Bridge expansion joint intelligent monitoring and early warning method and system

Abstract

This invention discloses an intelligent monitoring and early warning method and system for bridge expansion joints, comprising: collecting multi-dimensional data such as stress, strain, displacement, temperature, humidity, and load frequency through a distributed sensor array, transmitting the data to a big data analysis platform for classification, storage, and correlation mapping; calculating damage variables by calling a fatigue damage evolution model; extrapolating load time-series response using a dynamic load response prediction model; and performing feature matching and anomaly determination using an intelligent abnormal deformation diagnosis model. This invention generates early warning information including damage level, deformation trend, and risk range by integrating model output results, achieving multi-dimensional data linkage analysis and early warning signal output, providing comprehensive and accurate technical support for the health monitoring of bridge expansion joints, and improving the scientific nature and timeliness of operation and maintenance decisions.

Description

Technical Field

[0001] This invention relates to the field of bridge expansion joint monitoring technology, and in particular to an intelligent monitoring and early warning method and system for bridge expansion joints. Background Technology

[0002] With the continuous growth of traffic volume and the gradual extension of bridge service life, expansion joints, as key functional components in bridge structures that adapt to temperature changes, load effects, and foundation settlement, are subjected to repeated stress under complex working conditions for extended periods. Their health directly affects the overall structural safety and traffic stability of the bridge. Traditional monitoring methods rely heavily on manual inspections and single-point data collection, making it difficult to comprehensively capture the stress state, deformation trends, and damage evolution of expansion joints. Meanwhile, the demand for real-time and precise monitoring and early warning in bridge operation and maintenance is increasingly urgent. Against this backdrop, there is a pressing need to construct a monitoring and early warning system that can integrate multi-source monitoring information and achieve intelligent analysis throughout the entire process. Through systematic data collection, analysis, and judgment, potential risks to expansion joints can be identified in a timely manner, providing scientific support for bridge maintenance decisions.

[0003] Existing technologies suffer from two significant drawbacks: First, data acquisition and processing lack systematic coordination. Most solutions only monitor a single type of parameter, failing to achieve comprehensive integration of stress-related data, environmental impact data, and structural deformation data. Furthermore, the lack of an efficient correlation mechanism during data transmission and storage makes it difficult to characterize the actual working state of expansion joints from multiple dimensions. Second, the accuracy of risk identification and early warning is insufficient. Existing methods fail to fully consider the structural characteristics of expansion joints and the differences in service environments. They lack dynamic adaptability in judging damage development trends, and the early warning thresholds are relatively fixed, unable to be adjusted in real time according to changes in actual monitoring data. This results in delayed early risk identification, making it difficult to meet the refined needs of bridge operation and maintenance. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of the existing technology, the present invention provides a method and system for intelligent monitoring and early warning of bridge expansion joints.

[0005] The technical solution adopted in this invention is an intelligent monitoring and early warning method for bridge expansion joints, comprising the following steps: S1, collecting multi-dimensional monitoring data of bridge expansion joints, including stress amplitude, strain accumulation, geometric displacement, ambient temperature and humidity, and load frequency, through a distributed sensor array; S2, transmitting the collected multi-dimensional monitoring data to a bridge expansion joint big data analysis platform for data classification, storage, and correlation mapping; S3, calling the expansion joint fatigue damage evolution model to calculate damage variables based on stress amplitude, strain accumulation, and load frequency; S4, using a dynamic load response prediction model to integrate geometric displacement and ambient temperature and humidity data to perform load time-series response deduction; S5, using an abnormal deformation intelligent diagnosis model to perform feature matching and anomaly judgment on the fatigue damage calculation results and load response deduction data; S6, the bridge expansion joint big data analysis platform integrates the model output results to generate monitoring and early warning information including damage level, deformation trend, and risk range, completing multi-dimensional data linkage analysis and early warning signal output.

[0006] Furthermore, the fatigue damage evolution model expression for the expansion joint is as follows: ,in, The fatigue damage degree of the expansion joint at time t; This refers to the stress amplitude. The cumulative load frequency at time t; This is the load influence factor; The fatigue accumulation index; The fatigue coefficient of the material; The initial elastic modulus of the expansion joint substrate; This is the temperature influence coefficient; Let t be the ambient temperature. For strain damage weight; for Constantly accumulating responses; This is the damage attenuation coefficient.

[0007] Furthermore, the expression for the dynamic load response prediction model is: ,in, The value of the dynamic load response at time t; For response amplification factor; Let t be the geometric displacement at time t; The load vibration coefficient; The frequency of the load application; The cumulative load frequency at time t; This is the secondary influence coefficient of temperature; Let t be the ambient temperature. for Geometric displacement at any given moment; This refers to the multi-point fusion coefficient. Let be the geometric displacement of the i-th measuring point at time t; The weight coefficient for the i-th measurement point; This represents the total number of measurement points.

[0008] Furthermore, the expression for the intelligent diagnostic model for abnormal deformation is: ,in, This is an abnormal diagnostic value; The degree of fatigue damage at time t; This is the initial damage threshold; The standard deviation of the damage; For damage anomaly weights; The load response value at time t; For reference response value; The standard deviation of the response value; In response to abnormal weights; Weighted by deformation rate; For time t and Displacement difference at time; For time intervals; This represents the damage acceleration factor.

[0009] Furthermore, the data integration model expression of the bridge expansion joint big data analysis platform is as follows: ,in, This is a comprehensive analysis index for the platform; The weights for the m-th type of damage data; The data represents the m-th type of damage at time t; Number of damaged data types; The weights for the nth type of response data; The data represents the nth type of response at time t; Number of response data types; Weights for abnormal diagnosis; This is the diagnostic value for the anomaly at the k-th measuring point; This represents the total number of measurement points. This is the data covariance correction factor; Let be the covariance between the damage level and the response value.

[0010] Furthermore, the threshold calculation expression for intelligent monitoring and early warning of bridge expansion joints is as follows: ,in, The threshold is a dynamic early warning threshold; Basic warning threshold; This is the damage impact coefficient; The degree of fatigue damage at time t; The response impact coefficient; The load response value at time t; This is the maximum allowed response value; The coefficient representing the influence of temperature change; For time t and Temperature difference over time; For time intervals.

[0011] Further, step S3 includes the following sub-steps: S31, selecting stress amplitude, cumulative strain, load frequency, and material elastic modulus parameters to construct a basic dataset of fatigue damage for expansion joints, sorting it by time series and associating it with corresponding ambient temperature data; S32, setting initial conditions for damage evolution, inputting the basic dataset into the fatigue damage evolution model of expansion joints, and obtaining the damage increment at each time node through iterative calculation; S33, combining the structural characteristics of expansion joints, spatially mapping the damage increment to clarify the damage differences in different parts; S34, integrating the damage data and spatial distribution information at each time node to form a global fatigue damage evolution sequence, providing basic data support for anomaly diagnosis.

[0012] Further, S4 includes the following sub-steps: S41, extracting geometric displacement, ambient temperature and humidity, load frequency, and structural natural frequency parameters to construct a dynamic load response input dataset and establish the correlation mapping relationship between parameters; S42, importing the input dataset into the dynamic load response prediction model, setting the time step and simulation period, and performing load response time series calculations under multiple scenarios; S43, performing spatiotemporal calibration on the simulation results, adjusting the model output deviation in conjunction with historical monitoring data to ensure that the response value matches the actual working conditions; S44, generating a dataset including the load response trend for future periods, clarifying the time node and numerical range of the response peak, and providing a reference basis for anomaly judgment.

[0013] Further, S5 includes the following sub-steps: S51, collecting fatigue damage evolution sequences and load response trend datasets, extracting damage mutation features, response peak features, and deformation rate features, and constructing an anomaly diagnosis feature library; S52, setting feature matching thresholds and anomaly judgment rules, inputting the feature library data into the intelligent diagnostic model for abnormal deformation, and performing feature similarity calculation and anomaly probability assessment; S53, classifying the anomaly probability values ​​output by the model, and classifying the anomaly level in conjunction with the structural safety level of the expansion joint; S54, outputting diagnostic results including anomaly features, anomaly level, and corresponding location information, and synchronously transmitting them to the bridge expansion joint big data analysis platform for integration.

[0014] A smart monitoring and early warning system for bridge expansion joints is disclosed. This system, applied to a smart monitoring and early warning method for bridge expansion joints, includes: a multi-dimensional sensor data acquisition unit, used to deploy a distributed sensor array to collect and transmit parameters such as expansion joint stress amplitude, cumulative strain, geometric displacement, ambient temperature and humidity, and load frequency; a data classification, storage, and transmission unit, which receives data output from the multi-dimensional sensor data acquisition unit, stores it according to parameter type, and sends it to subsequent units via an encrypted transmission protocol; a model calculation and processing unit, which incorporates a fatigue damage evolution model, a dynamic load response prediction model, and an intelligent abnormal deformation diagnosis model for expansion joints, receives the classified and stored data, and performs calculations; a big data comprehensive analysis unit, connected to the model calculation and processing unit, integrates the calculation results for correlation analysis and threshold comparison, and generates a comprehensive analysis report; a smart early warning signal generation unit, which receives the report output by the big data comprehensive analysis unit and generates corresponding early warning signals based on the anomaly level; and an early warning signal output and linkage unit, which receives the signal from the smart early warning signal generation unit, outputs it through multiple channels, and establishes linkage with the bridge operation and maintenance management system for precise push and response activation of early warning information.

[0015] Beneficial Effects: This invention proposes an intelligent monitoring and early warning method and system for bridge expansion joints. It comprehensively collects multi-dimensional data such as stress, strain, displacement, ambient temperature and humidity, and load frequency through a distributed sensor array. A big data analysis platform is used to classify, store, and map the data. A dedicated model is then used for damage calculation, load response simulation, and anomaly detection. Finally, early warning information including damage level, deformation trend, and risk range is generated and output in a coordinated manner, forming a full-process intelligent monitoring system. This invention overcomes the shortcomings of existing technologies, such as single data collection and lack of system coordination. Through deep integration and correlation analysis of multi-source data, it comprehensively depicts the working state and damage evolution of expansion joints. It also addresses the problems of lagging risk identification and insufficient accuracy in early warning in traditional methods. Through dynamically adaptable analysis logic and a hierarchical judgment mechanism, it achieves accurate capture of early minor damage and potential anomalies. Simultaneously, the dynamically adjusted early warning logic adapts to different service conditions, forming a full-link collaboration from data collection and analysis to early warning output. This provides scientific and efficient technical support for bridge operation and maintenance, significantly improving the refinement and intelligence of expansion joint health management. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the overall process of the method of the present invention.

[0017] Figure 2 This is a flowchart of method step S3 of the present invention; Figure 3 This is a flowchart of method step S4 of the present invention; Figure 4 This is a flowchart of step S5 of the method of the present invention; Figure 5 This is a diagram showing the system unit composition of the present invention. Detailed Implementation

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] like Figure 1 As shown, an intelligent monitoring and early warning method for bridge expansion joints includes the following steps: S1, collecting multi-dimensional monitoring data of bridge expansion joints, including stress amplitude, strain accumulation, geometric displacement, ambient temperature and humidity, and load frequency, through a distributed sensor array; S2, transmitting the collected multi-dimensional monitoring data to a bridge expansion joint big data analysis platform for data classification, storage, and correlation mapping; S3, calling the expansion joint fatigue damage evolution model to calculate damage variables based on stress amplitude, strain accumulation, and load frequency; S4, using a dynamic load response prediction model to integrate geometric displacement and ambient temperature and humidity data to perform load time-series response deduction; S5, using an intelligent abnormal deformation diagnosis model to perform feature matching and anomaly judgment on the fatigue damage calculation results and load response deduction data; S6, the bridge expansion joint big data analysis platform integrates the model output results to generate monitoring and early warning information including damage level, deformation trend, and risk range, completing multi-dimensional data linkage analysis and early warning signal output.

[0020] Step S1 involves deploying a sensor array consisting of 50-80 distributed sensors at key stress-bearing locations, deformation-sensitive areas, and environmental monitoring points along the bridge expansion joint to achieve comprehensive acquisition of multi-dimensional monitoring data. The sensor array is evenly distributed at 20-30 cm intervals along the expansion joint body, anchorage structure, and connections to adjacent beams. The stress sensors are piezoelectric sensors with a range of 0-250 MPa; the strain sensors are fiber optic strain gauges with an accuracy of ±1 με; the displacement sensors are laser displacement sensors with a resolution of 0.01 mm; the temperature and humidity sensors are digital sensors with a measurement range of -40℃ to 85℃ and humidity of 0-100% RH; and the load counter is integrated below the stress-bearing surface of the expansion joint. During the data acquisition process, stress amplitude was collected in real time at a frequency of 10Hz, recording the instantaneous stress changes of the expansion joint under vehicle loads; strain accumulation was continuously collected at a frequency of 5Hz, and the strain superposition value per unit time was obtained through cumulative calculation; geometric displacement was collected in the longitudinal, transverse, and vertical directions of the expansion joint, and displacement data was recorded every 500 milliseconds; ambient temperature and humidity were collected at a frequency of 1Hz, and the dynamic changes of temperature and humidity were recorded synchronously; the load application frequency was counted in real time by a load counter, and a count was completed for each effective load application detected. All data collected by the sensors were converted into standard digital signals by a signal conditioning module to ensure the accuracy and consistency of data acquisition, providing comprehensive and reliable raw data support for subsequent analysis. This step, through the collaborative work of multiple types of sensors and high-frequency data acquisition, achieves comprehensive and seamless monitoring of the working status of the expansion joint, ensuring the integrity and timeliness of the monitoring data.

[0021] Step S2 involves transmitting multi-dimensional monitoring data, including stress amplitude, cumulative strain, geometric displacement, ambient temperature and humidity, and load frequency, collected by the distributed sensor array, to the bridge expansion joint big data analysis platform via industrial Ethernet, 4G / 5G wireless communication modules, and fiber optic communication. During data transmission, TCP / IP protocol is used for data encapsulation, with the transmission rate controlled between 10-100 Mbps to ensure real-time data transmission without loss. The bridge expansion joint big data analysis platform has a built-in distributed database with a capacity of 1000-2000 GB. After receiving the data, it first categorizes it by parameter type: stress amplitude data is stored in the mechanical parameter database, cumulative strain data in the deformation parameter database, geometric displacement data in the displacement parameter database, ambient temperature and humidity data in the environmental parameter database, and load frequency data in the load parameter database. After classification and storage, the platform performs a correlation mapping based on timestamps, establishing a correspondence between stress amplitude, cumulative strain, geometric displacement, ambient temperature and humidity, and load frequency at the same monitoring time, forming a multi-dimensional data correlation table indexed by time. Simultaneously, the platform performs integrity checks on the stored data, eliminating invalid and abnormal data, and supplementing missing data using interpolation methods to ensure data continuity and availability. This step, through efficient data transmission methods and scientific storage and correlation mechanisms, achieves orderly management of multi-source monitoring data, providing structured and standardized data support for subsequent model calculations and ensuring the smooth progress of the entire monitoring and early warning process.

[0022] Step S3 invokes the pre-built expansion joint fatigue damage evolution model in the system, using the stress amplitude, strain accumulation, and load frequency data (classified, stored, and mapped in Step S2) as core input parameters to calculate damage variables. During the calculation, effective peak values ​​are first extracted from the stress amplitude data, and stress values ​​greater than 50 MPa and lasting longer than 10 milliseconds are selected as effective stress amplitudes for calculation. For strain accumulation data, the strain superposition value per unit time is summarized to obtain the total strain accumulation within the monitoring period. The load frequency is calculated using a cumulative counting method, counting the total number of load applications during the monitoring period. In model calculation, the initial damage contribution value caused by the load is first calculated based on the product relationship between stress amplitude and load frequency. Then, the initial damage contribution value is corrected by considering the continuous influence of strain accumulation, comprehensively taking into account the synergistic effect of stress concentration and strain accumulation. The calculation process sets 100-200 calculation steps, each corresponding to 5-10 minutes of monitoring data. Damage variable values ​​at each time point are obtained through iterative calculation at each step size. Simultaneously, by combining the physical properties and structural parameters of the expansion joint material, the intermediate results in the calculation process are verified to ensure that the damage variable calculation results conform to actual engineering laws. This step, through precise calculation by a dedicated model, achieves a quantitative assessment of the fatigue damage state of the expansion joint, clarifies the degree and trend of damage development, provides core data support for subsequent load response extrapolation and anomaly determination, and lays the foundation for the health status assessment of bridge expansion joints.

[0023] Step S4 utilizes a dynamic load response prediction model to deeply integrate the geometric displacement data obtained in Step S2 with environmental temperature and humidity data to achieve load time-series response simulation. During the simulation, longitudinal, lateral, and vertical displacement data are first extracted from the geometric displacement data. The maximum, minimum, and variation amplitude of displacement in each direction are statistically analyzed to establish a displacement change feature library. Environmental temperature and humidity data are summarized hourly, calculating the average temperature, average humidity, temperature change rate, and humidity change rate within each hour. The model uses the longitudinal displacement as the core input parameter, and the lateral and vertical displacements as auxiliary correction parameters. Environmental temperature indirectly affects the load response by influencing the elastic modulus of the expansion joint material, while environmental humidity adjusts the response results by affecting the friction coefficient of the material. The simulation period is set to 24-72 hours, with simulation nodes divided into 1-hour intervals. Based on the correlation between load and displacement, temperature, and humidity in historical monitoring data, and combined with the model's built-in response prediction algorithm, the load response value for each future moment is calculated node by node. Simultaneously, historical load response data is introduced to calibrate the simulation results. By comparing the deviations between historical simulation values ​​and actual monitoring values, the correction coefficients in the model are adjusted to ensure the accuracy of the simulation results. This step, through multi-parameter fusion and time-series simulation, predicts the load response state of the expansion joint in advance over a period of time, clarifies the deformation trend under load, provides a predictive basis for the identification of abnormal deformation, and improves the foresight of monitoring and early warning.

[0024] Step S5 uses an intelligent diagnostic model for abnormal deformation to perform comprehensive feature matching and anomaly determination on the fatigue damage calculation results obtained in Step S3 and the load response simulation data from Step S4. First, features are extracted from the fatigue damage calculation results to identify abrupt changes, growth rates, and cumulative damage values ​​of the damage variables, establishing a damage feature vector. For the load response simulation data, features such as response peak value, response duration, and response change trend are extracted to construct a response feature vector. The model inputs the damage feature vector and the response feature vector into the feature matching module, using a similarity calculation algorithm to compare and analyze the degree of fit between the two sets of vectors. A similarity threshold of 0.7-0.9 is set; when the similarity is below the threshold, an anomaly risk is initially determined. Subsequently, based on the structural safety standards for expansion joints, different levels of damage and response thresholds are set, and the damage variable values ​​and load response values ​​are compared with the corresponding thresholds to clarify the severity of the anomaly. During the anomaly determination process, the correlation between damage development trends and load response is comprehensively considered. When the damage variable continues to increase and the load response value exceeds a threshold, it is determined to be a deterministic anomaly. When the damage variable changes abruptly but the load response is normal, it is determined to be a suspected anomaly, requiring further monitoring and verification. When the load response is abnormal but the damage variable shows no significant change, environmental factors or measurement errors are investigated. Finally, a determination result including anomaly type, anomaly location, anomaly degree, and anomaly occurrence probability is generated, providing a direct basis for subsequent early warning information generation. This step, through precise feature matching and scientific determination rules, achieves accurate identification of abnormal deformation, ensuring the reliability of monitoring and early warning.

[0025] Step S6 integrates the anomaly judgment results from Step S5, the fatigue damage calculation results from Step S3, and the load response simulation data from Step S4 through the bridge expansion joint big data analysis platform. This generates monitoring and early warning information including damage level, deformation trend, and risk range, completing multi-dimensional data linkage analysis and early warning signal output. The platform first standardizes the output results of each model, classifying damage variable values ​​into three levels: mild damage, moderate damage, and severe damage, based on the ranges of 0-0.3, 0.3-0.7, and 0.7-1.0. The deformation trend, based on the load response simulation data, predicts the direction and magnitude of deformation over the next 24-72 hours, clarifying the development trend of deformation. The risk range, combining damage level and deformation trend, is divided into low-risk, medium-risk, and high-risk ranges. The low-risk range corresponds to mild damage with a stable deformation trend, the medium-risk range corresponds to moderate damage or an intensifying deformation trend, and the high-risk range corresponds to severe damage or deformation about to exceed the safe range. During data linkage analysis, the platform comprehensively considers the synergistic effects of multiple parameters, such as stress amplitude, cumulative strain, geometric displacement, environmental temperature and humidity, and load frequency, analyzing the correlation between each parameter and damage level, deformation trend, and risk range to ensure the comprehensiveness and accuracy of early warning information. When outputting early warning signals, the platform's built-in signal generation module converts the monitoring and early warning information into standardized early warning signals, which are then output through various methods, including audible and visual alarm devices, SMS notifications, platform push notifications, and linkage with the operation and maintenance management system. High-risk early warning signals trigger continuous alarms from audible and visual alarm devices and send SMS notifications to operation and maintenance personnel immediately. Medium-risk early warning signals are output simultaneously through platform push notifications and SMS notifications, while low-risk early warning signals are only recorded and displayed on the platform, ensuring that operation and maintenance personnel receive early warning information in a timely manner and take appropriate measures.

[0026] Preferably, the fatigue damage evolution model expression for the expansion joint is: ,in, The fatigue damage degree of the expansion joint at time t; This refers to the stress amplitude. The cumulative load frequency at time t; This is the load influence factor; The fatigue accumulation index; The fatigue coefficient of the material; The initial elastic modulus of the expansion joint substrate; This is the temperature influence coefficient; Let t be the ambient temperature. For strain damage weight; for Constantly accumulating responses; This is the damage attenuation coefficient.

[0027] Specifically, the fatigue damage evolution model for expansion joints is based on fatigue cumulative damage theory and material mechanical properties. It incorporates the multi-factor coupling influence law during the service life of expansion joints. First, it establishes the basic correlation between load and damage using Miner's linear cumulative theory. Then, it introduces the influence function of temperature on material properties to correct the elastic modulus parameter. Simultaneously, it considers the time decay effect of strain accumulation and uses an integral form to characterize the continuous contribution of strain to damage, ultimately forming a multi-parameter coupled damage evolution expression. The model is established based on the fact that expansion joint damage is caused by the combined effects of load impact, strain accumulation, and temperature influence. It requires a segmented modeling and then fusion approach to comprehensively cover the damage inducing factors. The load influence coefficient ranges from 1.2 to 2.5, the fatigue accumulation index is set to 0.6 to 1.0 based on the expansion joint material type, the material fatigue resistance coefficient is determined through material testing to be 3.5 to 6.8, the initial elastic modulus is taken as 200-220 according to the base material standard value, the temperature influence coefficient is 0.001-0.003, the strain damage weight is set to 0.3-0.7, and the damage decay coefficient is 0.01-0.05. During implementation, the values ​​of each coefficient are first calibrated through experiments. Then, stress amplitude, load frequency, ambient temperature and cumulative strain data are collected in real time and substituted into the model for iterative calculation at each time node to obtain the fatigue damage degree at different times. This model can quantitatively reflect the synergistic effect of load accumulation, strain superposition and temperature change on expansion joint damage, providing a scientific basis for accurately assessing the fatigue state of expansion joints and solving the problem that single-factor modeling cannot fully depict the damage evolution.

[0028] Preferably, the expression for the dynamic load response prediction model is: ,in, The value of the dynamic load response at time t; For response amplification factor; Let t be the geometric displacement at time t; The load vibration coefficient; The frequency of the load application; The cumulative load frequency at time t; This is the secondary influence coefficient of temperature; Let t be the ambient temperature. for Geometric displacement at any given moment; This refers to the multi-point fusion coefficient. Let be the geometric displacement of the i-th measuring point at time t; The weight coefficient for the i-th measurement point; This represents the total number of measurement points.

[0029] Specifically, the dynamic load response prediction model is based on structural dynamics theory and combines the coupled response law of expansion joint geometric displacement and environmental factors. First, a linear relationship between geometric displacement and load response is established. Then, a load vibration correction term is introduced to reflect the influence of the periodic action of the load. A temperature quadratic term is added to correct the nonlinear influence of temperature on material stiffness. Finally, the spatial adaptability of the model is improved by a multi-point data fusion term, forming a multi-dimensional parameter coupled prediction expression. The core logic of the model is that the load response of expansion joints is affected by geometric deformation, load characteristics, ambient temperature, and differences in measuring points. Accurate prediction is achieved through component modeling and weighted fusion. The response amplification factor is set to 1.1-1.8 based on the structural stiffness of the expansion joint, the load vibration factor is set to 0.2-0.5, the load frequency is set to 5-15 based on the actual traffic load characteristics, the temperature secondary influence coefficient is set to 0.0008-0.002, the multi-measuring point fusion coefficient is 0.8-0.95, the measuring point weight coefficient is determined by the analytic hierarchy process, the sum of the weights of each measuring point is 1, and the total number of measuring points is set to 5-12 based on the length of the expansion joint. During implementation, the coefficients are first calibrated through field tests, and then geometric displacement, ambient temperature, load frequency and displacement data of each measuring point are collected in real time. The data are then input into the model for time-series simulation according to the set time step. The prediction results are updated every 10-30 minutes. This model can comprehensively consider the influence of multiple spatiotemporal factors on the load response, improve the accuracy and foresight of the response prediction, and provide reliable load response benchmark data for subsequent anomaly judgment.

[0030] Preferably, the expression for the intelligent diagnostic model for abnormal deformation is: ,in, This is an abnormal diagnostic value; The degree of fatigue damage at time t; This is the initial damage threshold; The standard deviation of the damage; For damage anomaly weights; The load response value at time t; For reference response value; The standard deviation of the response value; In response to abnormal weights; Weighted by deformation rate; For time t and Displacement difference at time; For time intervals; This represents the damage acceleration factor.

[0031] Specifically, the intelligent diagnostic model for abnormal deformation is based on feature matching theory and statistical analysis methods. Combining the correlation between damage state and load response, it first constructs damage anomaly and response anomaly terms through deviation standardization. Then, it introduces a deformation rate term to reflect the dynamic trend of deformation, while considering the accelerating effect of damage on the deformation rate. An exponential function is used to correct the deformation rate weight, ultimately forming a multi-dimensional feature fusion-based anomaly diagnostic expression. The model is established based on the principle that abnormal deformation requires the coordinated judgment of damage state, load response, and deformation rate. A single indicator is insufficient to avoid misjudgment; therefore, weighted fusion is necessary for comprehensive identification. The damage anomaly weight is set at 0.4-0.6, the response anomaly weight at 0.3-0.5, the deformation rate weight at 0.2-0.4, the damage acceleration coefficient at 0.5-1.2, and the initial damage threshold is set at 0.2-0.3 according to the expansion joint safety level. The standard deviation of damage degree and response value are determined through historical data statistical analysis, and the time interval is set at 5-10 minutes according to the monitoring frequency. During implementation, thresholds and standard deviations are first determined based on historical normal working condition data. Then, fatigue damage degree, load response value and displacement change data are collected in real time and substituted into the model to calculate abnormal diagnostic values. Abnormality is determined by comparing with the set diagnostic thresholds. This model can comprehensively reflect the abnormal characteristics of damage, response and deformation, improve the ability to identify early subtle abnormalities, and provide core judgment basis for accurate early warning.

[0032] Preferably, the data integration model expression of the bridge expansion joint big data analysis platform is as follows: ,in, This is a comprehensive analysis index for the platform; The weights for the m-th type of damage data; The data represents the m-th type of damage at time t; Number of damaged data types; The weights for the nth type of response data; The data represents the nth type of response at time t; Number of response data types; Weights for abnormal diagnosis; This is the diagnostic value for the anomaly at the k-th measuring point; This represents the total number of measurement points. This is the data covariance correction factor; Let be the covariance between the damage level and the response value.

[0033] Specifically, the data integration model of the bridge expansion joint big data analysis platform is based on multi-source data fusion theory and statistical analysis methods. First, it integrates different types of damage and response data through weighted summation. Then, it introduces a maximum value term for anomaly diagnosis to highlight key anomaly information. Finally, it supplements a covariance correction term for damage and response to eliminate analytical bias caused by data correlation, forming a comprehensive analytical expression. The core logic of the model is that the big data platform needs to integrate the output results of multiple models and achieve comprehensive evaluation through multi-dimensional data correlation analysis. It needs to consider both data weight allocation and correlation correction. The weights of damage and response data are determined by expert scoring, with the sum of all weights being 1. The number of damage data types is set to 3-5 categories according to monitoring dimensions, and the number of response data types is set to 4-6 categories. The anomaly diagnosis weight is set to 0.3-0.5, and the data covariance correction coefficient is set to 0.1-0.3. The total number of measuring points is consistent with the number of measuring points used in the model calculation. During implementation, the weight coefficients are first determined through expert review, and then various types of damage data, response data, and abnormal diagnostic values ​​are collected. The data are then comprehensively calculated according to the model formula to generate the platform's comprehensive analysis index. This model can systematically integrate effective information from multiple sources of data, eliminate the limitations of single data evaluation, provide a comprehensive analysis basis for subsequent early warning threshold comparison, and enhance the data analysis and decision support capabilities of the big data platform.

[0034] Preferably, the threshold calculation expression for the intelligent monitoring and early warning of bridge expansion joints is as follows: ,in, The threshold is a dynamic early warning threshold; Basic warning threshold; This is the damage impact coefficient; The degree of fatigue damage at time t; The response impact coefficient; The load response value at time t; This is the maximum allowed response value; The coefficient representing the influence of temperature change; For time t and Temperature difference over time; For time intervals.

[0035] Specifically, the intelligent monitoring and early warning threshold model for bridge expansion joints is based on dynamic threshold theory and risk classification principles. Using a basic early warning threshold as a benchmark, it introduces damage impact correction terms, response impact correction terms, and temperature change impact correction terms. Through linear superposition, the threshold is dynamically adjusted to form an early warning threshold expression adapted to different working conditions. The model is established because fixed early warning thresholds are insufficient to adapt to changes in working conditions during the service life of expansion joints. Dynamic adjustments are needed based on damage status, load response, and environmental changes to ensure the accuracy and relevance of the early warning. The basic early warning threshold is set at 0.6-0.8 according to the structural safety standards for expansion joints; the damage impact coefficient is set at 0.3-0.5; the response impact coefficient is set at 0.2-0.4; the temperature change impact coefficient is set at 0.001-0.002; and the maximum allowable response value is 1.5-2.0 times the rated response value based on the design limit of the expansion joint. The time interval is consistent with the monitoring data acquisition cycle, set at 5-10 minutes. During implementation, the basic early warning threshold and coefficients are first determined based on the expansion joint design parameters and safety standards. Then, fatigue damage, load response, and temperature change data are collected in real time and substituted into the model to dynamically calculate the early warning threshold. The threshold is then compared with the platform's comprehensive analysis index. This model can dynamically adjust the early warning standard according to the actual working conditions, avoiding misjudgment or missed judgment caused by fixed thresholds, improving the accuracy and adaptability of the early warning, and providing a scientific threshold basis for the generation of intelligent early warning signals.

[0036] Preferred, such as Figure 2 As shown, step S3 includes the following sub-steps: S31, selecting stress amplitude, cumulative strain, load frequency, and material elastic modulus parameters to construct a basic dataset of fatigue damage for expansion joints, sorting it by time series and associating it with corresponding ambient temperature data; S32, setting initial conditions for damage evolution, inputting the basic dataset into the fatigue damage evolution model of expansion joints, and obtaining the damage increment at each time node through iterative calculation; S33, combining the structural characteristics of expansion joints, spatially mapping the damage increment to clarify the damage differences in different parts; S34, integrating the damage data and spatial distribution information at each time node to form a global fatigue damage evolution sequence, providing basic data support for anomaly diagnosis.

[0037] Specifically, step S3 achieves precise quantitative calculation of fatigue damage in expansion joints through four sub-steps. S31: Core parameters such as stress amplitude, cumulative strain, load frequency, and material elastic modulus are selected from multi-dimensional monitoring data. A basic dataset is constructed with data nodes every 5 minutes. Simultaneously, each data node is linked to the corresponding ambient temperature data to ensure temporal consistency and integrity, providing structured and highly correlated input data for subsequent model calculations. S32: Initial conditions are set, such as an initial damage value of 0 for the expansion joint and a material elastic modulus threshold of 200 to 220. The constructed basic dataset is input into the expansion joint fatigue damage evolution model node by node in chronological order. An iterative calculation method with a step size of 2 minutes is used to solve for the damage increment at each time node, dynamically tracking the development process of damage from initiation to accumulation. S33: Combining the expansion joint structural design drawings, the spatial distribution characteristics of key components such as the main body, anchorage zone, and connection joints are clarified. The damage increment at each time node is mapped and allocated according to the corresponding spatial location, quantifying the differences in damage degree in different parts and accurately locating the concentrated damage area. S34 integrates damage data from all nodes according to time series and combines it with spatial distribution information to form a fatigue damage evolution sequence covering the entire domain and all time periods. This sequence fully presents the temporal pattern and spatial distribution characteristics of damage development, providing comprehensive and continuous damage state data support for subsequent intelligent diagnosis of abnormal deformation. It ensures that abnormal judgment is based on a complete damage evolution background, greatly improving the reliability and accuracy of diagnostic results.

[0038] Preferred, such as Figure 3 As shown, step S4 includes the following sub-steps: S41, extracting geometric displacement, ambient temperature and humidity, load frequency, and structural natural frequency parameters to construct a dynamic load response input dataset and establish the correlation mapping relationship between parameters; S42, importing the input dataset into the dynamic load response prediction model, setting the time step and simulation period, and performing load response time series calculations under multiple scenarios; S43, performing spatiotemporal calibration on the simulation results, adjusting the model output deviation in conjunction with historical monitoring data to ensure that the response value matches the actual working conditions; S44, generating a dataset including the load response trend for future periods, clarifying the time node and numerical range of the response peak, and providing a reference for anomaly judgment.

[0039] Specifically, step S4 achieves accurate time-series extrapolation of dynamic load response through four sub-steps. S31 extracts key parameters such as geometric displacement, ambient temperature and humidity, load frequency, and structural natural frequency from multi-dimensional monitoring data. Data subsets are divided according to parameter type, and correlation analysis is used to establish the mapping relationship between parameters, clarifying the influence weight of different parameters on the load response. The weight of geometric displacement is set to 0.4 to 0.6, ambient temperature and humidity to 0.2 to 0.3, load frequency to 0.1 to 0.2, and structural natural frequency to 0.05 to 0.1, constructing a dynamic load response input dataset including parameter correlation information. S42 splits the input dataset into 10-minute time steps, imports it into the dynamic load response prediction model, and sets a 48-hour extrapolation period. Boundary conditions are set for different scenarios such as traffic flow and temperature changes, and time-series load response calculations are performed step-by-step to obtain preliminary extrapolation results at each time step. S43 retrieves concurrent data from the historical monitoring database and compares it with the preliminary projection results. It calculates the deviation value and establishes a deviation correction model. Based on the deviation distribution characteristics, it adjusts the influence coefficients in the model and performs spatiotemporal calibration of the projection results to ensure that the response value matches the actual working conditions by more than 90%. S44 integrates the calibrated projection data to generate a dataset including load response values, trends, and peak nodes for each future time period. This clarifies the fluctuation range and extreme values ​​of the load response in different time periods, providing accurate load response benchmark data for subsequent abnormal deformation diagnosis and facilitating the rapid identification and location of early anomalies.

[0040] Preferred, such as Figure 4 As shown, step S5 includes the following sub-steps: S51, collecting fatigue damage evolution sequences and load response trend datasets, extracting damage mutation features, response peak features, and deformation rate features, and constructing an anomaly diagnosis feature library; S52, setting feature matching thresholds and anomaly judgment rules, inputting the feature library data into the intelligent diagnostic model for abnormal deformation, and performing feature similarity calculation and anomaly probability assessment; S53, classifying the anomaly probability values ​​output by the model, and classifying the anomaly level in conjunction with the structural safety level of the expansion joint; S54, outputting diagnostic results including anomaly features, anomaly level, and corresponding location information, and synchronously transmitting them to the bridge expansion joint big data analysis platform for integration.

[0041] Specifically, step S5 is refined into four sub-steps to achieve accurate identification and level determination of abnormal deformation. S51 collects the fatigue damage evolution sequence generated in step S3 and the load response trend dataset from step S4. A feature extraction algorithm is used to extract damage mutation features, response peak features, and deformation rate features. Damage mutation features include mutation amplitude and duration; response peak features include peak size and frequency of occurrence; and deformation rate features include instantaneous rate and average rate. An abnormality diagnostic feature library with multi-dimensional features is constructed. S52 sets the feature matching threshold to 0.75 to 0.85, formulates three-level anomaly judgment rules, inputs the feature library data into the intelligent diagnostic model for abnormal deformation, and compares and analyzes the degree of fit between measured features and standard features using a feature similarity calculation algorithm. Simultaneously, anomaly probability assessment is performed, outputting anomaly probability values ​​between 0 and 1. S53 classifies anomalies according to the anomaly probability values: below 0.3 indicates no anomaly, 0.3 to 0.7 indicates suspected anomaly, and above 0.7 indicates definitive anomaly. Combined with the structural safety level of the expansion joint, the risk level corresponding to each anomaly level is further clarified. The S54 output includes diagnostic results for abnormal features, abnormal levels, corresponding locations, and risk levels. These results are synchronously transmitted to the bridge expansion joint big data analysis platform via a high-speed data transmission channel for integration. This step-by-step process enables multi-feature collaborative judgment of abnormal deformation, effectively reducing the probability of misjudgment and missed judgment, and providing accurate and comprehensive judgment basis for the subsequent generation of early warning information.

[0042] like Figure 5 As shown, a bridge expansion joint intelligent monitoring and early warning system is applied to a bridge expansion joint intelligent monitoring and early warning method. The system includes: a multi-dimensional sensor data acquisition unit, used to deploy a distributed sensor array to collect and transmit parameters such as expansion joint stress amplitude, cumulative strain, geometric displacement, ambient temperature and humidity, and load frequency; a data classification, storage, and transmission unit, which receives data output from the multi-dimensional sensor data acquisition unit, stores it according to parameter type, and sends it to subsequent units via an encrypted transmission protocol; a model calculation and processing unit, which incorporates an expansion joint fatigue damage evolution model, a dynamic load response prediction model, and an abnormal deformation intelligent diagnosis model, receives the classified and stored data, and performs calculation processing; a big data comprehensive analysis unit, connected to the model calculation and processing unit, integrates the calculation results for correlation analysis and threshold comparison, and generates a comprehensive analysis report; an intelligent early warning signal generation unit, which receives the report output by the big data comprehensive analysis unit and generates corresponding early warning signals based on the anomaly level; and an early warning signal output and linkage unit, which receives the signal from the intelligent early warning signal generation unit, outputs it through multiple channels, and establishes linkage with the bridge operation and maintenance management system to accurately push and initiate responses to early warning information.

[0043] This invention integrates various scalar and vector parameters for unified calculation, achieving collaborative computation through parameter standardization, dimensional consistency transformation, and binding of physical meaning correlations. Scalar parameters, such as stress amplitude, ambient temperature, and load frequency, possess clear numerical quantification characteristics and can be directly used as the basic input for the formulas. Vector parameters, such as geometric displacement and cumulative strain, are converted into scalar form by extracting their magnitude, component amplitude, or cumulative effect value, eliminating dimensional differences. Simultaneously, the various coefficients introduced in the formulas (such as temperature influence coefficient, damage weight, and measurement point weight) essentially establish a bridge between parameters with different physical meanings. For example, in the dynamic load response prediction model, the amplitude of the vector geometric displacement is bound to the scalar ambient temperature through a temperature second-order influence coefficient, preserving the spatial characteristic quantification of displacement while incorporating the influence of temperature on material properties. In the intelligent diagnosis model for abnormal deformation, the magnitude of the vector deformation rate is correlated with the scalar damage degree through an exponential function, reflecting the dynamic coupling relationship between the two. Furthermore, all parameters revolve around the core physical processes of expansion joint damage evolution, load response, or anomaly diagnosis, possessing inherent logical connections. The formulas organically integrate these parameters through algebraic operations, integration, summation, and other methods, ensuring both the rigor of mathematical calculations and the comprehensive characterization of the contribution of multi-dimensional parameters to the target results (damage degree, response value, diagnostic value), ultimately achieving unified calculation and accurate output of cross-type parameters.

[0044] A method and system for intelligent monitoring and early warning of bridge expansion joints is proposed. This system constructs a multi-source data collaborative and end-to-end intelligent monitoring framework. Through distributed sensor deployment, it comprehensively collects multi-dimensional parameters of the expansion joint, including stress, deformation, environmental factors, and loads. These parameters are then categorized, stored, and mapped using a big data platform. A dedicated model enables progressive damage evolution calculation, load response simulation, and abnormal deformation diagnosis. This design completely changes the traditional monitoring model of single-parameter acquisition and isolated data processing, achieving deep integration of multi-dimensional information. It can fully depict the entire lifecycle evolution of the expansion joint from minor damage to abnormal deformation, significantly improving the completeness and relevance of monitoring data.

[0045] Meanwhile, this method and system effectively overcome the shortcomings of existing technologies, such as lagging risk identification and insufficient accuracy in early warning, through dynamically adaptable analysis logic and a hierarchical early warning mechanism. By employing multi-model collaborative calculations and fully integrating the structural characteristics of expansion joints with changes in the service environment, it achieves accurate detection of early, subtle damage. Furthermore, by dynamically adjusting the early warning threshold, it adapts to the actual needs under different working conditions, avoiding misjudgments or missed detections caused by fixed thresholds. The efficient linkage and end-to-end collaboration of all system units ensure rapid response from data acquisition and analysis to early warning output, providing accurate and timely decision support for bridge operation and maintenance, and significantly improving the intelligence and refinement of expansion joint health management.

[0046] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0047] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent 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 bridge expansion joint intelligent monitoring and early warning method, characterized in that, Includes the following steps: S1 collects multi-dimensional monitoring data on the stress amplitude, cumulative strain, geometric displacement, ambient temperature and humidity, and load frequency of bridge expansion joints through a distributed sensor array. S2 transmits the collected multi-dimensional monitoring data to the bridge expansion joint big data analysis platform for data classification, storage, and correlation mapping. S3, invoke the fatigue damage evolution model of expansion joint, and calculate damage variables based on stress amplitude, strain accumulation and load frequency; S4 utilizes a dynamic load response prediction model, integrating geometric displacement and environmental temperature and humidity data, to perform load time-series response deduction. S5 uses an intelligent diagnostic model for abnormal deformation to perform feature matching and anomaly determination on fatigue damage calculation results and load response extrapolation data. S6 integrates the model output results from the bridge expansion joint big data analysis platform to generate monitoring and early warning information including damage level, deformation trend and risk range, and completes multi-dimensional data linkage analysis and early warning signal output. 2.The bridge expansion joint intelligent monitoring and early warning method according to claim 1, characterized in that, The expression for the fatigue damage evolution model of the expansion joint is: in, The fatigue damage degree of the expansion joint at time t; This refers to the stress amplitude. The cumulative load frequency at time t; This is the load influence factor; The fatigue accumulation index; The fatigue coefficient of the material; The initial elastic modulus of the expansion joint substrate; This is the temperature influence coefficient; Let t be the ambient temperature. For strain damage weight; for Constantly accumulating responses; This is the damage attenuation coefficient.

3. The intelligent monitoring and early warning method for bridge expansion joints according to claim 1, characterized in that, The expression for the dynamic load response prediction model is: in, The value of the dynamic load response at time t; For response amplification factor; Let t be the geometric displacement at time t; The load vibration coefficient; The frequency of the load application; The cumulative load frequency at time t; This is the secondary influence coefficient of temperature; Let t be the ambient temperature. for Geometric displacement at any given moment; This refers to the multi-point fusion coefficient. Let be the geometric displacement of the i-th measuring point at time t; The weight coefficient for the i-th measurement point; This represents the total number of measurement points.

4. The intelligent monitoring and early warning method for bridge expansion joints according to claim 1, characterized in that, The expression for the intelligent diagnostic model for abnormal deformation is: in, This is an abnormal diagnostic value; The degree of fatigue damage at time t; This is the initial damage threshold; The standard deviation of the damage; For damage anomaly weights; The load response value at time t; For reference response value; The standard deviation of the response value; In response to abnormal weights; Weighted by deformation rate; For time t and Displacement difference at time; For time intervals; This represents the damage acceleration factor.

5. The intelligent monitoring and early warning method for bridge expansion joints according to claim 1, characterized in that, The expression for the data integration model of the bridge expansion joint big data analysis platform is as follows: in, This is a comprehensive analysis index for the platform; The weights for the m-th type of damage data; The data represents the m-th type of damage at time t. Number of damaged data types; The weights for the nth type of response data; The data represents the nth type of response at time t; Number of response data types; Weights for abnormal diagnosis; This is the diagnostic value for the anomaly at the k-th measuring point; This represents the total number of measurement points. This is the data covariance correction factor; Let be the covariance between the damage level and the response value.

6. The intelligent monitoring and early warning method for bridge expansion joints according to claim 1, characterized in that, The threshold calculation expression for intelligent monitoring and early warning of bridge expansion joints is as follows: in, Dynamic early warning threshold; Basic warning threshold; This is the damage impact coefficient; The degree of fatigue damage at time t; The response impact coefficient; The load response value at time t; This is the maximum allowed response value; The coefficient representing the influence of temperature change; For time t and Temperature difference over time; For time intervals.

7. The intelligent monitoring and early warning method for bridge expansion joints according to claim 1, characterized in that, S3 includes the following steps: S31. Select stress amplitude, strain accumulation, load frequency and material elastic modulus parameters to construct a basic dataset of fatigue damage of expansion joints, sort it by time series and associate it with the corresponding ambient temperature data. S32, set the initial conditions for damage evolution, input the basic dataset into the expansion joint fatigue damage evolution model, and obtain the damage increment at each time node through iterative calculation; S33, combining the structural characteristics of expansion joints, spatially distributes the damage increment to clarify the damage differences in different parts; S34 integrates damage data and spatial distribution information at various time points to form a global fatigue damage evolution sequence, providing basic data support for anomaly diagnosis.

8. The intelligent monitoring and early warning method for bridge expansion joints according to claim 1, characterized in that, S4 includes the following steps: S41, extract geometric displacement, ambient temperature and humidity, load frequency and structural natural frequency parameters, construct dynamic load response input dataset, and establish the correlation mapping relationship between parameters; S42, import the input dataset into the dynamic load response prediction model, set the time step and simulation period, and carry out load response time series calculations under multiple scenarios; S43, perform spatiotemporal calibration on the simulation results, and adjust the model output deviation by combining historical monitoring data to ensure that the response value matches the actual working conditions; S44 generates a dataset that includes the load response trend for future periods, clearly defining the time nodes and numerical ranges of response peaks, providing a reference for anomaly detection.

9. The intelligent monitoring and early warning method for bridge expansion joints according to claim 1, characterized in that, S5 includes the following steps: S51, collect fatigue damage evolution sequences and load response trend datasets, extract damage mutation features, response peak features and deformation rate features, and construct an abnormal diagnostic feature library; S52, set the feature matching threshold and anomaly judgment rules, input the feature library data into the abnormal deformation intelligent diagnosis model, and perform feature similarity calculation and anomaly probability assessment. S53, classify the abnormal probability values ​​output by the model, and classify the abnormal level in combination with the structural safety level of the expansion joint; S54 outputs diagnostic results including abnormal characteristics, abnormal level, and corresponding location information, which are simultaneously transmitted to the bridge expansion joint big data analysis platform for integration.

10. A smart monitoring and early warning system for bridge expansion joints, characterized in that, This system is applied to the intelligent monitoring and early warning method for bridge expansion joints as described in claim 1, comprising: The multi-dimensional sensing data acquisition unit is used to deploy a distributed sensor array to collect and transmit parameters such as expansion joint stress amplitude, cumulative strain, geometric displacement, ambient temperature and humidity, and load frequency. The data classification, storage and transmission unit receives the output data from the multi-dimensional sensor data acquisition unit, stores it according to parameter type, and sends it to subsequent units through an encrypted transmission protocol. The model calculation and processing unit has a built-in expansion joint fatigue damage evolution model, dynamic load response prediction model and abnormal deformation intelligent diagnosis model. It receives classified and stored data and performs calculation and processing. The big data comprehensive analysis unit is connected to the model calculation and processing unit to integrate the calculation results for correlation analysis and threshold comparison, and generate a comprehensive analysis report. The intelligent early warning signal generation unit receives the report output by the big data comprehensive analysis unit and generates a corresponding early warning signal according to the anomaly level. The early warning signal output and linkage unit receives signals from the intelligent early warning signal generation unit, outputs them through multiple channels, and establishes linkage with the bridge operation and maintenance management system to accurately push early warning information and initiate responses.

Images