Bridge displacement remote intelligent monitoring method and system and computer equipment

Abstract

The invention relates to the technical field of data processing, and discloses a bridge displacement remote intelligent monitoring method and system and computer equipment. The method comprises the following steps: setting a multi-stage trigger threshold to perform grouped polling sampling to obtain displacement data; adopting dynamic weight fusion to denoise and identify abnormities, and utilizing a Kriging method to generate a digital twinborn model; extracting features to classify and distinguish environment and structure response displacement; calculating a deviation index to estimate fatigue damage to form a health report; calculating a risk entropy value and setting an early warning generation disposal scheme; and evaluating improvement rate update configuration to complete closed-loop optimization. According to the method, the environment data and the historical monitoring data are fused, a self-adaptive sampling strategy and an intelligent analysis processing mechanism are established, displacement changes caused by environment factors and structural factors are accurately distinguished, and accurate risk quantification and automatic decision support are carried out.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to a method, system and computer device for remote intelligent monitoring of bridge displacement. Background Art

[0002] As a key infrastructure of the transportation system, the safety status of bridges is directly related to people's lives, property safety and social and economic development. Traditional bridge monitoring methods mainly rely on manual inspections and regular detections, making it difficult to detect potential dangers in bridge structures in a timely manner. With the development of sensor technology and communication technology, bridge health monitoring systems have gradually been applied to actual projects, realizing continuous monitoring of key parts of bridges. Currently, common bridge monitoring systems mainly include various types such as strain monitoring, displacement monitoring, vibration monitoring and environmental parameter monitoring. These systems deploy various sensors at key parts of the bridge to collect structural response data, and after data transmission and processing, evaluate and warn the health status of the bridge.

[0003] However, the existing bridge displacement monitoring technologies have a series of deficiencies: First, the sensor sampling strategy is fixed and cannot be dynamically adjusted according to the actual state of the bridge, resulting in energy waste or missed key data; Second, the analysis and processing of displacement data mainly rely on simple statistics and threshold comparison, lacking effective distinction between environmental factors and structural factors, and it is difficult to accurately identify real structural anomalies; Third, the early warning decision lacks a scientific and quantitative risk assessment mechanism, often relying on empirical judgment, with insufficient accuracy and timeliness; Fourth, the monitoring system parameters are fixed and cannot be adaptively adjusted according to the monitoring effect and environmental changes, making it difficult to meet the long-term monitoring requirements; Finally, there is a lack of a complete closed-loop feedback mechanism, and it is impossible to continuously optimize the monitoring strategy based on the actual intervention effect, and the system performance is difficult to improve. Summary of the Invention

[0004] This application provides a method, system and computer device for remote intelligent monitoring of bridge displacement, which is used to establish an adaptive sampling strategy and an intelligent analysis and processing mechanism by integrating environmental data and historical monitoring data, accurately distinguish displacement changes caused by environmental factors and structural factors, and perform accurate quantification of risks and automatic decision support.

[0005] In a first aspect, the present application provides a method for remotely and intelligently monitoring bridge displacement. The method for remotely and intelligently monitoring bridge displacement includes: grouping and polling sensors at key structural parts of the bridge by setting multi-level trigger thresholds to obtain original displacement data packets; according to the original displacement data packets, using dynamic weight fusion technology to denoise and identify outliers for displacement data at different positions and of different types, and at the same time using Kriging interpolation method to construct a three-dimensional displacement field to generate a real-time digital twin model of bridge displacement; based on the real-time digital twin model of bridge displacement, extracting displacement spatio-temporal features and performing pattern classification processing to distinguish environmental response displacement and structural response displacement; based on the environmental response displacement and structural response displacement, calculating a displacement deviation index and performing fatigue damage cumulative estimation to form a bridge health status report; according to the bridge health status report, calculating a displacement risk entropy value and setting a warning level to generate a hierarchical disposal plan; calculating an improvement rate index for the displacement change after the implementation of the hierarchical disposal plan, updating the monitoring parameter configuration table, and completing the monitoring closed-loop optimization.

[0006] In a second aspect, the present application provides a system for remotely and intelligently monitoring bridge displacement. The system for remotely and intelligently monitoring bridge displacement includes:

[0007] A sampling module, configured to group and poll sensors at key structural parts of the bridge by setting multi-level trigger thresholds to obtain original displacement data packets;

[0008] An identification module, configured to use dynamic weight fusion technology to denoise and identify outliers for displacement data at different positions and of different types according to the original displacement data packets, and at the same time use Kriging interpolation method to construct a three-dimensional displacement field to generate a real-time digital twin model of bridge displacement;

[0009] A classification module, configured to extract displacement spatio-temporal features and perform pattern classification processing based on the real-time digital twin model of bridge displacement to distinguish environmental response displacement and structural response displacement;

[0010] An estimation module, configured to calculate a displacement deviation index and perform fatigue damage cumulative estimation based on the environmental response displacement and structural response displacement to form a bridge health status report;

[0011] A grading module, configured to calculate a displacement risk entropy value and set a warning level to generate a hierarchical disposal plan according to the bridge health status report;

[0012] A calculation module, configured to calculate an improvement rate index for the displacement change after the implementation of the hierarchical disposal plan, update the monitoring parameter configuration table, and complete the monitoring closed-loop optimization.

[0013] A third aspect of the present invention provides a computer device, comprising: a memory and at least one processor, wherein instructions are stored in the memory; the at least one processor invokes the instructions in the memory to cause the computer device to execute the above-mentioned remote intelligent monitoring method for bridge displacement.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium, in which instructions are stored, and when it runs on a computer, it causes the computer to execute the above-mentioned remote intelligent monitoring method for bridge displacement.

[0015] In the technical solution provided by this application, by setting multi-level trigger thresholds to group and poll the sensors at key structural parts of the bridge for sampling, intelligent scheduling of data collection is realized, which not only ensures the capture of key data but also optimizes the energy usage efficiency; the dynamic weight fusion technology is used to denoise and identify outliers for displacement data at different positions and types, and combined with the Kriging interpolation method to construct a three-dimensional displacement field, generating a real-time digital twin model of bridge displacement, which not only improves the data quality but also realizes the leap from discrete monitoring points to a continuous displacement field, providing a data basis for global analysis; by extracting the spatio-temporal characteristics of displacement and performing pattern classification processing to distinguish environmental response displacement and structural response displacement, the core problem that cannot distinguish the influence of environmental factors and structural factors in traditional monitoring is solved, greatly improving the accuracy of anomaly identification; based on the environmental response displacement and structural response displacement, the displacement deviation index is calculated and the fatigue damage accumulation is estimated to form a bridge health status report, transforming the qualitative judgment into a quantitative assessment and providing a scientific basis for decision-making; according to the bridge health status report, the displacement risk entropy value is calculated and the early warning level is set to generate a hierarchical disposal plan, realizing the transformation from passive response to active early warning and enhancing the risk control ability; the improvement rate index of the displacement change after the implementation of the hierarchical disposal plan is calculated, the monitoring parameter configuration table is updated, and the monitoring closed-loop is optimized, establishing a complete feedback mechanism, enabling the monitoring system to self-learn and continuously evolve. It is particularly emphasized that this solution fully considers the application characteristics of artificial intelligence algorithms in the field of bridge monitoring, such as using the dynamic weight fusion technology to adaptively process sensor data at different positions and types, constructing a continuous displacement field through the Kriging spatial interpolation algorithm, using spatio-temporal feature extraction and pattern classification algorithms to distinguish environmental and structural factors, and quantifying the system uncertainty based on the entropy value theory. These algorithm features are deeply combined with the professional requirements of bridge displacement monitoring, and are optimized for the characteristics of high-dimensional, multi-source, and non-linear displacement data, significantly improving the accuracy, intelligence level, and adaptability of monitoring. Description of the Drawings

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0017] Figure 1 It is a schematic diagram of an embodiment of the bridge displacement remote intelligent monitoring method in the embodiments of the present application;

[0018] Figure 2 It is a schematic diagram of an embodiment of the bridge displacement remote intelligent monitoring system in the embodiments of the present application;

[0019] Figure 3 It is a schematic block diagram of the structure of the computer device in the embodiments of the present invention. Detailed implementation manners

[0020] The embodiments of the present application provide a bridge displacement remote intelligent monitoring method, system and computer device. The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and accompanying drawings of the present application are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0021] For ease of understanding, the following describes the specific process of the embodiments of the present application. Please refer to Figure 1 An embodiment of the bridge displacement remote intelligent monitoring method in the embodiments of the present application includes:

[0022] Step S101: Group and poll the sensors at the key structural parts of the bridge by setting multi-level trigger thresholds to obtain the original displacement data packet;

[0023] Step S102: According to the original displacement data packet, use the dynamic weight fusion technology to denoise and identify outliers for displacement data at different positions and types, and at the same time use the Kriging interpolation method to construct a three-dimensional displacement field to generate a real-time bridge displacement digital twin model;

[0024] Step S103: Extract the spatio-temporal features of the displacement based on the real-time digital twin model of bridge displacement, and perform pattern classification to distinguish between environmental response displacement and structural response displacement;

[0025] Step S104: Calculate the displacement deviation index based on the environmental response displacement and the structural response displacement, and estimate the cumulative fatigue damage to form a bridge health status report;

[0026] Step S105: Calculate the displacement risk entropy value according to the bridge health status report, set the warning level, and generate a hierarchical disposal plan;

[0027] Step S106: Calculate the improvement rate index for the displacement change after the implementation of the hierarchical disposal plan, update the monitoring parameter configuration table, and complete the monitoring closed-loop optimization.

[0028] It can be understood that the execution entity of this application can be a remote intelligent monitoring system for bridge displacement, or a terminal or a server. Specifically, it is not limited here. This application example is described by taking the server as the execution entity as an example.

[0029] Specifically, the sensors at key structural parts of the bridge are grouped and polled for sampling by setting multi-level trigger thresholds. Specifically, based on the historical stress distribution map of the bridge, the deployment positions of the sensors are determined, monitoring points are arranged for key parts such as the main girder, pier, bearing and cable tower, the sensors are divided into a displacement sensor group, an inclination sensor group and a strain sensor group according to the functional type, and different sampling time windows are allocated to form a sensor polling table. The multi-level trigger thresholds include a low-level threshold, a medium-level threshold and a high-level threshold, corresponding to low-frequency sampling parameters, medium-frequency sampling parameters and high-frequency sampling parameters respectively. The trigger thresholds are dynamically adjusted according to the temperature, humidity and wind speed information contained in the real-time environmental data to generate an environmental compensation threshold coefficient. The grouped polling sampling is performed according to the polling table and the environmental compensation threshold coefficient, the physical parameters are collected and encrypted and marked to form an original displacement data packet. According to the obtained original displacement data packet, the dynamic weight fusion technology is used to denoise and identify outliers for displacement data at different positions and types, and at the same time the Kriging interpolation method is used to construct a three-dimensional displacement field. First, the multi-source displacement data is processed by multi-scale time alignment to generate a time-synchronized displacement matrix, the time-domain statistical features are calculated by the sliding window technology, and a displacement fluctuation feature set is constructed. Based on the displacement fluctuation feature set, the Mahalanobis distance method is used to identify outliers, and the outliers are removed to form a displacement effective data set. According to the structural importance and data quality index of each monitoring point, the dynamic fusion weight coefficient is calculated, and the displacement effective data set is weighted. The weighted data is input into the Kriging spatial interpolation processor, and the displacement values at non-monitoring point positions are calculated through semi-variogram analysis and optimal linear unbiased estimation. The displacement data of all monitoring points and interpolation points are mapped to the bridge three-dimensional structure model to form a real-time digital twin model of bridge displacement.

[0030] Based on the real-time digital twin model of bridge displacement, extract the spatio-temporal characteristics of displacement and conduct pattern classification to distinguish environmental response displacement and structural response displacement. Extract the displacement time series from the displacement digital twin model, construct a multi-dimensional displacement feature vector, perform time-domain and frequency-domain transformations on it, and obtain the displacement waveform feature set and frequency component table. Match the displacement waveform feature set with the real-time meteorological data in time series, calculate the displacement-temperature sensitivity coefficient and displacement-wind speed correlation degree, and construct the separation matrix of environmental factor influence. Decompose the original displacement data through the separation matrix of environmental factor influence, separate the displacement into environmental response displacement and structural response displacement, and analyze the spatial distribution characteristics of the two types of displacements to form a displacement pattern feature library. Based on the environmental response displacement and structural response displacement, calculate the displacement deviation index and estimate the fatigue damage accumulation to form a bridge health status report. Compare the environmental response displacement with the normal range values in the historical environmental response database and calculate the environmental deviation rate. By segmenting the structural response displacement with time windows, extract the displacement amplitude sequence and frequency change characteristics to form a structural displacement feature table. According to the environmental deviation rate and the structural displacement feature table, construct a comprehensive weight matrix to quantify the displacement anomaly degree of each monitoring point and generate a displacement deviation index. Conduct rain-flow counting on the structural response displacement, count the occurrence frequency and cycle times of the displacement amplitude, and form a displacement stress spectrum. According to the mapping relationship between the displacement stress spectrum and the material fatigue curve, calculate the fatigue damage values of each key component according to the principle of linear damage accumulation and construct a damage distribution map. Combine the displacement deviation index and the damage distribution map with the importance rating of the bridge structure, and generate a bridge health status report through multi-level weight superposition calculation.

[0031] According to the bridge health status report, calculate the displacement risk entropy value and set the warning level, and generate a hierarchical disposal plan. Extract the displacement deviation index and damage distribution parameters from the bridge health status report to establish a risk assessment basic table. Conduct uncertainty quantification analysis on the data in the risk assessment basic table, calculate the displacement volatility and trend indicators of each monitoring point, and form the risk probability distribution of the monitoring points. According to the dispersion degree and fluctuation range of the risk probability distribution of the monitoring points, construct the displacement risk entropy value to quantify the uncertainty level of the overall bridge system. Divide the displacement risk entropy value into first-level warning, second-level warning, and third-level warning according to the preset interval to determine the warning level. For different warning levels, match the corresponding intervention measure sets from the pre-plan knowledge base, including speed limit and load limit measures, inspection and reinforcement measures, and emergency control measures. Combine the historical response characteristics of the bridge and the surrounding environmental conditions, sort the intervention measure sets by priority and optimize the combination to generate a hierarchical disposal plan.

[0032] Calculate the improvement rate index for the displacement change after the implementation of the hierarchical disposal plan, update the monitoring parameter configuration table, and complete the optimization of the monitoring closed-loop. Record the bridge displacement data before and after the implementation of the hierarchical disposal plan, and construct a comparison dataset of intervention effects. Conduct a time series comparison on the comparison dataset of intervention effects, calculate the percentage change in displacement amplitude and the reduction rate of abnormal points, and form a displacement improvement quantification table. According to the displacement improvement quantification table, combined with the structural importance weights of each monitoring point, calculate the comprehensive improvement rate index. Evaluate and rank the effectiveness of different types of intervention measures based on the improvement rate index, and establish an evaluation library for the effects of intervention measures. According to the historical data in the evaluation library for the effects of intervention measures, adjust the sensor sampling strategy, trigger threshold, and warning level determination criteria, and update the monitoring parameter configuration table. Apply the updated monitoring parameter configuration table to the next monitoring cycle, start a new round of data collection and analysis process, and complete the optimization of the monitoring closed-loop.

[0033] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0034] (1) Determine the sensor deployment locations based on the bridge historical stress distribution map, and arrange monitoring points for key parts such as the main girder, bridge pier, bearing, and cable tower;

[0035] (2) Divide the sensors into a displacement sensor group, an inclination sensor group, and a strain sensor group according to their functional types, and allocate corresponding sampling time windows to form a sensor polling table;

[0036] (3) Set a three-level trigger threshold value set for each sensor group, namely a low-level threshold, a medium-level threshold, and a high-level threshold, and configure low-frequency sampling parameters, medium-frequency sampling parameters, and high-frequency sampling parameters correspondingly;

[0037] (4) Dynamically adjust the trigger thresholds of each sensor group according to the temperature, humidity, and wind speed information included in the real-time environmental data, and generate an environmental compensation threshold coefficient;

[0038] (5) Perform grouped polling sampling according to the sensor polling table and the environmental compensation threshold coefficient, and collect the physical parameters of each monitoring point;

[0039] (6) Encrypt and mark the collected physical parameters, label the time stamp and location information, and package them to form the original displacement data packet.

[0040] Specifically, the historical stress distribution map refers to the graphical representation of the stress concentration areas of various parts of the bridge obtained through finite element analysis or the accumulation of measured data, which includes the stress distribution of the bridge under various load conditions. By analyzing the historical stress distribution map, areas with large stress gradients and obvious changes are identified as key monitoring areas, and monitoring points are arranged for the main girder, pier, bearing, and cable tower, which are components that bear important loads. The main girder is the main load-bearing component that spans across rivers or valleys, the pier is the vertical component that supports the main girder, the bearing is the structure that connects the main girder and the pier, and the cable tower is the tall component that supports the main cable in a suspension bridge. The arrangement of monitoring points follows the principle of "highlighting key points and covering comprehensively", with dense points arranged in stress concentration areas and appropriate points arranged in conventional areas to form a grid-like coverage structure. The sensors are classified into a displacement sensor group, an inclination sensor group, and a strain sensor group according to their functional types. The displacement sensor group mainly monitors the displacement changes of various parts of the bridge, the inclination sensor group monitors the inclination angle changes of bridge components, and the strain sensor group monitors the internal stress and strain state of components. The purpose of grouping is to optimize the data acquisition efficiency, and sensors in the same group have similar data characteristics and processing requirements. Different sampling time windows are allocated according to the importance of each group of sensors and the data change frequency to form a sensor polling table. The sensor polling table is a data acquisition time schedule that specifies the sampling order and duration of each group of sensors on the time axis, avoiding data communication conflicts and peak energy consumption caused by simultaneous sampling.

[0041] A three-level trigger threshold value set is set for each sensor group, namely the low-level threshold, the medium-level threshold, and the high-level threshold. The threshold refers to the displacement or strain change amount that activates different sampling frequencies. The low-level threshold corresponds to daily minor changes, the medium-level threshold corresponds to obvious but non-urgent changes, and the high-level threshold corresponds to significant changes that require immediate attention. These thresholds are respectively configured with low-frequency sampling parameters, medium-frequency sampling parameters, and high-frequency sampling parameters. The low-frequency sampling parameters are usually set to about 1Hz and are suitable for static monitoring; the medium-frequency sampling parameters are set to about 10Hz and are suitable for quasi-dynamic monitoring; the high-frequency sampling parameters are set to 50 - 100Hz and are suitable for dynamic response monitoring. The core of the three-level trigger mechanism is to dynamically adjust the data acquisition frequency according to the degree of change in the bridge state, which not only ensures the integrity of the data but also optimizes the use of system resources.

[0042] Dynamically adjusting the trigger threshold according to real-time environmental data is one of the innovative points of this method. The real-time environmental data includes temperature, humidity, and wind speed information, and these environmental factors will directly affect the natural displacement of the bridge. By establishing a correlation model between environmental factors and displacement changes, the trigger threshold is dynamically adjusted to generate an environmental compensation threshold coefficient. The environmental compensation threshold coefficient is an adjustment factor used to correct the basic threshold according to the current environmental conditions. For example, in high-temperature weather, the bridge will naturally expand and the displacement will increase. At this time, the trigger threshold should be increased to avoid misjudging the normal displacement caused by temperature as abnormal. When calculating the environmental compensation threshold coefficient, the temperature influence coefficient, humidity influence coefficient, and wind speed influence coefficient are comprehensively considered to perform weighted adjustment on the basic threshold.

[0043] Performing grouped polling sampling according to the sensor polling table and the environmental compensation threshold coefficient is the specific implementation step of data acquisition. Grouped polling sampling means activating different groups of sensors in sequence for data acquisition according to a preset schedule, rather than collecting all data simultaneously. The sampling process first checks the current environmental conditions, applies the environmental compensation threshold coefficient to correct the trigger threshold, and then determines the sampling frequency according to the corrected threshold. When collecting the physical parameters of each monitoring point, the original data such as displacement, inclination change rate, and strain value are recorded. Encrypting and marking the collected physical parameters is a key step in data security and management. The data encryption uses the AES encryption algorithm to protect the security of data transmission, and the marking process adds a timestamp and location information to each group of data. The timestamp accurately records the specific time of data acquisition, and the location information includes the spatial coordinates of the monitoring point and the information of the component to which it belongs. Finally, the encrypted and marked data is packaged in a unified format to form an original displacement data packet, preparing for subsequent analysis and processing.

[0044] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0045] (1) Perform multi-scale time alignment processing on the multi-source displacement data in the original displacement data packet to generate a time-synchronized displacement matrix;

[0046] (2) Calculate the time-domain statistical features of the time-synchronized displacement matrix through the sliding window technique to construct a displacement fluctuation feature set;

[0047] (3) Based on the displacement fluctuation feature set, use the Mahalanobis distance method to identify abnormal points, and form a displacement valid data set after removing the outliers;

[0048] (4) Calculate the dynamic fusion weight coefficient according to the structural importance and data quality index of each monitoring point, and perform weighted processing on the displacement valid data set;

[0049] (5) Input the weighted displacement effective data set into the Kriging spatial interpolation processor, and calculate the displacement values at non-monitored points through semi-variogram analysis and optimal linear unbiased estimation.

[0050] (6) Map the displacement data of all monitored points and interpolated points to the three-dimensional bridge structure model to form the real-time bridge displacement digital twin model.

[0051] Specifically, perform multi-scale time alignment processing on the multi-source displacement data in the original displacement data packet to generate a time-synchronized displacement matrix. Multi-source displacement data refers to data recorded by different types of sensors (displacement sensors, inclination sensors, strain sensors) at different sampling frequencies, and these data are not completely consistent in time stamps. The multi-scale time alignment processing uses the dynamic time warping (DTW) algorithm to map the data points on different time series to a unified time axis to solve the problem of time asynchrony. The specific approach is to use the data with the highest sampling frequency as the benchmark, and perform data point filling on the data with low sampling frequency through linear or spline interpolation methods to align all data in the time dimension. The time-synchronized displacement matrix is a two-dimensional data structure, where the rows represent different time points, the columns represent different monitoring positions, and each element in the matrix represents the displacement value at a specific time point and a specific position. Calculate the time-domain statistical features of the time-synchronized displacement matrix through the sliding window technique to construct a displacement fluctuation feature set. The sliding window technique refers to setting a window with a fixed length on the time series, and the window slides on the time axis according to a preset step size, and calculates the statistical features of the data in each window. The window length is usually set to be able to contain the duration of a complete displacement cycle. For example, for the displacement caused by vehicle load, the window length can be set to 10 - 30 seconds. The sliding step size determines the density of feature extraction. The smaller the step size, the denser the features, but the greater the computational amount. Calculate the statistical features of the data in each sliding window, including mean, standard deviation, peak value, peak-to-valley ratio, waveform factor, etc., to form a displacement fluctuation feature set. The displacement fluctuation feature set is a set of feature vectors that describe the variation characteristics of displacement data in the time domain, and contains both static displacement information and dynamic fluctuation characteristics.

[0052] Based on the displacement fluctuation feature set, use the Mahalanobis distance method to identify outliers, and form a displacement effective data set after removing the outliers. The Mahalanobis distance method is an outlier detection method that considers the data distribution characteristics and can effectively identify outliers in a multi-dimensional feature space. When calculating the Mahalanobis distance, the correlation between features is considered, which is particularly applicable to displacement data because the displacements at different positions of the bridge often have strong correlations. Calculate the Mahalanobis distance between each data point and the overall distribution center of the features, and mark the points whose distance exceeds a preset threshold (usually set to 3 times the distribution standard deviation) as outliers and remove them. After removing the outliers, the remaining data forms a displacement effective data set, which represents the displacement state of the bridge under normal conditions.

[0053] According to the structural importance and data quality indicators of each monitoring point, calculate the dynamic fusion weight coefficient and perform weighted processing on the effective displacement data set. Structural importance refers to the importance of the bridge component where the monitoring point is located in the overall structure, which is usually determined by structural engineers based on the force conditions and functional roles of the components. The data quality indicator is a quantitative parameter for measuring the reliability of monitoring data, including signal-to-noise ratio, stability index, and integrity score. The calculation formula for the dynamic fusion weight coefficient is:

[0054] W ij = α·SI i + β·DQ ij + γ·TC ij

[0055] where, W ij represents the weight coefficient of the i-th monitoring point at the j-th moment; SI i represents the structural importance index of the i-th monitoring point, with a value range of 0 - 1; DQ ij represents the data quality indicator of the i-th monitoring point at the j-th moment, with a value range of 0 - 1; TC ij represents the time-varying credibility of the i-th monitoring point at the j-th moment, reflecting the reliability trend of the data changing over time; α, β, and γ are the weight factors of structural importance, data quality, and time-varying credibility respectively, and satisfy α + β + γ = 1. Apply the calculated weight coefficient to the effective displacement data set for weighted processing, emphasizing the contributions of important monitoring points and high-quality data, and suppressing the interference of secondary or low-quality data.

[0056] Input the weighted effective displacement data set into the Kriging spatial interpolation processor, and calculate the displacement values at non-monitoring point locations through semi-variogram analysis and optimal linear unbiased estimation. Kriging spatial interpolation is a geostatistical method suitable for interpolating data with spatial correlation. In bridge displacement monitoring, due to the limitation of the number of monitoring points, only displacement data at limited locations can be obtained, and through Kriging interpolation, the continuous displacement field of the entire bridge structure can be estimated. Semi-variogram analysis is the prerequisite for Kriging interpolation. By calculating the displacement value differences between monitoring points at different distances, a displacement spatial correlation model is established. Optimal linear unbiased estimation is the core of Kriging interpolation. Based on the displacement values of known monitoring points and the semi-variogram, the displacement estimation value at any non-monitoring point location is calculated. The advantage of Kriging interpolation is that it takes into account both the influence of spatial distance and the directionality and anisotropy of the data, and is particularly suitable for structures with complex geometries such as bridges.

[0057] The displacement data of all monitoring points and interpolation points are mapped to the three-dimensional structural model of the bridge to form a real-time bridge displacement digital twin model. The three-dimensional structural model of the bridge is a virtual bridge representation in the format of computer-aided design (CAD) or building information model (BIM), which contains the geometric shape, material properties and component relationships of the bridge. In the process of mapping the displacement data to the three-dimensional model, the spatial correspondence between the monitoring points and the model nodes is first established, and then the measured displacement values and the interpolated displacement values are assigned to the corresponding model nodes. Finally, the displacement is magnified and displayed to intuitively show the deformation state of the bridge. The real-time bridge displacement digital twin model is a digital mapping of the physical bridge in the virtual space. It not only reflects the current displacement state of the bridge, but also supports historical displacement playback and future displacement prediction, providing an intuitive and effective tool for bridge health monitoring.

[0058] Taking a suspension bridge as an example, the original displacement data package contains data from 48 monitoring points, with a sampling frequency of 10 Hz for the main cable, 5 Hz for the suspenders, and 2 Hz for the bridge deck. Through multi-scale time alignment, all data are aligned to a time axis with a sampling frequency of 10 Hz, forming a time-synchronized displacement matrix containing 48 columns (representing 48 monitoring points). A 30-second sliding window is set, sliding across the time axis in 10-second steps. Statistical features such as the mean, standard deviation, and peak value are calculated for each window to construct a displacement fluctuation feature set. The Mahalanobis distance method is used to identify outliers. The third monitoring point on the main cable exhibited abnormal jitter during a certain period. The Mahalanobis distance value reached 4.2 times the threshold, and it was marked as an outlier and removed. Based on the structural analysis results, the structural importance index of the monitoring points in the main cable anchorage area was set to 0.9, the mid-span point to 0.8, and other less important locations to 0.6. The dynamic fusion weight coefficient was calculated by combining the signal-to-noise ratio and data integrity score of each point. The final weight coefficient for the monitoring points in the main cable anchorage area was 0.85. Kriging interpolation was applied to the weighted displacement data. By analyzing the semivariogram relationship between the monitoring points, displacement values were calculated for approximately 1,000 non-monitoring points on the bridge, forming a detailed displacement field. Finally, all displacement data was mapped to a pre-built 3D bridge model to generate a digital twin model that intuitively displays displacement deformation, supporting engineers in remote real-time monitoring and analysis.

[0059] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0060] (1) extracting displacement time series from the real-time bridge displacement digital twin model and constructing a multi-dimensional displacement feature vector;

[0061] (2) performing time domain and frequency domain transformation on the multidimensional displacement feature vector to obtain a displacement waveform feature set and a frequency component table;

[0062] (3) Perform time-series matching on the displacement waveform feature set and real-time meteorological data, and calculate the displacement-temperature sensitivity coefficient and displacement-wind speed correlation degree;

[0063] (4) Based on the displacement-temperature sensitivity coefficient and displacement-wind speed correlation degree, construct an environmental factor influence separation matrix;

[0064] (5) Perform decomposition operation on the original displacement data through the environmental factor influence separation matrix, and separate the displacement into the environmental response displacement and the structural response displacement;

[0065] (6) Analyze the spatial distribution characteristics of the environmental response displacement and the structural response displacement to form a displacement pattern feature library.

[0066] Specifically, extract the displacement time series from the real-time bridge displacement digital twin model to construct a multi-dimensional displacement feature vector. The real-time bridge displacement digital twin model is the digital mapping of the bridge physical structure in the virtual space, containing the complete displacement information of each monitoring point and interpolation point. When extracting the displacement time series, a fixed time window (usually 24 hours) is set for each key node, and the continuous displacement records within this window are extracted to form a time-displacement data pair. The multi-dimensional displacement feature vector refers to the high-dimensional vector formed by combining the displacement data of different monitoring points within the same time window. Each dimension represents the displacement value of a monitoring point, and the length of the vector is equal to the number of monitoring points. This vector representation not only retains the spatial distribution information of the displacement but also includes the time continuity characteristics. Perform time-domain and frequency-domain transformations on the multi-dimensional displacement feature vector to obtain the displacement waveform feature set and the frequency component table. The time-domain transformation refers to the feature extraction operation performed on the original time series, including calculating statistical quantities (mean, variance, skewness, kurtosis), trend analysis (linear regression coefficient, interval maximum change rate), and morphological features (peak factor, crest factor), etc. The frequency-domain transformation is to convert the time series into the frequency-domain representation through the fast Fourier transform (FFT) to analyze the frequency composition of the displacement signal. The displacement waveform feature set is the result of the time-domain transformation, describing the morphological characteristics of the displacement changing with time; the frequency component table is the result of the frequency-domain transformation, showing the energy distribution of the displacement signal at different frequencies. The frequency component table is particularly important because the displacement caused by environmental factors (such as temperature, wind speed) usually shows as low-frequency signals, while the self-vibration of the structure shows as specific high-frequency resonance peaks.

[0067] Perform time-series matching on the displacement waveform feature set and real-time meteorological data, and calculate the displacement-temperature sensitivity coefficient and displacement-wind speed correlation degree. The real-time meteorological data includes environmental parameters such as temperature, humidity, wind speed, and wind direction recorded at the monitoring station. The time-series matching is to align the displacement data and meteorological data according to the time stamp to establish the corresponding relationship between the data. The calculation formula for the displacement-temperature sensitivity coefficient is:

[0068]

[0069] Among them, S DT (p) represents the sensitivity coefficient of the displacement of monitoring point p to temperature; D(p, t) represents the displacement value of monitoring point p at time t; represents the average displacement of monitoring point p; T(t) represents the temperature value at time t; represents the average temperature; T represents the total number of time points. The calculation of the displacement-wind speed correlation degree uses the cross-correlation coefficient:

[0070]

[0071] Among them, R DW (p) represents the correlation coefficient of the displacement of monitoring point p and the wind speed; W(t) represents the wind speed value at time t; represents the average wind speed. These two coefficients quantify the response degree of the displacement to environmental factors and provide a basis for subsequent separation of environmental factors.

[0072] Based on the displacement-temperature sensitivity coefficient and the displacement-wind speed correlation degree, an environmental factor influence separation matrix is constructed. The environmental factor influence separation matrix is a mathematical tool used to decompose the original displacement into the part caused by environmental factors and the part of the structure's own response. When constructing this matrix, first determine the main environmental impact factors (usually including temperature, wind speed, humidity, etc.), and then establish a response model of the displacement to these factors. Each row in the environmental factor influence separation matrix corresponds to a monitoring point, each column corresponds to an environmental factor, and the matrix element represents the sensitivity of a specific monitoring point to a specific environmental factor.

[0073] Through the environmental factor influence separation matrix, the original displacement data is decomposed and calculated to separate the displacement into the environmental response displacement and the structural response displacement. The decomposition calculation is based on the following model:

[0074] D(p, t) = D E (p, t) + D S (p, t)

[0075] Among them, D(p, t) represents the original displacement value of monitoring point p at time t; D E (p, t) represents the environmental response displacement; D S (p, t) represents the structural response displacement. The calculation formula for the environmental response displacement is:

[0076]

[0077] Among them, D E (p, t) represents the environmental response displacement of monitoring point p at time t; S DT$(p)$ represents the displacement-temperature sensitivity coefficient of monitoring point $p$; $T(t)$ represents the temperature value at time $t$; $T$ ref represents the reference temperature, usually taken as the historical average; $R$ DW $(p)$ represents the displacement-wind speed correlation coefficient of monitoring point $p$; $K$ W represents the normalization coefficient of wind speed, used to convert the influence of wind speed into displacement; $W(t)$ represents the wind speed value at time $t$; $W$ ref represents the reference wind speed, usually taken as the historical average; $R$ DH $(p)$ represents the displacement-humidity correlation coefficient of monitoring point $p$; $K$ H represents the normalization coefficient of humidity, used to convert the influence of humidity into displacement; $H(t)$ represents the humidity value at time $t$; $H$ ref represents the reference humidity, usually taken as the historical average.

[0078] $D$ S $D(p,t)=D(p,t)-D$ E $(p,t)$

[0079] This decomposition method can effectively distinguish the normal displacement caused by environmental factors and the abnormal displacement caused by structural state changes. Analyze the spatial distribution characteristics of the environmental response displacement and the structural response displacement to form a displacement pattern feature library. The spatial distribution characteristic analysis refers to studying the distribution law of displacement at different parts of the bridge, including synchronism, symmetry, and propagation characteristics, etc. For the environmental response displacement, analyze its distribution pattern on the entire bridge to verify whether it conforms to the overall expansion and contraction characteristics caused by temperature changes or the lateral deformation characteristics caused by wind loads. For the structural response displacement, focus on analyzing the local abnormal deformation area and the deformation propagation path to detect potential structural problems. The displacement pattern feature library is a knowledge library formed by extracting and classifying typical displacement patterns from historical data, containing displacement distribution characteristic templates under different types of loads, providing a reference basis for subsequent displacement cause identification.

[0080] Taking a cable-stayed bridge as an example, the displacement time series of three types of key monitoring points, namely the main cable, the top of the tower, and the bridge deck, are extracted from its real-time digital twin model of bridge displacement, and a multi-dimensional feature vector containing the displacement values of 32 monitoring points is constructed. Time-domain analysis is performed on the feature vector, and it is calculated that the average displacement of the midpoint of the bridge deck in 24 hours is 22.8 mm, the standard deviation is 3.5 mm, and the peak-to-valley ratio is 1.8. Frequency-domain analysis shows that the main frequency components of the displacement at this point are concentrated at 0.01 Hz (corresponding to diurnal temperature changes) and 0.25 Hz (corresponding to the influence of wind load). By matching the displacement waveform characteristics with the meteorological station data in time series, the displacement-temperature sensitivity coefficient of the midpoint of the bridge deck is calculated to be 1.2 mm / °C, indicating that for every 1°C increase in temperature, the displacement increases by approximately 1.2 mm; the displacement-wind speed correlation degree at this point is 0.78, indicating a strong correlation between displacement and wind speed. Based on the sensitivity coefficients and correlation degrees of all monitoring points, an environmental factor influence separation matrix is constructed. The first column of the matrix is the temperature sensitivity coefficient of each point, and the second column is the wind speed correlation degree. The original displacement data is decomposed through this matrix, and it is determined that 17.6 mm of the displacement of the midpoint of the bridge deck on that day is due to temperature changes (environmental response displacement), and 5.2 mm is due to the structural state itself (structural response displacement). By comparing and analyzing the environmental response displacements and structural response displacements of different monitoring points, it is found that the environmental response displacements show a symmetric distribution that increases from both ends to the middle in the longitudinal direction of the bridge, which conforms to the expansion characteristics caused by uniform temperature changes; while the structural response displacements increase significantly in the area near the right main cable, showing an asymmetric distribution, indicating that there may be structural abnormalities in this area. These analysis results are recorded in the displacement pattern feature library.

[0081] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0082] (1) Compare the environmental response displacement with the normal range values in the historical environmental response database, and calculate the environmental deviation rate;

[0083] (2) By segmenting the structural response displacement with time windows, extract the displacement amplitude sequence and frequency change characteristics to form a structural displacement feature table;

[0084] (3) According to the environmental deviation rate and the structural displacement feature table, construct a comprehensive weight matrix to quantify the displacement abnormality degree of each monitoring point and generate the displacement deviation index;

[0085] (4) Perform rainflow counting on the structural response displacement, count the occurrence frequency and cycle times of the displacement amplitude to form a displacement stress spectrum;

[0086] (5) According to the mapping relationship between the displacement stress spectrum and the material fatigue curve, calculate the fatigue damage values of each key component according to the principle of linear damage accumulation, and construct a damage distribution map;

[0087] (6) Combine the displacement deviation index and the damage distribution map with the importance rating of the bridge structure, and generate the bridge health status report through multi-level weight superposition calculation.

[0088] Specifically, compare the environmental response displacement with the normal range values in the historical environmental response database to calculate the environmental deviation rate. The environmental response displacement refers to the normal displacement changes caused by environmental factors such as temperature, wind speed, and humidity. The historical environmental response database is a collection of environmental response displacement data accumulated through long-term monitoring, which contains the normal change ranges of bridge displacements under different environmental conditions. The environmental deviation rate is a quantitative indicator that measures the degree to which the current environmental response displacement deviates from the historical normal range. The calculation method is to divide the difference between the current environmental response displacement and the displacement mean value under the same historical environmental conditions by the historical standard deviation to obtain the standardized deviation degree. The larger the environmental deviation rate, the higher the degree of current environmental response abnormality, which may reflect a decrease in the applicability of the environmental factor influence model or a change in the response characteristics of the bridge to the environment. By segmenting the structural response displacement with time windows, extract the displacement amplitude sequence and frequency change characteristics to form a structural displacement feature table. The structural response displacement refers to the displacement caused by the change of the bridge structure's own state and is not affected by environmental factors. Time window segmentation is to divide the continuous time series into segments of a fixed length for local analysis. The displacement amplitude sequence refers to the maximum value, minimum value, and their difference of the displacement within each time window, which reflects the intensity characteristics of the structural displacement. The frequency change characteristics are obtained by performing spectral analysis on the displacement data within each window to extract the main frequency components and their energy distributions, which reflect the dynamic characteristics of the structure. The structural displacement feature table is a data structure organized by time windows, which records the amplitude characteristics and frequency characteristics of each window and provides a comprehensive description of the structural response displacement over time.

[0089] Based on the environmental deviation rate and the structural displacement feature table, construct a comprehensive weight matrix to quantify the displacement abnormality degree of each monitoring point and generate a displacement deviation index. The comprehensive weight matrix is a mathematical tool used to integrate abnormal indicators from different sources and assign corresponding importance weights to different indicators. The construction process first determines the evaluation index system, including the environmental deviation rate, displacement amplitude abnormality degree, frequency feature deviation degree, etc.; then set weight coefficients for each indicator, and the weight allocation considers the reliability, sensitivity, and importance of the indicators; finally, multiply each indicator value by the weight and accumulate to obtain a comprehensive score. The displacement deviation index is an overall assessment of the displacement abnormality degree after comprehensively considering environmental and structural factors. The higher the index value, the greater the degree of displacement deviation from the normal state, which requires key attention.

[0090] The rain-flow counting method is applied to the structural response displacement to count the occurrence frequency and cycle times of displacement amplitudes, forming a displacement stress spectrum. The rain-flow counting method is a commonly used cycle counting method in fatigue analysis, applicable to the processing of irregular load histories, and can effectively identify the complete cycles and half-cycles in the load. The processing process includes: rotating the displacement time history diagram by 90 degrees so that the time axis is vertically downward; imagining the path of raindrops flowing from each displacement peak-valley point and determining the flow termination condition according to specific rules; recording the starting point and ending point of each flow to form a cycle, and its amplitude is the absolute value of the displacement difference between the starting point and the ending point. The displacement stress spectrum is the statistical result of the cycle times of displacements with different amplitudes, usually represented in the form of a two-dimensional matrix. The rows represent the displacement amplitude intervals, the columns represent the displacement average value intervals, and the matrix elements represent the cycle times under the corresponding conditions. The displacement stress spectrum is a bridge connecting displacement monitoring data and fatigue damage assessment, providing basic data for subsequent fatigue analysis.

[0091] According to the mapping relationship between the displacement stress spectrum and the material fatigue curve, the fatigue damage values of each key component are calculated according to the principle of linear damage accumulation, and a damage distribution map is constructed. The material fatigue curve (S-N curve) describes the number of cycles that the material can withstand under different stress levels and is an important characterization of the material's fatigue characteristics. Converting displacement to stress requires establishing a mapping through structural mechanics relationships. The commonly used method is to calculate the corresponding stress level according to the displacement value through a finite element model or a simplified mechanical model. The principle of linear damage accumulation (Miner's criterion) is a classic theory of fatigue damage accumulation, and its core formula is:

[0092]

[0093] where D f represents the cumulative fatigue damage value; n ab represents the cycle times in the amplitude interval a and the average value interval b in the displacement stress spectrum; N ab represents the total number of cycles that the material can withstand under the corresponding stress condition, determined by the S-N curve; A represents the total number of amplitude intervals; B represents the total number of average value intervals. When the cumulative fatigue damage value Df reaches 1, it is theoretically considered that the component has fatigue failure. The damage distribution map is a visual representation of mapping the calculated fatigue damage values onto the bridge structure model, intuitively showing the distribution of fatigue damage in each part of the bridge and helping to identify high-risk fatigue areas.

[0094] Combining the displacement deviation index and the damage distribution map with the bridge structure importance rating, a bridge health status report is generated through multi-level weight superposition calculation. The bridge structure importance rating is a classification of the importance of each component of the bridge in the overall structure, usually determined by structural engineers based on factors such as the function, stress condition, and failure consequences of the components. Multi-level weight superposition is a hierarchical analysis method that integrates three indicators at different levels, namely the displacement deviation index, fatigue damage value, and structural importance, according to a certain weight relationship to form a comprehensive evaluation result. The bridge health status report is the final evaluation result, including the overall health level of the bridge, the status evaluation of key components, the identification of abnormal positions, and maintenance suggestions, etc.

[0095] Taking a cross-sea bridge as an example, the main span of the bridge is 240 meters, and health assessment is carried out through the displacement data of 20 monitoring points. First, the environmental response displacement obtained by monitoring is compared with the historical database, and it is found that the environmental deviation rate in the main cable anchorage area is 2.3, exceeding the warning threshold of 2.0, indicating that the response characteristics of this area to the environment may have changed. Then, the structural response displacement is segmented with a 30-minute time window, and the displacement amplitude and frequency characteristics of each window are extracted. It is found that the energy ratio in the 5-8Hz frequency band in the main cable anchorage area has increased from 15% in history to 27%, indicating that the structural dynamic characteristics have changed significantly. According to the changes in the environmental deviation rate and frequency characteristics, a comprehensive weight matrix is constructed, where the environmental deviation weight is 0.4 and the frequency characteristic weight is 0.6. The calculated displacement deviation index of the main cable anchorage area is 0.78, which is in a moderately abnormal state. The rain-flow counting process is carried out on the displacement data of the main cable anchorage area to obtain the number of cycles under different combinations of amplitudes and averages, forming a displacement stress spectrum. Applying the principle of linear damage accumulation and combining with the S-N curve of the main cable material, the calculated cumulative fatigue damage value of this area is 0.65, which is already close to the warning line of 0.7. Considering that the structural importance rating of the main cable anchorage area is Class A (the highest level), after comprehensively evaluating the displacement deviation index, fatigue damage value, and structural importance, the generated bridge health status report marks this area as a concern that needs to be prioritized for disposal, and recommends increasing the detection frequency and carrying out targeted reinforcement and maintenance.

[0096] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0097] (1) Extract the displacement deviation index and damage distribution parameters from the bridge health status report to establish a bridge risk assessment basic table;

[0098] (2) Conduct uncertainty quantification analysis on the data in the risk assessment basic table, calculate the displacement volatility and trend indicators of each monitoring point, and form a risk probability distribution of the monitoring points;

[0099] (3) Construct the displacement risk entropy value according to the discrete degree and fluctuation range of the risk probability distribution of the monitoring points, and quantify the uncertainty level of the overall bridge system;

[0100] (4) Divide the displacement risk entropy value into first-level warning, second-level warning, and third-level warning according to a preset interval, and determine the warning level;

[0101] (5) For different warning levels, match the corresponding set of intervention measures from the pre-plan knowledge base, including speed limit and load limit measures, inspection and reinforcement measures, and emergency control measures;

[0102] (6) Combine the historical response characteristics of the bridge and the surrounding environmental conditions, rank the priority and combine and optimize the set of intervention measures to generate the hierarchical disposal plan.

[0103] Specifically, extract the displacement deviation index and damage distribution parameters from the bridge health status report to establish a bridge risk assessment basic table. The displacement deviation index is a comprehensive indicator that quantifies the degree to which the displacement of each monitoring point deviates from the normal state; the damage distribution parameter describes the distribution of fatigue damage in different parts of the bridge. The risk assessment basic table is a structured data table, with each row representing a monitoring point, and the columns including attributes such as location information, displacement deviation index, fatigue damage value, and structural importance level, providing a data basis for subsequent risk assessment. In actual operation, the data in the health status report is extracted according to a preset format and organized into a standardized table structure for easy next-step analysis and processing. Conduct uncertainty quantification analysis on the data in the risk assessment basic table, calculate the displacement volatility and trend indicators of each monitoring point, and form the risk probability distribution of the monitoring points. Uncertainty quantification analysis is an important means to evaluate the reliability and risk degree of data. In bridge monitoring, two aspects of characteristics are mainly concerned: displacement volatility and trend. The displacement volatility index is characterized by calculating statistical quantities such as the standard deviation and coefficient of variation of the displacement data within a certain time window, reflecting the discrete degree of the displacement data; the trend index fits the change trend line of the displacement data and calculates its slope and acceleration, reflecting the direction and rate of displacement change. The risk probability distribution of the monitoring points refers to estimating the possible displacement range and its probability of each monitoring point in a future period based on historical data and the current state, usually represented by a probability density function. In the calculation process, a probability model is established for the displacement time series of each monitoring point, considering the historical distribution characteristics of the data and the recent change trend, and the risk probability distribution is generated through statistical inference methods.

[0104] According to the discreteness and fluctuation range of the risk probability distribution of the monitoring points, a displacement risk entropy value is constructed to quantify the uncertainty level of the overall bridge system. The displacement risk entropy value draws on the concept of information entropy and is used to measure the uncertainty degree of the monitoring data. When calculating, first discretize the risk probability distribution of each monitoring point into several intervals, calculate the probability value of each interval, and then, according to the information entropy formula, sum the product of the probability of each interval and its logarithm and take the negative value to obtain the single-point entropy value. The overall displacement risk entropy value of the bridge is obtained by weighted averaging the entropy values of each monitoring point, and the weights take into account the structural importance and location representativeness of the monitoring points. The higher the displacement risk entropy value, the greater the uncertainty of the bridge displacement state and the higher the risk level. During the data processing process, attention should be paid to the normalization of the probability distribution and the rationality of the interval division to ensure the accuracy of the entropy value calculation.

[0105] Divide the displacement risk entropy value into first-level warning, second-level warning, and third-level warning according to the preset intervals to determine the warning level. Warning classification is a common method in risk management, and different response measures are taken according to the severity of the risk. In bridge displacement monitoring, the displacement risk entropy value is usually divided into three intervals: the first-level warning (attention level) corresponds to the situation where the entropy value is below the warning line but close to it, indicating that the bridge state begins to show abnormalities and the monitoring frequency needs to be increased; the second-level warning (warning level) corresponds to the situation where the entropy value exceeds the warning line but does not reach the danger line, indicating that the bridge state has obvious abnormalities and restrictive measures need to be implemented and maintenance needs to be arranged; the third-level warning (emergency level) corresponds to the situation where the entropy value exceeds the danger line, indicating that the bridge state is severely abnormal and there are potential safety hazards, and emergency intervention measures need to be taken immediately. The division thresholds of the warning levels are usually determined according to historical monitoring data and expert experience, and there will be differences for different types and importance of bridges. For different warning levels, match the corresponding set of intervention measures from the pre-plan knowledge base, including speed limit and load limit measures, inspection and reinforcement measures, and emergency control measures. The pre-plan knowledge base is a collection of emergency plans established by the bridge management department based on historical cases and professional knowledge, which contains the disposal plans for various abnormal situations. The speed limit and load limit measures are applicable to the first-level warning situation, such as reducing the vehicle speed limit, prohibiting overweight vehicles from passing, etc.; the inspection and reinforcement measures are applicable to the second-level warning situation, such as detailed inspection of specific parts, reinforcement and repair of local components, etc.; the emergency control measures are applicable to the third-level warning situation, such as temporarily closing the bridge, implementing traffic diversion, emergency repair, etc. The matching process uses a rule-based reasoning method to screen out the applicable set of measures from the knowledge base according to factors such as the current warning level, abnormal location, and abnormal type.

[0106] Combined with the historical response characteristics of the bridge and the surrounding environmental conditions, prioritize and optimize the combination of intervention measures to generate a hierarchical disposal plan. The historical response characteristics refer to the response performance and disposal effects of the bridge in past similar situations, and the surrounding environmental conditions include external factors such as traffic flow, meteorological conditions, and geographical location. The priority ranking considers multiple factors: the effectiveness of the measures (evaluated based on historical cases), implementation cost, implementation difficulty, scope of influence, etc. The combination optimization is to select a combination of measures with strong complementarity and the best overall effect from the candidate measures to avoid redundancy or conflicts. The hierarchical disposal plan is a detailed action guidance document that includes the basis for judging the warning level, the specific implementation methods of various measures, division of responsibilities, time arrangement, and effect evaluation criteria. The generation process of the plan uses a multi-criteria decision-making method, comprehensively considering technical feasibility, economic rationality, and social impact to form a systematic response strategy.

[0107] Take a certain highway bridge as an example to illustrate the entire processing flow: During the daily monitoring of the bridge, the displacement deviation indexes near the middle of the main span and the right support are 0.75 and 0.82 respectively, and the fatigue damage values are 0.42 and 0.56 respectively. First, input these data together with the location information and the structural importance level (the middle of the main span is Class A, and the support is Class B) into the risk assessment basic table. Then, conduct an uncertainty analysis on the historical displacement data of the two key points. The calculated displacement volatility index in the middle of the main span is 0.35 (indicating relatively stable), and the trend index is 0.12 (slight upward trend); the displacement volatility index of the right support is 0.62 (large fluctuations), and the trend index is 0.28 (obvious upward trend). Based on these indexes, generate the risk probability distributions of the two points through a statistical model, and calculate that the entropy value of the middle of the main span is 0.68, and the entropy value of the right support is 0.85. Considering the structural importance weight, calculate that the overall displacement risk entropy value of the bridge is 0.74, which is located in the preset secondary warning interval (0.7 - 0.9). According to the secondary warning level, match the corresponding measure set from the pre-plan knowledge base, including restricting the total vehicle weight not to exceed 25 tons, reducing the traffic speed to 40 km / h, increasing the monitoring frequency to twice a day, arranging a special inspection of the right support, etc. Considering that this bridge is an important transportation artery, located in an area with large traffic flow but with diversion roads, and the weather forecast shows no strong winds or rains in the next three days, the finally optimized hierarchical disposal plan is determined as: Implement the weight limit and speed limit measures, and at the same time arrange a detailed inspection of the right support during the low-traffic period from 2 am to 5 am the next day, prepare necessary reinforcement materials, and increase the monitoring frequency to once every 8 hours to continuously observe the displacement change trend.

[0108] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0109] (1) Record the bridge displacement data before and after the implementation of the hierarchical disposal plan, and construct a comparison dataset of intervention effects;

[0110] (2) Conduct time-series comparison on the comparison dataset of intervention effects, calculate the percentage change in displacement amplitude and the reduction rate of abnormal points, and form a displacement improvement quantification table;

[0111] (3) According to the displacement improvement quantification table, combined with the structural importance weights of each monitoring point, calculate the comprehensive improvement rate index;

[0112] (4) Based on the improvement rate index, rank and evaluate the effectiveness of different types of intervention measures, and establish an intervention measure effect evaluation library;

[0113] (5) According to the historical data in the intervention measure effect evaluation library, adjust the sensor sampling strategy, trigger threshold, and early warning level determination criteria, and update the monitoring parameter configuration table;

[0114] (6) Apply the updated monitoring parameter configuration table to the next monitoring cycle, start a new round of data collection and analysis process, and complete the monitoring closed-loop optimization.

[0115] Specifically, continuously optimize the monitoring system to form a closed-loop feedback mechanism. First, record the bridge displacement data before and after the implementation of the hierarchical disposal plan, and construct a comparison dataset of intervention effects. The hierarchical disposal plan is a combination of intervention measures for different early warning levels, such as speed limit and load limit, inspection and reinforcement, or emergency control. The comparison dataset of intervention effects is a structured data set that contains displacement data within the same time span before and after the intervention, and is used to evaluate the effectiveness of the intervention measures. During the construction process, the exact time point of the intervention implementation needs to be recorded to ensure the comparability of the data before and after, and external factors such as environmental conditions and traffic load during the intervention period need to be recorded for conditional correction in subsequent analysis. The data set adopts a two-dimensional time-space table structure, where the rows represent different time points, the columns represent different monitoring positions, and the cells record the corresponding displacement values, environmental parameters, and external conditions.

[0116] Perform a time series comparison on the intervention effect comparison dataset, calculate the percentage change in displacement amplitude and the reduction rate of abnormal points, and form a displacement improvement quantification table. Time series comparison is the core method for quantitatively evaluating the intervention effect, by calculating the displacement difference at the same monitoring points before and after the intervention under similar environmental conditions. The percentage change in displacement amplitude reflects the degree of reduction in the absolute displacement. The calculation method is to divide the difference between the displacement amplitude after the intervention and the displacement amplitude before the intervention by the displacement amplitude before the intervention, and then multiply by 100%. The reduction rate of abnormal points measures the improvement degree of the displacement abnormal phenomenon. The calculation method is to calculate the change ratio of the number of data points exceeding the normal threshold within the same length of time before and after the intervention. The displacement improvement quantification table is a systematic record of the comparative analysis results, including the percentage change in displacement amplitude, the reduction rate of abnormal points and other quantitative indicators of each monitoring point, providing basic data for the overall effect evaluation.

[0117] According to the displacement improvement quantification table, combined with the structural importance weights of each monitoring point, calculate the comprehensive improvement rate index. The structural importance weight is a weight coefficient determined according to the importance degree of the bridge component where the monitoring point is located in the overall structure. Usually, it is evaluated by structural engineers based on the functional role, stress condition and failure consequence of the component. When calculating the comprehensive improvement rate index, multiply the improvement index of each monitoring point by the corresponding structural importance weight, and then sum to obtain the weighted average. This process ensures that the improvement status of important components has a greater influence in the overall evaluation, and more accurately reflects the improvement effect of the intervention measures on the overall safety of the bridge. The comprehensive improvement rate index is a value between 0 and 1. The larger the value, the more significant the intervention effect.

[0118] Based on the improvement rate index, rank and evaluate the effectiveness of different types of intervention measures, and establish an intervention measure effect evaluation library. Multiple dimensions are considered in the ranking evaluation process: improvement rate index, implementation cost, implementation difficulty and intervention duration. For each type of intervention measure (such as speed limit and load limit, inspection and reinforcement, emergency control, etc.), collect the evaluation indicators of historical application cases, calculate the comprehensive score, and form the effectiveness ranking. The intervention measure effect evaluation library is a knowledge base that records the application effects of different types of intervention measures under various bridge types, problem types and environmental conditions, including quantitative evaluation indicators and description of applicable conditions. The evaluation library combines structured data with text description, facilitating quick retrieval and application recommendation.

[0119] Adjust the sensor sampling strategy, trigger thresholds, and early warning level determination criteria according to the historical data in the intervention measure effectiveness evaluation library, and update the monitoring parameter configuration table. The monitoring parameter configuration table is a set of key parameters that control the operation of the monitoring system, including the sampling frequencies of various sensors, trigger conditions, data processing parameters, and early warning judgment criteria, etc. The adjustment process is based on a feedback learning mechanism, analyzing the correlation between monitoring parameters and effects in historical intervention cases to identify the optimal parameter settings. For example, if it is found that a certain type of anomaly can be better captured by sampling at a specific frequency in the early stage, then increase that frequency accordingly; if a certain threshold setting results in too many false alarms, then appropriately raise the threshold. The adjustment strategy takes into account both monitoring effectiveness and system resource consumption, seeking the best balance point. The updated configuration table is managed through a version control system to ensure the traceability of parameter adjustments.

[0120] Apply the updated monitoring parameter configuration table to the next monitoring cycle, start a new round of data collection and analysis process, and complete the optimization of the monitoring closed-loop. The new round of monitoring uses the updated parameter settings, specifically strengthening the monitoring intensity of key locations and time periods, optimizing the data processing algorithm parameters, and improving the accuracy of anomaly detection. The optimization of the monitoring closed-loop means that through continuous iteration of the "monitoring - analysis - intervention - evaluation - adjustment" cycle, continuously improving the monitoring methods and parameter settings, enabling the monitoring system to gradually adapt to the structural characteristics and environmental conditions of a specific bridge, and achieving precise monitoring and efficient early warning.

[0121] Taking a suspension bridge as an example to illustrate the whole process: After implementing the hierarchical disposal plan for the secondary warning of this bridge (including a weight limit of 30 tons, a speed limit of 40 km / h, and main cable reinforcement), the displacement data for 7 days before and 7 days after the intervention were recorded, forming a comparison dataset of the intervention effects. Through time series comparison and analysis, it was calculated that the displacement amplitude at the midpoint of the main cable decreased from the original 125 mm to 78 mm, and the percentage change was -37.6%; the displacement at the top of the main tower decreased from 85 mm to 62 mm, and the percentage change was -27.1%; the number of outliers (data points exceeding 3 times the standard deviation) decreased from an average of 12 per day before the intervention to an average of 3 per day, and the reduction rate was 75%. These data were recorded in the displacement improvement quantification table. Considering that the structural importance weights of the midpoint of the main cable and the top of the main tower are 0.6 and 0.4 respectively, the comprehensive improvement rate index was calculated as 0.6×37.6% + 0.4×27.1% = 33.4%. Comparing this disposal case with historical cases, it was found that in similar situations, the combined measures of weight limit + speed limit + reinforcement were significantly more effective than the measures of simply weight limit or speed limit. Accordingly, the relevant entries in the intervention measure effect evaluation library were updated. According to the evaluation results, the sensor parameters were adjusted: the first-level warning trigger threshold at the midpoint of the main cable was adjusted from a displacement exceeding 90 mm to exceeding 85 mm, the sampling frequency during the peak traffic period was increased, and at the same time, the wind speed trigger threshold at the top of the main tower was adjusted from 8 m / s to 7 m / s. These adjustments were recorded in the updated monitoring parameter configuration table. After the new monitoring cycle was started, the monitoring system operated according to the updated parameter settings, achieving earlier identification of potential risks and more accurate warnings, and completing a complete monitoring closed-loop optimization.

[0122] The above described the bridge displacement remote intelligent monitoring method in the embodiment of the present application. Next, the bridge displacement remote intelligent monitoring system in the embodiment of the present application will be described. Please refer to Figure 2 , an embodiment of the bridge displacement remote intelligent monitoring system in the embodiment of the present application includes:

[0123] A sampling module, configured to perform grouped polling sampling on the sensors of the key structural parts of the bridge by setting multiple trigger thresholds to obtain the original displacement data packet;

[0124] An identification module, configured to perform denoising and outlier identification on the displacement data of different positions and types according to the original displacement data packet by using the dynamic weight fusion technology, and at the same time use the Kriging interpolation method to construct a three-dimensional displacement field to generate a real-time bridge displacement digital twin model;

[0125] A classification module, configured to extract the displacement spatio-temporal features and perform pattern classification processing according to the real-time bridge displacement digital twin model to distinguish the environmental response displacement and the structural response displacement;

[0126] An estimation module, configured to calculate a displacement deviation index and perform fatigue damage accumulation estimation based on environmental response displacement and structural response displacement, and form a bridge health status report;

[0127] A grading module, configured to calculate a displacement risk entropy value and set a warning level according to the bridge health status report, and generate a grading disposal plan;

[0128] A calculation module, configured to calculate an improvement rate index for the displacement change after the implementation of the grading disposal plan, update the monitoring parameter configuration table, and complete the monitoring closed-loop optimization.

[0129] Through the collaborative cooperation of the above-mentioned various components, by setting multi-level trigger thresholds to perform grouped polling sampling on the sensors at key structural parts of the bridge, intelligent scheduling of data collection is achieved, which not only ensures the capture of key data but also optimizes the energy usage efficiency; the dynamic weight fusion technology is used to denoise and identify outliers for displacement data of different positions and types, and combined with the Kriging interpolation method to construct a three-dimensional displacement field, generating a real-time digital twin model of bridge displacement, which not only improves the data quality but also realizes the leap from discrete monitoring points to a continuous displacement field, providing a data basis for global analysis; by extracting the spatio-temporal characteristics of displacement and performing pattern classification processing to distinguish environmental response displacement and structural response displacement, the core problem that environmental factors and structural factors cannot be distinguished in traditional monitoring is solved, greatly improving the accuracy of anomaly identification; based on environmental response displacement and structural response displacement, calculate the displacement deviation index and perform fatigue damage accumulation estimation, form a bridge health status report, transform qualitative judgment into quantitative evaluation, and provide a scientific basis for decision-making; according to the bridge health status report, calculate the displacement risk entropy value and set the warning level, generate a grading disposal plan, realize the transformation from passive response to active warning, and enhance the risk control ability; calculate the improvement rate index for the displacement change after the implementation of the grading disposal plan, update the monitoring parameter configuration table, complete the monitoring closed-loop optimization, and establish a complete feedback mechanism, enabling the monitoring system to self-learn and continuously evolve. It is particularly worth emphasizing that this solution fully considers the application characteristics of artificial intelligence algorithms in the field of bridge monitoring, such as using the dynamic weight fusion technology to adaptively process sensor data of different positions and types, constructing a continuous displacement field through the Kriging spatial interpolation algorithm, using spatio-temporal feature extraction and pattern classification algorithms to distinguish environmental and structural factors, and quantifying the system uncertainty based on the entropy value theory. These algorithm features are deeply combined with the professional requirements of bridge displacement monitoring, and are optimized and designed for the characteristics of high-dimensional, multi-source, and non-linear displacement data, significantly improving the accuracy, intelligence level, and adaptability of monitoring.

[0130] Refer to Figure 3 , in the embodiment of the present invention, a computer device is further provided. This computer device can be a server, and its internal structure can be as Figure 3As shown in the figure. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0131] Those skilled in the art can understand that Figure 3 the structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0132] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0133] Those of ordinary skill in the art can understand that all or part of the processes in the above embodiment methods can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments. Among them, any reference to memory, storage, database, or other media provided by the present invention and used in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0134] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, systems, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0135] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs.

[0136] As described above, the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A remote intelligent monitoring method for bridge displacement, characterized in that, The remote intelligent monitoring method for bridge displacement includes: Group polling sampling is performed on sensors at key structural parts of the bridge by setting multi-level trigger thresholds to obtain the original displacement data packet. Based on the original displacement data packet, the dynamic weight fusion technology is used to denoise and identify outliers for displacement data at different positions and types. Meanwhile, the Kriging interpolation method is used to construct a three-dimensional displacement field to generate a real-time digital twin model of bridge displacement. According to the real-time digital twin model of bridge displacement, the displacement spatio-temporal features are extracted and pattern classification processing is carried out to distinguish environmental response displacement and structural response displacement. Based on the environmental response displacement and structural response displacement, the displacement deviation index is calculated and fatigue damage accumulation estimation is carried out to form a bridge health status report. According to the bridge health status report, the displacement risk entropy value is calculated and the warning level is set to generate a hierarchical disposal plan. The improvement rate index of the displacement change after the implementation of the hierarchical disposal plan is calculated, the monitoring parameter configuration table is updated, and the monitoring closed-loop optimization is completed.

2. The remote intelligent monitoring method for bridge displacement according to claim 1, characterized in that, The group polling sampling is performed on sensors at key structural parts of the bridge by setting multi-level trigger thresholds to obtain the original displacement data packet, including: Based on the historical stress distribution map of the bridge, the sensor deployment positions are determined, and monitoring points are arranged for key parts such as the main girder, pier, bearing, and cable tower. The sensors are divided into a displacement sensor group, an inclination sensor group, and a strain sensor group according to their functional types, and corresponding sampling time windows are assigned to form a sensor polling table. A three-level trigger threshold value set is set for each sensor group, namely the low-level threshold, the medium-level threshold, and the high-level threshold, and the low-frequency sampling parameters, the medium-frequency sampling parameters, and the high-frequency sampling parameters are correspondingly configured. According to the temperature, humidity, and wind speed information included in the real-time environmental data, the trigger thresholds of each sensor group are dynamically adjusted to generate an environmental compensation threshold coefficient. According to the sensor polling table and the environmental compensation threshold coefficient, group polling sampling is performed to collect the physical parameters of each monitoring point. The collected physical parameters are encrypted and marked with a time stamp and location information, and then packed to form the original displacement data packet.

3. The remote intelligent monitoring method for bridge displacement according to claim 1, characterized in that, Based on the original displacement data packet, the dynamic weight fusion technology is used to denoise and identify outliers for displacement data at different positions and types. Meanwhile, the Kriging interpolation method is used to construct a three-dimensional displacement field to generate a real-time digital twin model of bridge displacement, including: Multi-scale time alignment processing is performed on the multi-source displacement data in the original displacement data packet to generate a time-synchronized displacement matrix. The time-domain statistical features are calculated for the time-synchronized displacement matrix through the sliding window technology to construct a displacement fluctuation feature set. Based on the displacement fluctuation feature set, the Mahalanobis distance method is used to identify outliers, and the outliers are removed to form a displacement effective data set. According to the structural importance and data quality index of each monitoring point, the dynamic fusion weight coefficient is calculated, and the displacement effective data set is weighted. The weighted displacement effective data set is input into the Kriging spatial interpolation processor, and the displacement values at non-monitoring point positions are calculated through semi-variogram analysis and optimal linear unbiased estimation. Map the displacement data of all monitoring points and interpolation points to the 3D bridge structure model to form the real-time bridge displacement digital twin model.

4. The remote intelligent monitoring method for bridge displacement according to claim 1, wherein Based on the real-time bridge displacement digital twin model, extract the spatio-temporal characteristics of displacement and perform pattern classification to distinguish environmental response displacement and structural response displacement, including: Extract the displacement time series from the real-time bridge displacement digital twin model and construct a multi-dimensional displacement feature vector; Perform time-domain and frequency-domain transformations on the multi-dimensional displacement feature vector to obtain a displacement waveform feature set and a frequency component table; Match the displacement waveform feature set with the real-time meteorological data in time series, and calculate the displacement-temperature sensitivity coefficient and the displacement-wind speed correlation; Based on the displacement-temperature sensitivity coefficient and the displacement-wind speed correlation, construct a separation matrix for the influence of environmental factors; Perform decomposition operations on the original displacement data through the separation matrix for the influence of environmental factors to separate the displacement into the environmental response displacement and the structural response displacement; Analyze the spatial distribution characteristics of the environmental response displacement and the structural response displacement to form a displacement pattern feature library.

5. The remote intelligent monitoring method for bridge displacement according to claim 1, characterized in that Based on the environmental response displacement and the structural response displacement, calculate the displacement deviation index and perform fatigue damage accumulation estimation to form a bridge health status report, including: Compare the environmental response displacement with the normal range values in the historical environmental response database and calculate the environmental deviation rate; Segment the structural response displacement by time window, extract the displacement amplitude sequence and frequency change characteristics to form a structural displacement feature table; According to the environmental deviation rate and the structural displacement feature table, construct a comprehensive weight matrix to quantify the displacement abnormality degree of each monitoring point and generate the displacement deviation index; Perform rain-flow counting on the structural response displacement, count the occurrence frequency and cycle times of the displacement amplitude to form a displacement stress spectrum; Based on the mapping relationship between the displacement stress spectrum and the material fatigue curve, calculate the fatigue damage values of each key component according to the principle of linear damage accumulation and construct a damage distribution map; Combine the displacement deviation index and the damage distribution map with the bridge structure importance rating, and calculate through multi-level weight superposition to generate the bridge health status report.

6. The remote intelligent monitoring method for bridge displacement according to claim 1, characterized in that According to the bridge health status report, calculate the displacement risk entropy value and set the warning level to generate a hierarchical disposal plan, including: Extract the displacement deviation index and damage distribution parameters from the bridge health status report to establish a bridge risk assessment basic table; Perform uncertainty quantification analysis on the data in the risk assessment basic table, calculate the displacement volatility and trend indicators of each monitoring point to form a risk probability distribution of the monitoring points; According to the dispersion degree and fluctuation range of the risk probability distribution of the monitoring points, construct the displacement risk entropy value to quantify the uncertainty level of the overall bridge system; Divide the displacement risk entropy value into a first-level warning, a second-level warning and a third-level warning according to a preset interval to determine the warning level; For different warning levels, match the corresponding intervention measure sets from the pre-plan knowledge base, including speed limit and load limit measures, repair and reinforcement measures, and emergency control measures; Combined with the historical response characteristics of the bridge and the surrounding environmental conditions, prioritize and combinatorially optimize the set of intervention measures to generate the hierarchical disposal plan.

7. The remote intelligent monitoring method for bridge displacement according to claim 1, wherein, Calculate the improvement rate index for the displacement change after the implementation of the hierarchical disposal plan, update the monitoring parameter configuration table, and complete the monitoring closed-loop optimization, including: Record the bridge displacement data before and after the implementation of the hierarchical disposal plan, and construct a comparison dataset of intervention effects; Perform a time series comparison on the comparison dataset of intervention effects, calculate the percentage change in displacement amplitude and the reduction rate of abnormal points, and form a displacement improvement quantification table; According to the displacement improvement quantification table, combined with the structural importance weights of each monitoring point, calculate the comprehensive improvement rate index; Based on the improvement rate index, rank and evaluate the effectiveness of different types of intervention measures, and establish an intervention measure effect evaluation library; According to the historical data in the intervention measure effect evaluation library, adjust the sensor sampling strategy, trigger threshold, and early warning level determination criteria, and update the monitoring parameter configuration table; Apply the updated monitoring parameter configuration table to the next monitoring cycle, start a new round of data collection and analysis process, and complete the monitoring closed-loop optimization.

8. A remote intelligent monitoring system for bridge displacement, which is used to implement the remote intelligent monitoring method for bridge displacement as described in any one of claims 1-7, is characterized in that, The bridge displacement remote intelligent monitoring system includes: A sampling module for group polling sampling of sensors at key structural parts of the bridge by setting multiple trigger thresholds to obtain the original displacement data packet; An identification module for denoising and outlier identification of displacement data at different positions and types using the dynamic weight fusion technique based on the original displacement data packet, and simultaneously constructing a three-dimensional displacement field using the Kriging interpolation method to generate a real-time bridge displacement digital twin model; A classification module for extracting displacement spatio-temporal features and performing pattern classification processing based on the real-time bridge displacement digital twin model to distinguish environmental response displacement and structural response displacement; An estimation module for calculating the displacement deviation index and performing fatigue damage accumulation estimation based on the environmental response displacement and structural response displacement to form a bridge health status report; A grading module for calculating the displacement risk entropy value and setting the early warning level according to the bridge health status report to generate a hierarchical disposal plan; A calculation module for calculating the improvement rate index for the displacement change after the implementation of the hierarchical disposal plan, updating the monitoring parameter configuration table, and completing the monitoring closed-loop optimization.

9. A computer device, characterized in that, It includes a memory and a processor, and the memory stores a computer program that can run on the processor. It is characterized in that when the processor executes the computer program, it implements the bridge displacement remote intelligent monitoring method according to any one of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, the processor is caused to execute the bridge displacement remote intelligent monitoring method according to any one of claims 1 to 7.

Images