Bridge bearing capacity evaluation method under random vehicle flow

Abstract

This invention discloses a method for assessing the load-bearing capacity of bridges under random traffic flow, belonging to the technical field of bridge load-bearing capacity assessment. This method improves the accuracy and efficiency of the assessment by extracting the bridge deflection influence line and calculating the compliance matrix through traffic flow statistics, Monte Carlo simulation, finite element analysis, and signal processing. Employing non-destructive testing technology, this method provides real-time health status of the bridge. By analyzing dynamic response data during heavy vehicle travel and combining it with the deflection influence line model, the method predicts the bridge's load-bearing limit and potential damage areas, providing a scientific basis for bridge safety assessment and maintenance. This invention is characterized by high efficiency, accuracy, and strong applicability, avoiding the problems of long testing times, traffic disruption, and large workload associated with traditional static load tests. It is suitable for the load-bearing capacity assessment and condition monitoring of small and medium-span bridges.

Description

Technical Field

[0001] This invention relates to the field of bridge load-bearing capacity assessment technology, specifically a method for assessing the load-bearing capacity of bridges under random traffic flow. Background Technology

[0002] After a period of large-scale construction, my country's bridge industry has entered a long phase of operation and maintenance. As bridges age, all will eventually enter their aging and maintenance period, resulting in a massive workload for inspection and evaluation. Bridge load-bearing capacity assessment is crucial for ensuring their operational safety. Traditionally, assessment methods primarily rely on static load testing and visual inspection. Static load testing measures the static deflection and strain values ​​at the most unfavorable location on the bridge, calculating the deflection or strain verification coefficient to assess its load-bearing capacity. However, static load testing suffers from problems such as heavy-load static placement, traffic disruption, high testing costs, long testing cycles, and low efficiency. It may also cause secondary damage to the bridge during the test, making it unsuitable for large-scale and high-frequency use. For bridges in poor condition, static load testing carries certain risks when the loading efficiency is high; when the loading efficiency is low, it fails to meet the specifications. In contrast, dynamic load testing offers advantages such as high testing efficiency, low cost, and short cycle time. Using dynamic load testing instead of static load testing to assess bridge condition can accurately reflect the bridge's operational status. Bridge load-bearing capacity assessment is key to ensuring the safe operation of bridges. Traditional methods typically rely on static load tests or simplified models, but these cannot accurately reflect the bridge response under the dynamic effects of random traffic flow. Existing techniques sometimes use influence lines or compliance matrices for evaluation, but these often require manual measurement or assumptions about load distribution, leading to low efficiency and large errors. Furthermore, extracting quasi-static components from the dynamic response is frequently affected by noise, impacting accuracy. Therefore, a method is needed to quickly and accurately assess the bridge's load-bearing capacity under random traffic flow. Summary of the Invention

[0003] To address the aforementioned issues, this invention provides a method for assessing the load-bearing capacity of bridges under random traffic flow. By integrating traffic flow statistics, Monte Carlo simulation, finite element analysis, and signal processing, this invention achieves automatic extraction of the bridge deflection influence line and calculation of the compliance matrix, thereby improving the accuracy and efficiency of the assessment.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for assessing the bearing capacity of bridges under random traffic flow, specifically including the following steps: S1. Collect actual traffic flow data through weighing sensors and cameras; The actual traffic flow data includes, but is not limited to, vehicle axle load, speed, and spacing. S2. Based on actual traffic flow data, random traffic flow samples are generated using the Monte Carlo method. This invention utilizes statistical laws to generate random traffic flow samples based on Monte Carlo simulation, and then transforms the traffic flow samples into loads acting on the bridge.

[0005] S3. Construct a bridge model, apply random traffic loads to the bridge deck, perform transient dynamic analysis, and output the bridge deflection time history response signal. This invention establishes a bridge model using ANSYS APDL or other finite element software, applies random traffic loads to the model, and calculates the bridge's dynamic response (such as the deflection time history response signal u(t)). To improve the stability of subsequent decomposition, u(t) can be preprocessed as necessary.

[0006] S4. Based on the bridge deflection time history response signal, after performing preprocessing operations, extract the IMF components through EMD decomposition, and output the IMF components and residuals. The EMD decomposition and extraction of IMF components specifically involves identifying extreme points, constructing envelopes, calculating local means, extracting detail components, executing stop criteria, updating residuals, and iterative judgment. Specifically, the following steps are included: S4.1 Input the bridge deflection time history response signal, and after preprocessing, output the preprocessed deflection time history response signal and the initial residual signal; The preprocessing operations specifically include detrending, noise reduction filtering, and initialization; Specifically, the detrending term in this invention involves removing linearly or slowly changing trend terms from the bridge deflection time history response signal to prevent them from being misidentified as low-frequency components in EMD decomposition.

[0007] Linear regression is used to detect trend components in the signal. When the absolute value of the t-statistic of the trend term is greater than 2.0 or the coefficient of determination R² is greater than 0.05, a significant trend is identified and removed. The specific steps are as follows: 1) Perform linear regression fitting on the signal: ,in, For the intercept term, For the slope term, It is a time variable; 2) Calculate the t-statistic of the slope: ,in The standard error of the slope; 3) Calculate the coefficient of determination: Where SSE is the residual sum of squares and SST is the total sum of squares.

[0008] 4) When or At that time, subtract the fitted trend from the original signal: ; 5) When the trend is not significant, retain the original signal without processing.

[0009] In this invention, the noise reduction filtering uses wavelet threshold denoising for preliminary smoothing to suppress the influence of high-frequency noise on extreme point identification.

[0010] The wavelet thresholding method is used for signal preprocessing. Specifically, the Symlets 4 (sym4) wavelet basis function is used for 5-level decomposition, and the Stein unbiased risk estimation threshold rule is used to perform soft thresholding on the detail coefficients of each level.

[0011] The specific steps are as follows: 1) The Symlets 4 (sym4) wavelet is selected as the basis function. This wavelet has approximately symmetry and tight support characteristics, a support length of 7, and a fourth-order vanishing moment. It achieves a good balance between phase distortion and time-frequency resolution, and is particularly suitable for processing transient signals such as bridge dynamic response.

[0012] 2) Determining the number of decomposition layers: based on the signal sampling frequency. The number of decomposition layers is determined by the main frequency components of the bridge structure. , among them f c The highest frequency component that needs to be retained. For a typical bridge, f s =100Hz, f c =10Hz, the calculated Y=3, but considering the noise frequency band distribution and engineering practice experience, this invention adopts a 5-level decomposition, and the corresponding frequency bands are divided as follows: approximation coefficient A5: 0-1.5625 Hz, detail coefficient D5: 1.5625-3.125 Hz, detail coefficient D4: 3.125-6.25Hz, detail coefficient D3: 6.25-12.5 Hz, detail coefficient D2: 12.5-25 Hz, detail coefficient D1: 25-50 Hz.

[0013] 3) Threshold rules and processing: Stein unbiased risk estimation (rigrsure) is used as the threshold rule. This rule adaptively determines the threshold size based on the principle of minimizing the risk function.

[0014] Initialization: Set the residual signal = IMF serial number j=1.

[0015] S4.2 Based on the preprocessed deflection time history response signal and the initial residual signal, the IMF component is extracted by EMD decomposition and then the IMF component is output. S4.2.1 Identifying extreme points: For the current signal (t)(Initial time) h j,0 (t)= (identify all its local maxima and local minima), where, (t) represents the k-th digit of the j-th IMF. Signal of the first iteration.

[0016] S4.2.2 Constructing the envelope: Use cubic spline interpolation to interpolate the maxima and minima respectively to form the upper envelope. (t) and lower envelope (t).

[0017] S4.2.3 Calculate the local mean: Calculate the instantaneous mean of the upper and lower envelopes to obtain the local mean function, expressed as follows: m(t) = .

[0018] S4.2.4 Extracting detail components: Subtracting the local mean from the current signal to obtain a new signal. (t) = h j,k 1 (t) m(t) .

[0019] S4.2.5, Criteria for Suspension: Judging Candidate IMFs (t) Whether it meets the two conditions of the IMF, specifically including: (1) Over the entire signal length, the number of extreme points is equal to or at most differs from the number of zero-crossing points by one.

[0020] (2) At any time, the mean of the upper envelope defined by the local maxima and the lower envelope defined by the local minima is zero.

[0021] If the conditions are met, then define the IMF component. (t)= (t) represents the j-th IMF component; otherwise, (t) Repeat the EMD decomposition and IMF component extraction steps as new data until the conditions are met.

[0022] S4.2.6 Update Residuals: Subtract the newly obtained IMF from the current residuals to obtain the new residuals. (t)= (t) (t).

[0023] S4.2.7, Loop Check: Check residuals (t) Whether it has become a monotonic function or contains only one extreme point (indicating that the IMF cannot be further decomposed). If not, let j = j + 1 and return to step 2 to start extracting the next IMF component.

[0024] S5. Based on the IMF components and the bridge fundamental frequency, the main frequency is analyzed by FFT. After screening out the IMF components with frequencies lower than the bridge structure fundamental frequency, the signal is reconstructed to obtain the quasi-static time history response. In this invention, the IMF components c1(t), c2(t), ..., obtained from EMD decomposition are... (t) Frequency arranged from high to low. To extract the quasi-static response, IMF components with frequencies lower than the fundamental frequency of the bridge structure need to be selected.

[0025] Specifically, the following steps are included: S5.1, the IMF components obtained from the decomposition ( t Perform a Fast Fourier Transform (FFT) to obtain the analytic signal, and then calculate the instantaneous frequency. The dominant frequency can be taken as the statistical mode or median of the instantaneous frequency.

[0026] S5.2 Setting the frequency threshold: Frequency threshold It can be set as the first-order vertical fundamental frequency of the bridge. of Times (in this invention) The value range is from 1.2 to 1.5, that is... This tolerance factor This is to ensure that all low-frequency components related to quasi-static effects are included.

[0027] S5.3 Filtering and Reconstruction Based on Frequency Thresholds: Filter and reconstruct all main frequencies ≤ The quasi-static time history response of the bridge is obtained by superimposing the IMF components. (t), the expression is as follows:

[0028] in, (t) represents the response mainly caused by the gravity of the moving vehicle after filtering out high-frequency dynamic interference.

[0029] S6. Based on the quasi-static response, through time-domain to spatial-domain conversion, signal interception and data alignment, the aligned static response is output. Specifically, the following steps are included: S6.1 Time-domain to spatial-domain conversion: via the relation x=v t, converts the time variable t on the time axis into the vehicle's position coordinate x on the bridge, where v is the vehicle speed.

[0030] S6.2 Signal Capture and Data Alignment: Capture the signal during the time a single vehicle crosses the bridge. (x) Data. To ensure the comparability of influence lines extracted from multiple events, alignment is required starting from a reference point. In this invention, the arrival of the vehicle at the bridgehead is taken as the reference point, and multiple vehicle passage data are aligned to obtain the aligned static response.

[0031] S7. Based on the aligned static response, after constructing the deflection influence line expression by using the multi-segment basis function iterative fitting method IFCM, construct the response-deflection influence line relationship equation and solve for the output deflection influence line. S7.1 Based on the aligned static response, the deflection influence line of the bridge is extracted using the multi-segment basis function iterative fitting method (IFCM): Let the expression for the deflection influence line be:

[0032] In the formula, It is the first undetermined coefficient. It is the second undetermined coefficient. It is the third undetermined coefficient. It is the first Coordinates within the segment range, For segment numbers, This represents the total number of segments in the influence line.

[0033] S7.2. Based on the deflection influence line, establish the matrix relationship between vehicle axle load and bridge response; Taking a three-axle vehicle as an example, the actual response of this invention can be expressed as:

[0034] In the formula, It is the quasi-static response of the bridge. These are the influence lines to be extracted. It is the number of points on the influence line. It is the first zero vector. It is the second zero vector, and its size is determined by the vehicle distance and the segment length. This is the weighting coefficient for the first axle load of the vehicle. This is the weighting coefficient for the second axle load of the vehicle. The weighting coefficient for the third axle load of the vehicle; the deflection influence line IL can be composed of multiple basis functions, as shown in the following expression:

[0035] in, ,and It is the first The coefficient vector of the segment can be solved using the constructed basis function equations. The point number within the segment; S8. Based on the deflection influence line, calculate the flexibility matrix and estimate the static load deflection, and output the bridge flexibility matrix and estimated static load deflection. S8.1 Assemble the deflection influence lines into a deflection influence line matrix and calculate the scaling factor. The calculation expression is as follows:

[0036] In the formula, Let n be the value at the nth point on the kth deflection influence line of the bridge. For matrix The value at the nth point on the kth deflection influence line, where K is the number of traffic flow samples and N is the number of bridge degrees of freedom; After calculating the scaling factor, the bridge flexibility matrix is ​​solved, and the expression is as follows:

[0037] In the formula, To construct a new matrix, G deflection influence lines were selected at points G in the bridge. for The transpose of the matrix; S8.2. Estimate the static load deflection of the bridge based on the bridge flexibility matrix: Convert the actual static load into an equivalent nodal load vector P, and multiply the flexibility matrix with the load vector P to obtain the estimated static load deflection.

[0038] S9. Based on the estimated static load deflection and theoretical model deflection, the bearing capacity is evaluated by calculating the deflection verification coefficient, and the bearing capacity evaluation conclusion is obtained, thus completing the bridge bearing capacity evaluation method under random traffic flow. The bridge deflection verification coefficient is calculated using the estimated static load deflection and the static load deflection obtained from the theoretical model established by Midas Civil software: when When this occurs, it indicates that the actual load-bearing capacity of the bridge meets the design requirements; when This indicates that the actual load-bearing capacity of the bridge does not meet the design requirements.

[0039] Compared with the prior art, the present invention provides a method for evaluating the bearing capacity of bridges under random traffic flow, which has the following beneficial effects: (1) This invention collects actual traffic flow information, generates random traffic flow samples using statistical laws and Monte Carlo simulation, and converts them into bridge loads. Then, it obtains the dynamic response through finite element simulation. Subsequently, it extracts the bridge deflection influence line using signal processing technology (EMD decomposition), and finally calculates the flexibility matrix and static load deflection. Then, it extracts the bridge deflection influence line using the multi-segment basis function iterative fitting method (IFCM). Finally, it obtains the bridge flexibility matrix using the physical relationship between the influence line matrix and the bridge flexibility matrix. This effectively solves the problem that previous tests required a large number of measurement points and sensors to obtain the bridge flexibility matrix. It makes good use of dynamic testing to replace static load testing to obtain the flexibility matrix, reducing workload and test costs, and making bridge inspection fast and efficient.

[0040] (2) This invention obtains the flexibility matrix and then calculates the index bridge deflection verification coefficient. To assess the load-bearing capacity of bridges, static tests are transformed into dynamic tests, and the usual "qualitative evaluation by dynamic testing" is transformed into "quantitative evaluation by static testing." This effectively solves the problems of long testing time, large workload, high cost, high risk, and the need to disrupt traffic in traditional static load tests. Attached Figure Description

[0041] Figure 1 This is a flowchart of a method for evaluating the load-bearing capacity of bridges under random traffic flow according to the present invention.

[0042] Figure 2 This is a flowchart of the dynamic evaluation process for bridge load-bearing capacity based on random traffic flow, as described in this invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] Please see Figure 1 and Figure 2 A method for assessing the load-bearing capacity of bridges under random traffic flow includes the following steps: Traffic flow information collection: Install sensors (such as weighing sensors and cameras) on the target bridge to collect actual traffic flow data (such as vehicle axle load, speed, and vehicle distance) for at least one week to obtain statistical patterns. Monte Carlo simulation: Based on the collected data, random traffic flow samples are generated using computer programs (such as MATLAB). In this embodiment, three sets of traffic flow samples were generated to ensure statistical representativeness. Finite element simulation: A fine finite element model of the bridge is built using ANSYS APDL. Random traffic loads are applied to the bridge deck, and transient dynamic analysis is performed. The deflection time history response u(t) of the key points of the bridge (abutment L=0m, mid-span L=10m, 1 / 4L=5m, 3 / 4L=15m, and bridge tail L=20m) is output. Signal processing: Implement the EMD decomposition algorithm in MATLAB. Specific steps include: Extreme point identification and envelope fitting are performed on u(t); The IMF components are obtained through iterative decomposition, and the frequencies are analyzed using FFT. The stopping condition is that the main frequency of the IMF is lower than the fundamental frequency of the bridge (which can be obtained through modal analysis). The quasi-static response is reconstructed, and the deflection influence line is extracted using piecewise polynomial fitting. Compliance matrix calculation: Assemble multiple sets of deflection influence lines into a matrix, calculate the scaling factor C, and then solve for the compliance matrix [F]. Static load deflection estimation: Generate a nodal load vector P based on the actual static load (such as the design load), calculate the static load deflection, and compare it with the theoretical value to obtain a verification coefficient. A verification coefficient close to 1 indicates accurate evaluation.

[0045] This embodiment takes a simply supported beam with a single span of L=20m as an example to illustrate the specific details of each step: The fundamental frequency of the bridge was obtained through modal analysis: the first-order vertical frequency. =3.0Hz, frequency threshold ,coefficient α Set to 1.2; Three sets of random traffic flow samples were obtained by analyzing traffic flow data and generating random traffic flow through Monte Carlo simulation.

[0046] Sample 1: Vehicle 1 (weight 160 kN, speed 20 m / s, appearance time t = 0 s) Vehicle 2 (weight 120 kN, speed 18 m / s, appearance time t = 2.5 s) Vehicle 3 (weight 200 kN, speed 22 m / s, appearance time t = 5.0 s) Sample 2: Vehicle 1 (weight 80 kN, speed 25 m / s, appearance time t=0 s) Vehicle 2 (weight 240 kN, speed 15 m / s, appearance time t = 3.0 s) Sample 3: Vehicle 1 (weight 120 kN, speed 20 m / s, appearance time t = 0 s) Vehicle 2 (weight 160 kN, speed 18 m / s, appearance time t = 1.8 s) Vehicle 3 (weight 200 kN, speed 20 m / s, appearance time t = 4.2 s) Transient dynamic analysis was performed using ANSYS APDL with a time step of Δt = 0.01s and a total duration of T = 10s. The original deflection time history response at the mid-span position is shown in Table 1. Table 1: Original Deflection Time History Data

[0047] EMD decomposition process (taking t=4.0s as an example). First, local maxima and minima are identified in the original deflection time history response u(t). Here, the mid-span position is still used as an example, as shown in Table 2: Table 2: Extreme Point Identification Table

[0048] Fit the envelope using cubic spline interpolation; Calculate the local mean (taking t=4.0s as an example): s(t)=(10.158+9.832) / 2=9.995mm; Extract the first IMF component (taking t=4.0s as an example): h1=10.021-9.995=0.026mm; Iterate the above process using MATLAB until... (t) satisfies the IMF conditions, and after 4 iterations, a qualified result is obtained. (t). After EMD decomposition, the original signal was decomposed into four IMF components and one residual, as shown in Table 3; Table 3: EMD Decomposition Results

[0049] The frequency was analyzed using FFT, and the following results were obtained, as shown in Table 4. Table 4: IMF Frequency Analysis Table;

[0050] Reconstructing the quasi-static response: (t) = IMF4(t) + residual(t). The quasi-static response results are shown in Table 5. Table 5: Quasi-static response reconstruction table

[0051] Extracting the deflection influence line (taking sample 1 (vehicle weight 160KN, vehicle speed 20m / s) as an example): First, perform a time-domain to spatial-domain transformation x=vt=20t, then normalize the quasi-static response IL(x)= (x) / p= (x) / 160. The influence line data is extracted and shown in Table 6; Table 6: Example Table of Influence Line Extraction

[0052] Repeat the above process to extract influence lines for the three groups of samples at five measuring points (abutment, 1 / 4 span, mid-span, 3 / 4 span, and bridge tail).

[0053] Assembly Influence Line Matrix: In this embodiment, 5 measuring points were selected (x=0m, 5m, 10m, 15m, 20m), and the resulting deflection influence line matrix is ​​shown in Table 7. Table 7: Deflection Influence Line Matrix

[0054] according to The scaling factor was calculated, where K = 3 (number of traffic flow samples) and N = 5 (number of measurement points) × 5 (number of locations) = 25. The calculated scaling factor C = 0.33480 / 75 = 0.004464. according to Solve for the bridge flexibility matrix: Given (Theoretical value), the calculation process is as follows: , , , Finally, the expression for the bridge flexibility matrix is ​​obtained through eigenvalue decomposition:

[0055] Based on the obtained bridge flexibility matrix, the static load deflection of the bridge is estimated (taking measuring point 3 as an example): P= , ; According to the "Specification for Testing and Evaluation of Bearing Capacity of Highway Bridges" (JTG / TJ21-2011), the deflection verification coefficient of the bridge is calculated from the static load deflection obtained above and the static load deflection obtained through the theoretical model (taking measuring point 3 at the mid-span as an example): the theoretical deflection at this point is known to be 8.684 mm (obtained from the calculation formula for a simply supported beam). The deflection estimated by the above process is 15.142 mm, and the relative error between the two is... The calculated deflection verification coefficient is 1.744.

[0056] The above example uses three sets of random traffic flow samples and basic EMD decomposition, representing a simplified demonstration of the method flow. Its purpose is to clearly demonstrate the operational process of each step. To prove the accuracy of this method in practical applications, this embodiment adds an optimized example, employing the following reasonable configuration: Sample size: 10 random traffic flow samples; Signal decomposition: Empirical Empirical Mode Decomposition (EEMD, ensemble order 100). Measurement point density: 9 measurement points (0m, 2.5m, 5m, 7.5m, 10m, 12.5m, 15m, 17.5m, 20m); Matrix solving: Regularized least squares method (regularization parameter is 0.01) The optimized results of the new concept extraction and the static load deflection estimation are shown below: Table 8: Influence line extraction results of the optimized case (taking the mid-span position as an example)

[0057] Table 9: Influence Line Matrix of Optimization Examples

[0058] The scaling factor C was calculated to be 6.3696 / 450 = 0.014155. Solving by eigenvalue decomposition Based on the obtained bridge flexibility matrix, the static deflection of the bridge (taking the mid-span as an example) was estimated, and the results are shown in the table below: Table 10: Static Load Deflection Estimation Results

[0059] The theoretical deflection at this point is known to be 8.684 mm (obtained from the formula for simply supported beams). The deflection estimated by the above process is 8.971 mm, and the relative error between the two is... The calculated deflection verification coefficient is 1.033.

[0060] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the load-carrying capacity of a bridge under random vehicle flow, characterized in that, Includes the following steps: S1. Collect actual traffic flow data through weighing sensors and cameras; The actual traffic flow data includes vehicle axle load, speed, and spacing; S2. Based on actual traffic flow data, random traffic flow samples are generated using the Monte Carlo method. S3. Based on random traffic flow samples, a bridge model is constructed. By applying random traffic flow loads to the bridge deck, transient dynamic analysis is performed, and the bridge deflection time history response signal is output. S4. Based on the bridge deflection time history response signal, after performing preprocessing operations, extract the IMF components through EMD decomposition, and output the IMF components and residuals. The EMD decomposition and extraction of IMF components specifically involves identifying extreme points, constructing envelopes, calculating local means, extracting detail components, executing stop criteria, updating residuals, and iterative judgment. S5. Based on the IMF components and the bridge fundamental frequency, the main frequency is analyzed by FFT. After screening out the IMF components with frequencies lower than the bridge structure fundamental frequency, the signal is reconstructed to obtain the quasi-static time history response. S6. Based on the quasi-static response, through time-domain to spatial-domain conversion, signal interception and data alignment, the aligned static response is output. S7. Based on the aligned static response, after constructing the deflection influence line expression by using the multi-segment basis function iterative fitting method IFCM, construct the response-deflection influence line relationship equation and solve for the output deflection influence line. S8. Based on the deflection influence line, calculate the flexibility matrix and estimate the static load deflection, and output the bridge flexibility matrix and estimated static load deflection. S9. Based on the estimated static load deflection and theoretical model deflection, the bearing capacity is assessed by calculating the deflection verification coefficient, and the bearing capacity assessment conclusion is obtained, thus completing the method for assessing the bearing capacity of bridges under random traffic flow.

2. The method for assessing the bearing capacity of a bridge under random traffic flow according to claim 1, characterized in that, In the signal processing, step S4 specifically includes the following steps: S4.1 Input the bridge deflection time history response signal, and after preprocessing, output the preprocessed deflection time history response signal and the initial residual signal; The preprocessing operations specifically include detrending, noise reduction filtering, and initialization; The detrending term specifically refers to removing linearly or slowly changing trend terms from the bridge deflection time history response signal; The noise reduction filtering uses wavelet threshold denoising for preliminary smoothing; wherein the wavelet basis function is Symlets 4, the decomposition level is 5, and the threshold rule uses Stein unbiased risk estimation. S4.2 Based on the preprocessed deflection time history response signal and the initial residual signal, the IMF component is extracted by EMD decomposition and then the IMF component is output.

3. The method for assessing the bearing capacity of a bridge under random traffic flow according to claim 2, characterized in that, In the signal processing, step S4.2 specifically includes the following steps: S4.2.1 Identify extreme points: When the current signal is initially set, identify all local maxima and local minima; S4.2.2 Constructing the envelope: Use cubic spline interpolation to interpolate the maximum and minimum points respectively to form the upper and lower envelopes; S4.2.3 Calculate the local mean: Find the instantaneous mean of the upper and lower envelopes to obtain the local mean function; S4.2.4 Extracting detail components: Subtracting the local mean from the current signal to obtain a new signal; S4.2.5, Criteria for Determination: This criterion determines whether a candidate IMF meets two conditions for becoming an IMF, specifically including: (1) The difference between the number of extreme points and the number of zero crossings over the entire signal length is within a preset threshold range, where the preset threshold is 1; (2) At any given time, the mean of the upper envelope defined by the local maxima and the lower envelope defined by the local minima is zero; S4.2.6 Update Residuals: Subtract the newly obtained IMF from the current residuals to obtain the new residuals; S4.2.7, Loop Check: Check whether the residual has become a monotonic function or contains only one extreme point.

4. The method for assessing the bearing capacity of a bridge under random traffic flow according to claim 1, characterized in that, In the signal processing, step S5 specifically includes the following steps: S5.1 Perform a fast Fourier transform on the IMF components to obtain the analytical signal and calculate the instantaneous frequency; S5.

2. Based on the preset multiple of the first-order vertical fundamental frequency of the beam, set the frequency threshold; S5.2 Filtering and reconstruction based on frequency thresholds.

5. The method for assessing the bearing capacity of a bridge under random traffic flow according to claim 1, characterized in that, In the signal processing, step S6 specifically includes the following steps: S6.1 Time Domain to Spatial Domain Conversion: Using the relationship between speed and time, the time variable on the time axis is converted into the vehicle's position coordinates on the bridge; S6.2 Signal capture and data alignment: Capture quasi-static time history response data during the single-vehicle bridge crossing period.

6. The method for assessing the bearing capacity of a bridge under random traffic flow according to claim 1, characterized in that, In the signal processing, step S7 specifically includes the following steps: S7.1 Based on the aligned static response, the deflection influence line of the bridge is extracted using the multi-segment basis function iterative fitting method; S7.

2. Based on the deflection influence line, establish the matrix relationship between vehicle axle load and bridge response.

7. The method for assessing the bearing capacity of a bridge under random traffic flow according to claim 1, characterized in that, In the signal processing, step S8 specifically includes the following steps: S8.1 Assemble the deflection influence lines into a deflection influence line matrix, calculate the scaling factor, and then solve for the bridge flexibility matrix. S8.2 Calculate the estimated static load deflection based on the bridge flexibility matrix; specifically, convert the actual static load into an equivalent nodal load vector, and multiply the bridge flexibility matrix with the load vector to obtain the estimated static load deflection.

8. The method for assessing the bearing capacity of a bridge under random traffic flow according to claim 1, characterized in that, In the signal processing, the carrying capacity assessment in S9 specifically includes: when When this occurs, it indicates that the actual load-bearing capacity of the bridge meets the design requirements; when This indicates that the actual load-bearing capacity of the bridge does not meet the design requirements.

9. The method for assessing the bearing capacity of a bridge under random traffic flow according to claim 1, characterized in that, The deflection verification coefficient is the ratio of the estimated static load deflection to the theoretical static load deflection.

Images