A method and system for early warning of hanger fracture in a through-arch bridge

Abstract

This invention discloses a method and system for early warning of hanger fracture in through-arch bridges, relating to the field of arch bridge structural health monitoring technology. The invention uses a deployed sensor array to collect real-time stress data, vibration frequency data, and environmental temperature and humidity data of the hangers. The collected data is preprocessed to remove outliers and noise. Based on the preprocessed data, the stress change rate and vibration frequency offset of the hangers are calculated, and the monitoring data is corrected using an environmental temperature and humidity compensation model. The corrected characteristic parameters are compared with preset early warning thresholds, and a multi-parameter fusion decision algorithm is used to determine the health status of the hangers, generating corresponding early warning levels and triggering early warning signals. This invention achieves accurate and real-time early warning of hanger fracture risk, effectively improving the reliability and accuracy of early warnings, providing a scientific basis for bridge maintenance, and reducing the probability of safety accidents.

Description

Technical Field

[0001] This invention relates to the field of arch bridge structural health monitoring technology, and in particular to a method and system for early warning of hanger fracture in a through-arch bridge. Background Technology

[0002] Through-arch bridges are widely used in bridge engineering due to their aesthetically pleasing design, economical cost, and good spanning capacity. As a key load-bearing component of through-arch bridges, the suspenders play a crucial role in transferring bridge deck loads to the arch ribs, and their performance directly affects the overall structural safety of the bridge. However, during long-term service, suspenders are susceptible to load fatigue, corrosion, and environmental erosion, leading to performance degradation and even fracture failure, seriously threatening bridge traffic safety.

[0003] Existing technologies for early warning of boom fracture mainly include stress monitoring, vibration monitoring, and visual inspection. Stress monitoring acquires boom stress data by deploying stress sensors; however, traditional stress monitoring does not fully consider the interference of environmental temperature and humidity on the measurement results, leading to insufficient monitoring accuracy. Vibration monitoring assesses the condition based on the correlation between boom vibration frequency and stiffness; however, early warning based on a single vibration parameter has low reliability and is prone to false alarms or missed alarms. Visual inspection relies on manual inspection or drone photography, making real-time monitoring difficult and subject to significant environmental limitations.

[0004] Furthermore, existing early warning systems mostly rely on single-parameter threshold judgments, lacking multi-parameter fusion decision-making mechanisms. The classification of early warning levels is vague, failing to accurately reflect the actual health status of the suspenders. Simultaneously, some systems suffer from significant response delays and insufficient data transmission stability, making it difficult to meet the needs of real-time bridge structure monitoring. For example, Chinese invention patent CN112326891A discloses a bridge suspender health monitoring method that relies solely on stress changes for early warning judgment, neglecting the influence of vibration characteristics and environmental factors, resulting in limited accuracy. Chinese invention patent CN113533245B proposes a suspender monitoring system that uses wired transmission, making installation and maintenance inconvenient and lacking a dynamic threshold calibration mechanism, leading to a decline in early warning reliability after long-term use.

[0005] Therefore, there is an urgent need to develop a method and system for early warning of suspender fracture in mid- and under-deck arch bridges that can integrate multiple parameters, eliminate environmental interference, and achieve accurate real-time early warning, so as to solve the shortcomings of existing technologies. Summary of the Invention

[0006] In view of this, the present invention provides a method and system for early warning of suspender fracture in a mid-bearing arch bridge. By constructing a suspender cable force influence matrix, a small number of sensors can cover the entire bridge for monitoring. By combining interference elimination and cable force reconstruction technology with multi-algorithm fusion, the accuracy of early warning is improved. At the same time, dynamic iterative optimization ensures long-term service adaptability, thus achieving efficient early warning of suspender fracture.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: A method for early warning of hanger fracture in a through-arch bridge includes the following steps: S1 Finite Element Model Calibration and Reference State Establishment Steps: Based on the design data and measured data, establish and correct the finite element model of the arch bridge to ensure that the error between the finite element model and the measured data is less than the first preset value; collect the cable force data of the hangers under multiple working conditions in a healthy state, and establish the reference dead load cable force library of the hangers for the whole bridge after preprocessing; S2 Steps for constructing the influence matrix of suspender cable force: Apply a unit cable force variation to each suspender using a calibration model, analyze its impact on the cable force of each suspender, and obtain initial influence coefficients; normalize the initial influence coefficients to construct a standardized influence matrix. M Its element M ij Quantified the first i The effect of the change in the cable force of the first shunt on the second j The response relationship of the root rod; S3 Sensor Optimization and Quantity Determination Steps: Based on the standardized influence matrix M The system filters the element with the largest absolute value in each row, identifies the key shunts that have a significant impact on cable force, uses a greedy algorithm to reduce the number of sensors, and determines the optimal sensor arrangement scheme with the cable force reconstruction error being less than or equal to the second preset value as a constraint. The system then installs and calibrates the sensors at the selected locations. S4 Real-time monitoring and parameter extraction steps: Collect vibration and strain signals of the boom, and extract at least one state parameter of the boom based on wavelet packet decomposition; S5 Dynamic Mapping and Early Warning Decision-Making Steps: Compare the state parameters with preset thresholds, and directly query the pre-stored early warning mapping table to determine the corresponding early warning level based on the comparison results; S6 Environmental Coupling Correction Step: Combine real-time collected environmental temperature and humidity data, and dynamically correct the warning level through a temperature-humidity coupling correction factor; S7 Warning Output and Display Steps: Output warning signals according to the corrected warning level and display them in real time on the monitoring platform.

[0008] Optionally, the S1 finite element model calibration and baseline state establishment steps of the above method include the following steps: Based on bridge design drawings, material parameters and on-site measured data, S101 established a finite element model of a mid-bearing arch bridge. The model was corrected by adjusting the node constraints and material elastic modulus parameters so that the error between the model calculation results and the measured data was controlled within 5%. S102 Collect multiple sets of suspender cable force data under different environments and traffic load levels under the bridge's healthy state, remove outliers and take the statistical mean to establish a benchmark dead load cable force library for the entire bridge's suspenders.

[0009] The above method, optionally, includes the following steps in S2 for constructing the influence matrix of the suspender cable force: Based on the calibrated finite element model, S201 applies a unit cable force change to each hanger of the entire bridge in sequence, calculates the influence of the sequential application of the unit cable force change on the cable force of all other hangers, and obtains the initial influence coefficient; S202 normalizes the initial influence coefficients to form a standardized influence matrix. M , where matrix elements M ij Indicates the first i When the unit cable force changes in the first shunt, the first... j The cable force response coefficient of the shunt.

[0010] Optionally, the steps for optimizing the arrangement and determining the number of S3 sensors as described above include the following: S301 Extracts the element with the maximum absolute value in each row of the influence matrix and filters out the key suspenders that have a significant impact on the cable force of other suspenders; Based on the sparsity of the influence matrix, S302 uses a greedy algorithm combined with cable force reconstruction accuracy verification to gradually reduce the number of candidate sensors. When the cable force reconstruction error exceeds 3%, the reduction stops, and the minimum number of sensors and their placement are determined. S303 installs the cable force sensor at the selected location, completes on-site calibration, and ensures data acquisition accuracy of ±1%.

[0011] Optionally, the S4 real-time monitoring and parameter extraction steps described above may include the following: The S401 collects cable force data, environmental sensor data, and traffic load monitoring data from sensors in real time and synchronously. S402 Establish an environment-load-cable force coupled interference model, and use a combination algorithm of multiple linear regression and Kalman filtering to separate and eliminate the influence of dynamic interference factors on cable force, and extract the dead load cable force; S403 uses wavelet threshold denoising to smooth and denoise the constant load cable force data, removes abnormal data, and obtains a reliable measured constant load cable force sequence.

[0012] Optionally, the S5 dynamic mapping and early warning decision-making steps described above may include the following: S501 subtracts the reference value of the corresponding suspender in the reference constant load cable force library from the measured constant load cable force to obtain the cable force change values ​​ΔF1, ΔF2, ..., ΔF of the sensor-equipped suspender. n , n For the number of sensors; S502: Using the influence matrix as a constraint, establish the cable force change reconstruction equation, and introduce L1 regularization to obtain the cable force change value ΔF of the sensorless boom. n+1 , ..., ΔF m , m This represents the total number of hangers for the entire bridge. S503: Perform residual analysis on the reconstructed changes in the cable forces of the entire bridge. If all residuals are less than 3%, the reconstruction is valid; otherwise, return to step S4 to reprocess the data.

[0013] Optionally, the S6 environment coupling correction steps described above may include the following: S601 uses the combined amplitude and duration of cable force changes to determine the damage status, avoiding false alarms caused by a single fluctuation; S602 triggers graded early warnings based on changes in cable tension: mild warning, moderate warning, and severe warning.

[0014] The above method can be used as an optional, mild warning: cable tension change of 5% to 10% and lasting for ≥6 hours; Moderate warning: Cable tension change of 10%–20% or a single sudden change of ≥5%; Severe warning: Cable tension change value >20% or multiple adjacent booms are abnormally synchronized.

[0015] The above method, optionally, includes the following steps for S7 warning output and display: S701 collects cable force data under healthy conditions at fixed time intervals, and updates the benchmark dead load cable force library in combination with bridge maintenance records; S702 feeds back actual damage / maintenance data to the finite element model and re-corrects the influence matrix; S703 iteratively calibrates the graded early warning thresholds based on long-term early warning records and actual damage conditions.

[0016] A cable breakage early warning system for a mid-deck arch bridge, used in any of the above-described methods for early warning of cable breakage in mid-deck arch bridges, includes a pre-preparation module, an influence matrix construction module, a sensor arrangement module, a data processing module, a cable force reconstruction module, a graded early warning module, and a dynamic update module, which are connected in sequence. The pre-preparation module is used to establish and calibrate the finite element model and construct the reference dead load cable force library for the entire bridge hangers; The influence matrix construction module is used to generate a standardized influence matrix of suspender cable forces based on the calibrated finite element model. The sensor placement module is used to determine the minimum number of sensors and their placement locations through sensitivity analysis, and to complete the sensor installation and calibration. The data processing module is used to collect multi-source data in real time, remove interference factors and extract constant load cable force, and perform data smoothing and noise reduction processing. The cable force reconstruction module is used to calculate the cable force change value of the sensor-equipped boom, reconstruct the cable force change value of the sensorless boom, and verify the accuracy. The graded early warning module is used to trigger graded early warnings based on the change value and duration of cable force, and to push corresponding response measures. The dynamic update module is used to periodically update the benchmark constant load cable force library, correct the influence matrix, and calibrate the early warning threshold.

[0017] As can be seen from the above technical solution, compared with the prior art, the present invention provides a method and system for early warning of suspension rod fracture in a mid-bearing arch bridge, which has the following beneficial effects: (1) By constructing the influence matrix of the cable force of the suspenders and optimizing the sensor layout, a small number of sensors can cover the monitoring of the suspenders of the entire bridge, which greatly reduces the monitoring cost and data processing pressure; (2) A multivariate linear regression combined with Kalman filtering algorithm is used to eliminate multi-source interference, and L1 regularization is combined to achieve accurate reconstruction of cable force, thereby improving the accuracy of damage identification and early warning. (3) Setting up a graded early warning mechanism and introducing dynamic iterative optimization can adapt to the long-term service performance degradation of the structure, reduce the false early warning rate and the missed early warning rate, and realize advanced, real-time and reliable early warning of suspender fracture, providing strong protection for bridge operation safety. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0019] Figure 1 This invention discloses a flowchart of a method for early warning of hanger fracture in a through-arch bridge; Figure 2 This is a structural block diagram of a suspension rod fracture early warning system for a mid-to-lower-bearing arch bridge disclosed in this invention; Figure 3This is a schematic diagram of the sensor placement location in Embodiment 1 of the present invention; Figure 4 This is a schematic diagram of the fluctuation amplitude of cable force data before and after interference removal in Embodiment 1 of the present invention. Detailed Implementation

[0020] 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.

[0021] In this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0022] See Figure 1 As shown, this invention discloses a method for early warning of hanger fracture in a through-arch bridge, comprising the following steps: S1 Finite Element Model Calibration and Reference State Establishment Steps: Based on the design data and measured data, establish and correct the finite element model of the arch bridge to ensure that the error between the finite element model and the measured data is less than the first preset value; collect the cable force data of the hangers under multiple working conditions in a healthy state, and establish the reference dead load cable force library of the hangers for the whole bridge after preprocessing; S2 Steps for constructing the influence matrix of suspender cable force: Apply a unit cable force variation to each suspender using a calibration model, analyze its impact on the cable force of each suspender, and obtain initial influence coefficients; normalize the initial influence coefficients to construct a standardized influence matrix. M Its element M ij Quantified the first i The effect of the change in the cable force of the first shunt on the second j The response relationship of the root rod; S3 Sensor Optimization and Quantity Determination Steps: Based on the standardized influence matrix MThe system filters the element with the largest absolute value in each row, identifies the key shunts that have a significant impact on cable force, uses a greedy algorithm to reduce the number of sensors, and determines the optimal sensor arrangement scheme with the cable force reconstruction error being less than or equal to the second preset value as a constraint. The system then installs and calibrates the sensors at the selected locations. S4 Real-time monitoring and parameter extraction steps: Collect vibration and strain signals of the boom, and extract at least one state parameter of the boom based on wavelet packet decomposition; S5 Dynamic Mapping and Early Warning Decision-Making Steps: Compare the state parameters with preset thresholds, and directly query the pre-stored early warning mapping table to determine the corresponding early warning level based on the comparison results; S6 Environmental Coupling Correction Step: Combine real-time collected environmental temperature and humidity data, and dynamically correct the warning level through a temperature-humidity coupling correction factor; S7 Warning Output and Display Steps: Output warning signals according to the corrected warning level and display them in real time on the monitoring platform.

[0023] Furthermore, the S1 finite element model calibration and baseline state establishment steps specifically include the following steps: Based on bridge design drawings, material parameters and on-site measured data, S101 established a finite element model of a mid-bearing arch bridge. The model was corrected by adjusting the node constraints and material elastic modulus parameters so that the error between the model calculation results and the measured data was controlled within 5%. S102 Collect multiple sets of suspender cable force data under different environments and traffic load levels under the bridge's healthy state, remove outliers and take the statistical mean to establish a benchmark dead load cable force library for the entire bridge's suspenders.

[0024] Furthermore, the on-site measured data in S101 includes the cable tension and structural displacement data under healthy conditions; the environmental parameters in S102 include temperature and wind speed, and the traffic load level includes traffic flow and vehicle weight.

[0025] Furthermore, the steps for constructing the influence matrix of the suspender cable force in S2 specifically include the following: Based on the calibrated finite element model, S201 applies a unit cable force change to each hanger of the entire bridge in sequence, calculates the influence of the sequential application of the unit cable force change on the cable force of all other hangers, and obtains the initial influence coefficient; S202 normalizes the initial influence coefficients to form a standardized influence matrix. M , where matrix elements M ij Indicates the first i When the unit cable force changes in the first shunt, the first... j The cable force response coefficient of the shunt.

[0026] Furthermore, the steps for optimizing the layout and determining the number of S3 sensors specifically include the following: S301 Extracts the element with the maximum absolute value in each row of the influence matrix and filters out the key suspenders that have a significant impact on the cable force of other suspenders; Based on the sparsity of the influence matrix, S302 uses a greedy algorithm combined with cable force reconstruction accuracy verification to gradually reduce the number of candidate sensors. When the cable force reconstruction error exceeds 3%, the reduction stops, and the minimum number of sensors and their placement are determined. S303 installs the cable force sensor at the selected location, completes on-site calibration, and ensures data acquisition accuracy of ±1%.

[0027] Furthermore, the cable force sensor in S303 can be a vibrating wire type or a fiber optic type sensor.

[0028] Furthermore, the S4 real-time monitoring and parameter extraction steps specifically include the following: The S401 collects cable force data, environmental sensor data, and traffic load monitoring data from sensors in real time and synchronously. S402 Establish an environment-load-cable force coupled interference model, and use a combination algorithm of multiple linear regression and Kalman filtering to separate and eliminate the influence of dynamic interference factors on cable force, and extract the dead load cable force; S403 uses wavelet threshold denoising to smooth and denoise the constant load cable force data, removes abnormal data, and obtains a reliable measured constant load cable force sequence.

[0029] Furthermore, the environmental sensor data in S401 includes temperature, wind speed, and humidity data, while the traffic load monitoring data includes traffic flow and vehicle weight data.

[0030] Furthermore, the S5 dynamic mapping and early warning decision-making steps specifically include the following: S501 subtracts the reference value of the corresponding suspender in the reference constant load cable force library from the measured constant load cable force to obtain the cable force change values ​​ΔF1, ΔF2, ..., ΔF of the sensor-equipped suspender. n , n For the number of sensors; S502: Using the influence matrix as a constraint, establish the cable force change reconstruction equation, and introduce L1 regularization to obtain the cable force change value ΔF of the sensorless boom. n+1 , ..., ΔF m , m This represents the total number of hangers for the entire bridge. S503: Perform residual analysis on the reconstructed changes in the cable forces of the entire bridge. If all residuals are less than 3%, the reconstruction is valid; otherwise, return to step S4 to reprocess the data.

[0031] Furthermore, the S6 environment coupling correction steps specifically include the following: S601 uses the combined amplitude and duration of cable force changes to determine the damage status, avoiding false alarms caused by a single fluctuation; S602 triggers graded early warnings based on changes in cable tension: mild warning, moderate warning, and severe warning.

[0032] Further, a mild warning is issued: cable tension changes by 5% to 10% for ≥6 hours; Moderate warning: Cable tension change of 10%–20% or a single sudden change of ≥5%; Severe warning: Cable tension change value >20% or multiple adjacent booms are abnormally synchronized.

[0033] Furthermore, the corresponding response measures for each warning level in S602 are as follows: a mild warning triggers routine inspection reminders, a moderate warning triggers key monitoring and on-site testing, and a severe warning triggers emergency warnings and traffic restrictions / closures.

[0034] Furthermore, the S7 warning output and display steps specifically include the following: S701 collects cable force data under healthy conditions at fixed time intervals, and updates the benchmark dead load cable force library in combination with bridge maintenance records; S702 feeds back actual damage / maintenance data to the finite element model and re-corrects the influence matrix; S703 iteratively calibrates the graded early warning thresholds based on long-term early warning records and actual damage conditions.

[0035] Specifically, a fixed time interval of 6 months can be selected.

[0036] and Figure 1 Corresponding to the method shown, the present invention also discloses an early warning system for the fracture of the hanger of a mid-bearing arch bridge, used for... Figure 1 The diagram illustrates a method for early warning of hanger fracture in a through-arch bridge. (See attached structural block diagram.) Figure 2 As shown, it includes a pre-processing module for data connection, an influence matrix construction module, a sensor placement module, a data processing module, a cable force reconstruction module, a graded early warning module, and a dynamic update module. The pre-preparation module is used to establish and calibrate the finite element model and construct the reference dead load cable force library for the entire bridge hangers; The influence matrix construction module is used to generate a standardized influence matrix of suspender cable forces based on the calibrated finite element model. The sensor placement module is used to determine the minimum number of sensors and their placement locations through sensitivity analysis, and to complete the sensor installation and calibration. The data processing module is used to collect multi-source data in real time, remove interference factors and extract constant load cable force, and perform data smoothing and noise reduction processing. The cable force reconstruction module is used to calculate the cable force change value of the sensor-equipped boom, reconstruct the cable force change value of the sensorless boom, and verify the accuracy. The graded early warning module is used to trigger graded early warnings based on the change value and duration of cable force, and to push corresponding response measures. The dynamic update module is used to periodically update the benchmark constant load cable force library, correct the influence matrix, and calibrate the early warning threshold.

[0037] Example 1: The bridge in this embodiment has a calculated main arch span of 300m and a rise-to-span ratio of 1 / 5. A total of 48 hangers are installed throughout the bridge, using high-strength parallel steel wire bundles with an elastic modulus E = 2.0 × 10⁻⁶. 5 MPa, standard cross-sectional area A = 0.012 m².

[0038] The specific steps are as follows: S1: Finite element model calibration and baseline state establishment S101: Based on the bridge design drawings and material parameters, and combined with the measured data of the cable force of 48 suspenders and the mid-span displacement of the main arch under healthy conditions, a refined finite element model was established. The model was corrected by adjusting parameters such as the arch foot constraint stiffness and the elastic modulus of the suspenders. Finally, the error between the cable force and displacement data calculated by the model and the measured data was controlled within 3.2%, which met the requirements for model validity. S102: Cable force data of suspenders were continuously collected for 3 months (covering typical temperatures in spring, summer, autumn and winter, different wind speeds and peak / off-peak traffic loads), and a total of 1200 valid data sets were obtained. After removing 28 outliers using the 3σ criterion, the statistical mean was taken to establish a baseline dead load cable force library for all 48 suspenders of the bridge. The baseline cable force of suspender #1 is 1250kN, the baseline cable force of suspender #2 is 1280kN, ..., the baseline cable force of suspender #48 is 1260kN.

[0039] S2: Construct the influence matrix of the suspender cable force

[0040] S201: Based on the calibrated finite element model, unit cable force changes of +10kN and -10kN are applied sequentially to 48 booms. The influence of each boom's cable force change on the remaining 47 booms is calculated, resulting in an initial influence coefficient matrix of 48×48. For example, when boom #1 is subjected to a +10kN cable force change, the cable force of boom #2 increases by 0.8kN, boom #3 increases by 0.3kN, ..., boom #48 increases by 0.2kN. S202: Normalize the initial influence coefficients to obtain the standardized influence matrix M. 48×48 Matrix element Mij ∈[0,1], where M 12 =0.08 (meaning that when the unit cable force of the #1 suspender changes, the cable force response coefficient of the #2 suspender is 0.08).

[0041] S3: Sensor Optimization and Quantity Determination

[0042] S301: Extract the element with the largest absolute value in each row of the influence matrix and select 9 key suspenders: 1#, 6#, 12#, 18#, 24#, 30#, 36#, 42#, and 48# (the influence of the cable force changes of these suspenders on other suspenders is greater than 0.05). S302: Based on a greedy algorithm, the number of candidate sensors is gradually reduced. When the number of sensors is reduced from 9 to 6 (retaining 1#, 12#, 24#, 30#, 36#, and 48#), the cable force reconstruction error is 2.8% (≤3%). When further reduced to 5 sensors, the reconstruction error increases to 3.5% (>3%). Therefore, the minimum number of sensors is determined to be 6, and their placement is at the aforementioned 6 key support rods (e.g., ...). Figure 3 (as shown) S303: Install a vibrating wire tension sensor at the selected location, complete on-site calibration, and the sensor acquisition accuracy is ±0.8%, which meets the design requirements.

[0043] S4: Measured Data Acquisition and Interference Removal

[0044] S401: Real-time synchronous acquisition of cable force data (sampling frequency 10Hz) from 6 monitoring booms, temperature (-5℃~35℃), wind speed (0~8m / s), humidity data from environmental sensors, and traffic flow (50~300 vehicles / hour) and vehicle weight data from traffic monitoring equipment. S402: Establish a coupled interference model of "temperature-wind speed-traffic flow-cable tension", and use a combination algorithm of "multiple linear regression + Kalman filtering" to separate interference factors, extracting the dead load cable tension from the measured cable tension data; such as Figure 4 As shown, the fluctuation range of cable force data before interference removal was ±8%, and the fluctuation range of dead load cable force after interference removal was reduced to ±1.5%. S403: Wavelet threshold denoising method is used to smooth and denoise the constant load cable force data, and three spike noise data are removed to obtain a reliable measured constant load cable force sequence.

[0045] S5: Reconstruction of the variation value of the cable tension of the entire bridge suspenders

[0046] S501: Subtract the baseline value of the corresponding hanger in the baseline dead load cable force library from the measured dead load cable force to obtain the cable force change values ​​of the 6 monitored hangers: ΔF1 = +65kN (Hanger #1), ΔF 12 =+58kN (12# boom), ΔF24 =+72kN (24# boom), ΔF 30 =+45kN (30# boom), ΔF 36 =+52kN (36# boom), ΔF 48 =+60kN (48# boom); S502: Using the standardized influence matrix as a constraint, establish the cable force change reconstruction equation, introduce L1 regularization to solve, and obtain the cable force change values ​​of the remaining 42 sensorless booms, where the cable force change value of boom #7 is +48kN, boom #8 is +32kN, ..., boom #47 is +25kN; S503: Residual analysis was performed on the force variation values ​​of the 48 suspension rods obtained from the reconstruction. The residual range was 0.5% to 2.8%, all of which were less than 3%, so the reconstruction was deemed valid.

[0047] S6: Tiered Early Warning and Response

[0048] S601: Calculate the proportion of the change amplitude of each suspender cable force to the reference cable force, and judge based on the duration of the change: The change amplitude of the cable force of suspender #7 is 48 / 1270≈3.8% (<5%), no warning is given; the change amplitude of suspender #1 is 65 / 1250=5.2%, and the duration has reached 8 hours; S602: Triggering a mild warning, the system automatically pushes a "routine inspection reminder" to the bridge management platform, and arranges personnel to check the status and appearance of the No. 1 hanger sensor for damage.

[0049] S7: Model Dynamic Updates and Closed-Loop Iteration

[0050] S701: Six months later, cable force data under healthy conditions will be collected again. Combined with the daily maintenance records of the bridge during the period, the benchmark dead load cable force library will be updated, and the benchmark cable force of No. 1 hanger will be updated to 1245kN. S702: Feed back the verification results of this mild warning (sensor status is normal, boom has no obvious damage, cable force change is normal environmental adaptation) to the finite element model, and revise the influence matrix to update M12 to 0.078; S703: Based on the warning records, the duration threshold for mild warnings will be slightly adjusted from 6 hours to 8 hours to further reduce the false alarm rate.

[0051] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for early warning of hanger fracture in a through-arch bridge, characterized in that, Includes the following steps: S1 Finite Element Model Calibration and Reference State Establishment Steps: Based on the design data and measured data, establish and correct the finite element model of the arch bridge to ensure that the error between the finite element model and the measured data is less than the first preset value; Collect cable force data of suspenders under multiple working conditions in a healthy state, and establish a reference dead load cable force library for the entire bridge suspenders after preprocessing; S2 Steps for constructing the influence matrix of suspender cable force: Use the calibration model to apply a unit cable force change to each suspender, analyze its influence on the cable force of each suspender, and obtain the initial influence coefficient; The initial influence coefficients were normalized to construct a standardized influence matrix. M Its element M ij Quantified the first i The effect of the change in the cable force of the first shunt on the second j The response relationship of the root rod; S3 Sensor Optimization and Quantity Determination Steps: Based on the standardized influence matrix M The algorithm filters the element with the largest absolute value in each row, identifies the key shunts that have a significant impact on cable force, uses a greedy algorithm to reduce the number of sensors, and determines the optimal sensor arrangement scheme with the cable force reconstruction error being less than or equal to the second preset value as a constraint. The sensors are then installed and calibrated at the selected locations. S4 Real-time monitoring and parameter extraction steps: Collect vibration and strain signals of the boom, and extract at least one state parameter of the boom based on wavelet packet decomposition; S5 Dynamic Mapping and Early Warning Decision-Making Steps: Compare the state parameters with preset thresholds, and directly query the pre-stored early warning mapping table to determine the corresponding early warning level based on the comparison results; S6 Environmental Coupling Correction Step: Combine real-time collected environmental temperature and humidity data, and dynamically correct the warning level through a temperature-humidity coupling correction factor; S7 Warning Output and Display Steps: Output warning signals according to the corrected warning level and display them in real time on the monitoring platform.

2. The method for early warning of hanger fracture in a through-arch bridge according to claim 1, characterized in that, The S1 finite element model calibration and baseline state establishment steps specifically include the following steps: Based on bridge design drawings, material parameters and on-site measured data, S101 established a finite element model of a mid-bearing arch bridge. The model was corrected by adjusting the node constraints and material elastic modulus parameters so that the error between the model calculation results and the measured data was controlled within 5%. S102 Collect multiple sets of suspender cable force data under different environments and traffic load levels under the bridge's healthy state, remove outliers and take the statistical mean to establish a benchmark dead load cable force library for the entire bridge's suspenders.

3. The method for early warning of hanger fracture in a through-arch bridge according to claim 2, characterized in that, The steps for constructing the influence matrix of the suspender cable force in S2 include the following: Based on the calibrated finite element model, S201 applies a unit cable force change to each hanger of the entire bridge in sequence, calculates the influence of the sequential application of the unit cable force change on the cable force of all other hangers, and obtains the initial influence coefficient; S202 normalizes the initial influence coefficients to form a standardized influence matrix. M , where matrix elements M ij Indicates the first i When the unit cable force changes in the first shunt, the first... j The cable force response coefficient of the shunt.

4. The method for early warning of hanger fracture in a through-arch bridge according to claim 3, characterized in that, The steps for optimizing the layout and determining the number of S3 sensors specifically include the following: S301 Extracts the element with the maximum absolute value in each row of the influence matrix and filters out the key suspenders that have a significant impact on the cable force of other suspenders; Based on the sparsity of the influence matrix, S302 uses a greedy algorithm combined with cable force reconstruction accuracy verification to gradually reduce the number of candidate sensors. When the cable force reconstruction error exceeds 3%, the reduction stops, and the minimum number of sensors and their placement are determined. S303 installs the cable force sensor at the selected location, completes on-site calibration, and ensures data acquisition accuracy of ±1%.

5. The method for early warning of hanger fracture in a through-arch bridge according to claim 4, characterized in that, The S4 real-time monitoring and parameter extraction steps specifically include the following: The S401 collects cable force data, environmental sensor data, and traffic load monitoring data from sensors in real time and synchronously. S402 Establish an environment-load-cable force coupled interference model, and use a combination algorithm of multiple linear regression and Kalman filtering to separate and eliminate the influence of dynamic interference factors on cable force, and extract the dead load cable force; S403 uses wavelet threshold denoising to smooth and denoise the constant load cable force data, removes abnormal data, and obtains a reliable measured constant load cable force sequence.

6. The method for early warning of hanger fracture in a through-arch bridge according to claim 5, characterized in that, The S5 dynamic mapping and early warning decision-making steps specifically include the following: S501 subtracts the reference value of the corresponding suspender in the reference constant load cable force library from the measured constant load cable force to obtain the cable force change values ​​ΔF1, ΔF2, ..., ΔF of the sensor-equipped suspender. n , n For the number of sensors; S502: Using the influence matrix as a constraint, establish the cable force change reconstruction equation, and introduce L1 regularization to obtain the cable force change value ΔF of the sensorless boom. n+1 , ..., ΔF m , m This represents the total number of hangers for the entire bridge. S503: Perform residual analysis on the reconstructed changes in the cable forces of the entire bridge. If all residuals are less than 3%, the reconstruction is valid; otherwise, return to step S4 to reprocess the data.

7. A method for early warning of hanger fracture in a through-arch bridge according to claim 6, characterized in that, The S6 environment coupling correction steps specifically include the following: S601 uses the combined amplitude and duration of cable force changes to determine the damage status, avoiding false alarms caused by a single fluctuation; S602 triggers graded early warnings based on changes in cable tension: mild warning, moderate warning, and severe warning.

8. A method for early warning of hanger fracture in a through-arch bridge according to claim 7, characterized in that, Mild warning: Cable tension changes by 5% to 10% and last for ≥6 hours; Moderate warning: Cable tension change of 10%–20% or a single sudden change of ≥5%; Severe warning: Cable tension change value >20% or multiple adjacent booms are abnormally synchronized.

9. A method for early warning of hanger fracture in a through-arch bridge according to claim 8, characterized in that, The S7 warning output and display steps specifically include the following: S701 collects cable force data under healthy conditions at fixed time intervals, and updates the benchmark dead load cable force library in combination with bridge maintenance records; S702 feeds back actual damage / maintenance data to the finite element model and re-corrects the influence matrix; S703 iteratively calibrates the graded early warning thresholds based on long-term early warning records and actual damage conditions.

10. A pre-warning system for the fracture of a suspension rod in a mid-bearing arch bridge, characterized in that, The method for early warning of suspender fracture of a mid-bearing arch bridge according to any one of claims 1-9 includes a pre-preparation module, an influence matrix construction module, a sensor arrangement module, a data processing module, a cable force reconstruction module, a graded early warning module, and a dynamic update module that are connected in sequence. The pre-preparation module is used to establish and calibrate the finite element model and construct the reference dead load cable force library for the entire bridge hangers; The influence matrix construction module is used to generate a standardized influence matrix of suspender cable forces based on the calibrated finite element model. The sensor placement module is used to determine the minimum number of sensors and their placement locations through sensitivity analysis, and to complete the sensor installation and calibration. The data processing module is used to collect multi-source data in real time, remove interference factors and extract constant load cable force, and perform data smoothing and noise reduction processing. The cable force reconstruction module is used to calculate the cable force change value of the sensor-equipped boom, reconstruct the cable force change value of the sensorless boom, and verify the accuracy. The graded early warning module is used to trigger graded early warnings based on the change value and duration of cable force, and to push corresponding response measures. The dynamic update module is used to periodically update the benchmark constant load cable force library, correct the influence matrix, and calibrate the early warning threshold.

Images