagation neural network; generating the detection result for the bridge deck based on the target load grade and the target action times, thereby improving the detection efficiency and accuracy of the bridge deck.

Description

Technical Field

[0001] The present invention relates to the technical field of reinforced concrete bridge deck detection, and in particular to a reinforced concrete bridge deck detection method, a reinforced concrete bridge deck detection device, an electronic device and a computer-readable storage medium. Background Art

[0002] As traffic volume also increases dramatically, small and medium-sized bridges may face more serious risks of vehicle overloading than larger bridges. The design of early concrete bridges was mostly based on the stress allowable method, while in recent years, the ultimate bearing capacity state has been used for design. High traffic volume and high load operation often cause reinforced concrete bridge decks to be under fatigue at high stress levels, and the risk of damage caused by fatigue cannot be ignored. In fact, after the bridge has been in operation for many years, various types of cracks often appear in reinforced concrete bridge decks. Longitudinal and transverse cracks in the bottom plate of the bridge deck due to high-cycle fatigue are a relatively common form. The generation of such cracks poses great hidden dangers and risks to the safe operation of the bridge. Therefore, a method is needed to detect such reinforced concrete bridge decks in service in a timely and rapid manner to provide technical support for repair decisions and risk control. Summary of the invention

[0003] The embodiments of the present invention provide a method, device, electronic device and computer-readable storage medium for detecting reinforced concrete bridge decks to overcome the above problems or at least partially solve the above problems.

[0004] The embodiment of the present invention discloses a method for detecting a reinforced concrete bridge deck, comprising:

[0005] Constructing a finite element calculation model for the bridge deck;

[0006] Determining a load classification for the finite element calculation model and a number of actions for the load classification;

[0007] Determining calculation parameters based on the load classification and the number of actions;

[0008] A data set is constructed using the calculation parameters; the data set is a high cycle fatigue nonlinear finite element calculation result data set; the data set includes deflection information for the bridge deck, average widths of cracks in the longitudinal and transverse directions of a number of units adjacent to the center after meshing the bridge deck, and a total proportion of cracked unit areas for bottom units;

[0009] Constructing a back propagation neural network using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area;

[0010] determining actual widths of cracks for the elements, and actual cracked mesh area percentages for the bridge deck, and measured deflection data;

[0011] Outputting a target load classification and a target number of actions for the target load classification based on the back propagation neural network according to the actual width of the unit crack, the actual cracking grid area percentage and the measured deflection data;

[0012] An inspection result for the bridge deck is generated based on the target load classification and the target number of actions.

[0013] Optionally, the step of constructing a finite element calculation model for the bridge deck comprises:

[0014] determining geometry and material properties for the bridge deck;

[0015] Generate finite element mesh;

[0016] A finite element computational model for the bridge deck is constructed based on the geometric shape, the material properties and the finite element mesh.

[0017] Optionally, the step of determining the calculation parameters based on the load classification and the number of actions includes:

[0018] Determine the target sampling algorithm;

[0019] A random sampling result is determined from the load classification and the number of actions based on the target sampling algorithm, and the random sampling result is determined as a calculation parameter.

[0020] Optionally, the step of constructing a back propagation neural network using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area includes:

[0021] Determine a minimum error function using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area;

[0022] A back propagation neural network is constructed based on the minimum error function.

[0023] Optionally, the step of constructing a data set using the calculation parameters and generating a simulation result based on the data set includes:

[0024] Determine the convergence criterion;

[0025] constraining a predetermined nonlinear solver using the convergence criterion;

[0026] A data set is constructed using the calculation parameters, and a simulation result is generated based on the data set by the nonlinear solver.

[0027] Optionally, the step of generating a detection result for the bridge deck based on the target load classification and the target number of actions includes:

[0028] Determining historical vehicle load information for the bridge deck according to the target load classification and the target number of actions;

[0029] Fatigue damage information for the bridge deck is determined based on the finite element calculation model using the historical vehicle load information.

[0030] Optionally, it also includes:

[0031] Determining future vehicle load information for the bridge deck according to the target load classification and the target number of actions;

[0032] The fatigue life information of the bridge deck is determined based on the finite element calculation model using the future vehicle load information and the fatigue damage information.

[0033] Optionally, the target sampling algorithm is a Monte Carlo random sampling algorithm or a Latin hypercube sampling algorithm.

[0034] Optionally, the target sampling algorithm is a Latin hypercube sampling algorithm.

[0035] Optionally, the mesh size of the finite element mesh is ten centimeters.

[0036] Optionally, it also includes:

[0037] A loading condition is determined, wherein the loading condition is a static load.

[0038] Optionally, it also includes:

[0039] A loading condition is determined, wherein the loading condition is a dynamic load.

[0040] The present invention also discloses a detection device for reinforced concrete bridge decks, comprising:

[0041] A finite element calculation model building module, used to build a finite element calculation model for the bridge deck;

[0042] A load data determination module, used to determine the load classification for the finite element calculation model and the number of actions for the load classification;

[0043] A calculation parameter determination module, used to determine the calculation parameters based on the load classification and the number of actions;

[0044] A simulation result generation module, used to construct a data set using the calculation parameters; the data set is a high cycle fatigue nonlinear finite element calculation result data set; the data set includes deflection information for the bridge deck, and average widths of cracks in the longitudinal and transverse directions of a number of units adjacent to the center after meshing the bridge deck, and a total proportion of cracked unit areas for bottom units;

[0045] A back propagation neural network construction module, used to construct a back propagation neural network using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area;

[0046] An actual data acquisition module, for determining an actual width of a unit crack, an actual cracked grid area percentage of the bridge deck, and measured deflection data;

[0047] A target data output module, configured to output a target load classification and a target number of actions for the target load classification based on the back propagation neural network according to the actual width of the unit crack, the actual cracking grid area percentage and the measured deflection data;

[0048] A test result generating module is used to generate a test result for the bridge deck based on the target load classification and the target number of actions.

[0049] The embodiment of the present invention further discloses an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0050] The memory is used to store computer programs;

[0051] The processor is used to implement the method described in the embodiment of the present invention when executing the program stored in the memory.

[0052] The embodiment of the present invention further discloses a computer-readable storage medium having instructions stored thereon, which, when executed by one or more processors, enables the processors to execute the method described in the embodiment of the present invention.

[0053] The embodiments of the present invention include the following advantages:

[0054] In an embodiment of the present invention, a finite element calculation model for the bridge deck is constructed; load classification for the finite element calculation model and the number of actions for the load classification are determined; calculation parameters are determined based on the load classification and the number of actions; a data set is constructed using the calculation parameters; the data set is a data set of high-cycle fatigue nonlinear finite element calculation results; the data set includes deflection information for the bridge deck, and the average width of cracks in the longitudinal and transverse directions of several units adjacent to the center after meshing the bridge deck, and the total proportion of cracked unit areas for the bottom units; the reverse is constructed using the calculation parameters, the deflection information, the average width and the total proportion of the cracked unit areas A back propagation neural network is provided; determining an actual width of a unit crack, an actual cracking grid area percentage of the bridge deck, and measured deflection data; outputting a target load grade and a target number of actions for the target load grade based on the back propagation neural network through the actual width of the unit crack, the actual cracking grid area percentage, and the measured deflection data; generating a detection result for the bridge deck based on the target load grade and the target number of actions, thereby reducing the subjective influence of the reinforced concrete crack assessment method, and at the same time, improving the accuracy and timeliness of the life prediction of in-service cracked reinforced concrete bridge deck components in the absence of historical vehicle load information. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a flow chart of steps of a reinforced concrete bridge deck detection method provided in an embodiment of the present invention;

[0056] Figure 2 It is a structural block diagram of a reinforced concrete bridge deck detection device provided in an embodiment of the present invention;

[0057] Figure 3 is a hardware structure block diagram of an electronic device provided in an embodiment of the present invention;

[0058] Figure 4 is a schematic diagram of a computer-readable medium provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0060] In practical applications, as the traffic volume also increases dramatically, small and medium-sized bridges may face more serious vehicle overloading risks than large bridges. The design of early concrete bridges was mostly based on the stress allowable method, while in recent years, the ultimate bearing capacity state has been used for design. High traffic volume and high load operation make reinforced concrete bridge decks often under fatigue at high stress levels, and the risk of damage caused by fatigue cannot be ignored. In fact, after many years of operation, reinforced concrete bridge decks often have various cracks. Longitudinal and transverse cracks in the bottom plate of the bridge deck due to high cycle fatigue are a common form. The generation of such cracks brings great hidden dangers and risks to the safe operation of the bridge. The existing non-destructive testing methods for bridges with cracks, the concrete bridge deck crack assessment method based on manual / image recognition, is based on qualitative judgment based on crack density, crack width and distribution law, which is more dependent on subjective judgment; or through the use of statistical methods based on actual measurement to quantitatively analyze the information such as the passing vehicles and load of the bridge deck, and use linear / nonlinear fatigue cumulative damage models for assessment. This method has a large workload and cannot quickly predict the remaining fatigue life of the reinforced concrete bridge deck in a timely manner.

[0061] Therefore, there is a need for a method that can timely and quickly predict the fatigue life of such reinforced concrete bridge decks in service and provide technical support for repair decisions and risk management.

[0062] The embodiment of the present invention uses graphic recognition to quickly analyze the measured crack width of reinforced concrete bridge decks and performs area ratio processing after gridding, as well as the bridge deck deflection deformation monitoring data, to reversely infer the historical vehicle load conditions of the in-service bridge decks, and uses the obtained vehicle load history as input to perform fatigue calculation on the reinforced concrete bridge decks using nonlinear finite elements. Based on the bridge deck deflection deformation control index, quantitative prediction of the remaining fatigue life of the in-service reinforced concrete bridge decks is achieved in the absence of historical vehicle load information.

[0063] Through the above methods, the subjective influence of the reinforced concrete crack assessment method can be reduced; in the absence of historical vehicle load information, the accuracy and timeliness of the life prediction of in-service cracked reinforced concrete bridge deck components can be improved, providing technical support for repair decisions and risk management.

[0064] Reference Figure 1 , shows a flowchart of a method for detecting reinforced concrete bridge decks provided in an embodiment of the present invention, which may specifically include the following steps:

[0065] Step 101, constructing a finite element calculation model for the bridge deck;

[0066] Finite Element Modeling (FEM) is a numerical simulation method used to solve partial differential equations in engineering and scientific problems. It works by breaking down complex problems into smaller, simpler parts called finite elements. Each finite element is defined by a set of nodes that are connected together to form a mesh. The partial differential equations governing the problem are then applied to each finite element and solved using numerical methods. Finally, the solutions from all the finite elements are combined to obtain an approximate solution to the entire problem.

[0067] The finite element calculation model has the following advantages:

[0068] Versatility: It can be used to solve various types of partial differential equations, including equations in solid mechanics, fluid mechanics, heat transfer, and electromagnetism.

[0069] Flexibility: Can handle complex geometries and boundary conditions.

[0070] Precision: The calculation precision can be increased as needed.

[0071] Finite element calculation models are used to build a model of reinforced concrete bridge decks. The model is used to simulate the behavior of bridge decks under different load levels and repetitions. The simulation results are used to train a BP (back propagation) neural network, which can be used to predict the remaining fatigue life of reinforced concrete bridge decks in service.

[0072] Step 102, determining a load classification for the finite element calculation model and a number of actions for the load classification;

[0073] For example,

[0074] The following factors should be considered when determining the load classification and number of actions for the finite element calculation model:

[0075] Load classification:

[0076] Specification requirements: Should comply with relevant specifications and standards, such as the American Association of Highway and Transportation Officials (AASHTO) "Long-Range Load Regulations" (LRFD).

[0077] Bridge type and purpose: Different bridge types and purposes have different load requirements. For example, highway bridges generally have higher loads than railway bridges.

[0078] Historical vehicle load data: If historical vehicle load data is available, it can be used to determine the load classification.

[0079] Engineering Judgment: The Engineer should use his or her experience and judgment to determine the load classifications.

[0080] Number of actions:

[0081] Fatigue Design: For fatigue design, the number of actions is usually related to the expected service life of the bridge. For example, if the bridge is designed to last 50 years, the number of actions can be set to 1 million.

[0082] Limit state design: For limit state design, the number of actions is usually related to the probability of occurrence of the extreme load event. For example, for a 100-year flood event, the number of actions can be set to 1.

[0083] Here are some specific methods for determining load classification and number of actions:

[0084] Statistics-based method: This method is based on statistical analysis of historical vehicle load data. For example, the mean, standard deviation and peak values ​​for different load levels can be calculated.

[0085] Probability-based method: This method uses probability theory to calculate the load levels and times of action. For example, the probability of exceeding a certain load level within a given time can be calculated.

[0086] Experience-Based Approach: This approach is based on the experience and judgment of engineers.

[0087] Step 103, determining calculation parameters based on the load classification and the number of actions;

[0088] Exemplarily, the calculation parameters including load classification and number of actions are determined by methods such as Monte Carlo random sampling or Latin hypercube sampling, which can generate load classification and number of actions that take into account uncertainties.

[0089] Step 104, constructing a data set using the calculation parameters; the data set is a high cycle fatigue nonlinear finite element calculation result data set; the data set includes deflection information for the bridge deck, average widths of cracks in the longitudinal and transverse directions of a number of units adjacent to the center after meshing the bridge deck, and a total proportion of cracked unit areas for bottom units;

[0090] High cycle fatigue: refers to the fatigue failure of materials under high cycles (i.e. many repeated loads). This failure usually occurs at a smaller stress level and is an important cause of failure of many engineering structures.

[0091] Nonlinear finite element method: It is a numerical calculation method used to simulate the nonlinear behavior of materials and structures under complex loading conditions. It can take into account factors such as nonlinear constitutive relations of materials, geometric nonlinearity and large deformation.

[0092] Calculation result data set: refers to the numerical results obtained through nonlinear finite element calculation. These results usually include node displacement, stress, strain and other information, which can be used to analyze the deformation, strength and fatigue life of the structure.

[0093] The high-cycle fatigue nonlinear finite element calculation result data set can be a set of data obtained through nonlinear finite element analysis, and these data can be used to evaluate the fatigue life of the structure under high-cycle loads.

[0094] The calculation parameters are used to express different load grades and the corresponding number of actions, and a data set of high-cycle fatigue nonlinear finite element calculation results is constructed to obtain the corresponding simulation results, including the mid-span deflection δ of the reinforced concrete bridge deck, that is, the deflection information; after the bridge deck is meshed, the average crack width w of the four units near the bottom center in the longitudinal and transverse directions of the bridge, and the total proportion p% (p=0~100) of the cracked unit area of ​​the bottom unit of the bridge deck; among them, w, p and δ are all simulation data, and their role is to construct the back propagation neural network.

[0095] Step 105, constructing a back propagation neural network using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area;

[0096] Step 106, determining the actual width of the unit crack, the actual cracked grid area percentage of the bridge deck, and the measured deflection data;

[0097] For example, when predicting the remaining fatigue life of reinforced concrete bridge decks in service, image recognition can be used to quickly analyze the actual crack width w1 at the bottom of the reinforced concrete bridge decks in service, that is, the actual width. After the bridge deck is gridded with a size of 10 cm, the area proportion p1 of the cracked grid is calculated, that is, the actual percentage of the cracked grid area, and the measured deflection data δ1 is obtained.

[0098] Step 107, outputting a target load classification and a target number of actions for the target load classification based on the back propagation neural network according to the actual width of the unit crack, the actual cracking grid area percentage and the measured deflection data;

[0099] For example, the actual width w1, the actual cracking grid area percentage p1 and the measured deflection data δ1 are used as input to obtain the target action times Ni corresponding to different target load grades Ti.

[0100] Step 108: Generate a test result for the bridge deck based on the target load classification and the target number of actions.

[0101] In an embodiment of the present invention, a finite element calculation model for the bridge deck is constructed; load classification for the finite element calculation model and the number of actions for the load classification are determined; calculation parameters are determined based on the load classification and the number of actions; a data set is constructed using the calculation parameters; the data set is a data set of high-cycle fatigue nonlinear finite element calculation results; the data set includes deflection information for the bridge deck, and the average width of cracks in the longitudinal and transverse directions of several units adjacent to the center after meshing the bridge deck, and the total proportion of cracked unit areas for the bottom units; the reverse is constructed using the calculation parameters, the deflection information, the average width and the total proportion of the cracked unit areas A back propagation neural network is provided; determining an actual width of a unit crack, an actual cracking grid area percentage of the bridge deck, and measured deflection data; outputting a target load grade and a target number of actions for the target load grade based on the back propagation neural network through the actual width of the unit crack, the actual cracking grid area percentage, and the measured deflection data; generating a detection result for the bridge deck based on the target load grade and the target number of actions, thereby reducing the subjective influence of the reinforced concrete crack assessment method, and at the same time, improving the accuracy and timeliness of the life prediction of in-service cracked reinforced concrete bridge deck components in the absence of historical vehicle load information.

[0102] Based on the above embodiment, a variant embodiment of the above embodiment is proposed. It should be noted that in order to make the description concise, only the differences from the above embodiment are described in the variant embodiment.

[0103] In an optional embodiment of the present invention, the step of constructing a finite element calculation model for the bridge deck comprises:

[0104] determining geometry and material properties for the bridge deck;

[0105] Generate finite element mesh;

[0106] constructing a finite element computational model for the bridge deck based on the geometric shape, the material properties and the finite element mesh;

[0107] The step of constructing a data set using the calculation parameters and generating simulation results based on the data set includes:

[0108] Determine the convergence criterion;

[0109] constraining a predetermined nonlinear solver using the convergence criterion;

[0110] A data set is constructed using the calculation parameters, and a simulation result is generated based on the data set by the nonlinear solver.

[0111] In a specific implementation, the embodiment of the present invention can generate simulation results in the following manner:

[0112] 1. Define geometry and material properties:

[0113] Geometry: Accurately represent the geometry of the bridge deck, including its size, thickness, and any associated features such as stiffeners or openings.

[0114] Material Properties: Define the material properties of concrete and reinforcement, including elastic modulus, Poisson's ratio, and yield strength.

[0115] 2. Generate finite element mesh:

[0116] Mesh size: The bridge deck is divided into small elements using a mesh size of approximately 10 cm. This mesh size should be sufficient to accurately capture the local stress and strain distribution while maintaining computational efficiency.

[0117] Element Type: Select the appropriate element type for the analysis. For concrete, solid elements are usually used, while for steel, beam or truss elements may be more appropriate.

[0118] Boundary Conditions: Apply appropriate boundary conditions to the model, such as fixed supports or roller supports, to represent the constraints of the bridge deck.

[0119] 3. Define loading conditions:

[0120] Static Loads: Apply any static loads acting on the bridge deck, such as deadweight or prestressing.

[0121] Dynamic Loads: If dynamic loads are considered, define the time history of the loads and the points at which they are applied.

[0122] It should be noted that, in the embodiment of the present invention, load classification and number of actions can be used as loading conditions.

[0123] 4. Select a nonlinear analysis solver:

[0124] Nonlinear Solver: Select a nonlinear solver that is capable of handling the nonlinear behavior of the material and the large deformations that may occur in the bridge deck.

[0125] Convergence Criteria: Set appropriate convergence criteria for the nonlinear solver to ensure accurate results.

[0126] 5. Perform nonlinear finite element analysis:

[0127] Solve Nonlinear Equations: Run the nonlinear finite element analysis solver to solve the nonlinear system of equations for displacements, stresses, and strains in the bridge deck.

[0128] Post-processing results: for example, deflection information of the bridge deck, average width of cracks in each bottom unit of the bridge deck, total proportion of cracked unit area in the bottom unit, etc.

[0129] Other notes:

[0130] Material Nonlinearity: Concrete and steel exhibit nonlinear stress-strain behavior, which should be considered in the analysis. This can be achieved using material models that capture the nonlinear behavior of the material.

[0131] Geometric nonlinearity: Large deformations of the bridge deck may lead to geometric nonlinearity, which should also be considered in the analysis. This can be achieved using a large deformation solver.

[0132] Contact and Friction: If there are any contact surfaces between different parts of the bridge deck, contact and friction should be simulated to account for the interaction between these surfaces.

[0133] In an optional embodiment of the present invention, the step of determining the calculation parameters based on the load classification and the number of actions includes:

[0134] Determine the target sampling algorithm;

[0135] A random sampling result is determined from the load classification and the number of actions based on the target sampling algorithm, and the random sampling result is determined as a calculation parameter.

[0136] In a specific implementation, the target sampling algorithm may include, but is not limited to, a Monte Carlo random sampling algorithm or a Latin Hypercube Sampling algorithm (LHS).

[0137] Using appropriate sampling techniques, such as Monte Carlo random sampling and Latin hypercube sampling, has the following benefits in determining the minimum error function and training BP neural networks:

[0138] 1. Improve computing efficiency:

[0139] Traditional methods, such as exhaustive or grid search, require traversing all possible parameter combinations to find the optimal solution, which is computationally intensive or even infeasible for high-dimensional problems.

[0140] The sampling technique only needs to randomly extract a finite number of sample points from the parameter space, and then use these sample points to estimate the optimal solution, which can greatly reduce the amount of calculation and improve the calculation efficiency.

[0141] 2. Improve fitting accuracy:

[0142] Traditional deterministic methods may get stuck in local optimal solutions and fail to find the global optimal solution.

[0143] The sampling technology can explore the parameter space more effectively and find the parameter combination closer to the global optimal solution, which can improve the fitting accuracy of the BP neural network and make it better predict the target value.

[0144] 3. Reduce uncertainty:

[0145] In practical applications, there are often many uncertain factors, such as parameter errors, measurement errors, etc.

[0146] Sampling techniques can reduce the uncertainty of the results by randomly sampling these uncertain factors to reflect their impact on the results.

[0147] 4. Simplified modeling:

[0148] Sampling technology can construct an approximate model by sampling data, which can simplify the modeling process and improve modeling efficiency.

[0149] The advantages of Monte Carlo random sampling are that it is simple to use and has low computational complexity.

[0150] The advantages of Latin hypercube sampling are high sampling efficiency and the ability to cover the parameter space more evenly.

[0151] In practical applications, appropriate sampling techniques can be selected according to specific circumstances. For example, if the amount of computation is less than a preset computation threshold, Monte Carlo random sampling can be used; if the required fitting accuracy is greater than a preset accuracy threshold, Latin hypercube sampling can be used.

[0152] In summary, the use of appropriate sampling techniques can improve computational efficiency, increase fitting accuracy, reduce uncertainty, and simplify modeling, and has broad application prospects in various fields.

[0153] In an optional embodiment of the present invention, the step of constructing a back propagation neural network using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area includes:

[0154] Determine a minimum error function using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area;

[0155] A back propagation neural network is constructed based on the minimum error function.

[0156] For example:

[0157] 1. Define the error function:

[0158] The error function represents the difference between the predicted value and the actual value, and is usually calculated using the mean square error (MSE):

[0159] MSE = (1 / N) * Σ[y_i - f(x_i)]^2

[0160] in:

[0161] N: total number of data points;

[0162] y_i: the actual value of the i-th data point;

[0163] f(x_i): the value of the i-th data point predicted by the BP neural network;

[0164] x_i: input vector of the ith data point, including deflection, average crack width and percentage of total cracked unit area;

[0165] 2. Training BP neural network:

[0166] The BP neural network is trained using the specified calculation parameters, deflection information, average width, and total crack unit area ratio. The training process minimizes the error function by adjusting the network weights and biases.

[0167] 3. Evaluate the performance of BP neural network:

[0168] After training is complete, the network performance is evaluated using an independent dataset that was not used for training. Evaluation indicators include prediction accuracy, goodness of fit, etc.

[0169] 4. Determine the minimum error function:

[0170] The network weights and biases are adjusted by an iterative optimization algorithm (such as gradient descent or Levenberg-Marquardt algorithm) until the MSE reaches a minimum, that is, the weight and bias combination that best fits the data is found.

[0171] 5. Verify the result:

[0172] The effectiveness of the minimum error function and BP neural network is verified using additional data or by comparison with established methods.

[0173] Detailed explanation:

[0174] 1. Define the error function:

[0175] Mean squared error (MSE) is a commonly used error function in regression tasks to predict continuous values. It measures the average squared difference between the predicted value and the actual value. The smaller the MSE value, the better the neural network fits the data.

[0176] 2. Training BP neural network:

[0177] BP neural network is an artificial neural network that learns from data by adjusting internal weights and biases. The training process involves inputting data (deflection, average crack width, and percentage of total cracked unit area) and corresponding target values ​​(actual deflection, average crack width, and percentage of total cracked unit area) into the network. The network then adjusts its weights and biases to minimize the error between the predicted value and the target value. This process is repeated until the network converges to a minimum error state.

[0178] 3. Evaluate the performance of BP neural network:

[0179] After training, the performance of the network should be evaluated using an independent dataset that was not used for training, as it ensures that the network can generalize to unseen data and avoid overfitting the training data. The evaluation should include metrics such as the accuracy of the network in predicting parameters relevant to new data points. This can be achieved by calculating metrics such as MSE or coefficient of determination (R^2) on the validation data.

[0180] 4. Determine the minimum error function:

[0181] The network weights and biases are adjusted until the MSE is minimized by an iterative optimization algorithm such as gradient descent or the Levenberg-Marquardt algorithm. These algorithms repeatedly update the weights and biases to reduce the error function. This process continues until the error function converges to a minimum, indicating that the network has found a set of weights and bias combinations that best fit the data.

[0182] 5. Verify the result:

[0183] Validate the effectiveness of the minimum error function and BP neural network using additional data or by comparing with established methods. This may involve testing the network on a completely new dataset or comparing its predictions with those of other modeling methods. It is important to ensure that the network performs well on a variety of data to ensure its reliability.

[0184] In the context of predicting fatigue life of in-service reinforced concrete bridge decks, the minimum error function and the trained BP neural network should accurately reflect the relationship between the input parameters (deflection, average crack width and total cracked unit area fraction) and the target value (fatigue life). This will allow reliable prediction of the remaining fatigue life of the bridge under different loading conditions.

[0185] In an optional embodiment of the present invention, the step of generating a detection result for the bridge deck based on the target load classification and the target number of actions includes:

[0186] Determining historical vehicle load information for the bridge deck according to the target load classification and the target number of actions;

[0187] Fatigue damage information for the bridge deck is determined based on the finite element calculation model using the historical vehicle load information.

[0188] Optionally, it also includes:

[0189] Determining future vehicle load information for the bridge deck according to the target load classification and the target number of actions;

[0190] The fatigue life information of the bridge deck is determined based on the finite element calculation model using the future vehicle load information and the fatigue damage information.

[0191] For example:

[0192] S1: A finite element calculation model for the bridge deck is established using a nonlinear method, where the size of the finite element grid is 10 cm, and the calculation parameters include different load levels T and the number of actions N of different load levels;

[0193] S2: Select an appropriate sampling method, such as Monte Carlo random sampling or Latin hypercube sampling (LHS), to establish a data set of reinforced concrete bridge decks containing the calculation parameters in S1, and obtain the corresponding calculation results, including the mid-span deflection δ of the reinforced concrete bridge deck, that is, the deflection information; the average width w of the cracks in each unit at the bottom of the bridge deck obtained by strain transformation, and the total proportion of cracked unit area p% (p=0~100);

[0194] S3: Use the parameters and results in S1 and S2 as the data training set to train the BP neural network and obtain the minimum error function of the BP neural network;

[0195] S4: When predicting the remaining fatigue life of reinforced concrete bridge decks in service, image recognition is used to quickly analyze the actual crack width w1 at the bottom of the reinforced concrete bridge decks in service, that is, the actual width. After the bridge deck is gridded with a size of 10 cm, the area proportion p1 of the cracked grid is calculated, that is, the actual cracked grid area percentage, and the measured deflection data δ1 is obtained. (w1, p1, δ1) is used as input to obtain the target fatigue number Ni corresponding to different target graded loads Ti.

[0196] S5: Taking (Ti, Ni) as the estimated historical vehicle load information of the in-service reinforced concrete bridge deck with unknown historical vehicle load information in the service life of a year, the finite element calculation model in S1 is applied to perform nonlinear calculation, and the fatigue damage information of the current in-service bridge deck is obtained, wherein the fatigue damage information is used to express the fatigue damage condition of the bridge deck.

[0197] S6: The historical vehicle load information and future vehicle load information for the bridge deck are determined by target load classification and target number of actions. Based on S5, nonlinear finite element calculations can be continued based on (Ti, Ni)~a as a prediction of historical loads and future vehicle loads until the bridge deflection or crack width exceeds the value specified in the specification (an indicator of fatigue damage to the bridge), thereby achieving quantitative prediction of the fatigue life of the bridge deck in service.

[0198] In an embodiment of the present invention, historical vehicle load information for the bridge deck is determined through the target load classification and the target number of actions; fatigue damage information for the bridge deck is determined based on the finite element calculation model through the historical vehicle load information, thereby achieving an assessment of the current fatigue damage state of the bridge deck in the absence of historical vehicle load information.

[0199] Furthermore, the embodiment of the present invention determines the future vehicle load information for the bridge deck through the target load classification and the target number of actions; and determines the fatigue life information for the bridge deck based on the finite element calculation model through the future vehicle load information and the fatigue damage information, thereby improving the accuracy and timeliness of the life prediction of in-service cracked reinforced concrete bridge deck components in the absence of historical vehicle load information.

[0200] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0201] Reference Figure 2 , shows a structural block diagram of a reinforced concrete bridge deck detection device provided in an embodiment of the present invention, which may specifically include the following modules:

[0202] A finite element calculation model building module 201 is used to build a finite element calculation model for the bridge deck;

[0203] A load data determination module 202, used to determine the load classification for the finite element calculation model and the number of actions for the load classification;

[0204] A calculation parameter determination module 203, used to determine the calculation parameters based on the load classification and the number of actions;

[0205] A simulation result generation module 204 is used to construct a data set using the calculation parameters; the data set is a high cycle fatigue nonlinear finite element calculation result data set; the data set includes deflection information for the bridge deck, average widths of cracks in the longitudinal and transverse directions of a number of units adjacent to the center after meshing the bridge deck, and a total proportion of cracked unit areas for bottom units;

[0206] A back propagation neural network construction module 205 is used to construct a back propagation neural network using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area;

[0207] An actual data acquisition module 206 is used to determine the actual width of the unit crack, the actual cracked grid area percentage of the bridge deck, and the measured deflection data;

[0208] A target data output module 207, configured to output a target load classification and a target number of actions for the target load classification based on the back propagation neural network according to the actual width of the unit crack, the actual cracking grid area percentage and the measured deflection data;

[0209] The detection result generating module 208 is used to generate the detection result for the bridge deck based on the target load classification and the target number of actions.

[0210] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0211] In addition, an embodiment of the present invention also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the various processes of the above-mentioned embodiment of the reinforced concrete bridge deck detection method are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0212] The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, each process of the above-mentioned embodiment of the reinforced concrete bridge deck detection method is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0213] Figure 3A schematic diagram of the hardware structure of an electronic device for implementing various embodiments of the present invention.

[0214] The electronic device 300 includes but is not limited to: a radio frequency unit 301, a network module 302, an audio output unit 303, an input unit 304, a sensor 305, a display unit 306, a user input unit 307, an interface unit 308, a memory 309, a processor 310, and a power supply 311. Those skilled in the art will appreciate that Figure 3 The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and the electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. In the embodiments of the present invention, the electronic device includes but is not limited to a mobile phone, a tablet computer, a laptop computer, a PDA, a vehicle-mounted terminal, a wearable device, and a pedometer.

[0215] It should be understood that in the embodiment of the present invention, the radio frequency unit 301 can be used for receiving and sending signals during information transmission or communication. Specifically, after receiving downlink data from the base station, it is sent to the processor 310 for processing; in addition, uplink data is sent to the base station. Generally, the radio frequency unit 301 includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc. In addition, the radio frequency unit 301 can also communicate with the network and other devices through a wireless communication system.

[0216] The electronic device provides users with wireless broadband Internet access through the network module 302, such as helping users to send and receive emails, browse web pages, and access streaming media.

[0217] The audio output unit 303 can convert the audio data received by the RF unit 301 or the network module 302 or stored in the memory 309 into an audio signal and output it as sound. Moreover, the audio output unit 303 can also provide audio output related to a specific function performed by the electronic device 300 (for example, a call signal reception sound, a message reception sound, etc.). The audio output unit 303 includes a speaker, a buzzer, a receiver, etc.

[0218] The input unit 304 is used to receive audio or video signals. The input unit 304 may include a graphics processor (GPU) 3041 and a microphone 3042. The graphics processor 3041 processes the image data of a static picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The processed image frame can be displayed on the display unit 306. The image frame processed by the graphics processor 3041 can be stored in the memory 309 (or other storage medium) or sent via the radio frequency unit 301 or the network module 302. The microphone 3042 can receive sound and can process such sound into audio data. The processed audio data can be converted into a format output that can be sent to a mobile communication base station via the radio frequency unit 301 in the case of a telephone call mode.

[0219] The electronic device 300 also includes at least one sensor 305, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor includes an ambient light sensor and a proximity sensor, wherein the ambient light sensor can adjust the brightness of the display panel 3061 according to the brightness of the ambient light, and the proximity sensor can turn off the display panel 3061 and / or the backlight when the electronic device 300 is moved to the ear. As a type of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in each direction (generally three axes), and can detect the magnitude and direction of gravity when stationary, which can be used to identify the posture of the electronic device (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; the sensor 305 can also include a fingerprint sensor, a pressure sensor, an iris sensor, a molecular sensor, a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, etc., which will not be repeated here.

[0220] The display unit 306 is used to display information input by the user or information provided to the user. The display unit 306 may include a display panel 3061, which may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0221] The user input unit 307 can be used to receive input digital or character information, and generate key signal input related to user settings and function control of the electronic device. Specifically, the user input unit 307 includes a touch panel 3071 and other input devices 3072. The touch panel 3071, also known as a touch screen, can collect user touch operations on or near it (such as operations performed by users using fingers, styluses, or any other suitable objects or accessories on or near the touch panel 3071). The touch panel 3071 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact point coordinates, and then sends it to the processor 310, receives the command sent by the processor 310 and executes it. In addition, the touch panel 3071 can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic waves. In addition to the touch panel 3071, the user input unit 307 may also include other input devices 3072. Specifically, other input devices 3072 may include but are not limited to a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which are not described in detail here.

[0222] Furthermore, the touch panel 3071 may be overlaid on the display panel 3061. When the touch panel 3071 detects a touch operation on or near it, it is transmitted to the processor 310 to determine the type of touch event. Then, the processor 310 provides corresponding visual output on the display panel 3061 according to the type of touch event. Figure 3 In the figure, the touch panel 3071 and the display panel 3061 are used as two independent components to realize the input and output functions of the electronic device. However, in some embodiments, the touch panel 3071 and the display panel 3061 can be integrated to realize the input and output functions of the electronic device, which is not limited here.

[0223] The interface unit 308 is an interface for connecting an external device to the electronic device 300. For example, the external device may include a wired or wireless headset port, an external power supply (or battery charger) port, a wired or wireless data port, a memory card port, a port for connecting a device with an identification module, an audio input / output (I / O) port, a video I / O port, a headphone port, etc. The interface unit 308 may be used to receive input (e.g., data information, power, etc.) from an external device and transmit the received input to one or more elements within the electronic device 300 or may be used to transmit data between the electronic device 300 and an external device.

[0224] The memory 309 can be used to store software programs and various data. The memory 309 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory 309 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0225] The processor 310 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire electronic device. By running or executing software programs and / or modules stored in the memory 309 and calling data stored in the memory 309, it performs various functions of the electronic device and processes data, thereby monitoring the electronic device as a whole. The processor 310 may include one or more processing units; preferably, the processor 310 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 310.

[0226] The electronic device 300 may also include a power source 311 (such as a battery) for supplying power to various components. Preferably, the power source 311 may be logically connected to the processor 310 through a power management system, thereby implementing functions such as charging, discharging, and power consumption management through the power management system.

[0227] In addition, the electronic device 300 includes some functional modules not shown, which will not be described in detail here.

[0228] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0229] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0230] like Figure 4 As shown, in another embodiment provided by the present invention, a computer-readable storage medium 401 is also provided, in which instructions are stored. When the computer-readable storage medium 401 is run on a computer, the computer executes the reinforced concrete bridge deck detection method described in the above embodiment.

[0231] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

[0232] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the embodiments of the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0233] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0234] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0235] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0236] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0237] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0238] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for detecting reinforced concrete bridge decks, characterized in that: include: Constructing a finite element calculation model for the bridge deck; Determining a load classification for the finite element calculation model and a number of actions for the load classification; The load classification is preset according to the bridge type, and the number of actions is preset according to the design service life of the bridge; Determining calculation parameters based on the load classification and the number of actions; constructing a data set using the calculation parameters; The data set is a data set of high cycle fatigue nonlinear finite element calculation results; the data set includes deflection information for the bridge deck, average widths of cracks in the longitudinal and transverse directions of a number of units adjacent to the center after meshing the bridge deck, and a total proportion of cracked unit areas for bottom units; Constructing a back propagation neural network using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area; determining actual widths of cracks for the elements, and actual cracked mesh area percentages for the bridge deck, and measured deflection data; Outputting a target load classification and a target number of actions for the target load classification based on the back propagation neural network according to the actual width of the unit crack, the actual cracking grid area percentage and the measured deflection data; An inspection result for the bridge deck is generated based on the target load classification and the target number of actions.

2. The method according to claim 1, characterized in that The step of constructing a finite element calculation model for the bridge deck comprises: determining geometry and material properties for the bridge deck; Generate finite element mesh; A finite element computational model for the bridge deck is constructed based on the geometric shape, the material properties and the finite element mesh.

3. The method according to claim 1, characterized in that The step of determining the calculation parameters based on the load classification and the number of actions comprises: Determine the target sampling algorithm; A random sampling result is determined from the load classification and the number of actions based on the target sampling algorithm, and the random sampling result is determined as a calculation parameter.

4. The method according to claim 1, characterized in that: The step of constructing a back propagation neural network using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area comprises: Determine a minimum error function using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area; A back propagation neural network is constructed based on the minimum error function.

5. The method according to claim 1 or 2, characterized in that: The step of constructing a data set using the calculation parameters and generating simulation results based on the data set includes: Determine the convergence criterion; constraining a predetermined nonlinear solver using the convergence criterion; A data set is constructed using the calculation parameters, and a simulation result is generated based on the data set by the nonlinear solver.

6. The method according to claim 1, characterized in that The step of generating the detection result for the bridge deck based on the target load classification and the target number of actions includes: Determining historical vehicle load information for the bridge deck according to the target load classification and the target number of actions; Fatigue damage information for the bridge deck is determined based on the finite element calculation model using the historical vehicle load information.

7. The method according to claim 6, characterized in that Also includes: Determining future vehicle load information for the bridge deck according to the target load classification and the target number of actions; The fatigue life information of the bridge deck is determined based on the finite element calculation model using the future vehicle load information and the fatigue damage information.

8. The method according to claim 3, characterized in that The target sampling algorithm is a Monte Carlo random sampling algorithm.

9. The method according to claim 3, characterized in that: The target sampling algorithm is a Latin hypercube sampling algorithm.

10. The method according to claim 2, characterized in that The mesh size of the finite element mesh is ten centimeters.

11. The method according to claim 2, characterized in that Also includes: A loading condition is determined, wherein the loading condition is a static load.

12. The method according to claim 2, characterized in that: Also includes: A loading condition is determined, wherein the loading condition is a dynamic load.

13. A detection device for reinforced concrete bridge deck, characterized in that: include: A finite element calculation model building module, used to build a finite element calculation model for the bridge deck; A load data determination module, used to determine the load classification for the finite element calculation model and the number of actions for the load classification; The load classification is preset according to the bridge type, and the number of actions is preset according to the design service life of the bridge; A calculation parameter determination module, used to determine the calculation parameters based on the load classification and the number of actions; A simulation result generation module, used to construct a data set using the calculation parameters; the data set is a high cycle fatigue nonlinear finite element calculation result data set; the data set includes deflection information for the bridge deck, and average widths of cracks in the longitudinal and transverse directions of a number of units adjacent to the center after meshing the bridge deck, and a total proportion of cracked unit areas for bottom units; A back propagation neural network construction module, used to construct a back propagation neural network using the calculation parameters, the deflection information, the average width and the total proportion of the crack unit area; An actual data acquisition module, for determining an actual width of a unit crack, an actual cracked grid area percentage of the bridge deck, and measured deflection data; A target data output module, configured to output a target load classification and a target number of actions for the target load classification based on the back propagation neural network according to the actual width of the unit crack, the actual cracking grid area percentage and the measured deflection data; A test result generating module is used to generate a test result for the bridge deck based on the target load classification and the target number of actions.

14. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; The memory is used to store computer programs; The processor is used to implement the method according to any one of claims 1 to 12 when executing the program stored in the memory.

15. A computer-readable storage medium having instructions stored thereon, which, when executed by one or more processors, cause the processors to perform the method according to any one of claims 1 to 12.

Images